On Thu, 4 Dec 2003, Vintage Computer Festival wrote: It has a link for "Learn More About the FAT File System" which takes you to : http://www.microsoft.com/mscorp/ip/tech/fathist.asp which is entitled "FAT File System: The Story Behind the Innovation" which starts off with "December 3, 2003 "The FAT file system was one of the first technologies developed by Microsoft. In 1976, Bill Gates created the initial version of FAT during a short hotel stay in Albuquerque, N.M. Intel incorporated FAT technology in an operating system it developed for the Intel 8086 chip. Microsoft purchased the rights to this system and then recoded the FAT file system as part of the first version of the Microsoft MS-DOS operating system." and then proceeds to blabber on about the file nickname (*) system recently added (Windoze 98) THAT is the least useful "history" that I have ever seen about it! "Bill gates created...Microsoft purchased the rights"???? and WHAT Intel OS was it? I don't know where billg got the linked list idea from. Patterson of Seattle Computer Products shared a booth with Microsoft at the West Coast Computer Faire, and was "inspired" by the use of it in the NEC or NCR "Microsoft Stand Alone BASIC" and used it in QDOS, which MICROS~1 bought, and resold (after completion and a few changes) as PC-DOS/MS-DOS. (*) It is NOT "long filenames". If it were, then they would stand alone, and be indexed BY the "long filenames". Instead, ALL indexing is done by the 8.3 name. EVERY file has an 8.3 name that is used for all access. In addition the user can refer to a file by a "nickname", that is then used to FIND the 8.3 filename. The "long filenames" are a SUPPLEMENTARY added name, and therefore a "nickname".

On Thu, 4 Dec 2003, William Donzelli wrote: I'm NOT objecting to "supplementing". (THIS TIME) I'm griping that in an entire page purporting to be the history, they provide only a couple of sentences, and those leave gaps so big that they don't make complete sense. (When saying that Bill Gates created it in a hotel room in Albuquerque NM (a somewhat less important detail), and then saying "purchased rights", they oughta explain why they would "purchase rights" to what they had created! Even "purchased rights to IMPROVEMENTS in order to make a better OS for YOU" would be a better "history") Then they have multiple paragraphs about one particular [kludged] modification to it. I would RATHER see tham "supplementing", than the few almost RANDOM unconnected snippets that they provide. I DO object to the Cringely CRAP, wherein he has Microsoft cold calling IBM to sell IBM on the idea of having an operating system! NEVER HAPPENED. (IBM went to Microsoft, and Bill Gates did a good job of peddling his stuff) and then Cringely starts preaching into the camera that business students should study his FABRICATED FICTION! (THAT is when he should F OFF) and the Killdall eulogy (which I really like) is done by people who loved and admired Gary, and don't SEE anything out of place about standing up IBM to "deliver some documentation manuals to Oakland" (not delegatable?) It also fails to mention some of the "culture clash" issues - IBM was "shocked" at the working environment at DR, and there were unsubstantiated reports that when the IBM people arrived (looking a bit out of place for that community, some DR people thought that it was a drug raid. -- Grumpy Ol' Fred cisin(a)xenosoft.com

On Thu, 4 Dec 2003, Tom Jennings wrote: I agree. <sigh> But there were some precursors and warnings, even before IBM got into it, such as the "serious" renaming of playful company names: Thinker Toys became Morrow Designs Kentucky Fried Computer became Northstar Intergalactic Digital Research became Digital Research Inc. Parasitic Engineering and Golemics ceased to be significant players, ... Computer companies hired advertising agencies, instead of hoky homemade ads, If you called Microsoft on the phone (206) 255-8080?, you had to know the last name of who you wanted. Software came in a shrink wrapped box, instead of a ziploc bag. The suits arrived.

On Thu, 4 Dec 2003, Vintage Computer Festival wrote: http://www.atarimagazines.com/creative/v10n11/30_The_first_decade_of_perso.… "As people started adding peripherals to their Altairs, the limited capacity of the power supply reared its ugly head. Hence, Howard Fulmer brought out a beefy power supply to replace the original Altair unit. Ed Roberts had been attacking the board compatible companies, calling them parasites, so in a burst of honesty, Fulmer called his company Parasitic Engineering." Golemics was Lee Felsenstein's actually, I was thinking of the OLD days, with a polaroid picture cut out and pasted on typed text

Ahhh... thanks for the memory. Some of my best hacking (*not* cracking, folks) was done while watching Bob Wilkins on Creature eatures... --Patrick :-)

Patrick, I know what you mean, I'd start off the evening working on my BBS code with some Star Trek the original series on Wpix at 11pm, some twighlight zone, then Creature.... Fright Night, ah, usually you could find some Dr. Who on an off-station and maybe if you were really lucky a re-run of Computer Chronicals or C|Net Computer News or similar, sometimes a Discover Science special on PBS.... Now with my Satellite I can't find nearly that much "get you through the night" TV shows.... Curt Patrick Rigney wrote:

Curt, no Tivo?!? I'm never at a loss for something I _care_ to watch, whatever the hour. --Patrick

On Thu, 4 Dec 2003, William Donzelli wrote: The big deal is that they do it. I don't care how many companies do it. They're all assholes then. -- Sellam Ismail Vintage Computer Festival ------------------------------------------------------------------------------ International Man of Intrigue and Danger http://www.vintage.org [ Old computing resources for business || Buy/Sell/Trade Vintage Computers ] [ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]

On Thu, 4 Dec 2003, Tom Jennings wrote: "86" not "89"? It did, indeed use "FAT" "86-DOS" earlier in the process, it had previously been known as QDOS ("Quick and Dirty OS") We're talking about different names of the same OS -- Grumpy Ol' Fred cisin(a)xenosoft.com

Am 4 Dec 2003 11:14 meinte Vintage Computer Festival: I think the trick roots deeper. First of all, it does not look like they are sueing anyone, it's only that they offer licences for their _source_code_ to handle FAT media. Now, if one buys that (and may be inclined to do so to get LFN (*1), he also agrees that his knowledge to handle such structures is from Microsoft, and therefore, whatever he will develop, he will never be able to get out of that, as long as his software supports FAT. it's a one way trap. Also, if enough people accept, it sets the standards for future court issues. What pisses me of most is that they want licence for media if they are formated with FAT. Beside that this is even more a joke than the Rambus scam, it sets the stage that not only software (and the underlaying algorythm) is protected by a patend but also the data generated by this algorythm! Let's just continue the idea: Assumeing MS had a patent for some parts of Word, under the above circumstances, every document written using Word, stored on a media will be the same ... so does MS own what I write, because I'm using their programm? I wouldn't realy say FAT is directly lifted from _one_ other OS, it's a nice development starting from CP/M and other sources (as the intenion of the original developers have been). Jup, that is the most ridicoulous par ... now the original MS-Basic had even FAT ... next thing they tell is that it also had a GUI and of course invented the Internet ... all in MS-Basic :) Gruss H. *1 - to me, LFNs are not realy a new idea - at least not the way they are implemented for FAT .... that's exactly the same scheme CP/M already used 20 years before to handle more than 16 data blocks: spread out a file entry over several directory entries. -- VCF Europa 5.0 am 01./02. Mai 2004 in Muenchen http://www.vcfe.org/

On Thu, 4 Dec 2003, Tom Jennings wrote: It was a Microsoft BASIC, that had some rudimentary file handling built in, including the capability of launching programs. The best known is Coco RS-DOS (6809), but there were earlier versions, on NEC8001 (Z-80) and some non-US Okidata machines (Z-80) The directory structure was on a middle track (35 or 40), with a one sector FAT (called a "GAT" "granule allocation table"). "Clusters" ("granules") were often half a track, and the table used 8 bit entries. The FPDEs included a file size and a starting block number.

On Fri, 5 Dec 2003, Tony Duell wrote: They probably didnt last long, either. --f

On Thu, 4 Dec 2003, SP wrote: I cant comment on that, since I am part of that procedure. Yeah, looks like they do :) The funny part is, that after careful reading, they indeed base their claims on patents for something *related* to FAT (the LFN addons to implement long(er) file names), but not FAT itself. As Sellam already pointed out, in those days, software was not patentable as such, and even copyrights were still worked on :) So.. the question remains: regardless of who *invented* the FAT disk structure (GAT, FAT, FAT16, MACDOS) principle, was it ever subject to a *license* of sorts? I very seriously doubt this. If this were so, anyone being subpoenad by MS *now* could claim belated charges (you cant let something happen for 20 years, and then suddenly go "oooh, darn, thats not right, sue the bastard!"), and be done with it. Microsoft will have to come up to me with better paperwork than those patents. (yes, I just sent them a FAX announcing the refusal to comply in my PDP-11 OS.) Cant remember who wrote this, but: they are not licensing their *code* to process FAT file systems, they intend to license the file system itself. Meaning not just the software that handles it (think any phone or camera, Cisco's IOS software, etc) but also the media that physically implements (SD/flash/MemoryStick card around) this great finding of the past.. Cheers, Fred -- Fred N. van Kempen, DEC (Digital Equipment Corporation) Collector/Archivist Visit the VAXlab Project at http://www.pdp11.nl/VAXlab/ Visit the Archives at http://www.pdp11.nl/ Email: waltje(a)pdp11.nl BUSSUM, THE NETHERLANDS / Sunnyvale, CA, USA

A non-text attachment was scrubbed... Name: not available Type: text Size: 750 bytes Desc: not available Url : http://www.classiccmp.org/pipermail/cctalk/attachments/20031205/5400d7df/at…

A non-text attachment was scrubbed... Name: not available Type: text Size: 1016 bytes Desc: not available Url : http://www.classiccmp.org/pipermail/cctalk/attachments/20031205/3e5a94c0/at…

Jerome Fine replies: Since RT-11 is still actively used, a set of DOCs can still help anyone who is interested. Assuming that RT11 actually means RT-11 based originally on DEC (Digital Equipment Corporation) PDP-11 computer systems, then a brief review of the RT-11 file structure might be helpful. (a) RT-11 does NOT use a FAT!!!!!!!!!!!!!!!!! (b) While block ONE for a file structured device does contain some additional directory information, it is not normally required for directory operations. (c) Block 6 is the start of the directory for ALL random access file structured media. Note that while block 1 contains a word that has 6 as the value that is nominally used to point to block 6, so many RT-11 programs ASSUME that block 6 is the start that it is no longer possible for any other block to be used. (d) The file structure is based on directory segments. (e) Each directory segment is EXACTLY 2 blocks in size or 512 words of 16 bits. (f) The RT-11 command to set-up an RT-11 file structure is "INITIALIZE dev:" and this command is totally separate from the FORMAT command which deals ONLY with the LLF (Low Level Format) of a disk drive or the media in the case of removable media (with certain exceptions that most users would not appreciate - including the fact that probably most media can't be FORMATed FORMAT command does NOT include setting up the RT-11 file structure. For DOS/Windows people, this usually causes GREAT confusion!!!! (g) All file space in an RT-11 directory is accounted for via entries within each directory segment - including areas that are NOT yet allocated that are noted in the status word as being just that - UNUSED. (h) Each freshly initialized RT-11 file structure has exactly one directory segment which contains exactly one file entry for all of the UNUSED blocks on that random access device. There are a maximum of 31 directory segments allowed (some RT-11 file structures for devices that are small such as floppy media allow for many fewer than 31 segments. (i) Each directory segment has 5 words at the start: o Maximum Number of Directory Segments o Next Directory Segment - Zero if the Last o Total Number of Directory Segments in Use - ONLY maintained in Directory Segment #1 o Number of Extra Bytes in Each File Entry o Starting Block Number of First Area which means the maximum device is 65535 blocks (RT-11 does NOT check to verify that this value is correct and follows immediately after the previous file entry in the linked list - thus it is possible to successfully "fool" RT-11, but this is STRONGLY against ALL rules and should not even be considered EXCEPT for READ ONLY MEDIA such as a CD - and then ONLY in special cases!) (j) Each File Entry is Composed of a Minimum of 7 words - extra WORDS are allowed o Status Word - File Entry Type (see k below) o 3 words of RAD50 file name (2) and type (1) o Number of Blocks in the File Entry which means that the maximum file is 65535 blocks o Used for TENTATIVE File Entries - used by TSX-PLUS for file times for PERMANENT files - proposed for use for Y10K dates o Creation Date (01-Jan-1972 up to 31-Dec-2099) o Extra WORDS are used by user programs as specified by Number of Extra Bytes (ALWAYS EVEN) in Each File Entry (k) The are (currently - and probably this will NEVER change) ONLY three types of space: o UNUSED - (1000) the space is available o TENTATIVE - (400) the space has been requested o PERMANENT - (2000) the space has been allocated o END OF FILE entries (4000 - only ONE word is then required for the whole file entry) (l) The maximum number of file entries in each directory segment is 72 [=(512-6)/7]. (m) All file entry blocks are contiguous by definition which means that all files are contiguous. Files can be "extended" only if the immediately following file entry is UNUSED and has sufficient blocks available - this is rare and can't be relied upon in most cases. Since the above is already TOO much detail, I will not mention how files are allocated by the USR (User Service Routine) along with how extra directory segments are created. Any questions are welcome! Sincerely yours, Jerome Fine -- If you attempted to send a reply and the original e-mail address has been discontinued due a high volume of junk e-mail, then the semi-permanent e-mail address can be obtained by replacing the four characters preceding the 'at' with the four digits of the current year.