knudsen@ihwpt.UUCP (mike knudsen) (04/07/86)
Phillips, Sony, and Microware (a not-quite-as-big SW house in Iowa that created/es OS9) have agreed on a standard for CD-ROM peripherals. The internal controller will be a 68000 running OS9/68K, presumably out of ROM (or booted from each CD disk maybe?). This was revealed in the latest MOTD, the newsletter of the National OS9 Users' Group. I don't know what percentage of CD-ROM machines this scheme will cover -- I think there are already some boxes out there. mike k "Facts? You can't post this in net.rumor!!"
jimomura@lsuc.UUCP (Jim Omura) (04/15/86)
Mike K. mentioned a CD-ROM standard based on OS-9 68K. In fact, it's
called CDI. Jerry Pournelle was very excited about it (see recent Infoworld
articles by him). We've had a special conference on CD-ROM this month on
BIX, but unfortunately, little has been said about CDI so far despite the
fact that two of us have asked about it. It may be because none of the
people closely involved with CDI showed up. I'm hoping that some of them
will before the end of the month. This summer is going to be very important
to the 68000 software industry.
Cheers! -- Jim O.
--
James Omura, Barrister & Solicitor, Toronto
ihnp4!utzoo!lsuc!jimomura
Byte Information eXchange: jimomura
(416) 652-3880jimm@amiga.UUCP (Jim Mackraz) (04/17/86)
In article <1185@lsuc.UUCP> jimomura@lsuc.UUCP (Jim Omura) writes: > > Mike K. mentioned a CD-ROM standard based on OS-9 68K. In fact, it's >called CDI. >Jerry Pournelle was very excited about it (see recent Infoworld >articles by him). >This summer is going to be very important to the 68000 software industry. > > Cheers! -- Jim O. >James Omura, Barrister & Solicitor, Toronto I can sense Jim's glee now that OS-9 is the darling of JerryP and the InfoWorld circles of journalism, by virtue of its incorporation in the CDI standard, but I have reservations and I would like to hear some net input. As I understand it, OS-9 will be used as an internal control operating system within a closed, appliance type of product (i.e., like a stereo component) which will use custom--and we can assume proprietary-- chips for graphics and audio. The question is: What advantage is held by a computer running OS-9 if the CDI standard becomes a success? [By the way, I am not nearly as skeptical or sarcastic as the following may sound, it just came out that way today, and I'll let it ride, OK?] I can only conjecture: 1- that your computer will be able to execute software on the CDI disks. I doubt this very strongly. 2- that your computer will be able to be used as a development station for CDI software. This will require many other things. 3- that your computer manufacturer is in bed with the right people at an early stage, if he/she hurries, as far as arranging 2, above. 4- that you will catch nice publicity from Jerry Pournelle, since you are on the wave of the future of 68000 software technology. 5- that you can learn to be a great consultant for people who build CDI systems (none in the US, i think) authoring systems, and so on, to say nothing of those people swayed by item 4. 6- that OS-9 is really great, independent of CDI, and that you come out ahead even without any of the above panning out. This item would have been omitted, given the specific "question", but I know what OS fanatics are capable of doing to one's mail burden. cheers, and good luck to OS-9, 68000's, CDI, and you all. jimm
crowl@rochester.ARPA (Lawrence Crowl) (04/18/86)
DEC is also pushing a standard for CD-ROMS. It is a file system based
on their Files-11 file system. See the April 1986 issue of Mini-Micro
Systems. My one comment is that DEC's [dir,dir,dir]file.ext naming
scheme is not as good as the Unix /dir/dir/dir/file.ext naming scheme.
It tends to cause problems for name aliasing schemes. You also have
to hunt down the [] keys.
How many companies are pushing their own file system as standard? How
about putting them all together in one room until they come up with a
standard? Well, the standards that come out of committees like this
are always baroque and unwieldy. How about having someone who is not
finacially interested but is technically competent design a system
incorporating the best features of the various proposed standards?
Or am I dreaming?
--
Lawrence Crowl 716-275-5766 University of Rochester
Computer Science Department
...!{allegra,decvax,seismo}!rochester!crowl Rochester, New York, 14627ralphw@ius2.cs.cmu.edu.UUCP (04/18/86)
In article <17347@rochester.ARPA> crowl@rochester.UUCP (Lawrence Crowl) writes: >DEC is also pushing a standard for CD-ROMS. It is a file system based >on their Files-11 file system. See the April 1986 issue of Mini-Micro >Systems. My one comment is that DEC's [dir,dir,dir]file.ext naming >scheme is not as good as the Unix /dir/dir/dir/file.ext naming scheme. >It tends to cause problems for name aliasing schemes. You also have >to hunt down the [] keys. The file naming scheme should be completely independent of the actual representation of files and internal filenames (which can use non-ASCII characters for internal delimiters) in the filesystem. A reasonable filesystem should allow for many naming systems (especially delimiters, including '[,]' (VMS), '/' (Unix, Apple ProDOS), '\' (MSDOS), or '<.>' (Tops-20), although typically only one user interface is provided. Hierarchical, tree-structured filesystems are 'in' now, but what about the more general digraph format, where a file or directory can be in more than one directory? (Unix sort of does this, but there is only one parent directory at each point in the tree, referenced by ..). If done properly this could solve the symbolic link semantics problem in Unix (cd through a link, then cd .. puts you somewhere you didn't expect to be). The user interface issues are more complicated, but that shouldn't prevent designers from building the digraph capability (my personal favorite) into the filesystem. -- - Ralph W. Hyre, Jr. Internet: ralphw@c.cs.cmu.edu (cmu-cs-c.arpa) Usenet: ralphw@mit-eddie.uucp Fido: Ralph Hyre at Net 129, Node 0 (Pitt-Bull) Phone: (412)CMU-BUGS
knudsen@ihwpt.UUCP (mike knudsen) (04/19/86)
> In article <1185@lsuc.UUCP> jimomura@lsuc.UUCP (Jim Omura) writes: > > > > Mike K. mentioned a CD-ROM standard based on OS-9 68K. In fact, it's > >called CDI. > > Cheers! -- Jim O. > >James Omura, Barrister & Solicitor, Toronto > > I can sense Jim's glee now that OS-9 is the darling of JerryP and the > InfoWorld circles of journalism, by virtue of its incorporation in > the CDI standard, but I have reservations and I would like to hear some > net input. > > As I understand it, OS-9 will be used as an internal control operating > system within a closed, appliance type of product (i.e., like > a stereo component) which will use custom--and we can assume proprietary-- > chips for graphics and audio. > > The question is: What advantage is held by a computer running OS-9 if > the CDI standard becomes a success? ... > 6- that OS-9 is really great, independent of CDI, and that you > come out ahead even without any of the above panning out. > jimm When I posted, I thought only of (6) above -- that this CD standard was a nice feather in the cap of OS9 and we could all brag about "knowing OS9 when." True, OS9 will be sealed up inside a box, and hardly anyone will know it's there. But I will, and so will you... mike k
dibble@rochester.ARPA (Peter C. Dibble) (04/21/86)
In article <821@ihwpt.UUCP>, knudsen@ihwpt.UUCP (mike knudsen) writes: > > In article <1185@lsuc.UUCP> jimomura@lsuc.UUCP (Jim Omura) writes: > > > > > > Mike K. mentioned a CD-ROM standard based on OS-9 68K. In fact, it's > > >called CDI. > > > Cheers! -- Jim O. > > >James Omura, Barrister & Solicitor, Toronto > > As I understand it, OS-9 will be used as an internal control operating > > system within a closed, appliance type of product (i.e., like > > a stereo component) which will use custom--and we can assume proprietary-- > > chips for graphics and audio. > > > > The question is: What advantage is held by a computer running OS-9 if > > the CDI standard becomes a success? CDI black boxes will certainly be sold, but I can't imagine Sony or Phillips (or any other company) passing up the chance to sell a powerful computer to CDI owners for the price of a floppy drive, keyboard, and some interface components. Perhaps the CDI player with OS-9 showing will cost $500 extra. For people who want a CDI player that should give the computer good price performance. .........: 680xx processor Excellent color graphics Excellent sound processing OS-9 (Multi-user Multi-tasking OS) Integrated CDI player Even if non-CDI OS-9 systems don't profit in any other way, lots of new users means lots of inexpensive software. Also, OS-9 supports networking. It should be easy to attach a minimal CDI computer to a host OS-9 system via network. ******* Peter Dibble@Rochester I'm not a bit unbiased about OS-9. If lots of people get OS-9 with their CDI players maybe some of them will buy my books!
jimomura@lsuc.UUCP (Jim Omura) (04/21/86)
In article <821@ihwpt.UUCP> knudsen@ihwpt.UUCP (mike knudsen) writes: >> >> As I understand it, OS-9 will be used as an internal control operating >> system within a closed, appliance type of product (i.e., like >> a stereo component) which will use custom--and we can assume proprietary-- >> chips for graphics and audio. >> >> The question is: What advantage is held by a computer running OS-9 if >> the CDI standard becomes a success? > ... >> 6- that OS-9 is really great, independent of CDI, and that you >> come out ahead even without any of the above panning out. >> jimm > >When I posted, I thought only of (6) above -- that this CD >standard was a nice feather in the cap of OS9 and we could all >brag about "knowing OS9 when." True, OS9 will be sealed up inside >a box, and hardly anyone will know it's there. >But I will, and so will you... mike k Mike, I've been trying to find out what level the CDI will see OS-9 and so far I haven't gotten an answer. On BIX, we are currently winding up a special conference with experts on CD-ROM, some of whom may know a bit about this, but my question seems to have been buried in the mass. If I remember, I'll try to ask the question again. I'll try to post what I find out at that time. In the meantime, we can't assume anything. Cheers! -- Jim O. -- James Omura, Barrister & Solicitor, Toronto ihnp4!utzoo!lsuc!jimomura Byte Information eXchange: jimomura (416) 652-3880
geoff@desint.UUCP (04/21/86)
In article <333@ius2.cs.cmu.edu> ralphw@ius2.cs.cmu.edu.UUCP writes: > Hierarchical, tree-structured filesystems are 'in' now, but what about the > more general digraph format, where a file or directory can be in more than one > directory? (Unix sort of does this, but there is only one parent directory > at each point in the tree, referenced by ..). This is an old idea that has been tried more than once, including in earlier versions of Unix. To my knowledge, every attempt on a read-write file system has been a dismal failure. The problem is that there are just too many ways it can bite you. Probably the most common is the way linked files bit Unix-er's: some poor schmuck does 'cp msg /bin/mail' to install msg as the default mailer, not realizing that in the process he just wiped out /bin/rmail. It is an open question whether this would be a problem in read-only file systems. BTW, in passing I must put in my 2 cents' worth on Files-11. I am all too intimately familiar with the design of Files-11, as well as a large number of comparative systems (file system design is one of my subspecialties). Files-11 has a remarkable number of serious design flaws, along with less-serious howlers like their general principle of "store all data in the least-manipulable format, especially if that takes up extra space". -- Geoff Kuenning {hplabs,ihnp4}!trwrb!desint!geoff
mc68020@gilbbs.UUCP (Tom Keller) (04/23/86)
In article <821@ihwpt.UUCP>, knudsen@ihwpt.UUCP (mike knudsen) writes: > > The question is: What advantage is held by a computer running OS-9 if > > the CDI standard becomes a success? > ... > > 6- that OS-9 is really great, independent of CDI, and that you > > come out ahead even without any of the above panning out. > > jimm > > When I posted, I thought only of (6) above -- that this CD > standard was a nice feather in the cap of OS9 and we could all > brag about "knowing OS9 when." True, OS9 will be sealed up inside > a box, and hardly anyone will know it's there. > But I will, and so will you... mike k In fact, OS9 was specifically designed to act not only as an interactive operating system, but as an embedded control environment. MicroWare made much of this capability in their early literature. I was operating OS9 back in 1980. -- Disclaimer: I hereby disclaim any and all responsibility for disclaimers. tom keller {ihnp4, dual}!ptsfa!gilbbs!mc68020 (* we may not be big, but we're small! *)
rb@ccird2 (04/26/86)
For those of you in Amiga, Mac, and ST In article <821@ihwpt.UUCP> knudsen@ihwpt.UUCP (mike knudsen) writes: >> In article <1185@lsuc.UUCP> jimomura@lsuc.UUCP (Jim Omura) writes: >> > >> > Mike K. mentioned a CD-ROM standard based on OS-9 68K. In fact, it's >> >called CDI. >> > Cheers! -- Jim O. >> >James Omura, Barrister & Solicitor, Toronto >> >> I can sense Jim's glee now that OS-9 is the darling of JerryP and the >> InfoWorld circles of journalism, by virtue of its incorporation in >> the CDI standard, but I have reservations and I would like to hear some >> net input. >> >> As I understand it, OS-9 will be used as an internal control operating >> system within a closed, appliance type of product (i.e., like >> a stereo component) which will use custom--and we can assume proprietary-- >> chips for graphics and audio. >> >> The question is: What advantage is held by a computer running OS-9 if >> the CDI standard becomes a success? > ... >> 6- that OS-9 is really great, independent of CDI, and that you >> come out ahead even without any of the above panning out. >> jimm > >When I posted, I thought only of (6) above -- that this CD >standard was a nice feather in the cap of OS9 and we could all >brag about "knowing OS9 when." True, OS9 will be sealed up inside >a box, and hardly anyone will know it's there. >But I will, and so will you... mike k After the fiascos with 5 1/4" disks, and 3",3 1/2",100mm,... disks, it looks like the drive makers have learned their lesson. It looks like they are trying to agree on one interface (SCSI), one size, one format, and one file system (OS-9). This reduces the risks to the drive makers, their customers, and the end users. At the same time, there are a wide range of areas for competition, such as rotational speed, caching, queuing, seek times, other speed related issues, and price. This compatibility is extra important when you consider that CD-ROMs will have to be "Printed" like records. Especially when you realize the quantity of information that will be stored. Things like an entire encyclopedia, net.* for 1 year, every public domain source ever published, all of Shakespears plays with illustrations, the index to the library of congress, the telephone directory for the state of New York,... The mind boggles. The point is that all of these things cost money to develope. Even 10% of the IBM market wouldn't cover it. All this before the first device is even released! My guess is that the BIGGEST reason for using OS-9 in CDI is that OS-9 is the ONLY system which has a TRULY sharable Remote File System. This not a file server like TCP/IP, FTP, or UUCP. Or a block server like certain other systems. This is a "remotely executed device driver". With it you can send an "open /usr/lib/libc.a" or whatever file name, and the treat it just like a standard block mode file. You don't have to know anything about how the files are structured, how they are stored, or what the format of the disk is. Just "read(3)" :-). Furthermore, the PC side can do any translations of ":,\,<,>,..." to a simple path format. From the little bit of information I DO have on it, it sounds like a VERY GOOD choice. In addition, the OS can be ROMed, MMU is not required, but is supported, and since it will probably run mostly the RFS server, heap compaction isn't a real strain. In addition, OS-9 has a VDI interface for presentation graphics, good network protocols, and lots of other good things. I can see it now, 68070, 2 256Kx4 rams, SCSI controller/adapter, and CD interface. SCSI/ethernet adapter makes it available to several hosts. SCSI on the hosts make direct connection as simple as "plugging in". Besides, you should see the royalty schedule for 10K copies of OS-9 :-). I wonder if any of the 68K box makers will "follow suit". :-). Note: followups to this article will be posted to net.micro.68k (No need to always keep the net cluttered with duplicates)