From dbk at opentext.com Thu Jan 1 11:49:12 1998 From: dbk at opentext.com (Dan Keith) Date: Thu Nov 3 15:58:31 2005 Subject: Oracle or Sybase available on Rhapsody? Message-ID: Does anyone know if Oracle or Sybase is planning on releasing a version of their database systems for Rhapsody? I would hope (and expect) that with Larry Ellison on the board at Apple, an Oracle port is assured, but nothing is certain these days. If the server side isn't available any time soon, how about client APIs that will let me access these databases (i.e., OCI for Oracle and dblib for Sybase)? bud ! Dan "Bud" Keith dbk@mcs.com ! "You never can tell what goes on down below. This pool might be bigger than you or I know." - Dr. Seuss From kherlihy at earthlink.net Thu Jan 1 12:59:27 1998 From: kherlihy at earthlink.net (kherlihy@earthlink.net) Date: Thu Nov 3 15:58:31 2005 Subject: Help with DR1 booting for PPC (newbie help) References: <199712302150.NAA11850@scv3.apple.com> Message-ID: <34AC03AD.6A74@earthlink.net> Thanks for the resposes from all of you who offered suggestions on my problem.. Incidentally, almost all of you who replied were correct, using: rootdev=sd0 made my problems go away. Thanks Again. -- Kyle kherlihy@earthlink.net ------------------------ From scott at cacti.org Thu Jan 1 13:11:25 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: Help with DR1 booting for PPC (newbie help) Message-ID: <199801012102.NAA10067@proxy3.ba.best.com> >Thanks for the resposes from all of you who offered suggestions on my >problem.. Incidentally, almost all of you who replied were correct, >using: > >rootdev=sd0 Just for reference, I had to specify: rootdev=sd1 to make my problems go away. :) Rhapsody is on the second internal SCSI drive. - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From wmeyer at mediaone.net Thu Jan 1 14:05:51 1998 From: wmeyer at mediaone.net (Will Meyer) Date: Thu Nov 3 15:58:31 2005 Subject: Yellow & Blue boxes sharing an IP address Message-ID: <199801012202.RAA19043@chmls05.mediaone.net> On 12/31/97 6:00 PM , Robert Fisher (rfisher@onr.com) wrote: >It seems to me that this is very feasible and straight forward. The blue >box is just a yellow box application. When a blue app tried to open an ip >port, the blue box would, in turn, try to open the same yellow ip port. If >a blue app tries to open a port that is in use by a yellow app, it'll get >the normal "port in use" error, because that's what the blue box will get >when it tries to open the port on behalf of the blue app. If a yellow app >tries to open a port that is in use by a blue app, it'll get the normal >"port in use" error because the blue box (a yellow app itself) has already >opened the port on behalf of that blue app. > >There's no need for "demultiplexing" when an inbound packet arrives. As is >usual, it's given to the yellow app that holds the port it came in on. If >that yellow app happens to be the blue box, then the blue box routes the >packet to the proper blue app. One of the problems is that the blue box is NOT just a YB app. The Blue Box is the real Mac OS running on Mach. It has an application interface to start it and prepare the front end to have access to it. Therefore the Mac OS has access to networking through the kernel in the same manner that YB does. Routing all networking through YB would kill performence and would also push BB farther from the real Mac OS. > >No changes need be made to the yellow box networking stack. All the changes >would be within the blue box itself. Blue box apps might suffer some loss >in performance, but yellow apps wouldn't. (And, assuming loss in >performance is necessary for such a feature, better to make the blue apps >suffer.) > >If you ping the machine, the yellow box will reply. After all, the yellow >box is the real OS; the blue box is merely a compatibility environment. (I >imagine it wouldn't be difficult to set up the blue box to respond to pings >instead. Are pings handled by a server deamon or by the network stack >itself? In any case, I don't see much point in doing that.) > >Obviously you couldn't run two servers on the same port (one in blue and >one in yellow). It only makes sense that that would fail if you are sharing >a single ip address between the two environments. Trying to do that would >be as nonsensical as running two servers on the same port on any other OS. > One of the solutions that is being brought up over and over on this thread is the possibility of having an IP Masq scheme withine the box to distribute the TCP packets between OSes. AFAIK, IP Masq works by using an alternate TCP port to send requests from the hidden machines. What if one of the networking options was to have all Mac services transcribed up, say, 20000 ports, transparent to the BB? In other words a Mac ftp server would be active on port 20021 and Personal Web Sharing on port 20080. I don't know enough about TCP/IP to know the possibility of implementing this, but if it worked it would let the Mac have it's own servers and clients and have them be seperate to virtually all YB operations while not doing any hiding of packets one enviroment to another. >Of course, you'd also want the -option- of both environments having >separate ip addresses if desired, but I don't know how feasible that is. > Are there currently options in a YB app or control panel to change options for the BB? You could put a networking enu in there to change between compatibility modes. That may be too complex for the average artist-turned-DTPer who can't remember which menu is the Apple menu, but it would be the optimal solution with the large range of rolls that Rhapsody is tryting to acheve. Would my solution work? What TCP services would it be likely to foul up? |\----/----/\/\-----------------------------------------| | \/\/ill / \eyer | | | wmeyer@mediaone.net | http://wmeyer.ne.mediaone.net | |_______________________________________________________| | "If I see one more cutesy, 'thought inspiring' quote | | in an email .sig, I am going to go postal!" | | -Will Meyer | |-------------------------------------------------------| |----BEGIN GEEK CODE BLOCK----- | | Version: 3.1 | | GCS/CC d? s++: a--- C+++>++++ ULX>+++$ P--- L+>+++ | | E W+++>$ N+ o? K- w--- !O M++$ V-- PS@ PE Y+ PGP | | t-- 5 X+ R- tv b+ DI+++ D- G e h r-->+ y? | |-----END GEEK CODE BLOCK------ | |-------------------------------------------------------| From domenici at onramp.net Thu Jan 1 16:14:18 1998 From: domenici at onramp.net (Matt Domenici) Date: Thu Nov 3 15:58:31 2005 Subject: Yellow & Blue boxes sharing an IP address Message-ID: <199801020011.SAA09561@mailhost.onramp.net> Actually- Blue is really a Yellow app, with some special features. ;-) It's closer to an app from my perspective (it's a process that can be killed without disturbing the rest of the system, the system isn't dependent upon it for services, etc.) - Matt Will Meyer (wmeyer@mediaone.net) said: >On 12/31/97 6:00 PM , Robert Fisher (rfisher@onr.com) wrote: > >>It seems to me that this is very feasible and straight forward. The blue >>box is just a yellow box application. When a blue app tried to open an ip >>port, the blue box would, in turn, try to open the same yellow ip port. If >>a blue app tries to open a port that is in use by a yellow app, it'll get >>the normal "port in use" error, because that's what the blue box will get >>when it tries to open the port on behalf of the blue app. If a yellow app >>tries to open a port that is in use by a blue app, it'll get the normal >>"port in use" error because the blue box (a yellow app itself) has already >>opened the port on behalf of that blue app. >> >>There's no need for "demultiplexing" when an inbound packet arrives. As is >>usual, it's given to the yellow app that holds the port it came in on. If >>that yellow app happens to be the blue box, then the blue box routes the >>packet to the proper blue app. > >One of the problems is that the blue box is NOT just a YB app. The Blue >Box is the real Mac OS running on Mach. It has an application interface >to start it and prepare the front end to have access to it. Therefore >the Mac OS has access to networking through the kernel in the same manner >that YB does. Routing all networking through YB would kill performence >and would also push BB farther from the real Mac OS. >> >>No changes need be made to the yellow box networking stack. All the changes >>would be within the blue box itself. Blue box apps might suffer some loss >>in performance, but yellow apps wouldn't. (And, assuming loss in >>performance is necessary for such a feature, better to make the blue apps >>suffer.) >> >>If you ping the machine, the yellow box will reply. After all, the yellow >>box is the real OS; the blue box is merely a compatibility environment. (I >>imagine it wouldn't be difficult to set up the blue box to respond to pings >>instead. Are pings handled by a server deamon or by the network stack >>itself? In any case, I don't see much point in doing that.) >> >>Obviously you couldn't run two servers on the same port (one in blue and >>one in yellow). It only makes sense that that would fail if you are sharing >>a single ip address between the two environments. Trying to do that would >>be as nonsensical as running two servers on the same port on any other OS. >> >One of the solutions that is being brought up over and over on this >thread is the possibility of having an IP Masq scheme withine the box to >distribute the TCP packets between OSes. AFAIK, IP Masq works by using >an alternate TCP port to send requests from the hidden machines. What if >one of the networking options was to have all Mac services transcribed >up, say, 20000 ports, transparent to the BB? In other words a Mac ftp >server would be active on port 20021 and Personal Web Sharing on port >20080. I don't know enough about TCP/IP to know the possibility of >implementing this, but if it worked it would let the Mac have it's own >servers and clients and have them be seperate to virtually all YB >operations while not doing any hiding of packets one enviroment to >another. > >>Of course, you'd also want the -option- of both environments having >>separate ip addresses if desired, but I don't know how feasible that is. >> >Are there currently options in a YB app or control panel to change >options for the BB? You could put a networking enu in there to change >between compatibility modes. That may be too complex for the average >artist-turned-DTPer who can't remember which menu is the Apple menu, but >it would be the optimal solution with the large range of rolls that >Rhapsody is tryting to acheve. > >Would my solution work? What TCP services would it be likely to foul up? > > > >|\----/----/\/\-----------------------------------------| >| \/\/ill / \eyer | | >| wmeyer@mediaone.net | http://wmeyer.ne.mediaone.net | >|_______________________________________________________| >| "If I see one more cutesy, 'thought inspiring' quote | >| in an email .sig, I am going to go postal!" | >| -Will Meyer | >|-------------------------------------------------------| >|----BEGIN GEEK CODE BLOCK----- | >| Version: 3.1 | >| GCS/CC d? s++: a--- C+++>++++ ULX>+++$ P--- L+>+++ | >| E W+++>$ N+ o? K- w--- !O M++$ V-- PS@ PE Y+ PGP | >| t-- 5 X+ R- tv b+ DI+++ D- G e h r-->+ y? | >|-----END GEEK CODE BLOCK------ | >|-------------------------------------------------------| > > +----------------------------------------------------------------------+ | Matt Domenici | | domenici@onramp.net | +----------------------------------------------------------------------+ From 9512087 at luc.ac.be Thu Jan 1 16:42:40 1998 From: 9512087 at luc.ac.be (Jasper) Date: Thu Nov 3 15:58:31 2005 Subject: Yellow & Blue boxes sharing an IP address In-Reply-To: <199801012202.RAA19043@chmls05.mediaone.net> Message-ID: Hello all, On Thu, 1 Jan 1998, Will Meyer wrote: > On 12/31/97 6:00 PM , Robert Fisher (rfisher@onr.com) wrote: > > >It seems to me that this is very feasible and straight forward. The blue > >box is just a yellow box application. When a blue app tried to open an ip > >port, the blue box would, in turn, try to open the same yellow ip port. If > >a blue app tries to open a port that is in use by a yellow app, it'll get > >the normal "port in use" error, because that's what the blue box will get > >when it tries to open the port on behalf of the blue app. If a yellow app > >tries to open a port that is in use by a blue app, it'll get the normal > >"port in use" error because the blue box (a yellow app itself) has already > >opened the port on behalf of that blue app. > > > >There's no need for "demultiplexing" when an inbound packet arrives. As is > >usual, it's given to the yellow app that holds the port it came in on. If > >that yellow app happens to be the blue box, then the blue box routes the > >packet to the proper blue app. > > One of the problems is that the blue box is NOT just a YB app. The Blue > Box is the real Mac OS running on Mach. It has an application interface > to start it and prepare the front end to have access to it. Therefore > the Mac OS has access to networking through the kernel in the same manner > that YB does. Routing all networking through YB would kill performence > and would also push BB farther from the real Mac OS. Routing all networking through YB might seem to kill performance,... but I did some testing here: I ftp-ed some files (average file size: 3,5Mb - binary) from an alpha machine to my PPC 9500/150Mhz with 80Mb RAM on 10base-T: theoretical maximum throughput 1.25Mb/s (not counting ftp-headers and so on). Speeds that I got exceeded 1.1 Mb/sec,... I was really impressed (compare with MacOS-based FTP-deamons). So I don't think people are going to use the blue box as a server. So I am convinced that routing the BB networking through the YB IS the best solution. Doing this MIGHT cause _some_ performance penalties, but the solution is more beautiful and will also encourage people to use the YB for what it's meant to be,... IMHO > >No changes need be made to the yellow box networking stack. All the changes > >would be within the blue box itself. Blue box apps might suffer some loss > >in performance, but yellow apps wouldn't. (And, assuming loss in > >performance is necessary for such a feature, better to make the blue apps > >suffer.) > > > >If you ping the machine, the yellow box will reply. After all, the yellow > >box is the real OS; the blue box is merely a compatibility environment. (I > >imagine it wouldn't be difficult to set up the blue box to respond to pings > >instead. Are pings handled by a server deamon or by the network stack > >itself? In any case, I don't see much point in doing that.) > > > >Obviously you couldn't run two servers on the same port (one in blue and > >one in yellow). It only makes sense that that would fail if you are sharing > >a single ip address between the two environments. Trying to do that would > >be as nonsensical as running two servers on the same port on any other OS. > > > One of the solutions that is being brought up over and over on this > thread is the possibility of having an IP Masq scheme withine the box to > distribute the TCP packets between OSes. AFAIK, IP Masq works by using > an alternate TCP port to send requests from the hidden machines. What if > one of the networking options was to have all Mac services transcribed > up, say, 20000 ports, transparent to the BB? In other words a Mac ftp > server would be active on port 20021 and Personal Web Sharing on port > 20080. I don't know enough about TCP/IP to know the possibility of > implementing this, but if it worked it would let the Mac have it's own > servers and clients and have them be seperate to virtually all YB > operations while not doing any hiding of packets one enviroment to > another. This is a networking problem: if Personal Web Sharing listens on port 20080, there's no problem: but are you going to tell everybody that when trying to connect to your machine they should connect to port 20080 instead of the normal 80 one? At this moment it's perfectly possible to run more then one httpd on different ports (8080 is also popular), but this means also specifying that port in the URL of each page. Not really kewl I guess. Whats more: with routing the tcp through the YB, this will be possible, without any nasty masquerading. You could have Webstar running in the blue box, and while you're switching to Apache on the yellow box, run that on a different port (8080 for example), do all the testing on it, and when finished: quit Webstar (thus free-ing port 80), and then running Apache on port 80 will do the job,... a nice transition, I'd say. This would also enhance the feeling the users get that the Blue Box is 'just a yellow box application', an impression that's really important as a concept to stick to I think. > >Of course, you'd also want the -option- of both environments having > >separate ip addresses if desired, but I don't know how feasible that is. > > > Are there currently options in a YB app or control panel to change > options for the BB? You could put a networking enu in there to change > between compatibility modes. That may be too complex for the average > artist-turned-DTPer who can't remember which menu is the Apple menu, but > it would be the optimal solution with the large range of rolls that > Rhapsody is tryting to acheve. > > Would my solution work? What TCP services would it be likely to foul up? I think Open Transport in the BB should be adapted with a 'connect through Rhapsody'-option: very clear and simple,... Optionally there could be an option to give it a seperate IP-adres, but I don't think it'd be necessary,... Greetz, Jasper Nuyens ;) From cmh at greendragon.com Thu Jan 1 17:38:43 1998 From: cmh at greendragon.com (Chris Hanson) Date: Thu Nov 3 15:58:31 2005 Subject: Yellow & Blue boxes sharing an IP address In-Reply-To: Message-ID: Why not just make the IP stack its own Mach task, like the BSD Unix environment, and have everything rely on it for IP networking services? In other words: +----------+----+----+ | Blue Box | | | + +-------+ | IP | | | Yellow Box | | +--+------------+----+ | Mach | +--------------------+ I mean, what's the point of having the BSD Unix single-server run IP networking compared to having IP networking run as its own task? The only real advantage I can think of for having BSD run the networking is that then you don't have Mach's messaging overhead, but I was under the impression that it supported optimizations for just this sort of thing. Besides, multi-server architectures and multi-processor hardware can work together nicely... From rcfa at cubiculum.com Thu Jan 1 18:46:58 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: enqueue GUI activated commands Message-ID: <9801020246.AA28754@kannix.cubiculum.com> I often wished that the NeXTSTEP/OPENSTEP Workspace would provide an option to enqueue commands given. Typical example: You copy a bunch of files to various places and then want to delete a parent directory that contains the source of these copy operations. Right now, you either have to use the CLI, or you have to wait until the copy operations are done, before you issue the delete command. What would be nice if after issuing a file operation, such as cmd-r, or dragging a folder elsewhere, an alert panel would pop up asking if the operation should be enqueued or executed in a separate thread. This simple change would allow a whole bunch of operations to be issued and completed as desired without additional user interaction or the need of keeping track of when one operation is completed before issuing the next one. Ronald From scott at cacti.org Thu Jan 1 21:07:10 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: enqueue GUI activated commands Message-ID: <199801020458.UAA04710@proxy3.ba.best.com> >I often wished that the NeXTSTEP/OPENSTEP Workspace would provide an option >to enqueue commands given. Typical example: You copy a bunch of files to >various places and then want to delete a parent directory that contains >the source of these copy operations. Granted, this isn't a gui, but.... to queue several commands, just seperate each with a ";" myhost> cp /Test_Directory/* /Test_Directory2/* ; rm -R /Test_Directory and to send several commands, each to a different thread, just end the command with a "&". myhost> cp /Test_Directory/* /Test_Directory2/* & myhost> cp rm -R /Test_Directory3 & - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From rcfa at cubiculum.com Thu Jan 1 21:46:56 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: enqueue GUI activated commands In-Reply-To: <9801020246.AA28754@kannix.cubiculum.com> References: <9801020246.AA28754@kannix.cubiculum.com> Message-ID: <9801020547.AA29135@kannix.cubiculum.com> > Granted, this isn't a gui, but.... > > to queue several commands, just seperate each with a ";" I'm fully aware of this. I have been using Unix for over 8 years now. Hence I wrote in my original post: > Right now, you either have to use the CLI [aka Command Line Interface or > Shell], or you have to wait until the copy operations are done, before you > issue the delete command. Sometimes however, the CLI is cumbersome, or I'm just too lazy to type or want the nice feedback of the Workspace Processes panel to monitor where things are at. I like to be prompted if symlinks are encountered, etc. I don't like to have to worry if files are moved over file system boundaries, etc. There are of course also plenty of cases when I would not want to miss the Shell, and where I'd ditch almost any sort of GUI in favor of a CLI. Greetings, Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From cake at cwr.uwa.edu.au Thu Jan 1 23:42:13 1998 From: cake at cwr.uwa.edu.au (David Cake) Date: Thu Nov 3 15:58:31 2005 Subject: Security was Re: Messing with standards: Just say NO! In-Reply-To: <01bd15d9$1c3ba780$0a01a8c0@pluto.rc-produktion.se> Message-ID: Replying to Robert and Ronald. Robert first. >Least Privilege solves that, even without the help of ACL's. Most Unix >kernels currently checks for uid 0 (root) and bypasses most permission >checks if the root user is running the process. Least Privilege replaces >this all-purpose super authority with discrete permission bits. There's >one permission for doing this, another for doing that, etc. Applications >can even add their own permissions. Creating a user with id 0, naming it >"root" and turning on all permissions for it sort of re-creates the root >user in such an environment, but the permission checks aren't bypassed any >longer. > Sounds like a great thing for them to be looking at adding, then. Count my vote in for Least Privelege. My problem is not with what is possible with Unix security - but with the standard unix way of doing security, parts of which are inadequate for Rhapsody (where you can't assume that a unix literate sysadmin is hanging around spending hours patching stuff), and other parts of which (like over-reliance on SUID) are inadequate for any Unix that wants to be thought of as secure. >All security information was stored in a metadata fork of each file. Yes, >this >was under Unix. I believe that HP-UX, AIX and others also store >their ACL's in >something that resembles the metadata fork. Should be no problem with HFS+ either. Ronald >> Which shouldn't mean 'shipping Unix systems with bad security flaws >> is OK, because true unix people will 'do-it-themselves' to fix the holes'. > >Bugs are bugs. I will not and do not defend those. However pretty much >all problems with Unix security are not due to the underlying model, but >due to bugs or due to the fact that there are insecure systems on the >same subnet But some practices either make it more likely that bugs will appear (like the SUID/buffer overrun combination) or make the bugs much more serious if they do appear (such as having services turned on by default). You need to look at things statistically - if a particular set of practices is a) likely to produce problems and b) used widely then statistically, there will be problems. The solution is to avoid that practice or change it so it is mcuh more likely to be safe, not to try and find and fix all the bugs. >What I defend here is the basic Unix security model, not the >flaws in the particular implementations thereof. The basic Unix security model is quite good if you compare it to any number of foolish things out there in the industry, like SMB or ActiveX. Which doesn't mean it can't substantially improved - and now, while Rhapsody has yet to become established, is the time to do so, by making sure that Rhapsody is a shining example of how to do unix security right, not yet another example of how to turn good security into mediocre by sloppy practices. >Not arguing they shouldn't. However, they shouldn't mess up the standard >APIs and the standard Unix security model while doing so. There is a distinction between 'standard'. meaning well-defined and widely available, and 'standard', meaning the way most people do it. Kerberos is a good example of something that fits the first definition, but no the second. And there is nothing wrong with Rhapsody aiming for a level of security that is higher than normal - indeed, I'd argue its essential, because it needs it more than most unices do. They should aim for best practice unix standards, not lowest common denominator unix standards. Which means that they should ship systems set up for security, not sysadmin convenience. And it means if there is a fairly standard alternative to the normal, yet fairly insecure, way most unices do things, then it should strongly be considered for Rhapsody. Like turning services OFF by default, like Kerberos, etc. >All in all, I don't think there is any major cause for worries in that >respect. I agree (surprisingly, Ronald, we seem mostly in agreement). And NeXTStep can certainly be MADE what I consider secure quite easily. But I'd still like to see more emphasis on out-of-box security. And if that means turning off services that Unix folks rely on, but most MacOS types don't use, then they should do it. Cheers David David Cake Certified Apple Engineer Centre for Water Research University of Western Australia From truong at nol.net Thu Jan 1 23:48:35 1998 From: truong at nol.net (Tuan Truong) Date: Thu Nov 3 15:58:31 2005 Subject: enqueue GUI activated commands In-Reply-To: <9801020547.AA29135@kannix.cubiculum.com> Message-ID: <3.0.5.32.19980102014835.007a6c80@nol.net> Can you bind a CMD-key to a script? Might be something to try with TickleServices. I'd imagine if a scripting environment was integrated into YB, you could record GUI events and make it a service. -Tuan From rcfa at cubiculum.com Fri Jan 2 00:11:40 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Security was Re: Messing with standards: Just say NO! In-Reply-To: References: Message-ID: <9801020811.AA29472@kannix.cubiculum.com> > >Not arguing they shouldn't. However, they shouldn't mess up the standard > >APIs and the standard Unix security model while doing so. > > There is a distinction between 'standard'. meaning well-defined and > widely available, and 'standard', meaning the way most people do it. > Kerberos is a good example of something that fits the first definition, but > no the second. And there is nothing wrong with Rhapsody aiming for a level > of security that is higher than normal - indeed, I'd argue its essential, > because it needs it more than most unices do. > They should aim for best practice unix standards, not lowest common > denominator unix standards. Which means that they should ship systems set > up for security, not sysadmin convenience. And it means if there is a > fairly standard alternative to the normal, yet fairly insecure, way most > unices do things, then it should strongly be considered for Rhapsody. Like > turning services OFF by default, like Kerberos, etc. The key issue from my perspective is that the standard APIs aren't broken. In other words, if I have a standard Unix program's source, I should be able to compile it and run it UNMODIFIED. I should not be in a position where I have to change anything about the way files or resources are accessed, how network connections are established, etc. in the code. Whatever security improvements are made, they have to be completely transparent to the code base and applications. > >All in all, I don't think there is any major cause for worries in that > >respect. > > I agree (surprisingly, Ronald, we seem mostly in agreement). Grin. ;-) > And > NeXTStep can certainly be MADE what I consider secure quite easily. But I'd > still like to see more emphasis on out-of-box security. And if that means > turning off services that Unix folks rely on, but most MacOS types don't > use, then they should do it. That should be an installation option, and one that the system remembers. It is one of the most obnoxious features about OS-MACH and NeXTSTEP that each time I do a system upgrade it wants to upgrade all my customized admin related files, such as /etc/disktab, /etc/crontab, /etc/services, /etc/swaptab, /usr/adm/daily, etc. etc. and I have to carefully review the upgrade plan to prevent that from happening. So if they turn off services by default, I at least don't want to have to worry about turning them back on after each system upgrade. Greetings, Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From david at onestep.co.uk Fri Jan 2 01:49:02 1998 From: david at onestep.co.uk (David Andrew Knight) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199712311855.KAA24020@scv1.apple.com> References: <199712311855.KAA24020@scv1.apple.com> Message-ID: <199801020951.JAA04062@onestep.co.uk> Hi, I was just about to post this and saw you were from Apple, maybe you would like to comment on the items below. You wrote: > Rhapsody will make a fine server platform. You may use it however you > wish. There is no "inappropriate use" of the platform. Rhapsody will make a fine, small, non-trusted (from a security point of view) server platform upon its release and may well mature into a more appropriate server platform over time. Its a matter of perspective (which I may have missed earlier in the thread), do you want it to serve a small office or a 200 user department, what services do you want to use it for, NetInfo, NFS, Web... > >Has Apple mentioned whether they plan on having Rhapsody compete with > >products like Solaris and WindowsNT (i.e., powerful workstation and server > >software) or is it more targeted towards single-user systems (i.e., > >graphics workstations, CAD, publishing). It would seem a shame for such a > >powerful system as Rhapsody's UNIX to be wasted as a client-only machine. > > There is no distinction (that I know of) between Rhapsody the client and > Rhapsody the server. Any Rhapsody machine can be either. It's not an NT > Workstation/Server kind of deal where some stuff is broken in one or the > other. There was some talk of Apple producing a server version which was a tuned version of the client product with bundled web server etc. I do not know if this is still the case. With the new layered file system you could see the server having a journal facility, larger maximum memory size, SMP > >I've looked through all the Rhapsody literature but I can't seem to pin > >down what the big picture is regarding Rhapsody. I strongly hope Rhapsody > >becomes both a robust single-user OS with a rich GUI, as well as a > >powerful, scalable server platform (with a pretty GUI to administer it, of > >course). > > This is currently the case. Is it ? Its not really scalable as there is no SMP support, still quite a small maximum memory (256MB ?), no resilience in terms failover controllers etc. Please understand none of this is aimed at criticizing Rhapsody, I just think that your words are a little to positive and without qualification in terms of Rhapsody based servers. As I said I may have missed something early in the thread but I would not risk, today, my client's server based operations to Rhapsody. --- Regards David Knight OneStep Solutions Plc | UK phone: 01702 426400 | Vendors of NS/OS 351 London Road | fax: 01702 551515 | MCCAs, Hardware Hadleigh | Int'l prefix: +44 1702 | Apps, Networks Essex | | ISDN, Training SS7 2BT | Email: david@onestep.co.uk | Maintenance England | (NeXTMail/MIME ok) | and Support From david at onestep.co.uk Fri Jan 2 02:08:12 1998 From: david at onestep.co.uk (David Andrew Knight) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801021010.KAA04134@onestep.co.uk> Whoops, That last posting was meant for wsanchez to comment on before I posted to the group. I knew I should have taken today off as well :-{ --- Regards David OneStep Solutions Plc From rcfa at cubiculum.com Fri Jan 2 02:30:55 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801020951.JAA04062@onestep.co.uk> References: <199801020951.JAA04062@onestep.co.uk> Message-ID: <9801021031.AA29801@kannix.cubiculum.com> > Rhapsody will make a fine, small, non-trusted (from a security point of > view) server platform upon its release and may well mature into a more > appropriate server platform over time. Its a matter of perspective (which I > may have missed earlier in the thread), do you want it to serve a small > office or a 200 user department, what services do you want to use it for, > NetInfo, NFS, Web... Rhapsody, the OS, can handle both, at least in it's final, tuned form that isn't bogged down by debugging code, etc. > There was some talk of Apple producing a server version which was a tuned > version of the client product with bundled web server etc. I do not know if > this is still the case. With the new layered file system you could see the > server having a journal facility, larger maximum memory size, SMP Tuning as in NT-Workstation vs. NT-Server, is only an issue in artificially cripple OSs. Whatever is tuneable about Rhapsody can be done by means of a user editable configuration file. SMP is planned for ALL Rhapsody versions, so there is nothing server specific about it. The maxium memory size is AFAIK a function of hardware restrictions and the fact that the OS runs on 32-bit hardware. If you really need a large address space, you need a 64-bit machine. Needless to say, I hope that Rhapsody will make a quick move to 64-bit hardware, be it 64-bit PPC, Alpha, Sparc or whatever Intel cooks up. All sorts of multimedia software will directly benefit from such a move. > Is it ? Its not really scalable as there is no SMP support, still quite a > small maximum memory (256MB ?), no resilience in terms failover controllers > etc. SMP is planned for all versions, but obviously, right now, a few other things are more important, like stability, BlueBox integration, etc. But it's coming. The 256MB I think, correct me if I'm wrong, are a PER PROCESS limit, not a limit of physical memory. The 256MB are, again correct me if I'm wrong, the result of using some of the 32 address bits for tagging information. A 64-bit hardware implementation of Rhapsody would take care of this. On the other hand, 256MB per process doesn't seem a major limitation for server applications. High-availability servers are a breed of their own. Not even most of SUN's servers fit that bill, and they are a very limited, highly specialized market. I would not expect Apple to go there anytime soon. (High availability is something with 99.95% uptime, or better. NT doesn't even come close to that. OS-MACH, although highly reliable, is not high-availability in that strict sense of the word. The 99.95% include system upgrades, etc. e.g. such servers must be upgradable while they are running and performing server functions, etc. Maybe once Apple uses a multi server MACH implementation, where various system services can be stopped, started and updated independently from the rest of the system, they may have a chance of entering this segment, if they even wish to do so). > Please understand none of this is aimed at criticizing Rhapsody, I just > think that your words are a little to positive and without qualification in > terms of Rhapsody based servers. As I said I may have missed something early > in the thread but I would not risk, today, my client's server based > operations to Rhapsody. I use an entirely OS-MACH based environment: dbserver, web server, file server, distributed apps, etc. with an uptime that makes any other normal OS go pale with envy. If the Rhapsody OS is anywhere near similar in stability, plus adds all the BSD4.4 enhancements, like bigger partitions, plus SMP, as is planned, then there is hardly a server task that this OS will not be up to. I would trust it more than Solaris and WinNT combined. Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From robert at rc-produktion.se Fri Jan 2 03:49:05 1998 From: robert at rc-produktion.se (Robert Claeson) Date: Thu Nov 3 15:58:31 2005 Subject: Security was Re: Messing with standards: Just say NO! Message-ID: <01bd1774$6d70dc00$0c00a8c0@toy.rc-produktion.se> Ronald C.F. Antony writes: >It is one of the most obnoxious features about OS-MACH and NeXTSTEP that >each time I do a system upgrade it wants to upgrade all my customized >admin related files, such as /etc/disktab, /etc/crontab, /etc/services, >/etc/swaptab, /usr/adm/daily, etc. etc. and I have to carefully review the >upgrade plan to prevent that from happening Many commercial Unix system vendors ship their upgrades with the timestamps of the files set to January 1 1980 or something like that. They then use a cpio option to not overwrite files newer than that. If there is a format upgrade to the file, it's also shipped as .new so that it gets installed and you can review it. Windows NT actually also does something similar, but they rely more on version numbers in binaries and libraries than on the timestamps of the files. From scott at cacti.org Fri Jan 2 08:31:30 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801021622.IAA11781@proxy4.ba.best.com> >> There was some talk of Apple producing a server version which was a tuned >> version of the client product with bundled web server etc. I do not know if >> this is still the case. With the new layered file system you could see the >> server having a journal facility, larger maximum memory size, SMP > >Tuning as in NT-Workstation vs. NT-Server, is only an issue in artificially >cripple OSs. Whatever is tuneable about Rhapsody can be done by means of >a user editable configuration file. SMP is planned for ALL Rhapsody >versions, >so there is nothing server specific about it. Yes -- I think this is a very important point. The functional differences between NT Workstation and NT Server are few -- except for the price tag. Remember that whole Netscape/Microsoft webserver fiasco? It's actually in the best interest of the consumer if Apple *doesn't* sell seperate server and workstation versions. However, then they won't be able to rip off consumers and they won't make as much money as MS does. :) - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From wsanchez at apple.com Fri Jan 2 10:42:26 1998 From: wsanchez at apple.com (Wilfredo Sanchez) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801021842.KAA12300@scv3.apple.com> >Rhapsody will make a fine, small, non-trusted (from a security point of >view) server platform upon its release and may well mature into a more >appropriate server platform over time. Its a matter of perspective (which I >may have missed earlier in the thread), do you want it to serve a small >office or a 200 user department, what services do you want to use it for, >NetInfo, NFS, Web... I'm not sure I see your point. Either it scales or it doesn't. I believe it will scale. It can also be as secure as any UNIX platform, and certainly more secure that NT. Rhapsody will do all of the above for any number of users. There's no "stop working if > 200 users" anywhere. You'll need to buy the hardware to support whatever load you need, and the system should be able to take it from there. This isn't *currntly* true, because we have a memery size limit right now, etc. But right now, we don't have a product, we have a prerelease for developers so they can get a feel for the system. If you want to use DR1 as a server, you loose. But we never said that would work. In fact, I will say it does not. >There was some talk of Apple producing a server version which was a tuned >version of the client product with bundled web server etc. I do not know if >this is still the case. With the new layered file system you could see the >server having a journal facility, larger maximum memory size, SMP I'm not sure I see why you think we'd "tune" the "server version" but not the "client version". Why bother with the distinction? Yes, I know NT has a client and a server, and Microsoft must know what they are doing. I'm sure they do, but I'm not clever enough to see their logic in making the distinction. >> This is currently the case. > >Is it ? Its not really scalable as there is no SMP support, still quite a >small maximum memory (256MB ?), no resilience in terms failover controllers >etc. Pardon me. I correct myself: "This is the plan, and we're on track." THEIR IS NOT PRODUCT RIGHT NOW. So I don't know what your point is. There is no SMP in the first developer release we shipped nine(?) months after we started the project. I don't see why that surprises you. Do we intend to implement SMP? Yes. That should be what you want. If you expected more, you are asking too much. I think, but maybe I'm nuts. :-) >Please understand none of this is aimed at criticizing Rhapsody, I just >think that your words are a little to positive and without qualification in >terms of Rhapsody based servers. As I said I may have missed something >early >in the thread but I would not risk, today, my client's server based >operations to Rhapsody. I don't see why you even considered it. Personally, I would not base my client's server ops on Windows 2010, either, but I thought that's a given. Today, there is no Rhapsody. The system doesn't even have a name yet. You seem to have grossy misinterpreted your license aggreement. You aren't licensed to deploy anything, anyway. We're workin' real hard. That I promise. -Fred --- Wilfredo Sanchez - wsanchez@apple.com - 408.974-5174 Apple Computer, Inc. - Rhapsody Core Operating Systems Group 2 Infinite Loop, Mail Stop 302-4K, Cupertino, CA 95014 From rcfa at cubiculum.com Fri Jan 2 11:52:35 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Security was Re: Messing with standards: Just say NO! In-Reply-To: <01bd1774$6d70dc00$0c00a8c0@toy.rc-produktion.se> References: <01bd1774$6d70dc00$0c00a8c0@toy.rc-produktion.se> Message-ID: <9801021952.AA01172@kannix.cubiculum.com> you wrote: > Ronald C.F. Antony writes: > > >It is one of the most obnoxious features about OS-MACH and NeXTSTEP that > >each time I do a system upgrade it wants to upgrade all my customized > >admin related files, such as /etc/disktab, /etc/crontab, /etc/services, > >/etc/swaptab, /usr/adm/daily, etc. etc. and I have to carefully review the > >upgrade plan to prevent that from happening > > Many commercial Unix system vendors ship their upgrades with the timestamps > of the files set to January 1 1980 or something like that. They then use a > cpio option to not overwrite files newer than that. If there is a format > upgrade to the file, it's also shipped as .new so that it gets > installed and you can review it. Windows NT actually also does something > similar, but they rely more on version numbers in binaries and libraries > than on the timestamps of the files. NeXT/Apple's upgrader software is CAPABLE of doing the right thing, it does so with all the others. For some reason though, the people who decide what the upgrade should do by default, decided that certain modified admin related files should be automatically reset to their virgin state, unless you explicitly choose the "plan upgrade" option and review the proposed changes, which take a lot of time... Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From paipai at ita.flashnet.it Fri Jan 2 13:30:27 1998 From: paipai at ita.flashnet.it (Paolo Di Francesco) Date: Thu Nov 3 15:58:31 2005 Subject: X and Rhapsody. Message-ID: <199801022027.VAA14434@ita.flashnet.it> I would know if Apple (or someone else) will port an X server to R. Whitout an X server, many "unix" apps will not run on R. or they must be ported to the R. window system. I think it's more convenient to port X than in single porting of every app. Some infos? Some ideas? #@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@# Reply to : paipai@ita.flashnet.it OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO Ciao Ciao Paolo Di Francesco _ ->B<- All Recycled Bytes Message ... ~ From rcfa at cubiculum.com Fri Jan 2 13:56:19 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: X and Rhapsody. In-Reply-To: <199801022027.VAA14434@ita.flashnet.it> References: <199801022027.VAA14434@ita.flashnet.it> Message-ID: <9801022156.AA01472@kannix.cubiculum.com> you wrote: > I would know if Apple (or someone else) will port an X server to R. > Whitout an X server, many "unix" apps will not run on R. or they must be > ported to the R. window system. I think it's more convenient to port X than > in single porting of every app. > Some infos? Some ideas? As far as I know, someone is in the process of porting X to Rhapsody. It needs to be said though, that porting things to the "Rhapsody window system" is more than just a new way of display. A native port will integrate, can take advantage of things like Services, Postscript printing, etc. So while an X server is a good stop-gap measure, as is the Blue Box, it is not a replacement for native ports. Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From paipai at ita.flashnet.it Fri Jan 2 17:11:23 1998 From: paipai at ita.flashnet.it (Paolo Di Francesco) Date: Thu Nov 3 15:58:31 2005 Subject: X and Rhapsody. In-Reply-To: <199801022139.WAA05526@bouleau.wanadoo.fr> Message-ID: <199801030007.BAA17962@ita.flashnet.it> > >I would know if Apple (or someone else) will port an X server to R. > >Whitout an X server, many "unix" apps will not run on R. or they must be > >ported > >to the R. window system. I think it's more convenient to port X than in > >single > >porting of every app. > >Some infos? Some ideas? > > > > There is an OpenStep X server. > YES! I think. I remember there was a NeXTStep server, and OS too. (public domain) If you have money, you can buy a commercial server... 8-) > Good start humm ? > #@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@# Reply to : paipai@ita.flashnet.it OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO Ciao Ciao Paolo Di Francesco _ ->B<- All Recycled Bytes Message ... ~ From paipai at ita.flashnet.it Fri Jan 2 17:11:23 1998 From: paipai at ita.flashnet.it (Paolo Di Francesco) Date: Thu Nov 3 15:58:31 2005 Subject: X and Rhapsody. In-Reply-To: References: <199801022027.VAA14434@ita.flashnet.it> Message-ID: <199801030007.BAA17950@ita.flashnet.it> > What I think would be great is if someone (maybe Apple) would make a set of > libs or better runtime objects which would allow exsisting X programs to > run without X but show up with rhapsody windows etc. "Run time" could be a possible solution, but it seems too difficult... Only Apple could do it, a free software project will never be able to do it: it's too specific (only Rhapsody, no linux, no freebsd, no ...etc) and time consuming. 8-( Too difficult! 8-( A public release of an X server it's avaible for NeXTStep (Openstep too, I think), but it's too old. This could be the way. I don't know which is needed to do an X server, so actually I don't know if it could be possible to port an Xfree86 (or something similar) server to the R. platform. Actually I don't have the time to do this, maybe in the next future...8-) > Since X was > originally designed as a protocol as best I can tell and not a GUI then > there are few human interface guide lines and that makes for inconsistant > uses of X across programs. Yes, but it must be not only a server, but something more: it must be able to communicate to the R. gui, so you could have two worlds: X and R. GUI. > This might make it difficult to do the X to Rhap windows shim. Hum... > I have heard that umich is doing some work on X for > rhapsody but to what end or extent I do not know. umich? What's umich? (university of Michigan?) It could be an idea: many people (student or slaves, same thing...8-)) ), high tecnology, etc... But we need NOW! I don't want to wait 5 years, or I'll use Linux ! 8-) #@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@# Reply to : paipai@ita.flashnet.it OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO Ciao Ciao Paolo Di Francesco _ ->B<- All Recycled Bytes Message ... ~ From kylep at nwlink.com Fri Jan 2 16:23:57 1998 From: kylep at nwlink.com (kylep) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <1328360681-56880633@csnw.com> Wilfredo Sanchez wrote: >I'm not sure I see why you think we'd "tune" the "server version" but not >the "client version". Why bother with the distinction? Yes, I know NT has >a client and a server, and Microsoft must know what they are doing. I'm >sure they do, but I'm not clever enough to see their logic in making the >distinction. Wasn't there a small scandal when someone published how to change a couple of lines in the registry on NT Client and have it turn into NT server (complete with the correct Server version splash screens)??? From paul at eisusa.com Sat Jan 3 10:29:37 1998 From: paul at eisusa.com (Paul Nicholson) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death - can't command-option-PR out of it. In-Reply-To: <1328360681-56880633@csnw.com> Message-ID: By now many of us have experienced the Black Screen of Death, for example if you try to boot to an external rhapsody disk that was not powered up. I've always been able to recover from this malady by judicious use of reset and command-option-PR to reset the parameter RAM. The last time I recovered from this I had to power off the 9500 in addition to zapping the PRAM. Well it happened again, and this time it refuses to recover. The machine sounds the bong chime when it powers up, but nothing happens after that. No monitor initialization, floppy, or SCSI activity - nothing. I've checked the monitor card in another machine (works great), disconnected the SCSI disks (the boot disk will boot a 7200), and given it a floppy to try to boot from (no access is even attempted). It's dead as a doornail. I've removed the lithium battery from the mother board, and pushed the little red pushbutton next to the ADB/parameter RAM chip - all to no avail. Perhaps I've suffered a hardware failure, but when the incident happened, I powered up the 9500 without turning on the Rhapsody disk, which is a sure fire way to invoke the black screen of death. I have a hard time accepting the odds that a hardware failure would occur coincidently that exactly mimics the black screen of death, but then what could be in the PRAM that won't go away with command-option-PR, battery removal, or pressing the mother board PRAM reset switch? Has anyone else seen this and were you able to recover? And here's a thought. Where is the machines Ethernet address stored? Is it possible that there is some non-volitile memory that is used to store the ethernet address and perhaps other stuff that has been erased, thereby preventing the machine from booting? And if so, then how do you recover? Paul From scott at cacti.org Sat Jan 3 15:00:13 1998 From: scott at cacti.org (scott) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death Message-ID: <199801032251.OAA27405@proxy4.ba.best.com> >Well it happened again, and this time it refuses to recover. The machine >sounds the bong chime when it powers up, but nothing happens after that When I first had this problem, my first instinct was to reset the pram as well. However, when it happens to me now, I follow these steps (until one works): 1. Verify the caps lock key is *not* down 2. Soft reset the machine the black screen (this usually solves it) 3. Soft reset + pram reset 4. Hard power off the machine and turn back on 10 seconds later. At black screen, soft reset. 5. Hard power off machine. Power on holding command-option-p-r down. 6. Keep soft reseting and holding down command-option-p-r 7. Try booting from a MacOS 7.6 or 8 CD (holding down C) >And here's a thought. Where is the machines Ethernet address stored? As far as I know, this is etched into the ethernet device. I don't think it can be erased (correct me if I'm wrong). - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From jblumel at mailhost.tcs.tulane.edu Sat Jan 3 15:17:40 1998 From: jblumel at mailhost.tcs.tulane.edu (John Blumel) Date: Thu Nov 3 15:58:31 2005 Subject: Submission: WebWriter v0.1 Message-ID: <199801032317.RAA00235@wymple.gs.net> Just submitted to ftp.peak.org WebWriter v0.1 PPC & Intel binaries WebWriter is a freeware HTML tagging editor for Rhapsody. This WebWriter.app is not related in any way to Robert Bose's WebWriter.app for NeXTStep. The name comes instead from an OS/2 based program, WebWriter/2 that I developed from 1995-1997. I have, however, corresponded with Robert Bose to make sure that he had no objections to me using the name and he has indicated that he does not. Any resemblance between my WebWriter and his is purely coincidental. WebWriter, in its current form supports the HTML 3.2 specification and also includes support for some commonly used extensions such as FRAMEs. Currently this support is available only as menu options for inserting tags, attributes and entities. Future releases should be more 'user friendly' and include prompt dialogs to help out with tag and attribute syntax, drag and drop support, and other niceties. I'm releasing this version now so that those who prefer a tag based editor and would like to create HTML documents using a native Rhapsody tool will have something to work with. John Blumel jblumel@omi.tulane.edu Assistant Director (504) 599-6773 Office of Medical Informatics Tulane University Medical Center From flight at thefly.mathi.uni-heidelberg.de Fri Jan 2 03:56:42 1998 From: flight at thefly.mathi.uni-heidelberg.de (Gregor Hoffleit) Date: Thu Nov 3 15:58:31 2005 Subject: Security was Re: Messing with standards: Just say NO! In-Reply-To: <9712310807.AA23250@kannix.cubiculum.com> References: <9712310807.AA23250@kannix.cubiculum.com> Message-ID: <9801021156.AA00363@thefly.mathi.uni-heidelberg.de> "Ronald C.F. Antony" wrote: you wrote: > >> Plug-in authentication architecture > > > >Already in NeXTSTEP/OPENSTEP, just not documented and widely > >used. > > > Evangelise evangelise, Apple. Better yet, howabout if Apple > actually used it, allowing an option of more powerful encryption > than crypt as part of a standard install? And how about packet > level encryption, like IPSEC? All these things run up against retarded US legislation. It's not even that Apple Europe could do these things, since they are still considered subsidiaries of a US company and the parent would be held liable for trying to circumvent export laws. Until some US computer companies smarten up and move their headquarters elsewhere, and declare the US branches as subsidiaries, there will be not much change in this respect. Apple would have a hell to deal with all the various versions of the software and the interoperability problems between strong US versions and weak international versions. It gets worse: even sales to non-US citizens within the US is considered export, thus you would have to go with the passport to the local computer store in order to buy Rhapsody. In short it's not feasible. The only thing they can do, is take free implementations, port them, and put the diffs on some ftp site, such that people who are so inclined, can modify their systems themselves. Actually, this is exactly what PAM is about: Define an interface where third parties can plug in their authentification modules so that they will be used by all tools in need of authentification services. Solaris has it, Linux has it.. NeXTstep/OPENSTEP doesn't support it AFAIK, as well as RDR. Gregor --- | Gregor Hoffleit admin MATHInet / contact HeidelNeXT | | MAIL: Mathematisches Institut PHONE: (49)6221 56-5771 | | INF 288, 69120 Heidelberg / Germany FAX: 56-3812 | | EMAIL: flight@mathi.uni-heidelberg.de (NeXTmail) | -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 2127 bytes Desc: not available Url : /mailman/archive/macosx-talk/attachments/19980102/a0ce7278/attachment.bin From robert at rc-produktion.se Sun Jan 4 03:38:05 1998 From: robert at rc-produktion.se (Robert Claeson) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <01bd1905$38f2a980$0a01a8c0@pluto.rc-produktion.se> kylep writes: >Wasn't there a small scandal when someone published how to change a >couple of lines in the registry on NT Client and have it turn into NT >server (complete with the correct Server version splash screens)??? It was O'Reilly & Associates. I don't know about the splash screen, but NT Server comes with several apps that just aren't there in the Workstation package, so you don't get a complete NT Server. Otherwise they seem to be the same. Robert From david at onestep.co.uk Sun Jan 4 08:30:59 1998 From: david at onestep.co.uk (David Andrew Knight) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801021622.IAA11781@proxy4.ba.best.com> References: <199801021622.IAA11781@proxy4.ba.best.com> Message-ID: <199801041633.QAA06837@onestep.co.uk> Hi, You wrote: > >> There was some talk of Apple producing a server version which was a > >> tuned version of the client product with bundled web server etc. I do > >> not know if this is still the case. With the new layered file system you > >> could see the server having a journal facility, larger maximum memory > >> size, SMP > > > >Tuning as in NT-Workstation vs. NT-Server, is only an issue in > >artificially cripple OSs. Whatever is tuneable about Rhapsody can be done > >by means of a user editable configuration file. SMP is planned for ALL > >Rhapsody versions, > >so there is nothing server specific about it. > > Yes -- I think this is a very important point. The functional > differences between NT Workstation and NT Server are few -- except for > the price tag. Remember that whole Netscape/Microsoft webserver fiasco? > It's actually in the best interest of the consumer if Apple *doesn't* > sell seperate server and workstation versions. However, then they won't > be able to rip off consumers and they won't make as much money as MS > does. :) Bearing in mind my fauxpas with the previous email/posting I think I should explain what I meant by tuning, perhaps packaging would have been a better word to use. The server version would be 'tuned' or 'packaged' with a journalizing filesystem for faster system recovery, ie no long fsck, with disk performance tuned for NFS rather than say VM, etc. It would include 3rd party products for Web Servers, maybe LDAP/DHCP servers etc. All this adds value to the product and allows Apple to differentiate and charge more thus making more money from Companies who wish to buy the Apple Rhapsody badged 'Server' version. This is not a rip-off and there are many companies who have a preference for buying pre-packaged server suites of Operating system, web server, web-based admin tools etc. For those who wish to use free products that provide the same function they can buy the workstation version which hopefully Apple will not restrict in the licence agreement as other companies seem to do. --- Regards David OneStep Solutions Plc From robert at rc-produktion.se Sun Jan 4 08:49:55 1998 From: robert at rc-produktion.se (Robert Claeson) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <01bd1930$c8d9e650$0c00a8c0@toy.rc-produktion.se> David Andrew Knight writes: >Bearing in mind my fauxpas with the previous email/posting I think I should >explain what I meant by tuning, perhaps packaging would have been a better >word to use. > >The server version would be 'tuned' or 'packaged' with a journalizing >filesystem for faster system recovery, ie no long fsck, with disk performance >tuned for NFS rather than say VM, etc. It would include 3rd party products >for Web Servers, maybe LDAP/DHCP servers etc. Most Unix system vendors offer the ability to tune the operating system for specific uses (or the operating system does it by itself by finding it out from the kind of system it is running on). There's usually not a difference in features. Even a workstation can make great use of a journaling file system. The tuning is usually of the kind of opimising for quick interactive response vs file server vs database server use and the like. Robert From eharley at pacbell.net Sun Jan 4 09:05:29 1998 From: eharley at pacbell.net (Eric Harley) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: >It was O'Reilly & Associates. I don't know about the splash screen, but NT >>Server comes with several apps that just aren't there in the Workstation >>package, so you don't get a complete NT Server. Otherwise they seem to be >the >same. Yeah, but at work I run NT Workstation developing intranet apps with WebObjects. I have been running up against the wall with the limit of 10 connections when I do testing. Man, what a help that registry hack was! Wash, Lather, Rinse, Repeat -- Eric Harley eharley@pacbell.net From paul at eisusa.com Sun Jan 4 09:56:59 1998 From: paul at eisusa.com (Paul Nicholson) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death In-Reply-To: <199801032251.OAA27405@proxy4.ba.best.com> Message-ID: Thanks to all who replied regarding my black screen of death. >5. Hard power off machine. Power on holding command-option-p-r down. I looked at Apple's tech info to see what the proper key combination was for zappping PRAM and I found a reference that it was command-shift-P-R. I was using command-option-P-R. Neither of these sequences worked. I then tried command-control-P-R and was rewarded with monitior image. Removing the battery and resetting the PRAM with the red reset button next to the keyboard controller/PRAM/RTC chip does not work. Perhaps they use EEPROM on this chip, which I believe is based on the Motorola 6805, (a device that is available in variants with EEPROM). Scott Stevenson replied: >>And here's a thought. Where is the machines Ethernet address stored? > > As far as I know, this is etched into the ethernet device. I don't >think it can be erased (correct me if I'm wrong). Adapter cards usually use an 8 pin 256 bit serial EEPROM to hold the ethernet address. SInce the address is assigned by the card vendor and not the chip manufacturer, it is most easily programmed when the card is manufactured. Extra processing steps are required to add EEPROM or EPROM storage transistors and it is not done on ethernet controller chips, at least this is the case with the DEC fast ethernet chips. I don't think Apple puts EEPROM cells on the glue chips that support onboard ethernet and other functions. I suspect the address is stored off chip, but where? Paul Paul Nicholson Electronic Imaging Systems, Inc. TEL 805 532 1068 5148 Commerce Avenue, Unit F FAX 805 532 1065 Moorpark, CA 93021 U.S.A. paul@eisusa.com http://www.eisusa.com From rcfa at cubiculum.com Sun Jan 4 13:52:49 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801041633.QAA06837@onestep.co.uk> References: <199801041633.QAA06837@onestep.co.uk> Message-ID: <9801042152.AA07538@kannix.cubiculum.com> > Bearing in mind my fauxpas with the previous email/posting I think I should > explain what I meant by tuning, perhaps packaging would have been a better > word to use. > > The server version would be 'tuned' or 'packaged' with a journalizing > filesystem for faster system recovery, ie no long fsck, with disk > performance tuned for NFS rather than say VM, etc. It would include 3rd > party products for Web Servers, maybe LDAP/DHCP servers etc. Do you really think, that I would not want these features IMMEDIATELY in my client version of the OS, if Apple had them ready? Also, OS-MACH, if that's any indication, comes with full suite of admin tools, directory services, etc. The only web server Apple ships is Apache. Not only is it in my opinion the best web server availabla at any price, it is also free. > All this adds value to the product and allows Apple to differentiate and > charge more thus making more money from Companies who wish to buy the Apple > Rhapsody badged 'Server' version. This is not a rip-off and there are many > companies who have a preference for buying pre-packaged server suites of > Operating system, web server, web-based admin tools etc. Yes it is a rip off. I wish Apple well, but I don't want to pay for packaging and a different sticker, and I don't want Apple make other people who are too stupid to realize that they have been ripped of, be made to pay either. > OneStep Solutions Plc Hm, of course, if you are/plan on being an Apple reseller, this all makes sense: It's not about Apple making more money, but about you making more. More expensive product, more margin... Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From julian.morrison at virgin.net Sun Jan 4 17:22:17 1998 From: julian.morrison at virgin.net (Julian Morrison) Date: Thu Nov 3 15:58:31 2005 Subject: Compatibility? Message-ID: <34B035C8.CEE3449F@virgin.net> I'm thinking about getting DR1/Intel (a) Is this machine compatible? Which bits are unsupported? --begin machine description-- AMD K6 200 S3 Trio 64 V+ (2mb) 48mb RAM 850mb primary IDE HDD (with Win95 on) formatted vfat (not fat32) 2.5gb second IDE HDD (the one I would put Rhapsody on) sound blaster 16 Samsung SyncMaster 3Ne monitor Microsoft natural keyboard Psion Dacom Surfer voice modem Creative IDE quad-speed CDROM on the sound blaster IDE port (linux can't seem to recognize this) --end machine description-- (2) Is the boot loader safe to use for dual booting into Win95 and Rhapsody? (3) Is there any chance Rhapsody will screw up my win95 disk? (4) Will my win95 be able to communicate with Rhapsody at all? Can Rhapsody read vfat/fat partitions? -- Julian Morrison From scott at cacti.org Mon Jan 5 00:04:33 1998 From: scott at cacti.org (scott) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death Message-ID: <199801050755.XAA26513@proxy4.ba.best.com> >I looked at Apple's tech info to see what the proper key combination was >for zappping PRAM and I found a reference that it was command-shift-P-R. I >was using command-option-P-R. Neither of these sequences worked. I then >tried command-control-P-R and was rewarded with monitior image. The appearance of monitor image doesn't guarantee that "command-control-P-R" is the correct pram sequence. You should be able to tell because you'll hear the startup chime, hold the keys down, then hear the startup chime again. If you don't hear the chime, the PRAM probably wasn't cleared. Since the beginning of time, Command-Option-P-R has cleared PRAM. It's always worked for me -- but clearing the PRAM doesn't always fix the Rhapsody monitor problem. The keys may be different for non-Apple keyboards(?). - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From scott at cacti.org Mon Jan 5 00:09:41 1998 From: scott at cacti.org (scott) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801050800.AAA03120@proxy4.ba.best.com> >The server version would be 'tuned' or 'packaged' with a journalizing >filesystem for faster system recovery, ie no long fsck, with disk >performance >tuned for NFS rather than say VM, etc. It would include 3rd party products >for Web Servers, maybe LDAP/DHCP servers etc. The third party products bit I agree with, however the filesystem coponents shouldn't differ from package to package. If Apple figures out a way to make the filesystem faster, more stable, or able to recover quick -- it should be in both packages. The idea of "well, this is just a workstation, so it doesn't have to reboot as fast" seem instinctively wrong to me. This does seem like it would be a case of crippling the workstation/client version. Then you could conceivably get the response "if your reboots are taking too long -- just get the server version". Maybe that's just what I could see MS doing, though. Additionally software packages, though, like a professional webserver, personal oracle or whatever are quite logical to have in an "enterprise package". - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From scott at cacti.org Mon Jan 5 00:14:01 1998 From: scott at cacti.org (scott) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801050805.AAA01633@proxy3.ba.best.com> >Yeah, but at work I run NT Workstation developing intranet apps with >WebObjects. I have been running up against the wall with the limit of 10 >connections when I do testing. Man, what a help that registry hack was! Pirating software solves a lot of financial problems. :) As weird as it is for Microsoft to cripple the workstation version just to charge more for the server, I don't think it's ethical to hack the registry to allow more connections. If you need the server version, buy it. If not, don't buy it or find something else. The way we've been talking about this sounds as if the workstation version of NT is just a bug -- and the registry hack is a workaround fix. Since MS wrote the code (presumably), it's up to them how much they want to charge for it. (getting down off of soapbox) - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From sanguish at digifix.com Mon Jan 5 00:10:53 1998 From: sanguish at digifix.com (Scott Anguish) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death In-Reply-To: <199801050755.XAA26513@proxy4.ba.best.com> References: <199801050755.XAA26513@proxy4.ba.best.com> Message-ID: <199801050810.DAA24814@digifix.com> scott wrote: > >I looked at Apple's tech info to see what the proper key combination was > >for zappping PRAM and I found a reference that it was command-shift-P-R. I > >was using command-option-P-R. Neither of these sequences worked. I then > >tried command-control-P-R and was rewarded with monitior image. > > The appearance of monitor image doesn't guarantee that > "command-control-P-R" is the correct pram sequence. You should be able > to tell because you'll hear the startup chime, hold the keys down, then > hear the startup chime again. If you don't hear the chime, the PRAM > probably wasn't cleared. > Whoa.. really? I've had to reboot (that is turn the machine off and then on repeatedly) fairly often after shutting down directly from Rhapsody. I THOUGHT I had been clearing the PRAM when I was doing this using Command-Option P-R.. but I've never heard an additional chime.. This is on an Apple keyboard (whichever one comes with the 8600/250) > Since the beginning of time, Command-Option-P-R has cleared PRAM. It's > always worked for me -- but clearing the PRAM doesn't always fix the > Rhapsody monitor problem. > > The keys may be different for non-Apple keyboards(?). > Its funny, this has been the biggest headache for me with my installations of machines running Rhapsody to this point... Seems like a pretty small thing, but its dang annoying.. can't wait for a fix! :-) From scott at cacti.org Mon Jan 5 00:57:51 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death Message-ID: <199801050848.AAA26495@proxy3.ba.best.com> > I've had to reboot (that is turn the machine off and then on >repeatedly) fairly often after shutting down directly from Rhapsody. Here's what works for me fairly reliably: When you want to switch from MacOS to Rhapsody, click the caps lock key down, and choose "restart" from the special menu. As soon as Rhapsody begins to load, un-click the caps lock key. When you want to switch back to MacOS -- make sure the caps lock key is RELEASED. Choose "shutdown" from the File menu (or login screen). As the machine is shutting down -- that is, when you see the "Apple Rhapsody Mach Operating System" screen with all the text shooting across -- click down the caps lock key. Once the machine is powered off, turn it back on (with the key still locked down), and wait until you see the MacOS logo. At that point -- press the caps lock key again to un-lock it. BTW: About the PRAM chime - you won't hear it if you have something plugged into the headset jack. (I love that feature!) :) - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From david at onestep.co.uk Mon Jan 5 01:30:27 1998 From: david at onestep.co.uk (David Andrew Knight) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801050800.AAA03120@proxy4.ba.best.com> References: <199801050800.AAA03120@proxy4.ba.best.com> Message-ID: <199801050933.JAA07433@onestep.co.uk> Hi, +You wrote: > >The server version would be 'tuned' or 'packaged' with a journalizing > >filesystem for faster system recovery, ie no long fsck, with disk > >performance > >tuned for NFS rather than say VM, etc. It would include 3rd party products > >for Web Servers, maybe LDAP/DHCP servers etc. > > The third party products bit I agree with, however the filesystem > coponents shouldn't differ from package to package. If Apple figures out > a way to make the filesystem faster, more stable, or able to recover > quick -- it should be in both packages. > > The idea of "well, this is just a workstation, so it doesn't have to > reboot as fast" seem instinctively wrong to me. This does seem like it > would be a case of crippling the workstation/client version. Then you > could conceivably get the response "if your reboots are taking too long > -- just get the server version". Maybe that's just what I could see MS > doing, though. The problem is that a journal based filesystem whilst faster to recover and more resilient can be somewhat slower and steal more cpu cycles. It all depends on the approach, ie linear control or multi-process with pre-sorted sequential output and buffering, there are also options for journalling on paired disks which would not be possible on typical workstation configs (ie single disk) --- Regards David OneStep Solutions Plc From david at onestep.co.uk Mon Jan 5 01:29:35 1998 From: david at onestep.co.uk (David Andrew Knight) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <9801042152.AA07538@kannix.cubiculum.com> References: <9801042152.AA07538@kannix.cubiculum.com> Message-ID: <199801050932.JAA07427@onestep.co.uk> Hi, You wrote: > > All this adds value to the product and allows Apple to differentiate and > > charge more thus making more money from Companies who wish to buy the > > Apple Rhapsody badged 'Server' version. This is not a rip-off and there > > are many companies who have a preference for buying pre-packaged server > > suites of Operating system, web server, web-based admin tools etc. > > Yes it is a rip off. I wish Apple well, but I don't want to pay for > packaging and a different sticker, and I don't want Apple make other > people who are too stupid to realize that they have been ripped of, > be made to pay either. I did not say YOU had to pay more, I mentioned before that the freeware route should stay open with a non-restrictive licence on the workstation version. > Hm, of course, if you are/plan on being an Apple reseller, this all > makes sense: It's not about Apple making more money, but about you > making more. More expensive product, more margin... This is just plain insulting. Most of our corporate customers source their product direct or from the most discounted source. We then add value by... (no point is advertising and getting flamed). I advocated both approaches, a corporate approach which is what many of them expect AND WANT and the existing approach using Apache and other freely available software. I can understand people not wanting to be forced down a NT type approach with a higher cost server product and I agree. But many corporates want a higher value server package with named, supported (ie through a large corporate like say NetScape) bundled packages and these two approaches should not be mutually exclusive. --- Regards David OneStep Solutions Plc From rcfa at cubiculum.com Mon Jan 5 01:25:18 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801050800.AAA03120@proxy4.ba.best.com> References: <199801050800.AAA03120@proxy4.ba.best.com> Message-ID: <9801050925.AA09069@kannix.cubiculum.com> I think there is a big difference between an add-on package, and a differently branded server OS product. The former may be acceptable, the second isn't. There may be the Developer package, with the compiler, tools, headers, libs etc. There could be a Web package with WebObjects, the Apache server (preferrably with a native Apache module based WOAdaptor added), there could be a database package, etc. And that's ok. After all, not everybody needs all of this, nor does everyone want to waste the diskspace for all the stuff. On the other hand, nothing needs really that much space that these things wouldn't fit on a single CD-ROM or maybe DVD-ROM and couldn't be selectively installed by reasonably smart installation tools. But the idea of a different server version, that is priced differently and leaves the option of a crippled client version, etc. makes my gall come up. Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From nberch at db.lv Mon Jan 5 05:27:45 1998 From: nberch at db.lv (Nils Berzins) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801051124.NAA19810@saulite> > The third party products bit I agree with, however the filesystem >coponents shouldn't differ from package to package. If Apple figures out >a way to make the filesystem faster, more stable, or able to recover >quick -- it should be in both packages. They are moving Rhapsody to HFS+ (...I hope!). Then we will get rid at least of some UN*Xisms like fsck... Nils Berzins From chongt at bah.com Mon Jan 5 03:52:13 1998 From: chongt at bah.com (Chong Tim) Date: Thu Nov 3 15:58:31 2005 Subject: Internet Tools ? References: <9710151930.AA71518@torak.corp.apple.com> Message-ID: <34B0C96C.7BFC4202@bah.com> Is it me or is RDR/Intel can't print yet? Click Create in PrintManager.app, the thing just quit. TIA, TC From rcfa at cubiculum.com Mon Jan 5 04:58:01 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801050933.JAA07433@onestep.co.uk> References: <199801050933.JAA07433@onestep.co.uk> Message-ID: <9801051258.AA09540@kannix.cubiculum.com> > The problem is that a journal based filesystem whilst faster to recover and > more resilient can be somewhat slower and steal more cpu cycles. It all > depends on the approach, ie linear control or multi-process with pre-sorted > sequential output and buffering, there are also options for journalling on > paired disks which would not be possible on typical workstation configs (ie > single disk) While this is true, it's no argument for splitting up versions. OS-MACH right now supports Mac, MS-DOS, CD-ROM, BSD4.3, NFS file systems. Rhapsody will add to these BSD4.4, HFS+ and whatever other options they may choose, one of which could be a journalling file system. It is then up to the user to choose whatever they think fits their needs the best. Greetings, Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From rcfa at cubiculum.com Mon Jan 5 05:03:18 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801051124.NAA19810@saulite> References: <199801051124.NAA19810@saulite> Message-ID: <9801051303.AA09556@kannix.cubiculum.com> > They are moving Rhapsody to HFS+ (...I hope!). Then we will get rid at > least of some UN*Xisms like fsck... I severely doubt it. Unless HFS+ is journalling, it too, will need fsck, or an equivalent utility, if an unclean shutdown has happened. Note, that fsck is only run upon an unclean shutdown (system crash, power failure, etc.). If this is not the case in Rhapsody, then only because you're dealing with a system under development. OS-MACH e.g. does not run fsck each time it boots. It checks a clean-flag, which is set upon successful clean shutdown, and then skips fsck at boot time. Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From rcfa at cubiculum.com Mon Jan 5 04:54:36 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801050932.JAA07427@onestep.co.uk> References: <9801042152.AA07538@kannix.cubiculum.com> <199801050932.JAA07427@onestep.co.uk> Message-ID: <9801051254.AA09525@kannix.cubiculum.com> > I can understand people not wanting to be forced down a NT type approach > with a higher cost server product and I agree. But many corporates want a > higher value server package with named, supported (ie through a large > corporate like say NetScape) bundled packages and these two approaches > should not be mutually exclusive. I don't see what the point is of doing bundled packages. If someone wants e.g. a Netscape web server, they can go out and buy it. There isn't any more reason to bundle a particular word processor with the "client" version than there is to bundle a particular web server with the "server" version. I want to buy an OS. If I need additional sofware, be it web server, text processing or spreadsheet or development tools, I go and buy them as I need them. There is no need for Apple to introduce confusion to the Rhapsody OS market by segmenting it, other than to rip off people and charge more, along the lines MS did it. I hope Apple will stay away from this. As long as Apple doesn't write their own web server, there is no need for them to bundle anything but good freeware as a courtesy service. If MS, Netscape or any of the other server sources want to sell their product, or if anyone wants to buy their products, there are standard software distribution channels that take care of it. If a particular customer doesn't want to deal with the hassle of putting things together, then they should buy a preconfigured "bundle" from a VAR, but shouldn't expect Apple to get into that game. Last I checked, that's what system integrators and VARs are for. Ronald PS: Sorry if you were insulted by my previous post. It wasn't intended, but sometimes a little sarcastic comment presenting itself on a silver platter is hard to pass... ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From dwittenb at dome.weeg.uiowa.edu Mon Jan 5 06:23:01 1998 From: dwittenb at dome.weeg.uiowa.edu (daniel-wittenberg@uiowa.edu) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security Message-ID: I just wanted to see what people thought about having TCP wrappers and SSH security built into Rhapsody. I know I couldn't live without them as a Unix Admin, and as a Mac fan I'd love to see those as a standard package. I don't know if there is some legal reason for incorporating, but has anyone ported SSH yet? I know tcp wrappers will compile without too much trouble, though I haven't done much testing.. Dan -------------------------------------------------------------------------------- Daniel Wittenberg System Administrator University of Iowa Phone: (319) 335-6279 FAX: (319) 335-5505 daniel-wittenberg@uiowa.edu http://dan.weeg.uiowa.edu -------------------------------------------------------------------------------- From C.Ridd at isode.com Mon Jan 5 06:21:18 1998 From: C.Ridd at isode.com (Chris Ridd) Date: Thu Nov 3 15:58:31 2005 Subject: Blue TCP/IP, and Rhapsody UI (fwd) In-Reply-To: Your message of "Wed, 31 Dec 1997 02:36:42 EST." <9712310736.AA23179@kannix.cubiculum.com> Message-ID: <1495.884010078@isode.com> On Wed, 31 Dec 1997 02:36:42 EST, "Ronald C.F. Antony" wrote: > you wrote: > > Several people where trying to convince them to > > change the architecture - to instead, just implement STREAMS on the BSD > > side (which would probably take only weeks if you get the Mentat people to > > do it, who specialise in porting STREAMS to things, and have worked with > > Apple before on Open Transport), and then you can use the same stack for > > both, all you need to do is write lots and lots (and lots and lots) of shim > > code for the Blue box, so that each Open Transport call just calls the > > appropriate call for the Yellow networking stack. This may happen, but I > > think they were sticking with the original plan for the Premiere deadline, > > as they thought that due to the sheer quantity of shim code needed, it > > would take longer, while the other plan required only a small amount of > > shim code (to replace the Blue network device drivers). > > Not only would it require lots of shim code, it would probably also take > lots of changes to the BSD side of things, which is more or less completely > based on sockets. So you'd either have to create some translation there, too, That appears to be exactly what Sun did with Solaris - I believe Solaris uses streams internally and has sockets working on top of streams. The Solaris stream code came from Mentat as well, so there is obviously expertise out there to do this sort of stuff. So this doesn't sound necessary: > or you have to rewrite all the networking code on the BSD side to use streams. Chris From John_Welch at aer.com Mon Jan 5 01:56:11 1998 From: John_Welch at aer.com (John_Welch) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death In-Reply-To: <199801050848.AAA26495@proxy3.ba.best.com> Message-ID: I had another encounter with the Black Screen, (BLSOD). I had been attempting to get DR1 to run off a Jaz connected to a 3400, and after it refused to, (duh), I was getting these eveil black screens, even with the jaz disconnected. Luckily, I happened to look really close at the lcd, and could BARELY make out an open firmware, "cannot find bootr" message, (I think, this was amazingly faint). Rebooting with capslock held down gave me another OF message, that said to boot MacOS, type 'bye' and press . This worked and I got my smiley face. (thank god, 'cause I'm running it with 8.1b7 & hfs+ == No Norton or boot floppy) After a couple of tries, I got the command-option-P-R to work, (You have to be WAY quick with this keystroke!), and I'm back to normal. Hope this helps John On Mon, 5 Jan 1998 00:51:04 -0800 (GMT-0800) Scott Stevenson wrote: > > I've had to reboot (that is turn the machine off and then on > >repeatedly) fairly often after shutting down directly from Rhapsody. > > Here's what works for me fairly reliably: > > When you want to switch from MacOS to Rhapsody, click the caps lock key > down, and choose "restart" from the special menu. As soon as Rhapsody > begins to load, un-click the caps lock key. > > When you want to switch back to MacOS -- make sure the caps lock key is > RELEASED. Choose "shutdown" from the File menu (or login screen). As > the machine is shutting down -- that is, when you see the "Apple Rhapsody > Mach Operating System" screen with all the text shooting across -- click > down the caps lock key. Once the machine is powered off, turn it back on > (with the key still locked down), and wait until you see the MacOS logo. > At that point -- press the caps lock key again to un-lock it. > > BTW: About the PRAM chime - you won't hear it if you have something > plugged into the headset jack. (I love that feature!) :) > > - Scott > > ------ > Scott Stevenson > scott@cacti.org > http://www.cacti.org <-- Macintosh & Rhapsody Resources > ----------------------------------------- John Welch Email: John_Welch@aer.com AER Inc. A EEEEEEE RRR A A E R R A A E R R A A E R R A A E R R A A EEEEE RRRRR AAAAAAAAAAA E R R A A . EEEEEEEEEE . R R . INC. From dev.random at dev.random.nu Mon Jan 5 08:05:13 1998 From: dev.random at dev.random.nu (dev.random@dev.random.nu) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: Message-ID: I -tried- to port ssh, but without much luck because of the gnu precision math libs involved (which complained about the mutilated version of gcc included with NeXT, which probably carries over to now).. However, as far as legal things.. The only restriction I know is that USA residents must use the WEAKER version of SSH, which uses RSAREF (due to a copyright law). This is an excerpt from FreeBSD ports (/usr/ports/security/ssh/Makefile): fetch-depends: .if !defined(USA_RESIDENT) || ${USA_RESIDENT} != YES && ${USA_RESIDENT} != NO @echo @echo You must set variable USA_RESIDENT to YES if you are a USA @echo resident or NO otherwise. @echo If you are a USA resident you have to get the RSAREF2 @echo library \(RSA Inc. holds a patent on RSA and public key @echo cypto in general - using RSA implementations other than @echo RSAREF will violate the US patent law\) @echo and extract it to ${WRKSRC}. It would be neat if Rhapsody found some way to incorporate at least KTH-Kerberized Telnet (even if they dont have kerberos installed, to facilitate encrypted sessions to those running a kerberized telnetd) and or SSH as standard. BTW, FreeBSD ports would be great in Rhapsody :) just doubt anyone can get that kind of organization together easily.. "packages" would suffice (though please do not emulate the RPM method, as its proprietary & just plain ugly). While on neat things for Rhapsody, having rc.conf & login.conf equivalents in Rhapsody (as in other BSD's) would be an administrators dream. Coming from someone who administers 11 different OS's, I wish all of them had those two features. (ports & cvsup the other two). _________________________________________________________________ thomas stromberg % sysadmin(royal.institute.of.technology@haninge/stockholm) smtp(dev.random@dev.random.nu)%irc(devrandom)%talkd(random@dev.random.nu) "nothing is impossible, some things just take more lines of perl than others" On Mon, 5 Jan 1998, daniel-wittenberg@uiowa.edu wrote: > I just wanted to see what people thought about having TCP wrappers and SSH > security built into Rhapsody. I know I couldn't live without them as a > Unix Admin, and as a Mac fan I'd love to see those as a standard package. > > I don't know if there is some legal reason for incorporating, but has > anyone ported SSH yet? I know tcp wrappers will compile without too much > trouble, though I haven't done much testing.. > > > Dan From volfreak at macconnect.com Mon Jan 5 09:36:24 1998 From: volfreak at macconnect.com (Jay Hill) Date: Thu Nov 3 15:58:31 2005 Subject: Rhapsody Client/Laptop/NC and the eMate Message-ID: <3.0.32.19980105123623.00909a40@dsunix2.dsrd.ornl.gov> All, As one of the learning Rhapsody developers with an interest in education and network computing environment, I have a question or two about the direction Apple and others (3rd party developers) are headed for the client workstations. Since I am interested in the education side of the development world, I have been following the eMate with great interest (especially since the state of Texas is considering replacing all textbooks with laptops/eMates). I also recently read a *rumor* that Apple may be looking at AllegroLite (?) for running on the eMate and Newton which would allow it to run most MacOS apps (ala WinCE). Would this then run both OSs (Newton and Mac) or would AllegroLite have the handwriting recognition added? I have also read that there may be voice recognition software soon to be released (already?) for the NewtonOS (Dragon Systems or something similar). With that said, I see Apple strongly pushing the eMate for education and will probably be announcing an eMate for business (bMate?), which I think is great. However, since it is based on the NewtonOS, how will Rhapsody work with it? What are the plans/thoughts on handwriting and voice recognition software for Rhapsody? I think the handwriting recognition is one of the best features of the NewtonOS. With voice technology becoming more prevalent, this too will be an important requirement. As for the NC, most discussions are that it will be sans HD. I think that users will need some portable systems that have storage of some manner for current working documents. So, when developing applications for the education market, what should one look for for the workstations? An NC with/without an HD running Rhapsody client, an eMate with connectivity to Rhapsody, or some other combination (eMate running Rhapsody with hw and speech recognition)? Any thoughts or comments are most welcome. Thanks and what a great forum for discussion. Jaybird B-) "The longer the island of knowledge, the longer the shore of imagination." Unknown Alternate: volfreak@macconnect.com From george at apple.com Mon Jan 5 08:56:21 1998 From: george at apple.com (Scott George) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death Message-ID: <199801051653.IAA27902@scv1.apple.com> One of the causes of the black screen is a slow spinning hard drive. I have a 9500 that had some model of IBM drive that gives the black screen on every cold boot. The problem is the drive takes too long to become ready and the system crashes I guess. Anyway, the easiest fix I found is a control-flower-power (control-command-power) reboot. The drive does not spin down during a restart so the wait problem is not present. I have also replaced the drive with a Quantum 1.2 GB drive and have not had the problem since. Scott George Apple Computer, Inc. >I had another encounter with the Black Screen, (BLSOD). I had been >attempting >to get DR1 to run off a Jaz connected to a 3400, and after it refused to, >(duh), I was getting these eveil black screens, even with the jaz >disconnected. >Luckily, I happened to look really close at the lcd, and could BARELY make >out >an open firmware, "cannot find bootr" message, (I think, this was amazingly >faint). Rebooting with capslock held down gave me another OF message, that >said >to boot MacOS, type 'bye' and press . This worked and I got my >smiley >face. (thank god, 'cause I'm running it with 8.1b7 & hfs+ == No Norton or >boot >floppy) After a couple of tries, I got the command-option-P-R to work, (You >have to be WAY quick with this keystroke!), and I'm back to normal. > >Hope this helps > >John > > >On Mon, 5 Jan 1998 00:51:04 -0800 (GMT-0800) Scott Stevenson > >wrote: > >> > I've had to reboot (that is turn the machine off and then on >> >repeatedly) fairly often after shutting down directly from Rhapsody. >> >> Here's what works for me fairly reliably: >> >> When you want to switch from MacOS to Rhapsody, click the caps lock key >> down, and choose "restart" from the special menu. As soon as Rhapsody >> begins to load, un-click the caps lock key. >> >> When you want to switch back to MacOS -- make sure the caps lock key is >> RELEASED. Choose "shutdown" from the File menu (or login screen). As >> the machine is shutting down -- that is, when you see the "Apple Rhapsody >> Mach Operating System" screen with all the text shooting across -- click >> down the caps lock key. Once the machine is powered off, turn it back on >> (with the key still locked down), and wait until you see the MacOS logo. >> At that point -- press the caps lock key again to un-lock it. >> >> BTW: About the PRAM chime - you won't hear it if you have something >> plugged into the headset jack. (I love that feature!) :) >> >> - Scott >> >> ------ >> Scott Stevenson >> scott@cacti.org >> http://www.cacti.org <-- Macintosh & Rhapsody Resources >> > >----------------------------------------- >John Welch >Email: John_Welch@aer.com > >AER Inc. > >A EEEEEEE RRR >A A E R R >A A E R R >A A E R R >A A E R R >A A EEEEE RRRRR >AAAAAAAAAAA E R R >A A . EEEEEEEEEE . R R . INC. > > From anita at tenon.com Mon Jan 5 10:52:36 1998 From: anita at tenon.com (Anita Holmgren) Date: Thu Nov 3 15:58:31 2005 Subject: Apache for Rhapsody Available from Tenon Message-ID: In response to a number of requests, Tenon is making their Apache port for Rhapsody available. Apple has already provided a port of Apache, however because there are performance problems with the current Apple port, you may want to use Tenon's port as an interim measure until Apple refines their port. We've measured this 'vanilla' Apache at about 180 hits/second. This Apache should not be confused with WebTen, Tenon's enhanced Apache. The newest WebTen sustains more than 700 hits/second on MacOS. Tenon will have a version of WebTen for the Rhapsody premier release. This public domain version of Apache is designed for people who want to get started using Apache on Rhapsody immediately. The Rhapsody Apache port can be found at: -Anita Tenon Intersystems 805-963-6983 1123 Chapala Street anita@tenon.com Santa Barbara, CA 93101 http://www.tenon.com From dwittenb at dome.weeg.uiowa.edu Mon Jan 5 10:55:44 1998 From: dwittenb at dome.weeg.uiowa.edu (daniel-wittenberg@uiowa.edu) Date: Thu Nov 3 15:58:31 2005 Subject: 2 button mouse Message-ID: Has anyone tried using rhapsody with a 2 button mouse? such as a kensington? Dan -------------------------------------------------------------------------------- Daniel Wittenberg System Administrator University of Iowa Phone: (319) 335-6279 FAX: (319) 335-5505 daniel-wittenberg@uiowa.edu http://dan.weeg.uiowa.edu -------------------------------------------------------------------------------- From paipai at ita.flashnet.it Mon Jan 5 14:33:05 1998 From: paipai at ita.flashnet.it (Paolo Di Francesco) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <9801050925.AA09069@kannix.cubiculum.com> Message-ID: <199801052129.WAA05043@ita.flashnet.it> > I think there is a big difference between an add-on package, and a differently > branded server OS product. The former may be acceptable, the second isn't. > I think it too... 8-) The add-on package is more acceptable. About the file system... I think the best idea is to add some configuration files, so you can easly choose if you want a server or a workstation. > There may be the Developer package, with the compiler, tools, headers, libs > etc. There could be a Web package with WebObjects, the Apache server Or something developed by Apple. 8-) I'll buy this ipotetic CD only if Apple will certificate the apps, or more simply, if I can install everything in 5 minute, otherwise I'll use Linux. Yes it needs 10 days to install everything correctly, but it's free...8-) > (preferrably with a native Apache module based WOAdaptor added), there could > be a database package, etc. I add: - tools for system monitoring - tools for log/user monitoring - mail > And that's ok. After all, not everybody needs > all of this, nor does everyone want to waste the diskspace for all the stuff. to WASTE ???? NT is diskspace "wasted"... 8-DD If it's usefull it's not wasted! ^__^ > On the other hand, nothing needs really that much space that these things > wouldn't fit on a single CD-ROM or maybe DVD-ROM and couldn't be selectively > installed by reasonably smart installation tools. > smart installation and CD...8-) (manual too, see below...) > But the idea of a different server version, that is priced differently and > leaves the option of a crippled client version, etc. makes my gall come up. > I think the idea of a different version it's a COMMERCIAL idea, not a good idea... Another issue is security. Linux is not so secure if you don't REconfigure the entire system files. So I hope Apple will add some tool to configure the security of the system, and they MUST add a security manual (how to configure everything safely against attacks) in the server CD. I think this could be a "PLUS". So in "my server *ADD-ON* package" you'll find : 1) 1 or more CD(s) 2) paper and manuals about security and how to configure all services in 5/10 minutes. 3) Phone number/email (optional) for tecnical questions. 4) (very, very, optional...) on site assistance. #@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@# Reply to : paipai@ita.flashnet.it OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO Ciao Ciao Paolo Di Francesco _ ->B<- All Recycled Bytes Message ... ~ From paipai at ita.flashnet.it Mon Jan 5 14:33:05 1998 From: paipai at ita.flashnet.it (Paolo Di Francesco) Date: Thu Nov 3 15:58:31 2005 Subject: Objective C Message-ID: <199801052129.WAA05037@ita.flashnet.it> Will Apple modify the current syntax? Will Apple add some new features? #@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@#@# Reply to : paipai@ita.flashnet.it OoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoOoO Ciao Ciao Paolo Di Francesco _ ->B<- All Recycled Bytes Message ... ~ From paulmeyers at worldnet.att.net Mon Jan 5 13:40:12 1998 From: paulmeyers at worldnet.att.net (Paul Meyers) Date: Thu Nov 3 15:58:31 2005 Subject: Blue Box won't boot Message-ID: <19980105214225.AAA6510@[12.70.9.173]> I just received the Blue Box yesterday and installed it on my PowerCenter Pro 180, but whenever I try to run it, I get the grey startup screen and nothing else. No arrow, no happy mac. Could this be an incompatibility problem with the computer? Or do I need some special incantation to make it work? I tried it with the loadable HFS both enabled and disabled, from root and non-root. Thanks for any help. Paul Meyers paulmeyers@worldnet.att.net From rcfa at cubiculum.com Mon Jan 5 13:26:07 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: References: Message-ID: <9801052126.AA10626@kannix.cubiculum.com> > However, as far as legal things.. The only restriction I know is that USA > residents must use the WEAKER version of SSH, which uses RSAREF (due to a > copyright law). This is an excerpt from FreeBSD ports > (/usr/ports/security/ssh/Makefile): I think it's safe to say, that until US laws change, DROP ALL HOPE of Apple EVER shipping a version of Rhapsody that contains any sort of STRONG encryption based security. While a product may be legal outside the use, legal to IMPORT into the US, legal to USE in the US, it is currently still COMPLETELY ILLEGAL to REEXPORT it, even if it came originally from outside. As far as I know even the transport of encrypted INFORMATION across US borders is illegal! All companies that do that kind of stuff, like banks, etc. have special permissions. So just forget about it. The best Apple can possibly do, and even that has some potential legal implications, is to have Rhapsody done in such a way that endusers can plug in encryption based products like ssh, kerberos, etc. and have them "just work". Even that though would not be a supported product (since the modules are out of Apples QA's reach) and depends on the standard distributions being ported to Rhapsody. We can all dream, but for the time being, rally in front of the closes US embassy, or write to your congress man, if you want a change in this regard, and don't expect Apple to stick its head out, they have better things to do right now. Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From paul at eisusa.com Mon Jan 5 17:01:05 1998 From: paul at eisusa.com (Paul Nicholson) Date: Thu Nov 3 15:58:31 2005 Subject: Black Screen of Death In-Reply-To: <199801051653.IAA27902@scv1.apple.com> Message-ID: I've verified that if the caps lock key is down, Option-Command-P-R does not reset the PRAM. This black screen of death stuff should be submitted to the esteemed "Journal of Irreproducible Results". Paul From cake at cwr.uwa.edu.au Mon Jan 5 17:58:30 1998 From: cake at cwr.uwa.edu.au (David Cake) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: <9801052126.AA10626@kannix.cubiculum.com> Message-ID: >> However, as far as legal things.. The only restriction I know is that USA >> residents must use the WEAKER version of SSH, which uses RSAREF (due to a >> copyright law). This is an excerpt from FreeBSD ports >> (/usr/ports/security/ssh/Makefile): > >I think it's safe to say, that until US laws change, DROP ALL HOPE of >Apple EVER shipping a version of Rhapsody that contains any sort of >STRONG encryption based security. But weak encryption (provided it is clearly marked as being weak) is still much better than no encryption. I know that there is a school of thought that says 'weak encryption is worse than useless, because it gives you peace of mind that is unjustified', but to me this makes about as much sense as saying if you don't have a quality deadlock, its not worth locking your door. There are several reasons why even weak encryption is really good. 1) even weak encryption protects against trivial packet sniffing attacks. There are situations where people need to log into our network from LANS that I can almost guarantee have privately owned machines equipped with packet sniffers, outside all staff control (ie residential college LANs). I would feel MUCH happier if people logging on from there were forced to use weak encryption rather than unencrypted products (ie ssh rather than telnet). 2) even if Apple can only ship weak encryption, they can certainly ship products that have plug-in encryption that can easily be upgraded (like ssh). 3) weak encryption starts people off with good habits. If Apple shipped ssh, and encouraged sites to use it rather than telnet (say, by not turning telnet on by default), then everyone already has good habits. So when the strong encryption version becomes available, you don't need to convince everyone to use it, you just need to install it. >We can all dream, but for the time being, rally in front of the closes >US embassy, or write to your congress man, if you want a change in this >regard, and don't expect Apple to stick its head out, they have better >things to do right now. How practical is it for Apple to do the Sun thing, and simply make strong encryption available from a foreign subsidiary? In any case, products with weak encryption that is plug-in replaceable would be a great start. Cheers David David Cake Certified Apple Engineer Centre for Water Research University of Western Australia From jpb at technologist.com Mon Jan 5 20:57:43 1998 From: jpb at technologist.com (jpb) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: <9801052126.AA10626@kannix.cubiculum.com> Message-ID: At 6:49 PM -0500 1/5/98, Ronald C.F. Antony wrote: >I think it's safe to say, that until US laws change, DROP ALL HOPE of >Apple EVER shipping a version of Rhapsody that contains any sort of >STRONG encryption based security. I think pam may have been mentioned here already. If it hasn't, pam (available on linux and maybe solaris) is a system of plug in modules for user authentication - pam aware versions of login don't care whether the system is using radius or md5 passwords or even crappy old standard unix crypt. As far as ssh goes, there's no good reason >While a product may be legal outside the use, legal to IMPORT into the >US, legal to USE in the US, it is currently still COMPLETELY ILLEGAL >to REEXPORT it, even if it came originally from outside. Sadly, true. >As far as I know even the transport of encrypted INFORMATION across US >borders is illegal! All companies that do that kind of stuff, like >banks, etc. have special permissions. Not true. Everyone who connects to an http server (like stronghold) overseas would be breaking the law. Joe Block "Our government has kept us in a perpetual state of fear -- kept us in a continuous stampede of patriotic fervor -- with the cry of grave national emergency... Always there has been some terrible evil to gobble us up if we did not blindly rally behind it by furnishing the exorbitant sums demanded. Yet, in retrospect, these disasters seem never to have happened, seem never to have been quite real." -- General Douglas MacArthur, 1957 From jpb at technologist.com Mon Jan 5 20:47:39 1998 From: jpb at technologist.com (jpb) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: Message-ID: At 11:09 AM -0500 1/5/98, dev.random@dev.random.nu wrote: >BTW, FreeBSD ports would be great in Rhapsody :) just doubt anyone can get >that kind of organization together easily.. "packages" would suffice >(though please do not emulate the RPM method, as its proprietary & just >plain ugly). proprietary? rpm's source code is freely available, and I believe it's been ported to solaris and irix already. I haven't seen the FreeBSD packages, but rpm is certainly nicer to use than slackware's pkgs. Joe Block "That's our advantage at Microsoft; we set the standards and we can change them." -- Karen Hargrove, Microsoft (quoted in the Feb 1993 Unix Review editorial) From rcfa at cubiculum.com Tue Jan 6 00:07:12 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: References: Message-ID: <9801060807.AA12052@kannix.cubiculum.com> > >As far as I know even the transport of encrypted INFORMATION across US > >borders is illegal! All companies that do that kind of stuff, like > >banks, etc. have special permissions. > > Not true. Everyone who connects to an http server (like stronghold) > overseas would be breaking the law. Legal and enforced are two things. "Sodomy" (e.g. oral sex) is illegal in many states in the US as well, but it may not be actively enforced. It could be that these laws recently changed, but a few years ago this was certainly still the case, and I haven't read any news about law changes in this respect. I actually even remember a journal article that showed how ridiculous the law was: since IP packets can be arbitrarily routed, you may be sending encrypted information from a US site to another US site, yet break the law, since the packets pass a foreign country on their way. We could even get into more legal gray areas, if we ask ourselves how IP packets routed over satellites are looked at. Is a satellite within or outside the US territory? Does it depend on the position on the sky or on who owns it, etc... Just goes to show that laws don't necessarily make sense, and that there may be a place for civil disobedience.... Anyway, I think we start to drift well outside the scope of this group, so follow ups, please in private mail... Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From scott at cacti.org Tue Jan 6 04:01:04 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801061152.DAA11409@proxy3.ba.best.com> >I don't see what the point is of doing bundled packages. If someone wants >e.g. a Netscape web server, they can go out and buy it Three reasons to bundle 3rd party: 1. Cheaper overall price for the consumer (that's the general idea, anyway) 2. Ideal for retail outlets who would carry the Rhapsody cd, but little else 3. Prove that there are useful apps for the platform I don't think bundles have typically been associated with "ripping off" the consumer. Although I'm sure MS could think of a way. - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From scott at cacti.org Tue Jan 6 04:03:09 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801061154.DAA13248@proxy4.ba.best.com> >They are moving Rhapsody to HFS+ (...I hope!). Then we will get rid at >least of some UN*Xisms like fsck... I think it makes sense to maintain fs compatibility with other unix flavors. Also, HFS (dunno about HFS+) does have a fsck equivalent. When your Mac crashes -- notice how it takes a little longer to restart? Guess what it's doing. :) - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From scott at cacti.org Tue Jan 6 04:03:29 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:31 2005 Subject: Blue Box won't boot Message-ID: <199801061154.DAA13254@proxy4.ba.best.com> >problem with the computer? Or do I need some special incantation to make >it work? I tried it with the loadable HFS both enabled and disabled, from >root and non-root. Thanks for any help. What's the file path to Blue Box on your machine? - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From dwittenb at dome.weeg.uiowa.edu Tue Jan 6 06:24:32 1998 From: dwittenb at dome.weeg.uiowa.edu (daniel-wittenberg@uiowa.edu) Date: Thu Nov 3 15:58:31 2005 Subject: Boot files Message-ID: In what order does DR1 boot? While working last night I rebooted to check something on the Mac side, and when I rebooted, the /etc directory was gone. I don't know what happened to it, or the files, but they were gone. It would boot, just mounting / ro, and therefore I couldn't fix it. I got booted into single-user mode from the install CD, and then moved /usr/template/client/etc into the /etc directory, but now it hangs after mounting root. There are drive accesses like it's booting, then it just stops. Is there anyway I can recover this thing?? There are some files I don't have backups of... oh, and single-user mode doesn't work, or zapping the pram, I figure I'll have to boot from CD and edit some files manually, it's just a matter of which ones need to be changed, and what do I need to make sure is/isn't there. Thanks to anyone who can get this fixed... Dan -------------------------------------------------------------------------------- Daniel Wittenberg System Administrator University of Iowa Phone: (319) 335-6279 FAX: (319) 335-5505 daniel-wittenberg@uiowa.edu http://dan.weeg.uiowa.edu -------------------------------------------------------------------------------- From comrade at umich.edu Tue Jan 6 09:23:20 1998 From: comrade at umich.edu (Robert A. Decker) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security Message-ID: On Tue, Jan 6, 1998 3:22 AM, Ronald C.F. Antony wrote: > It could be that these laws recently changed, but a few years ago this > was certainly still the case, and I haven't read any news about law changes > in this respect. I know this is _way_ off-topic, but I just wanted to point out that laws are rarely removed from the books because they would have to be removed through the legistlative process. Our representatives have better things to do like cracking down on skateboarders, protecting us from immigrants, and putting chicken-inspector murderers to death. Some laws that weren't enforced where I grew up in Minnesota were: -You can't own a red vehicle (this color was reserved for the fire department). -No spitting on sidewalks. -No leering at women. -and of course, everyones favorite, no sodomy! rob -- Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #19: "I'm like a rough diamond that just needs a little polishing, that's all. And that's exactly what Handsome Boy Modeling School is going to do for me." -Chris Elliott in "The Prettiest Week of My Life" From comrade at umich.edu Tue Jan 6 09:26:19 1998 From: comrade at umich.edu (Robert A. Decker) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security Message-ID: On Tue, Jan 6, 1998 3:22 AM, Ronald C.F. Antony wrote: > Just goes to show that laws don't necessarily make sense, and that there may > be a place for civil disobedience.... Another off-topic post (my last!). One of my coworkers makes it a point to wear a t-shirt that shows some perl encryption code everytime he travels into Canada. He hasn't had any trouble yet (of course, the border guards wouldn't know what they're looking at). rob -- Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #16: "It's not your fault. I just thought it was one of those days where I was semi-solid and cars could just go right through me." -Chris Elliott From mmorris at sterinfo.com Tue Jan 6 15:48:38 1998 From: mmorris at sterinfo.com (Mark Morris) Date: Thu Nov 3 15:58:31 2005 Subject: Express Modem available? Message-ID: <34B2C2D6.B41D880A@sterinfo.com> Hello! I'm new to this list, so please excuse me if this is not an appropriate question. I've looked but haven't found any info on whether my Geoport TA is going to work for me under Rhapsody. Has/will Apple Express Modem be ported, or do I need to get a "real" modem? :-) Regards, Mark From petro at playboy.com Tue Jan 6 16:09:22 1998 From: petro at playboy.com (Christopher C. Petro) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: Message-ID: >> be a place for civil disobedience.... > > Another off-topic post (my last!). One of my coworkers makes it a point >to wear a t-shirt that shows some perl encryption code everytime he travels >into Canada. He hasn't had any trouble yet (of course, the border guards >wouldn't know what they're looking at). Despite that, Canada is the One Country it is legal to export Strong Crypto to. Is there anyone working on a port of, or a graphic interface to ssh? How 'bout PGP version 5+? From chad at pengar.com Tue Jan 6 16:21:03 1998 From: chad at pengar.com (Chad Leigh, Pengar Enterprises Inc and Shire.Net) Date: Thu Nov 3 15:58:31 2005 Subject: any vendors release PPC Rhapsody Video Drivers? Message-ID: Have any vendors released video drivers for PPC Rhapsody? I'd like to support the biggest baddest monitor I can find and my 2mb ATI from a 9500 won't do it... I'd like a new video card but Rhapsody support is a must. I'd like to support vendors that support Rhapsody. Chad --------------------------------------------------------------- Chad Leigh Pengar Enterprises, Inc and Shire.Net chad@pengar.com info@pengar.com info@shire.net Full service WWW services from just space to complete sites. Low cost virtual servers. DB integration. Tango. Email forwarding -- Permanent Email Addresses. POP3 and IMAP Email Accounts. mailto:info@shire.net for any of these. --------------------------------------------------------------- From rcfa at cubiculum.com Tue Jan 6 15:26:14 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:31 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801061152.DAA11409@proxy3.ba.best.com> References: <199801061152.DAA11409@proxy3.ba.best.com> Message-ID: <9801062326.AA14005@kannix.cubiculum.com> you wrote: > Three reasons to bundle 3rd party: > 1. Cheaper overall price for the consumer (that's the general idea, > anyway) > 2. Ideal for retail outlets who would carry the Rhapsody cd, but > little else > 3. Prove that there are useful apps for the platform > I don't think bundles have typically been associated with "ripping off" > the consumer. Although I'm sure MS could think of a way. The main reasons for bundles is to guarantee market dominance of one product over another. In concrete terms this means a lower initial cost of entry, (Hey it's a "sale", "special offer", etc.), but higher prices later on, once the position is established. Bundles are highly anticompetitive. And yes, MS knows that. That's one of the reasons why they invested 150M$ in Apple, such that IE is bundled with MacOS, rather than Netscape. You can also bet whose web server we'd be having in a bundle, if there ever will be one. And you can also bet, that the bundle will not last forever, and then prices will skyrocket, because the competition is out of business. Look at NT: it was sold as cheap UNIX alternative. People bought it, even though it sucked, because it seemed to save over Unix. Now that NT starts being entrenched, MS keeps rising the price, and since NT is proprietary, unlike UNIX, people are stuck with it and can't go to another vendor. So yes, bundles aren't associated with ripping off consumers, but people don't think rational and in economic terms, otherwise they would be. Bundles are a way of buying market share, it is like taking a credit card loan: you "save" now, you pay later. Greetings, Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From 9512087 at luc.ac.be Tue Jan 6 16:53:47 1998 From: 9512087 at luc.ac.be (Jasper) Date: Thu Nov 3 15:58:31 2005 Subject: rhapsody security In-Reply-To: Message-ID: Hello all, some words about the US export regulations: It is indeed illegal to export strong encription, Kerberos and PGP have also problems with that. But there's a known work-around that's applicated for PGP 5.0: It is illegal to export programs to outside the US. But it is NOT illegal to export a book with in it the source code of the program. So the full PGP 5.0 source code was printed in a font easy to do 'optical character recognition' on. It was exported outside the US, scanned and compiled. So there we have the international version of PGP 5.0 Check out http://www.pgpi.com There's also info on http://www.pgp.com (dough!) The MacOS version isn't finished yet, but the Windoze version is (what a shame). If Apple wants to, it IS possible to include strong encription into the OS, but it would require a 'workaround'. If Apple would do this, it would defenetly convince me of Apple's opposition to 'big brother'. ssh, pgp and Kerberos rule, if 'the network is the computer', then please let it be a SAFE network, Greetz, Jasper :) > >> be a place for civil disobedience.... > > > > Another off-topic post (my last!). One of my coworkers makes it a point > >to wear a t-shirt that shows some perl encryption code everytime he travels > >into Canada. He hasn't had any trouble yet (of course, the border guards > >wouldn't know what they're looking at). > > Despite that, Canada is the One Country it is legal to export > Strong Crypto to. > > Is there anyone working on a port of, or a graphic interface to > ssh? How 'bout PGP version 5+? The only graphical interface for SSH that I heared of until now, is "F-secure", the shareware client for MacOS and Windows. I'm buzy trying to compile ssh 4 Rhapsody :) Without nifty GUI tough. Encription for the masses! Jasper ;) From matthewl at cc1001520-a.sdotc.com Tue Jan 6 17:21:05 1998 From: matthewl at cc1001520-a.sdotc.com (Matthew Lehrian) Date: Thu Nov 3 15:58:31 2005 Subject: web and ftp through proxy Message-ID: <19980107052842093.AAA315@mozart> Hello, I'm writing because I've become entirly too frustrated with broken ftp on Rhapsody. Actually, ftp works, but I'm trying to get out through a proxy, which doesn't work. Ok, here's the setup. I've got an NT box with 2 NIC's, one connected to a cable modem, with a static internet-happy IP address, the other connected to my internal network, with a private IP address, no IP forwarding. I'm running WinGate 2.0 on NT for my proxy service (much nicer than MS Proxy Server). I'm running IIS with web and ftp services on the typical ports (80 and 21 respectively). Anyway, the proxy is set up to listen for internal ftp requests on port 8021 (and internal http requests on port 8080) and forward them out to the world. This works like a champ from my 2 Macs. The web proxy works from OmniWeb in Rhapsody, after setting up the proxy settings for http requests to be sent on port 8080. The only thing that doesn't work is when I try to hit the webserver on the NT box by name (iwww.sdotc.com, which is set up in my DNS as the internal IP address of the NT box). I can hit it by IP address, it just seems that because it's my internal domain that it doesn't go. Any ideas here? This might be a DNS issues, or a problem with the way I set up my proxy settings in OmniWeb. Second, ftp doesn't work at all. I can ftp to the NT box (iftp.sdotc.com) from the command line and the name resolves properly and everything. I can't ftp anywhere from OmniWeb. I tried to set up the ftp protocol in the proxy configuration panel of OmniWeb, but I wasn't able to get it to work. When I try to connect to ftp.apple.com, it says "". Can someone help me configure Rhapsody so that I can ftp from the command line and from OmniWeb, both internally and through the proxy? Thanks in advance. Oh, one more question. I know that I can do TCP/IP from the Blue Box via MacIP. Do any MacIP servers exists for either UNIX or NT? It would be nice to be able to run a MacIP proxy or something on my NT box, or maybe something that could even run in the Rhapsody portion of the same box that the Blue Box is running on, allowing TCP/IP connections via MacIP from Blue to Yellow, then using Rhapsody's TCP/IP to get to the rest of the world. Anyone? Thanks, again. Matthew Lehrian - matthewl@sdotc.com solutions.com - www.sdotc.com Cross Current Corporation - www.xcurrent.com From matthewl at cc1001520-a.sdotc.com Tue Jan 6 17:27:08 1998 From: matthewl at cc1001520-a.sdotc.com (Matthew Lehrian) Date: Thu Nov 3 15:58:31 2005 Subject: SMP and thread support Message-ID: <19980107053444140.AAA327@mozart> Hello, Anyone out there know how and when Rhapsody will provide SMP. I've got a dual-Pentium machine running NT and Rhapsody, but I'm pretty sure that Rhapsody is only using one engine. Since Mach supports SMP, does that mean that Mach will schedule tasks on both engines, or do I need a special version of the Mach? I would assume (probably incorrectly) that if I created an app that spins off non-GUI threads, that Mach would schedule them on either engine. Also, what type of threads does Mach use, kernel threads or user threads? Finally, MacWorks has a dual-engine upgrade card (2 200MHz 604e's) for $699.00. The price is pretty good. I know that in Mac OS, only certain apps will take advantage of the other engine, but in Rhapsody, what will happen? Will it even work? Will Rhapsody's SMP support the existing multi-engine Macs? Thanks in advance. Matthew Lehrian - matthewl@sdotc.com solutions.com - www.sdotc.com Cross Current Corporation - www.xcurrent.com From epeyton at cmg.FCNBD.COM Tue Jan 6 13:50:57 1998 From: epeyton at cmg.FCNBD.COM (Eric Peyton) Date: Thu Nov 3 15:58:32 2005 Subject: PDFConvert - Submission Message-ID: <199801070352.VAA00983@mercury.cmg.FCNBD.COM> All, I just submitted PDFConvert to peak (next-ftp.peak.org/pub/rhapsody). The readme is included but here is the low down. Eric ---------------------------------------------------------- PDFConvert.app Note ---- PPC users please read the why only intel section below. What is PDFConvert.app ---------------------- PDFConvert.app is a simple application that attempts to function as front end to pdftotext and pdftops from the xpdf package (See xpdf below). I assume that someone is porting a real pdf viewer to Rhapsody, this is just a stop gap measure to allow me to look at the docs. You can have it, and do with it as you please. xpdf ---- See the README.xpdf included with this distribution. xpdf is a pdf viewer for the x-windows system. It is freely available and distributed by Derek B. Noonburg at derekn@aimnet.com. The web page for xpdf is http://www.aimnet.com/~derekn/xpdf/. Derek has graciously allowed me to bundle a binary of pdftops and pdftotext with PDFConvert.app. Please visit Derek's sight and let him know you appreciate the software. pdftops and pdftotext --------------------- All PDFConvert.app does is provide a gui for running these applications (which do exactly as their names suggest, they convert pdf into either postscript or text). PDFConvert.app will also open pdf files with a double click in workspace, and then by default convert them to postscript and open them with your postscript viewer. These applications seem to handle the pdf docs on Rhapsody pretty well. This is a hack, this is only hack, I repeat this is a hack ---------------------------------------------------------- I know anothers are writing pdf viewers, (Hey omni where is OmniPDF for Intel?????). This is just a hack to allow me to see the .pdf files included with Rhapsody. If you wanna wait for the "real" viewers, fine by me, but don't send me flames saying that this app is slow, buggy, blew up my system or whatever. Distributing PDFConvert ----------------------- You have the source, if you wanna make it better, please feel free. All I ask is that you leave the README that I write alone and include both this README and the README.xpdf with each distribution. Please let me know if you are going to re-release PDFConvert and it would be nice of you to let Derek know as well. PDFConvert is Copyright 1998 Eric Peyton. xpdf, pdftops and pdftotext are copyright 1997 Derek B. Noonburg. Please distribute source with every release of PDFConvert.app. Please read the README.xpdf and follow Derek's restrictions on xpdf distribution. Thanks. Why only Intel -------------- I attempted to compile this both -arch ppc and -arch i386. The i386 compile went fine, however the ppc compile resulted in the following output Gfx.cc:222: Internal compiler error. Gfx.cc:222: Please submit a full bug report to `bug-next@next.com' I can find nothing wrong in the code that would cause this problem, however I rarely use c++ so I might not notice alot of things. I have submitted a bug report, with additional information. If a ppc user wants to go get the source from Derek's site and compile both pdftops and pdftotext, please submit them to me and I'll make sure that it gets into the distribution. epeyton@cmg.fcnbd.com Eric Peyton Openstep/Rhapsody Development (ONLY!!) From sanguish at digifix.com Tue Jan 6 20:28:11 1998 From: sanguish at digifix.com (Scott Anguish) Date: Thu Nov 3 15:58:32 2005 Subject: any vendors release PPC Rhapsody Video Drivers? In-Reply-To: References: Message-ID: <199801070428.XAA02073@digifix.com> "Chad Leigh, Pengar Enterprises Inc and Shire.Net" wrote: > Have any vendors released video drivers for PPC Rhapsody? > Nope. And they probably won't until Apple finalizes the DriverKit (or I/O Kit or whatever its called now) APIs and ships them out. > I'd like to support the biggest baddest monitor I can find and my 2mb ATI > from a 9500 won't do it... > > I'd like a new video card but Rhapsody support is a must. I'd like to > support vendors that support Rhapsody. > > Chad > > --------------------------------------------------------------- > Chad Leigh Pengar Enterprises, Inc and Shire.Net > chad@pengar.com info@pengar.com info@shire.net > Full service WWW services from just space to complete sites. > Low cost virtual servers. DB integration. Tango. > Email forwarding -- Permanent Email Addresses. POP3 and IMAP > Email Accounts. mailto:info@shire.net for any of these. > --------------------------------------------------------------- > From matthewl at cc1001520-a.sdotc.com Tue Jan 6 20:37:46 1998 From: matthewl at cc1001520-a.sdotc.com (Matthew Lehrian) Date: Thu Nov 3 15:58:32 2005 Subject: Rhapsody problems Message-ID: <19980107084522656.AAA260@mozart> Hello all, A couple things that I've noticed in Rhapsody... First thing, if I boot up Rhapsody and don't log in right away, the screen saver kicks in. I move the mouse, the machine comes back, and I can type in my username and password. Problem is, it clears the screen, and when it's about to paint the desktop... it doesn't. The screen just goes black and all I see is the spinning beach ball. I can still do things with the keyboard, including logging back out, as long as I can do it without needing to see anything. This appears to be screen saver related. Second, when I'm working in Project Builder, it seems that if I haven't done anything for a few minutes, then I try to do something, it hangs. I have to kill PB. This appears to be related to either the screen saver, or possible the PB's autosave. Also, if I have multiple windows open in FileManager, and I choose Update Viewers from the View menu, it usually mucks up the contents of the windows. It actually displays contents of incorrect directories. The last time I did this, I actually got a System Panic (first one in three months, more stable than my NT box, even in DR1). Has anyone else experienced any of these problems? Matthew From scott at cacti.org Tue Jan 6 23:36:49 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:32 2005 Subject: Rhapsody problems Message-ID: <199801070727.XAA13343@proxy3.ba.best.com> >[...]I can type in my username and password. Problem is, it clears the >screen, and when it's about to paint the desktop... it doesn't. The >screen just goes black and all I see is the spinning beach ball. I > >Second, when I'm working in Project Builder, it seems that if I >haven't done anything for a few minutes, then I try to do something, >it hangs. I have to kill PB.[snip] > >Also, if I have multiple windows open in FileManager, and I choose >Update Viewers from the View menu, it usually mucks up the contents >of the windows.[snip] PPC or Intel? How much ram? What model if it's a PowerMac? - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From scott at cacti.org Wed Jan 7 00:05:05 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:32 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801070756.XAA28154@proxy3.ba.best.com> >1) 1 or more CD(s) >2) paper and manuals about security and how to configure all services in 5/10 >minutes. Or like the Sun Netra servers, where you can configure all the services through a web browser. - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From Philippe.Robert at nice.ch Wed Jan 7 01:30:14 1998 From: Philippe.Robert at nice.ch (Philippe C.D. Robert) Date: Thu Nov 3 15:58:32 2005 Subject: SMP and thread support In-Reply-To: <19980107053444140.AAA327@mozart> References: <19980107053444140.AAA327@mozart> Message-ID: <9801070930.AA00465@nice.ch> You wrote: > Hello, > > Anyone out there know how and when Rhapsody will provide SMP. I've SMP will be in the Unified Release. Phil --- Philippe C.D. Robert, CS-student @ Uni Bern (Switzerland) OPENSTEP/Rhapsody Programmer Email:Philippe.Robert@nice.ch (NeXTMail & MIME accepted) Homepage:http://www.nice.ch/~phip From moetteli at monster.citeu.unige.ch Wed Jan 7 01:53:26 1998 From: moetteli at monster.citeu.unige.ch (Philip Motteli) Date: Thu Nov 3 15:58:32 2005 Subject: SMP and thread support In-Reply-To: <19980107053444140.AAA327@mozart> References: <19980107053444140.AAA327@mozart> Message-ID: <9801070953.AA00523@monster.citeu.unige.ch> > Anyone out there know how and when Rhapsody will provide SMP. I've > got a dual-Pentium machine running NT and Rhapsody, but I'm pretty > sure that Rhapsody is only using one engine. Since Mach supports > SMP, does that mean that Mach will schedule tasks on both engines, or > do I need a special version of the Mach? I would assume (probably > incorrectly) that if I created an app that spins off non-GUI threads, > that Mach would schedule them on either engine. As far as I know, at the moment Rhapsodies Mach is only capable of one Processor (e. g. type hostinfo in a shell). But Apple committed several times to SMP and it is expected to come in the final release of Rhapsody. When you then will start an application on a multiprocessing machine, the app will work automatically (sceduled by the kernel) on so many processors as processors or threads in the app are available. That means, every thread on one processor. Re Phil From rcfa at cubiculum.com Wed Jan 7 00:13:15 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801070756.XAA28154@proxy3.ba.best.com> References: <199801070756.XAA28154@proxy3.ba.best.com> Message-ID: <9801070813.AA15190@kannix.cubiculum.com> > Or like the Sun Netra servers, where you can configure all the services > through a web browser. This should be trivial. All of NetInfo is accessible through an object kit, the NetInfoKit. Since for WebObjects this is just yet another framework to dynamically load, it should be quite easy for anyone to create a WebObjects front-end to all of NetInfo. Needless to say, if we can do that, so can Apple, and I'd expect that sooner or later they will do it. They may even use an excellent web server like Apache, add some real apache modules to it, such that apache and WebObjects can transparently be configured and authenticate users through NetInfo. Just got to love object technology.... Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From matthewl at cc1001520-a.sdotc.com Wed Jan 7 04:45:14 1998 From: matthewl at cc1001520-a.sdotc.com (Matthew Lehrian) Date: Thu Nov 3 15:58:32 2005 Subject: Rhapsody problems Message-ID: <19980107165249531.AAA203@[10.31.71.20]> >>[...]I can type in my username and password. Problem is, it clears the >>screen, and when it's about to paint the desktop... it doesn't. The >>screen just goes black and all I see is the spinning beach ball. I >> >>Second, when I'm working in Project Builder, it seems that if I >>haven't done anything for a few minutes, then I try to do something, >>it hangs. I have to kill PB.[snip] >> >>Also, if I have multiple windows open in FileManager, and I choose >>Update Viewers from the View menu, it usually mucks up the contents >>of the windows.[snip] > > > PPC or Intel? How much ram? What model if it's a PowerMac? > > - Scott It's running on PPC, a PowerMac 7500 w/ 132 MHz 604 card and 96 MB RAM. No cache - when I upgrade to the 604, my cache started acting up :-( Matthew Matthew Lehrian - matthewl@sdotc.com Cross Current Corporation - www.xcurrent.com solutions.com - www.sdotc.com From comrade at umich.edu Wed Jan 7 07:29:06 1998 From: comrade at umich.edu (Robert A. Decker) Date: Thu Nov 3 15:58:32 2005 Subject: NSThread sleepUntilDate with NSCalendarDate? Message-ID: I'm having a weird problem using an NSCalendarDate to tell a thread to sleep until a certain date. If my NSCalendarDate is something like: 1998 1-7-98 10:00:00 -500 then it works fine (the thread sleeps until 10:AM on Jan 7). However, if my NSCalendarDate is something like: 1998 1-7-98 23:00:00 -500 then my thread will sleep until 11:AM. It subtracts twelve hours from the time, but then doesn't seem to be aware that it should go at 11 in the PM and not the AM. I don't think I should be having this problem because NSCalendarDate just inherits from NSDate. Does anyone know what I'm doing wrong? rob -- Programmer Analyst - Health Media Research Lab University of Michigan Comprehensive Cancer Center "Get A Life" quote #16: "It's not your fault. I just thought it was one of those days where I was semi-solid and cars could just go right through me." -Chris Elliott From dgmanning at earthlink.net Wed Jan 7 07:43:12 1998 From: dgmanning at earthlink.net (Dave Manning) Date: Thu Nov 3 15:58:32 2005 Subject: SMP and thread support In-Reply-To: <9801070953.AA00523@monster.citeu.unige.ch> Message-ID: This is probably ancient history, but wasn't NeXT considering a dual Motorola 88000 system as a replacement for the 68040 line in the early 1990s? Dave Dave Manning 5324 S. Broadway Circle #307 Englewood, Colorado 80110 (303) 761-0725 From ga at ed4u.com Wed Jan 7 10:40:41 1998 From: ga at ed4u.com (Gordon Apple) Date: Thu Nov 3 15:58:32 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <19980107184000.AAA9893@1Cust136.tnt1.redondo-beach.ca.da.uu.net> Did this list die on Jan 1? I haven't received any mail from it since Dec. 31 G. Apple ga@ed4u.com From scott at cacti.org Wed Jan 7 13:14:25 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:32 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? Message-ID: <199801072105.NAA11247@proxy4.ba.best.com> >And yes, MS knows that. That's one of the reasons why they invested 150M$ Uhhh... no. That was largley called "stealing QuickTime code". >in Apple, such that IE is bundled with MacOS, rather than Netscape. >You can also bet whose web server we'd be having in a bundle, if there >ever will be one. Dunno what you mean here. First, the idea that there isn't (or never will be) a webserver on Rhapsody is ludicrous. I've already had Apache running on Rhapsody PPC -- by far the most popular webserver. Secondly, I don't think Microsoft intends to port IIS to any flavor of unix. Maybe I'm missing your point? >And you can also bet, that the bundle will not last >forever, and then prices will skyrocket, because the competition is >out of business. Huh? >Now that NT starts >being entrenched, MS keeps rising the price, and since NT is proprietary, >unlike UNIX, people are stuck with it and can't go to another vendor. I think you're talking about two completely different things. I don't see how bundling software with Rhapsody is comparable to Microsoft selling NT cheap then raising the price. >Bundles are a way of buying market share, it is like taking a credit card >loan: you "save" now, you pay later. You are correct. Vendors will bundle software to get market share -- that's their interest in doing it. But I don't think that in itself is evil. Do you refuse to buy a new computer because it comes pre-bundled with software? Buying Rhapsody is very much like getting a new computer. If all else fails, you don't have to install the bundled software. - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From rcfa at cubiculum.com Wed Jan 7 16:32:21 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <199801072105.NAA11247@proxy4.ba.best.com> References: <199801072105.NAA11247@proxy4.ba.best.com> Message-ID: <9801080032.AA17430@kannix.cubiculum.com> you wrote: > >And yes, MS knows that. That's one of the reasons why they invested 150M$ > > Uhhh... no. That was largley called "stealing QuickTime code". > > >in Apple, such that IE is bundled with MacOS, rather than Netscape. > >You can also bet whose web server we'd be having in a bundle, if there > >ever will be one. > > Dunno what you mean here. First, the idea that there isn't (or never > will be) a webserver on Rhapsody is ludicrous. I've already had Apache > running on Rhapsody PPC -- by far the most popular webserver. Secondly, > I don't think Microsoft intends to port IIS to any flavor of unix. The "if there ever will be one" refers to the bundle, not to the web server. The fact that the justice department and the anti trust peope at the FTC are looking into the 150mm investment and the browser bundling clause that went with it, shows that the "QuickTime" stealing is half as important, and is made up to look more than it is. In fact, MS does not get access to ANY code, they got access to the patents, which means they would have to code it all from scratch. In the mean time, a proven version of Apple's QuickTime is selling for Win*. MS was after Netscapce at least as much as after Apple's patent portfolio. > >And you can also bet, that the bundle will not last > >forever, and then prices will skyrocket, because the competition is > >out of business. > > Huh? Yup. Once you penetrate the market with a product with proprietary extensions to standards, as MS does all the time, if they don't set the de-facto standard to begin with by sheer marketing muscle, you can then raise the price, since too many people have come to rely on the proprietary extensions to be able to switch. Just let them ship e.g. an MS web server. You can bet in a few years, if the FTC or justice dept. don't step in, NetScape will have bled to death. They already had to post their first quarterly loss, due to MS' "free" server and bundled IE. You can bet, that once Netscape is diminished to a negligible part of the market, MS will add all sorts of proprietary extensions to their web products, (they are trying to do this already with Java, which is why Sun is sueing them), and once people depend on these features, they will turning their free products into expensive ones, since people can't switch to a competitor easily anymore. Classic bait and switch. Hence I wish Apple will not bundle anything but a) their own software b) free software and let the market play on the rest. > >Now that NT starts > >being entrenched, MS keeps rising the price, and since NT is proprietary, > >unlike UNIX, people are stuck with it and can't go to another vendor. > > I think you're talking about two completely different things. I don't > see how bundling software with Rhapsody is comparable to Microsoft > selling NT cheap then raising the price. My point is, that with the current climate bundling a commercial web server, as was suggested, would most probably mean a MS product, since Apple and MS seem to be in bed together lately, while the Apple Netscape relations seem a bit cooled off. In that context, bundling a commercial web server means strengthening the MS monopoly, and their tactics of bait and switch: bait with low price/free products, then switch the pricing scheme and rip people off, once they became dependent on the proprietary "features". Apple should have no part in this. > >Bundles are a way of buying market share, it is like taking a credit card > >loan: you "save" now, you pay later. > > You are correct. Vendors will bundle software to get market share -- > that's their interest in doing it. But I don't think that in itself is > evil. Do you refuse to buy a new computer because it comes pre-bundled > with software? Buying Rhapsody is very much like getting a new computer. > > If all else fails, you don't have to install the bundled software. That's the old falacy: I don't have to buy MS Windows either, and I have successfully avoided it so far, but life becomes a lot more difficult, and the conscious consumer or corporate customer is in the minority. If something is bundled, the vast majority of people will use it, after all, it's "free"... MS wouldn't do anything more eagerly than to deplete competitor's coffres by forcing them to give away their core products for free, because MS can afford to do so. (Which is called dumping: finance one product under cost price with the profits from another.) I don't care about MS or Gates' wealth, but I do care about fair business practices and good products. The less Apple gets involved with MS, the better. For them and for us. Anyway, it's drifting well off topic... Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From scott at cacti.org Thu Jan 8 00:26:36 1998 From: scott at cacti.org (Scott Stevenson) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Rhapsody intentions) Message-ID: <199801080817.AAA13539@proxy3.ba.best.com> >In fact, MS does not get access to >ANY code, they got access to the patents, which means they would have to >code it all from scratch. In the mean time, a proven version of Apple's >QuickTime is selling for Win*. My friend, I believe you're a bit confused. Microsoft has no legal right to any of the QuickTime codebase or patents. The (alleged) story goes like this... When Apple did QuickTime for windows, it hired an outside agency to do much of the work for it. Coincidentally, Microsoft decided to hire the same agency to do work on Video for Windows (AVI), and well, you can figure out the rest. I don't know what you mean by "MS does not get access to ANY code, they got access to the patents". To the best of my knowledge, MS doesn't get access to any QuickTime anything. The "investment" was largely an out of court settlement for Microsoft taking pieces of QuickTime and calling them their own (how very Microsoft). >years, if the FTC or justice dept. don't step in, NetScape will have bled to >death. They already had to post their first quarterly loss, due to MS' "free" >server and bundled IE. But this doesn't explain how Apple bundling software with Rhapsody will cause some unnamed company to go out of business? >You can bet, that once Netscape is diminished to a negligible part of the >market, MS will add all sorts of proprietary extensions to their web >products, What do you mean "will"? Microsoft already has active server pages, frontpage's version of html, active X, their own implentation of JavaScript (JScript), their own version of Java, many MSIE-only tags... the list goes on and on. I won't argue that this is MS's intention. But the question remains -- if Apple decides to bundle a commercial webserver (say, Netscape Enterprise) with a particular distrubtion of Rhapsody, for example on an Apple Webserver machine -- how would that be a bad thing? I think you're really comparing two very different things: 1. Microsoft giving away Microsoft's webserver with Microsoft's OS 2. Apple giving away Netscape's (or whoever's) webserver with Apple OS How is Apple the bad guy here? >My point is, that with the current climate bundling a commercial web server, >as was suggested, would most probably mean a MS product This is ludicrous. Microsoft has no interest in porting IIS to a unix environment like Rhapsody. Short of Microsoft buying Apache and labeling as their own, this just ain't gonna happen. >since Apple and MS >seem to be in bed together lately, while the Apple Netscape relations seem >a bit cooled off Here's another reason that Apple may have chosen MSIE over Netscape -- Netscape's software crashes like crazy! MSIE for Mac is rock-solid. I think you're misunderstanding the MS/Apple relationship. Not all deals are black and white ... there's this grey area over here where most of the stuff goes on. If you haven't noticed, Apple and Micrsoft are *direct competitors* -- MacOS and Win95, Rhapsody and NT. The idea that they are sharing technologies is quite reasonable, but both have their guns pointed at their nemesis, both being very cautious. The two companies may wish to collaborate in certain areas where it's mutually beneficial, but they are not buddy-buddy. That's just standard marketing schmooze -- something Apple has not been doing nearly enough of. Make no mistake, there aren't Microsoft employees roaming Apple's buildings. >I don't care about MS or Gates' wealth, but I do care about fair business >practices and good products. The less Apple gets involved with MS, the >better. Don't be so sure. The media is obsessed with Microsoft right now. The company can do no wrong in their eyes (this may be due to some "persuasion"). My experience has been that quality and morales are completely for most analysts and journalists that I've come in contact with. If Apple refuses to work with Microsoft (like it has in the past), the media (and thus the obedient consumers) will punish Apple. And then we have no choice at all. >Anyway, it's drifting well off topic... I don't necessarily think so. The premise of "should Apple bundle apps with Rhapsody" is a valid one. As long as we do some downsizing on the content. :) - Scott ------ Scott Stevenson scott@cacti.org http://www.cacti.org <-- Macintosh & Rhapsody Resources From rcfa at cubiculum.com Thu Jan 8 02:33:38 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Rhapsody intentions) In-Reply-To: <199801080817.AAA13539@proxy3.ba.best.com> References: <199801080817.AAA13539@proxy3.ba.best.com> Message-ID: <9801081033.AA18750@kannix.cubiculum.com> you wrote: > >In fact, MS does not get access to > >ANY code, they got access to the patents, which means they would have to > >code it all from scratch. In the mean time, a proven version of Apple's > >QuickTime is selling for Win*. > > My friend, I believe you're a bit confused. Microsoft has no legal right > to any of the QuickTime codebase or patents. MS does not have rights to the code, but the 150M$ investment agreement included a cross-licensing agreement, whereby MS and Apple can freely use each others patented technologies, HOWEVER NOT their concrete implementations. These (the implementations) require probably additional licensing agreements, if they actually want those. However, nowhere did I even try to state that MS had access to Apple's code base. Hence your trying to run in open doors... Maybe my English is just so bad these days, that you keep misunderstanding me :-( Should get more sleep. > But this doesn't explain how Apple bundling software with Rhapsody will > cause some unnamed company to go out of business? It is more likely that whatever competitor's product is not bundled, will take a considerable beating. Now in concrete terms were talking about COMMERCIAL web servers. Given that Apache was explicitly left out of the bundle, since it was supposed to be targeted towards business people who want support, etc. (I take this assumption from the other posts in this thread), that leaves two serious alternatives: MS and Netscape Given the recent friendship between MS and Apple, and the cool atmosphere between Apple and Netscape, a bundling deal would strongly tend towards bundling some sort of MS server. If that were the case, it would seriously hurt Netscape in a potentially big market (so we hope for Rhapsody's sake). Given that Netscape is already in the red, due to MS bundling its own server for "free" with WinNT, bundling MS software with Rhapsody would drive even more nails into Netscape's coffin. Hence it would at least make it more likely for Netscape to go away. > But the question remains -- if Apple decides to bundle a commercial > webserver (say, Netscape Enterprise) with a particular distrubtion of > Rhapsody, for example on an Apple Webserver machine -- how would that be > a bad thing? > > I think you're really comparing two very different things: > > 1. Microsoft giving away Microsoft's webserver with Microsoft's OS > 2. Apple giving away Netscape's (or whoever's) webserver with Apple OS > > How is Apple the bad guy here? I think we have in so far completely different assumptions, as that I think if Apple were to bundle anything then a MS product, not an Netscape product, given the recent business climate between the various parties involved. Needless to say, if you exchange MS for Netscape in your own paragraph above, you will have no trouble answering your own question. > >My point is, that with the current climate bundling a commercial web > >server, as was suggested, would most probably mean a MS product > > This is ludicrous. Microsoft has no interest in porting IIS to a unix > environment like Rhapsody. Short of Microsoft buying Apache and labeling > as their own, this just ain't gonna happen. Wouldn't bet on it. Rhapsody will make a terrific webserver platform, particularly with all the media streaming etc. announced with QT3.0. If MS thinks they can deal Netscape a potentially deadly blow by doing so, they will have no problem porting their software to Rhapsody, particularly if they are interested in Apple's long-term survival. Also, MS has created a special Mac software division within the company. That division will surely have to deal with Rhapsody (i.e. Unix) sooner or later, which means they may as well work on a web server, while they are at it. > I think you're misunderstanding the MS/Apple relationship. Not all > deals are black and white ... there's this grey area over here where most > of the stuff goes on. If you haven't noticed, Apple and Microsoft are > *direct competitors* -- MacOS and Win95, Rhapsody and NT. Yes, but Apple is negligibly small, Netscape is not, in their respective market segments. MS needs Apple to stay alive, in order to prevent anti-trust litigation. Apple at the current level of market penetration, maybe even at a bit higher level, is something MS can live with. On the other hand, Netscape has a huge presence in the web market, directly on par with MS. That is a challenge and a threat. While MS can be chivalrous to Apple, they have no intent to be that with Netscape, at least not until Netscape is not down to a single digit percentage market share. In short, we just seem to have different ideas as to how far MS will go to strike Netscape. I think they will stop at nothing, you think the Unix OS is big enough a hurdle for them to stop chasing their arch rival in the internet market, which in turn is the only real threat to MS' desktop monopoly. I think MS is quite comfy with where Apple is and will be in the near future, since it's a cheap anti-trust insurance, and they still make money off selling productivity apps to Mac users, while they are not comfy at all with where Netscape is, particularly since they don't make a cent off of add-on products for Netscape. You think, MS considers Apple and Netscape as equally relevant competitors. The future will tell. Greetings, Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From cmh at greendragon.com Thu Jan 8 07:23:01 1998 From: cmh at greendragon.com (Chris Hanson) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Rhapsody In-Reply-To: <9801081033.AA18750@kannix.cubiculum.com> Message-ID: At 5:53 AM -0500 1/8/98, Ronald C.F. Antony wrote: >It is more likely that whatever competitor's product is not bundled, will >take a considerable beating. Now in concrete terms were talking about >COMMERCIAL web servers. Given that Apache was explicitly left out of the >bundle, since it was supposed to be targeted towards business people who >want support, etc. (I take this assumption from the other posts in this >thread), that leaves two serious alternatives: > MS and Netscape There's more commercial web servers in the world than just Microsoft and Netscape's. What about WebStar? That's been a rock-solid product on the Mac for a long time, with a good interface and API. I would be willing to bet that StarNine (QuarterDeck) will be porting it to Rhapsody -- if they're not already done -- and that *it* will be the webserver of choice for many of Rhapsody's users. (I.e. users that want a real human interface and none of the edit-a-text-file-in-vi crap that many Unix-derived servers use.) In fact, if Apple were to ship a Rhapsody webserver bundle I'd bet that WebStar is what they include... Apple's Personal Web Sharing is based on another web server from Maxum. Maybe they'll have a Rhapsody-based server too? I seem to recall that their commercial server is pretty darn speedy, and also has a good interface. Maybe Dave Winer'll port Frontier and Apple will bundle *that* -- it has a fully-functional HTTP server. (Though it suffers some of the same configuration problems that Unix servers do.) Don't be so locked in in your thinking to what already exists for Unix. There's a lot of great stuff on the Mac already that can become even greater when it's brought to Rhapsody. From gfin at psych.ualberta.ca Thu Jan 8 07:48:34 1998 From: gfin at psych.ualberta.ca (Gary Finley) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar Message-ID: <98Jan8.084840-0700_mst.51273-2+10@chrome.psych.ualberta.ca> Chris Hanson wrote: > What about WebStar? That's been a rock-solid product on the > Mac for a long time, with a good interface and API... > I would be willing to bet that StarNine (QuarterDeck) will be > porting it to Rhapsody... and that *it* will be the webserver of > choice for many of Rhapsody's users. Not if they have any interest in performance. I was looking at a ZD Labs review of Web servers a few weeks ago. I was interested in comparing Apache and IIS, but WebStar was included in the review. It stood out in that review as the worst server by a wide margin. "Dead last by a mile" is the way I'd sum it up. For those who are interested, the review is here: http://www.zdnet.com/zdimag/zdlabs/servers/index.html ---------------------------------------------- Gary Finley, Psychology Dept. Univ. of Alberta Network manager, Web manager, and postmaster. (403) 492-2834 gfin@psych.ualberta.ca http://web.psych.ualberta.ca/staff_bios/gary.finley.htmld/ From k97jm01 at cc.kzoo.edu Thu Jan 8 08:06:31 1998 From: k97jm01 at cc.kzoo.edu (Joshua Marker) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: <98Jan8.084840-0700_mst.51273-2+10@chrome.psych.ualberta.ca> Message-ID: > Chris Hanson wrote: > > > What about WebStar? That's been a rock-solid product on the > Not if they have any interest in performance. I was looking at a > ZD Labs review of Web servers a few weeks ago. I was interested in > comparing Apache and IIS, but WebStar was included in the review. > It stood out in that review as the worst server by a wide margin. > "Dead last by a mile" is the way I'd sum it up. > > For those who are interested, the review is here: > > http://www.zdnet.com/zdimag/zdlabs/servers/index.html Except they review Web* 1.31, which is ancient. Also, consider improvements in OT's handling of server-style connections. Web* 3.0, which is in beta, is supposed to be faster yet with OT 1.3. From gfin at psych.ualberta.ca Thu Jan 8 08:09:40 1998 From: gfin at psych.ualberta.ca (Gary Finley) Date: Thu Nov 3 15:58:32 2005 Subject: PC Mag Web server review Message-ID: <98Jan8.090941-0700_mst.51273-2+11@chrome.psych.ualberta.ca> Here's a pointer to another set of Web server tests (by PC Mag), showing results for many of the popular servers. It includes Apache, IIS, and WebStar: http://www8.zdnet.com/pcmag/features/webserver/iwsperformance.htm ---------------------------------------------- Gary Finley, Psychology Dept. Univ. of Alberta Network manager, Web manager, and postmaster. (403) 492-2834 gfin@psych.ualberta.ca http://web.psych.ualberta.ca/staff_bios/gary.finley.htmld/ From cmh at greendragon.com Thu Jan 8 10:04:12 1998 From: cmh at greendragon.com (Chris Hanson) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: <98Jan8.084840-0700_mst.51273-2+10@chrome.psych.ualberta.ca> Message-ID: At 10:50 AM -0500 1/8/98, Gary Finley wrote: >Not if they have any interest in performance. I was looking at a >ZD Labs review of Web servers a few weeks ago. I was interested in >comparing Apache and IIS, but WebStar was included in the review. >It stood out in that review as the worst server by a wide margin. >"Dead last by a mile" is the way I'd sum it up. This doesn't mean that WebStar on Rhapsody will also come in dead last in such a survey. Also, you haven't mentioned what the hardware differences between tested servers was. And there's also the fact that the vast majority of the MacOS is emulated 68k code, which means native apps will be slower than they otherwise could be. And someone already mentioned that the tested version of WebStar was ancient and that OT itself has been significantly improved since the test. Finally, there are quite a few people who are more interested in something that works well, works reliably, and has a good human interface. That's what Apple and the Macintosh are all about! Quite a few people are even willing to give up some performance for this, though *a good HI doesn't equal lower performance*. Those who want Apache for whatever reason should be able to install it and maintain it themselves, which is the whole point of basing Rhapsody on Unix. From paul at eisusa.com Thu Jan 8 10:03:46 1998 From: paul at eisusa.com (Paul Nicholson) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Rhapsody intentions) In-Reply-To: <199801080817.AAA13539@proxy3.ba.best.com> Message-ID: Scott Stevenson said, > Here's another reason that Apple may have chosen MSIE over Netscape -- >Netscape's software crashes like crazy! MSIE for Mac is rock-solid. You can say that again! Navigator is very unstable on my PowerBook540 with a MTBF of about 15 minutes. I wish Netscape would stop adding more bloated features, and just fix the foundations of the program so it would not crash. Navigator is also slower drawing to the screen than MSIE. Paul From klingler at Rt66.com Thu Jan 8 10:36:36 1998 From: klingler at Rt66.com (Dave Klingler) Date: Thu Nov 3 15:58:32 2005 Subject: PC Mag Web server review In-Reply-To: <98Jan8.090941-0700_mst.51273-2+11@chrome.psych.ualberta.ca> from "Gary Finley" at Jan 8, 98 08:11:25 am Message-ID: <199801081836.LAA12396@Rt66.com> Gary Finley said: > Here's a pointer to another set of Web server tests (by PC Mag), > showing results for many of the popular servers. It includes > Apache, IIS, and WebStar: > > http://www8.zdnet.com/pcmag/features/webserver/iwsperformance.htm > I found these graphs pretty interesting. I'd be interested in knowing the methodology for the tests and how the servers were set up in each case, because without that information the tests are sort of meaningless. I'm perfectly willing to accept that the MacOS platforms came out dead last, but it's the ones that came out on top that interest me. How many httpds did they run on the single and multi-processor machines, for instance? What sort of criteria did they use to compare machines with different processors? Why did the Sun and SGI systems only have two processors, for instance? Come to that, why was the Mac system an underpowered PowerMac 8500/180 with less memory than the other systems? When you get up to 12500 kB/sec, you're obviously running up against your network hardware and the numbers begin to get less meaningful, but I didn't see any mention of this facet of their testing. I learned a long time ago (1983, when most of the staff left) that the PC Magazine people could be less than conscientious, so I can't help wanting a little more information about their test results. My personal preference is Apache, of course, for its ease-of-use, high performance, and price. Most server admins seem to feel the same way. I wouldn't mind, however, giving IIS its due if the facts seemed to go that way. Dave Klingler Gondolin Software From dwittenb at dome.weeg.uiowa.edu Thu Jan 8 10:35:35 1998 From: dwittenb at dome.weeg.uiowa.edu (daniel-wittenberg@uiowa.edu) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's In-Reply-To: Message-ID: I can say that I leave netscape and my 8600 open for days at a time, and rarely have it crash. I still love netscape over MSIE, and has no problems with the LDAP stuff, navigator, newsgroups, etc... (and yes, I do other things with the machine such as PS, VPC, illustrator, quark, and usually they are all running at the same time) maybe my machine is just blessed or something :) Dan -------------------------------------------------------------------------------- Daniel Wittenberg System Administrator University of Iowa Phone: (319) 335-6279 FAX: (319) 335-5505 daniel-wittenberg@uiowa.edu http://dan.weeg.uiowa.edu -------------------------------------------------------------------------------- On Thu, 8 Jan 1998, Paul Nicholson wrote: > Scott Stevenson said, > > > Here's another reason that Apple may have chosen MSIE over Netscape -- > >Netscape's software crashes like crazy! MSIE for Mac is rock-solid. > > You can say that again! Navigator is very unstable on my PowerBook540 with > a MTBF of about 15 minutes. I wish Netscape would stop adding more bloated > features, and just fix the foundations of the program so it would not > crash. Navigator is also slower drawing to the screen than MSIE. > > Paul > > From klingler at Rt66.com Thu Jan 8 10:42:43 1998 From: klingler at Rt66.com (Dave Klingler) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: from "Chris Hanson" at Jan 8, 98 10:06:45 am Message-ID: <199801081842.LAA13340@Rt66.com> Chris Hanson wrote: > Finally, there are quite a few people who are more interested in something > that works well, works reliably, and has a good human interface. That's > what Apple and the Macintosh are all about! Quite a few people are even > willing to give up some performance for this, though *a good HI doesn't > equal lower performance*. Those who want Apache for whatever reason should > be able to install it and maintain it themselves, which is the whole point > of basing Rhapsody on Unix. Hmmmm. Whenever I see this, I grimace. NeXTStep, OpenStep and Rhapsody were never "based on Unix". A BSD4.3 emulator ships with the system. That's as far as it goes. Dave Klingler Gondolin Software From mbayme at hic.net Thu Jan 8 10:45:06 1998 From: mbayme at hic.net (Michael Bayme) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Message-ID: What can I say except that MSIE crashes on me all the time and Navigator 4.04 has been rock solid. Version 4.0 of outlook crashes randomly. Michael From jhendry at cmg.fcnbd.com Thu Jan 8 11:02:53 1998 From: jhendry at cmg.fcnbd.com (Jonathan Hendry) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar References: <199801081842.LAA13340@Rt66.com> Message-ID: <199801081903.NAA02820@isdnjhendry.cmg.fcnbd.com> Dave Klingler wrote: >Hmmmm. Whenever I see this, I grimace. NeXTStep, OpenStep and Rhapsody >were never "based on Unix". A BSD4.3 emulator ships with the system. That's >as far as it goes. Calling it an 'emulator' that 'ships with the system' is a pretty gross exaggeration. While the OpenStep docs do call it an emulator, the 4.3 functionality (or 4.4 in Rhapsody) is essential, and without it the machine wouldn't even boot. You can boot OpenStep without the GUI, Postscript, etc, but you cannot boot it without BSD 4.x. Your statement above makes the Unix support sound like BlueBox.app, or SoftPC, as if all the Unix is contained within Terminal.app, which it certainly is not. - Jon From gfin at psych.ualberta.ca Thu Jan 8 11:13:12 1998 From: gfin at psych.ualberta.ca (Gary Finley) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar/Rhapsody cont'd References: Message-ID: <98Jan8.121320-0700_mst.51273-2+3@chrome.psych.ualberta.ca> Chris Hanson said: > This doesn't mean that WebStar on Rhapsody will also come in dead > last in such a survey. No, of course not. In fact, I think that they say something about the non-multithreaded kernel being a big factor in the poor showing of WebStar in these tests. > Finally, there are quite a few people who are more interested in > something that works well, works reliably, and has a good human > interface.... Sure, if the speed of a WebStar/Rhapsody (or other server with good admin UI) is adequate for the purpose. In my case it would not likely be a contender, since I run a high traffic site. It'll be interesting to see how Tenon's Apache port runs under Rhapsody. The one on the Apple ftp site was terrible when I tested it. I think Anita said that Tenon will eventually release a version of their WebTen for Rhapsody too, although she told me that most of WebTen's impressive speed comes from their carefully tweaked IP stack, which I guess will not be an option for them in Rhapsody. -------------------------------------- Dave Klingler said: > I'm perfectly willing to accept that the MacOS platforms came out > dead last, but it's the ones that came out on top that interest me. > How many httpds did they run... Why did the Sun and SGI systems > only have two processors... Sure, lets not get into a flame war over benchmark numbers and configuration, that could take forever. I just wanted to point out to Chris that there is a real performance issue to be considered with WebStar. > My personal preference is Apache, of course, for its ease-of-use, > high performance, and price. Most server admins seem to feel the > same way. I wouldn't mind, however, giving IIS its due if the > facts seemed to go that way. OK, at the risk of re-starting the flame war that I just tried to avoid, I'll cite some numbers that I found interesting. I ran SGI's WebStone 2.0 against my Apache 1.2.4 server (under Linux) and IIS under NT 4.0. Briefly, I used 5 Web client machines and a small set of HTML files of 2, 5 and 10k size. Similar server hardware in both cases: P/166 machines with 32 meg RAM. The Linux box has wide SCSI, the NT box only EIDE. PCI netcards in both cases. Apache delivered around 80 hits/sec, IIS around 250. Noting that IIS doesn't do DNS lookups of client names for the access log, I turned that off in Apache and ran it again, getting 100 hits/sec. I was surprised by IIS's speed. It also held up well under load. I didn't see a single error, even when I set WebStone for 12 client processes per machine, so 60 virtual clients were hitting on it at the same time. I'm not too interesed in using IIS, since it is very feature-poor (doesn't even *log errors* for heaven's sake), but you've got to give MS a few marks for speed. The thing is fast. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 2862 bytes Desc: not available Url : /mailman/archive/macosx-talk/attachments/19980108/e0be0b0c/attachment.bin From rdreyer at math.berkeley.edu Wed Jan 7 19:28:29 1998 From: rdreyer at math.berkeley.edu (Renaud Dreyer) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's In-Reply-To: Your message of "Thu, 08 Jan 1998 10:25:47 PST." Message-ID: <199801081928.LAA18400@fhloston> > Scott Stevenson said, > > > Here's another reason that Apple may have chosen MSIE over Netscape -- > >Netscape's software crashes like crazy! MSIE for Mac is rock-solid. > > You can say that again! Navigator is very unstable on my PowerBook540 with > a MTBF of about 15 minutes. I wish Netscape would stop adding more bloated > features, and just fix the foundations of the program so it would not > crash. Navigator is also slower drawing to the screen than MSIE. What's ironic is that Netscape Communicator/Navigator (4.04) for MkLinux is extremely solid and lightning fast. *Much* faster than the MacOS version... I assume Netscape is working on a Rhapsody port? Ciao, Renaud Dreyer From mmorris at sterinfo.com Thu Jan 8 11:27:22 1998 From: mmorris at sterinfo.com (Mark Morris) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar References: <199801081903.NAA02820@isdnjhendry.cmg.fcnbd.com> Message-ID: <34B5289A.F131E0A5@sterinfo.com> IMHO the distinction here is at what level you're looking. Mach is not UNIX (as my Operating Systems professor made quite clear!), and Rhapsody is "based on" Mach. As I understand it (which could easily be very mistaken :-), the UNIX-ness is merely an implementation of the BSD API. (Have I oversimplified it?) Regards, Mark From gomezjr at iastate.edu Thu Jan 8 12:57:09 1998 From: gomezjr at iastate.edu (Matt Borgstrand) Date: Thu Nov 3 15:58:32 2005 Subject: DR1 Intel install problems!!! Message-ID: . . . I'm saying a prayer for plug n' play . . . ITMT, I finally got around to installing DR1 on my P166. WinNT4 on IDE and Rhapsody on SCSI (via Adaptec). Install goes fine but initial config is the problem: During initial config my built-in ATI Mach 64 video is detected and I went ahead and picked the default settings for IRQ, etc. Then I selected 640x480 to make sure I had a 256 color palette (to be on the safe side). Then on reboot, I get maybe 4 colors, tops! Most of the screen is blue or black and I'm stuck. Can't read any text on the screen to change config settings to save my life. Here's what chaps me: The Apple install manual says to type "config=default" at the boot: prompt to load general drivers so I can at least see the config panel to change stuff. But instead I get the error "Config can't find /private/Drivers/i386 . . . default.table". I boot in single user mode and the file is definitely there! BTW, I've taken out the IDE drive and shut down the BIOS entry so everything is strictly SCSI now. What gives?? BTW, I will eventually need to install the Rhapsody boot loader over my NTLDR on the IDE drive. Per Apple manual: enter "/usr/etc/disk -b /dev/rhd0h". It's not even there! Funny, but I found the "disk" utility in the /sbin dir. Anyone know what I'm REALLY supposed to enter? Lastly, where is there some good online Rhapsody tech libraries that should cover these "mistakes"? I can't find jack, just tons of Dev info. I guess this is the price you pay for having Apple "Associate" Developer status . . . Thanks to all and any who can answer these ???'s. -Matt Borgstrand Systems Support ------------------------------------------------------------------ Matt Borgstrand Systems Support International Institute of Theoretical and Applied Physics (IITAP) Iowa State University 123 Office & Lab Bldg. Ames, Iowa 50011 Office: (515) 294-8674 Main: (515) 294-3555 Fax: (515) 294-9933 ------------------------------------------------------------------ From jhendry at cmg.fcnbd.com Thu Jan 8 13:30:19 1998 From: jhendry at cmg.fcnbd.com (Jonathan Hendry) Date: Thu Nov 3 15:58:32 2005 Subject: Anyone using an AMD box? Message-ID: <199801082130.PAA03102@isdnjhendry.cmg.fcnbd.com> Has anyone used Rhapsody on an AMD K6 machine? Any opinions or experiences would be appreciated. Thanks, Jon From carlos at audiotel.com.mx Thu Jan 8 14:10:08 1998 From: carlos at audiotel.com.mx (Carlos Alberto Hernandez Hernandez) Date: Thu Nov 3 15:58:32 2005 Subject: are the same OpenStep and Rhapsody programing? Message-ID: <199801082209.QAA15388@huitzi.audiotel.com.mx> Hello all, are there any difference betwen OpenStep and Rhapsody programing? I wan to learn more about Rhapsody programing but I don't know if I can start with OpenStep. Thanks for any help, ________________________________________________ Carlos A. H. H. (carlos@audiotel.com.mx) Software Engineer Audiotel S.A. ________________________________________________ From robert at rc-produktion.se Thu Jan 8 13:49:54 1998 From: robert at rc-produktion.se (Robert Claeson) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Message-ID: <01bd1c7f$5a954c00$0a01a8c0@pluto.rc-produktion.se> Paul Nicholson writes: >Scott Stevenson said, >> Here's another reason that Apple may have chosen MSIE over Netscape -- >>Netscape's software crashes like crazy! MSIE for Mac is rock-solid. >You can say that again! Navigator is very unstable on my PowerBook540 with >a MTBF of about 15 minutes. I wish Netscape would stop adding more bloated >features, and just fix the foundations of the program so it would not >crash. Navigator is also slower drawing to the screen than MSIE. It's actually the same on every platform I've tried them on. MSIE (especially 4.0) is very solid, but Netscape (especially 4.0x) crashes like crazy. It's better if I try not to click like a madman sometimes, but hey, life's too short to wait for screen updates and web pages being loaded. On the other hand... I reinstalled Microsoft IIS 3 on a client's NT server today. It refused to let any user access any page at all, even though it was configured to permit any host to access it. This wasn't the first time it did like this, but at least I didn't have to reinstall the whole NT server like I had to two weeks ago. I still feel that IIS and ASP has ways to go before they become as stable as Apache and WebObjects. Robert From paul at eisusa.com Thu Jan 8 15:43:15 1998 From: paul at eisusa.com (Paul Nicholson) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's In-Reply-To: References: Message-ID: Hi Dan, >I can say that I leave netscape and my 8600 open for days at a time, and >rarely have it crash. I still love netscape over MSIE, and has no >problems with the LDAP stuff, navigator, newsgroups, etc... >(and yes, I do other things with the machine such as PS, VPC, illustrator, >quark, and usually they are all running at the same time) I like Netscape's user interface better than MSIE, and I have found that tables often will not format properly in MSIE. I'd use it in a blink if it would just run reliably. There's no love for MS here. >maybe my machine is just blessed or something :) Or mine is cursed! It runs better for me on PPC machines than on the 68040 PB540. Paul From rcfa at cubiculum.com Thu Jan 8 16:44:34 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: <199801081842.LAA13340@Rt66.com> References: <199801081842.LAA13340@Rt66.com> Message-ID: <9801090044.AA20582@kannix.cubiculum.com> > Hmmmm. Whenever I see this, I grimace. NeXTStep, OpenStep and Rhapsody > were never "based on Unix". A BSD4.3 emulator ships with the system. That's > as far as it goes. That's not true. There are a few basic things that were ripped out of the Unix kernel, and these are replaced with a MACH kernel. The rest is still BSD4.3 i.e. UNIX (You may of course start a religious war by claiming that "real" Unix is SysV and not BSD based, but let's keep that aside...). Everything but the core kernel is unix: BSD file system, all the tools, system services, networking, shell, compilers, web server, etc. As a matter of fact, even SUN ripped out the core of the SysV kernel and replaced it with a micro kernel, I think it was originally developed in France. So you might as well say that Solaris is not Unix, but has just a SysV emulator. For the same reason the MACH based Digital UNIX would not be Unix, etc. In short, just because NeXT added an OS abstracting object library that implements the OpenStep API, plus a proprietary, DPS based GUI and windowing system, doesn't make the system a non-UNIX system. Oh, and it was even sold as "UNIX for mere mortals" at some point in time.... Ask Steve :-) Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From rcfa at cubiculum.com Thu Jan 8 16:53:04 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: Unix Emulator...( was Re: WebStar) In-Reply-To: <199801081903.NAA02820@isdnjhendry.cmg.fcnbd.com> References: <199801081903.NAA02820@isdnjhendry.cmg.fcnbd.com> Message-ID: <9801090053.AA20613@kannix.cubiculum.com> The only thing that qualifies as a UNIX emulator would be the layers that NeXT added to YellowBox-NT/OpenStep-Enterprise, in order to make their tools work: a unix like shell, Mach messaging, gnu tools, etc. There is also a bit more Unix emulation for NT available, I hear, from Cygnus support. The NeXT MACH-OS and Rhapsody however are pretty much full blown UNIX. Apple may try to swipe that under the carpet, since they tried to give UNIX a bad name for years and used it as the anti-thesis to the MacOS, so they would have to surmount massive prejudice from current Mac users if they were to tell them: "Yes, it is UNIX, you heard correctly: WE, Apple, are trying to sell YOU Macuser, a UNIX BOX!" :-) Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From rcfa at cubiculum.com Thu Jan 8 16:36:08 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: PC Mag Web server review In-Reply-To: <199801081836.LAA12396@Rt66.com> References: <199801081836.LAA12396@Rt66.com> Message-ID: <9801090036.AA20560@kannix.cubiculum.com> > My personal preference is Apache, of course, for its ease-of-use, high > performance, and price. Most server admins seem to feel the same way. I > wouldn't mind, however, giving IIS its due if the facts seemed to go that > way. Check out the http://www.apacheweek.com site, or subscribe to the weekly newsletter. It's free. They always have a section on what the press writes about various web servers, and put in corrections. It's quite obnoxious to see, what the press does, and with how much bias the get away. A few examples: - "market share": often calculated either by units shipped: i.e. MS counts for one for each NT server package, even if the server isn't used. Apache counts zero, because they don't ship units. In some cases it's % of IS budget, again, Apache requires no IS budget, so again it gets 0%, or a bit above, for the few companies that try to put a number on the time required for installation. - feature list comparisons: based on MS features, e.g. Apache "misses" features that don't need to be part of Apache, since Unix provides them natively (e.g. ftp server, SMTP, etc.) with other programs. - performance measurements: incomparable hardware configurations, badly configured systems, old software versions. These tests have to be taken with more than a grain of salt. Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From rcfa at cubiculum.com Thu Jan 8 16:07:36 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: <98Jan8.084840-0700_mst.51273-2+10@chrome.psych.ualberta.ca> References: <98Jan8.084840-0700_mst.51273-2+10@chrome.psych.ualberta.ca> Message-ID: <9801090007.AA20488@kannix.cubiculum.com> While servers are being compared: for accurate information and corrections in respect to a constant stream of misrepresentation about Apache, as well as more realistic "market" share figures, check out:http://www.apacheweek.com Greetings, Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From rcfa at cubiculum.com Thu Jan 8 17:07:28 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: <34B5289A.F131E0A5@sterinfo.com> References: <34B5289A.F131E0A5@sterinfo.com> Message-ID: <9801090107.AA20649@kannix.cubiculum.com> you wrote: > IMHO the distinction here is at what level you're looking. Mach is not > UNIX (as my Operating Systems professor made quite clear!), and Rhapsody > is "based on" Mach. As I understand it (which could easily be very > mistaken <, the UNIX-ness is merely an implementation of the BSD API. > (Have I oversimplified it?) You're both right and you oversimplified it. MACH is only a microkernel. i.e. it does only part of an operating systems job: scheduling, memory management, and few other core services. The rest is missing. Without adding additional layers to it, MACH is close to useless. Also, MACH development was in a big part inspired to serve as a replacement for the very core of the UNIX system, since that core had become too cluttered over time. So on top of that MACH kernel, a BSD API is implemented, a OS personality in IBM speak. This API in turn is used in combination with the MACH API to implement the rest of the system. Even worse, since NeXT didn't choose a multiserver approach, in the end BSD and MACH stuff are lumped into a single kernel, which truely based on both. It's basically a chimera of an OS, but one that replaces UNIX coal with MACH diamonds, where needed.... Ronald ============================================================================== "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rcfa@cubiculum.com | NeXT-mail welcome From cake at cwr.uwa.edu.au Thu Jan 8 18:39:49 1998 From: cake at cwr.uwa.edu.au (David Cake) Date: Thu Nov 3 15:58:32 2005 Subject: Not Unix? was Re: WebStar In-Reply-To: <199801081842.LAA13340@Rt66.com> Message-ID: > Whenever I see this, I grimace. NeXTStep, OpenStep and Rhapsody >were never "based on Unix". A BSD4.3 emulator ships with the system. That's >as far as it goes. > I have to say that "based on Unix" is considerably closer to the truth than saying its an "emulator". An emulator makes it sound like a small part that that can easily be removed which is the complete opposite of the truth. Its pretty much a unix system with a Mach kernel and a big set of fantastic object oriented libraries on top. >IMHO the distinction here is at what level you're looking. Mach is not >UNIX (as my Operating Systems professor made quite clear!), and Rhapsody >is "based on" Mach. As I understand it (which could easily be very >mistaken :-), the UNIX-ness is merely an implementation of the BSD API. >(Have I oversimplified it?) Yes. Mach is not unix, but unix can be Mach - and Rhapsody (or OpenStep for Mach) is both Mach and Unix. Mach is not unix because Mach is only a kernel - and there is a lot more to unix than that. Exactly what you need to be considered unix is not defined clearly - but its clear that Rhapsody or OpenStep for Mach, which have the full set of unix commands, can have/ have a unix file system, and can run pretty much any unix software, are unix enough for most purposes. Of course, it doesn't include some things standard that are common in the unix world - X windows, for example - but these things are not necessary to be Unix. Its just that OpenStep/ Rhapsody has the novelty (for unix) of a truly useful GUI, so that you don't actually need to know cryptic unix commands in order to perform useful work. But its all there. Meanwhile, Ronald, while knowing his Unix, has some bizarre things to say about Apple (what a surprise) >Apple may try to swipe that under the carpet, since they tried to give >UNIX a bad name for years and used it as the anti-thesis to the MacOS, so >they would have to surmount massive prejudice from current Mac users if >they were to tell them: "Yes, it is UNIX, you heard correctly: WE, Apple, >are trying to sell YOU Macuser, a UNIX BOX!" :-) Not only is Apple specifically targeting Unix at any point news to me, it would seem pretty damn strange of them to poison the minds of Mac users against Unix considering that they have unix products and have had for years, both the old AU/X, and the current high end AIX servers, which Apple has been promoting widely as the best MacOS server solution. Plus, of course, the Apple support of MkLinux (not to mention that Apple also makes an X server). They have been trying to sell them as servers, true, not user machines - but that only indicates that they have some idea of what their business is. Of course, I doubt Apple are going to spend a lot of time talking about the Unixness of Rhapsody - because they want to concentrate on the areas where it is better than unix (windowing system, object oriented, multimedia, cool easy to use GUI). David Cake Certified Apple Engineer Centre for Water Research University of Western Australia From robert at rc-produktion.se Thu Jan 8 18:58:23 1998 From: robert at rc-produktion.se (Robert Claeson) Date: Thu Nov 3 15:58:32 2005 Subject: Bundling Software/Microsoft Invest. - (was Re: Apple's Message-ID: <01bd1caa$72c87600$0a01a8c0@pluto.rc-produktion.se> Paul Nicholson writes: >I like Netscape's user interface better than MSIE, and I have found that >tables often will not format properly in MSIE. I'd use it in a blink if it >would just run reliably. There's no love for MS here. Strange, I have the opposite experience... NS often won't format tables properly for me. From cake at cwr.uwa.edu.au Thu Jan 8 19:05:06 1998 From: cake at cwr.uwa.edu.au (David Cake) Date: Thu Nov 3 15:58:32 2005 Subject: Apple's Rhapsody intentions? Server, Client, or both? In-Reply-To: <9801080032.AA17430@kannix.cubiculum.com> Message-ID: replying to Ronald >In fact, MS does not get access to >ANY code, they got access to the patents, which means they would have to >code it all from scratch. The patent licencing was only half of the QuickTime dispute, and various other lawsuits. One of the less well known aspects of the Apple Microsoft deal was that Microsoft also gave Apple a large sum of money (though over several years, so that the amount in any given year is small enough that neither has to disclose it) in settlement of various lawsuits. Its not that the various Apple Microsoft lawsuits where not important, its that Apple won, and they both now consider them settled. I suspect Microsoft are to be allowed to continue to use the QuickTime code that is in Video for Windows. >My point is, that with the current climate bundling a commercial web server, >as was suggested, would most probably mean a MS product, since Apple and MS >seem to be in bed together lately, while the Apple Netscape relations seem >a bit cooled off. Which is very true. However, why bundle a commercial web server, when they can bundle Apache? Microsoft as yet don't have a Rhapsody web server, and I can't see that they would see much strategic advantage in developing one. A Rhapsody web server doesn't encourage anyone to use NT, and faces harsh competition from a free product. Plus the main selling point of IIS based solutions (all that active server page crap) faces extremely harsh competition from WebObjects. >My point is, that with the current climate bundling a commercial web server, >as was suggested, would most probably mean a MS product Nope. IIS is not a unix based product, it would require real effort to port, effort that has very little real payoff (and provides an alternative to NT for people locked into IIS). The most likely commercial web server to appear is Netscape (who already port to all sorts of obscure unixes) - but I doubt that will be bundled, except maybe to academic users, as sales of server products is a major source of revenue for Netscape. Its also quite likely that there will be some products from Star Nine, designed as an upgrade for WebStar users. But again, its their main revenue source, I can't see them bundling it. The bundled web server will almost certainly be Apache, and almost certainly not a commercial version. I guess they could pay Tenon enough to use their version, but I doubt they will. After all, its in Apple interests to provide a web server with Rhapsody, which will make many people happy. But its not in their interests to squash other web server developers, especially in favour of MS. Scott Stevenson > Here's another reason that Apple may have chosen MSIE over Netscape -- >Netscape's software crashes like crazy! MSIE for Mac is rock-solid. Yep. Thats why I use MSIE - Netscape crashed about five times as often. MSIE is very solid compared to Netscape for me (though its still my least stable app - except for MSIE 4, which has some serious problems). Other people, though, have found Netscape to be very stable. Cheers David David Cake Certified Apple Engineer Centre for Water Research University of Western Australia From jls at stl-online.net Thu Jan 8 19:22:38 1998 From: jls at stl-online.net (James Scott) Date: Thu Nov 3 15:58:32 2005 Subject: WebStar In-Reply-To: <9801090107.AA20649@kannix.cubiculum.com> References: <9801090107.AA20649@kannix.cubiculum.com> Message-ID: <9801090322.AA00598@stl-online.net> >> "ron" == Ronald C F Antony writes: ron> you wrote: >> IMHO the distinction here is at what level you're looking. Mach is >> not UNIX (as my Operating Systems professor made quite clear!), and >> Rhapsody is "based on" Mach. As I understand it (which could >> easily be very mistaken <, the UNIX-ness is merely an >> implementation of the BSD API. (Have I oversimplified it?) ron> You're both right and you oversimplified it. MACH is only a ron> microkernel. i.e. it does only part of an operating systems job: ron> scheduling, memory management, and few other core services. The ron> rest is missing. Without adding additional layers to it, MACH is ron> close to useless. Also, MACH development was in a big part ron> inspired to serve as a replacement for the very core of the UNIX ron> system, since that core had become too cluttered over time. Ronald puts it very well. Mach takes care of the most fundamental stuff - IPC, scheduling and memory management, mostly. It doesn't do everything we expect an OS to do - it relies on emulators to do that. It can run multiple emulators simultaneously. Shameless plug: read my "Meet Mach" article at http://www.stepwise.com/ for more info and links to still more info. JLS #James Scott "Idiots! I said buy $150 million of SNAPPLE!" - Attr. to Bill Gates From rcfa at cubiculum.com Thu Jan 8 19:23:43 1998 From: rcfa at cubiculum.com (Ronald C.F. Antony) Date: Thu Nov 3 15:58:32 2005 Subject: Not Unix? was Re: WebStar In-Reply-To: References: Message-ID: <9801090323.AA20965@kannix.cubiculum.com> > Meanwhile, Ronald, while knowing his Unix, has some bizarre things > to say about Apple (what a surprise) > >Apple may try to swipe that under the carpet, since they tried to give > >UNIX a bad name for years and used it as the anti-thesis to the MacOS, so > >they would have to surmount massive prejudice from current Mac users if > >they were to tell them: "Yes, it is UNIX, you heard correctly: WE, Apple, > >are trying to sell YOU Macuser, a UNIX BOX!" < I just happen to recall several full page ads run by Apple in past years. One e.g. was a two page spread: on one side you could read pseudo unix gibberish like: % ls -jdfs | grep #$852 > awk