Talk:Gary Kildall

Videos and articles
see http://www.commodore.ca/gallery/magazines/cpm/cpm.htm for video and articles about Gary

Did Tim Paterson have any rights to the OS he sold Bill Gates?
The only difference between Tim Paterson's CP/M, and Gary Kildall's CP/M was Tim Paterson wrote a patch for Gary Kildall's CPM so it would run on the 8088 chip. It is open to question whether Tim Paterson had any rights to the patched version of CP/M he wrote for Seattle Computer Products considering the fact he wrote the patch for Seattle Computer Products while he was their salaried employee. If Gary Kildall did not retain rights to the patched version of his software, then Seattle Computer Products should have owned the rights to it, not Tim Patterson.

Michael D. Wolok 07:20, 20 June 2006 (UTC)


 * One can't convert a program to run on a different processor by "writing a patch". Each processor family has different op codes.  Paterson's O/S was completely new. It simply shared some system call numbers for similar functions with CP/M, to make it easier to convert a CP/M program to QDOS.  The file system for QDOS was completely different, and actually borrowed the FAT concept from Microsoft's 1977 Disk BASIC. So Digital Research never had any connection with, or rights to it. There was also never a question that the rights belonged to SCP, not its employee. An employee's work always belongs to the employer unless it is explicitly addressed by an employment contract, and Paterson did not assert any such rights. --Blainster 15:02, 21 June 2006 (UTC)

Is Bill Gates the biggest software pirate that ever lived? If you want to know what kind of person Bill Gates really is read his own words below as they appear in the court record.

long post by user Wolok redacted because: (1) dupicating material across multiple articles (Talk:Bill Gates) is a form of spam, and (2) Wikipedia is about writing articles, not pursuing personal attacks. (If you wish to work on improving an article then please do so) --Blainster 22:13, 21 June 2006 (UTC)

The above is public record. --Michael D. Wolok 21:08, 21 June 2006 (UTC)

I don't know what kind of background you have in software, but it is in fact possible to reimplement, from scratch, a program that someone else has previously written. Gazpacho 09:30, 22 June 2006 (UTC)

PC-DOS legality
Wolok, or whoever made the edit on Nov. 18, please stop. Wikipedia isn't for publicizing conspiracy theories. Why don't you go read Gordon Eubanks testimony in US v. Microsoft, or his interview that is online, or the cited BusinessWeek article in which DR's lawyer at the time seems to have no recollection of a lawsuit? Gazpacho 08:53, 20 November 2006 (UTC)


 * Au contraire! A quick Google turned up this link (http://www.digitalresearch.biz/CPM.HTM) which contains a link to the entire sordid affair (http://www.maxframe.com/DR/Info/fullstory/fullstory.html). Apparently the lawsuit was brought by Caldera after it had purchased Novell (who had purchased DRI earlier) and was brought on more general grounds than simple theft [oh, like Micro$oft would never, ever do that - DOS 6.20 anyone?]. One only has to read one of the last court filings (http://www.maxframe.com/DR/Info/fullstory/last6responses.rel.html) to get an education on just how far Micro$oft will go to preserve their monopoly. And in case anyone under the age of 25 is wondering, this old electrical engineer remembers these events transpiring pretty much in the way described therein. JimScott 23:33, 20 November 2006 (UTC)


 * Jim, I'm well aware of the Caldera lawsuit and so is Wolok. It wasn't filed by DR, nor was any part of it litigated during Kildall's lifetime, nor did it go to trial, nor was there any gag order to obliterate knowledge of it. The reasons why the Nov. 18 edit cannot be accurate are endless. (What jurisdiction allows the liable party in a civil suit to get a gag order against the winner?) A specific claim is being made here, and it's pure urban legend and pulp fiction.


 * Researching the claim, I found that John Dvorak may have written about the supposed easter egg in 1996, referring to it in a "friend of a friend" context. The story was picked apart in a 1998 post to alt.folklore.urban, which led me to a 1995 message on the Cypherpunks mailing list (note, before Dvorak apparently wrote about it). According to those two messages, programmer Randy Cook might have used an easter egg to demonstrate that Tandy owed him royalties for TRS-DOS (an unrelated system) after he was fired. Whether this happened or not, it appears to be the origin of the CP/M story. Gazpacho 02:42, 21 November 2006 (UTC)


 * WADR, Gazpacho, I was only pointing out that there was a suit even though it was not filed by DR nor litigated during GK's lifetime. I apologise for not being clear on that; I see now that my addition was seen out of context. My intent was only to suggest that Wolok, et al, might wish to rework this portion of the article around something more substantial (like the actual suit as opposed to the urban legend you refer to). JimScott 16:29, 20 February 2007 (UTC)

Tonight while looking up info about Kildall, I happened upon a podcast that included Jerry Pournelle. Apparently he was Dvorak's unnamed source, and he specifically claims that Gary Kildall knew a command to display his name in DOS 1.0. Having looked over the CP/M-86 code, I don't believe Pournelle. Nonetheless, a source is a source, so I have added his claim at QDOS and rewritten the article here to reflect that there are open questions. I have also e-mailed Pournelle politely asking him to make public whatever documentation he has about the matter (and he does claim to have some.)

By the way, you can download a DOS 1.1 disk image by searching for "tk-dos11." It won't boot on any current machine, but you can still dump it and disassemble the code. According to Pournelle the infringement issues were not addressed until version 2. Where's that easter egg? Gazpacho 04:33, 29 November 2006 (UTC)


 * From Pournelle: "I have told this story about 20 times including in BYTE back in the 1980's. Enough is enough." Oh well. Gazpacho 18:04, 29 November 2006 (UTC)

GA on hold
This is a very nice article, well-written and seemingly comprehensive. All it needs is a few more citations to be GA quality. There are a few paragraphs without citations, and a couple of direct quotes from Gates and Kildall that need citations immediately after them. I am putting the article on hold to allow this to be dealt with, and will pass it once this is taken care of. MLilburne 18:03, 3 December 2006 (UTC)


 * I think you'll find that everything is now cited. Kildall's comment about Gates was quoted from his memoir in the Andrews article, which is also the source for the next sentence. Gazpacho 23:12, 3 December 2006 (UTC)


 * Thanks for making the changes so promptly. I'm now happy to pass it. MLilburne 11:25, 4 December 2006 (UTC)

the same?
Is Grolier Multimedia Encyclopedia the same as (or related to perhaps?) the "Grolier Electronic Encyclopedia" link that Mugsywwiii added?JimScott 16:29, 20 February 2007 (UTC)

The G.E.E. was not multimedia, it was text only (for reasons unrelated to storage capacity). Gazpacho 09:34, 28 February 2007 (UTC)

The IBM deal
Gary is purported to have been out flying his plane, missing a critical meeting with IBM.

While something like this likely happened on more than one occasion, it is not the cause of his missed opportunity. Nor is DOS the reason for the loss...

Anyone that owned an IBM AT remembers the boot screen, “AT Multiuser System”. Do any of you know what that meant? DOS is not multiuser, and save for a few specialized applications, the AT rarely provided a multiuser experience.

The operating system choices for the XT included CP/M-86, which IBM considered a premium offering, with DOS being a somewhat broken demo. The main reason for offering DOS was because DR would not come down on the price for CP/M-86, and would not sell it to IBM royalty-free. When IBM was developing the AT, using prototype 286 chips, they contracted with DR to produce the next generation OS to run on it.

The OS was a multiuser version of CP/M that used serial ports connected to terminals to support each of the users. It turned the AT into a machine that could support a whole office of users. It was a multitasking OS, and would have made a large impact on the computing world had it been released with the AT as planned.

DR got the OS working, using prototype AT systems supplied by IBM. IBM used the OS internally for prepackaging testing, and developed documentation to be shipped with the system. The complete package was assembled, and ready to go. Production of the AT was ramped up, and soon a warehouse of systems was ready to ship.

During QC testing, samples of the new systems were run through their paces, only to find that they were broken. It was discovered that Intel had changed to a production version of the 286 chip, which had a slightly different design to improve chip yields. This design change broke the protected mode instruction set, and prevented the OS from working without a major re-write.

Much finger pointing ensued. But being as DR's two largest customers were IBM and Intel, and this was just one project of dozens, it was all quietly forgotten when IBM decided to ship the AT with Dos, shelving the new OS for a future project that never came.

DR eventually shipped the OS as “Concurrent Dos” many years later. It was used in grocery store cash registers, as well as other OEM type systems.

DR was not threatened by Dos, since the new OS was so much better and more powerful. Dos was supposed to become unimportant in the wake of this new OS.

This story was related to me by a former DR employee, Michael Marking, who worked with me under Gary Kildall at Prometheus Light and Sound. --Georgedotcom 15:59, 21 March 2007 (UTC)


 * You were told what you were told, but it basically contradicts what authors who have researched the history in depth have found. Although there were probably engineers in IBM's PC division who preferred CP/M over DOS, as a business IBM was committed to Microsoft, from 1981 to 1991, as the operating system software partner (even though it would have supported other companies making OSes for the PC). At the same time, Microsoft was committed to keeping IBM committed. That was Steve Ballmer's job for most of the 1980s. So if you can find a printed source that says otherwise (such as an IBM press piece announcing a partnership with DRI), that would be interesting. Gazpacho 08:39, 26 March 2007 (UTC)
 * According to Wikipedia the Intel 80286 was announced in 1982, but I'm not sure about the exact release dates, given that the first processor stepping with reportedly fully usable protected mode (E-2) for DOS was not available before August 1985. (ToDo: Check old processor errata to find out more about the actual problems with earlier steppings.) According to the WP article, the IBM PC/AT saw its debut in mid 1984. If it actually shipped this early, it must have been shipped with a prior stepping. (ToDo: Does someone recall the exact 286 and AT release dates and the processor stepping in the original PC/AT?)
 * Digital Research had a real-time multi-user multi-tasking derivative of CP/M, named MP/M, available for 8080/Z80 and 8086/8088 CPUs, including a dual-processor variant. Documentation for MP/M-86 II (with BDOS 2.x) for Intel 8086/8088 processors has copyrights 1981. (ToDo: Check if MP/M-86 actually shipped in 1981 already, or not before 1982.)
 * The successor to MP/M-86 was named Concurrent CP/M-86 3.0 (with BDOS 3.0). The last version I know was Concurrent CP/M-86 3.1 (with BDOS 3.1). I'm not aware of a specific 286 version of Concurrent CP/M, although Concurrent CP/M-86 also ran on 286 machines, of course.
 * With the addition of the DOS emulator PCMODE, developed by Digital Research since August 1983, this operating system was renamed to Concurrent DOS 3.1 (also with BDOS 3.1) in March 1984.
 * A version of this operating system taking full advantage of the 286 protected mode existed as early as January 1985 and was scheduled to be shipped in April 1985 as Concurrent DOS 286 (IIRC with a BDOS 4.x kernel). Reportedly it worked on B-1 prototype steppings of the 286 (during development), but was found to be incompatible with C-1 steppings of the 286 in May 1985 when it was tested by unnamed industry customers. Digital Research had to talk to Intel to develop a fixed processor, which became available with the E-1 stepping in June/July? 1985, but it was too slow to be practically usable. This was finally solved in August 1985, when Intel's E-2 stepping of the 286 became available, which is officially the minimum requirement for Concurrent DOS 286 to run. This limitation also applied to FlexOS 286, a derivative of Concurrent DOS 286 developed since 1986 and marketed since January 1987. (For example, FlexOS 286 version 1.31 for IBM PC/AT machines was released in May 1987.)
 * What is interesting about Concurrent DOS 286 and FlexOS 286 in the context of the statements by the OP is the fact that they have been licensed by IBM. Named IBM 4680 OS version 1, IBM originally chosed DR Concurrent DOS 286 as the basis of their IBM 4680 computer for IBM Plant System products and Point-of-Sale terminals very early in 1986. (See for example: IBM selects Concurrent DOS-286 for PC AT retail system, Digital Research European Review, March 1986, Issue 18, p. 1, or IBM to use a DRI operating system by Melissa Calvo and Jim Forbes, InfoWorld, 1986-02-10, .) The last release of IBM 4680 OS has been version 4, before it was replaced by IBM 4690 version 1.
 * In July 1993, IBM announced the adoption of FlexOS version 2.32 as the basis of their IBM 4690 OS version 1, to be pre-released on 24 September 1993 and generally made available from 25 March 1994. (See: or .) IBM continues to maintain 4690 OS up to the present, with the most recent version in April 2010 being IBM 4690 OS version 6.2. --Matthiaspaul (talk) 17:32, 10 December 2011 (UTC)
 * Nevertheless, there have been a few other multi-user multi-tasking operating systems as well in 1984/1985, for example Microsoft's XENIX. XENIX 3 aka XENIX 286 existed in August 1984 and was shipped as SCO XENIX 286 in 1985. I don't know to which extent XENIX 286 actually utilized the protected mode, but given that it had no DOS compatibility, it certainly did not depend on certain processor properties which were essential for Digital Research's Concurrent DOS 286 to ensure binary compatibility with existing CP/M-86, MP/M-86, Concurrent CP/M-86, Concurrent DOS, and MS-DOS/PC DOS programs in a multitasking environment while taking full advantage of the 286 protected mode at the same time. While Intel developed the 286 E-2 stepping to solve the problems DRI was facing with the 286, they also issued a statement, that Digital Research attempted to use the processor in ways which were not originally intended by Intel. Presumably, backward compatiblity with DOS was no design goal for the 286 protected mode, only for the processor's real mode. The virtual 86 mode of the 386 was not available on the 286 - it appears reasonable to assume that it was at least inspired by this experience. --Matthiaspaul (talk) 20:19, 10 December 2011 (UTC)


 * I stumbled upon an article in a German computer magazine named Computerwoche, dated 1985-09-13: . In there they reported about IBM's withdrawl from previously announced plans to distribute Digital Research's Concurrent DOS 4.1 and GEM through IBM's own channels. IBM explained this with internal misunderstandings which had caused announcements being made before the actual contracts have been signed. A final decision would not have been reached at the time. Industry insiders in the USA were reported to see this in the light of a tighter cooperation with Microsoft and plans to establish Windows as a successor to IBM's TopView. --Matthiaspaul (talk) 23:31, 18 December 2011 (UTC)


 * As a follow-up regarding the IBM PC/AT as "AT Multiuser System" above, I have found a document titled "Evaluation of Proposals for an Electronic Data Processing System for the Papio Natural Resources District" by Donald F. Norris of the "Center for Applied Urban Research University of Nebraska", dated 1984-10-01. This document contains an addendum describing a (then) yet-to-be-released "AT Multiuser System" by IBM which was briefly pre-evaluated in this study as well. There is only one operating system mentioned in the context of this "AT Multiuser System": Microsoft's XENIX.
 * https://web.archive.org/web/20170108193204/http://digitalcommons.unomaha.edu/cgi/viewcontent.cgi?article=1246&context=cparpubarchives
 * --Matthiaspaul (talk) 20:02, 8 January 2017 (UTC)
 * These two articles also talk about Xenix only, not about Concurrent CP/M-286 or Concurrent DOS 286:
 * https://www.nytimes.com/1984/08/23/business/technology-the-multi-user-computers.html "Technology; The Multi-User Computers", David E. Sanger, 1984-08-23.
 * https://books.google.com/books?id=Hy8EAAAAMBAJ&pg=PA11&lpg=PA11 "IBM rolls out new PC" Infoworld, 1984-09-10, pages 10-11.
 * --Matthiaspaul (talk) 20:54, 24 January 2020 (UTC)
 * Some further research revealed a few mentionings of something (at least conceptually) named MP/M-286 (since 1982) and Concurrent CP/M-286 (1985), predecessors of what later would be released as Concurrent DOS 286 (1985) and FlexOS (since 1986) and licensed to IBM for use in their POS systems. This supports a lot of what the OP wrote, but still does not prove a deal regarding IBM's "AT Multiuser System" or that a "warehouse of systems" was already ready to ship.
 * --Matthiaspaul (talk) 11:54, 23 January 2020 (UTC)
 * Some more material:
 * https://books.google.com/books?id=kC4EAAAAMBAJ&pg=PA83&lpg=PA80&focus=viewport "Company Strategies Boomerang", InfoWorld, 1984-04-02, page 83: "DRI has also been very hard at developing other relationships with IBM and has just agreed to translate its languages for the IBM 3270/PC. In addition, rumors are flying that DRI is working on an operating system for a new IBM computer".
 * https://books.google.com/books?id=k1IbhUVEFgwC&pg=PA11&lpg=PA11 "IBM multiuser micro forecast - without Unix", Computerworld, 1984-01-16, page 11.
 * http://www.gaby.de/ekildall.htm "The true story or ... Gary Kildall, inventor of the smart phone", George Wenzel
 * http://dosmandrivel.blogspot.com/2007/08/is-dos-rip-off-of-cpm.html (Anonymous comment on 2012-12-09)
 * --Matthiaspaul (talk) 20:37, 24 January 2020 (UTC)

Prometheus Light and Sound
I worked for PLS for 3 years, having developed all of the hardware. We were designing a multi-function platform intended to be anything from a PBX to a cellphone. The cellphone was to have a LCD touchscreen with only software-defined iconic buttons. There was also a desktop version dubbed the “Intelliphone” that could support 1 or more sub-processors called PPUs. The system was intended to be used as building blocks for the development of complete communications systems employing every technology available at the time. Larger systems with dozens of PPUs would be used at central sites to provide digital telephony service not unlike today's digital cellphones. Smaller versions would reside on the desktop, kitchen counter, and pocket. The system could talk to your home appliances, and allow you to operate your home or office from a remote location, as well as provide a UUCP based store and forward system for moving email and other communications files between the various nodes. It also was to include TCP/IP support so that it could be used on the Internet as well, though the Internet was not yet available to the public. The intended audience was the business executive, and his grandmother. So it had to be powerful, and easy to use, at the same time.

The phone network was to be of hybrid design. At home, your handset acted as a cordless phone, using your home's land-line and Intelliphone to connect to the phone network. When away from home, the device became a cellphone, retaining it's phone number and functionality in a seamless manner. This design was based on cellphone minutes being expensive, so you wanted to use your home land-line as much as possible.

I was hired because of my electronics background, but I was found because I was developing a multi-user UNIX based bulletin board system designed to be a commercial portal to the communication matrix consisting of the Internet, the UUCP network, Fidonet, Bitnet, and others. The BBS was just a hobby at the time, but I considered it more than a hobby, it was my future. Eventually Internet access was available, for upwards of $700 per month for a 9600 baud dial-up connection. We smashed that price model, selling Internet access for $13 to $20 per month.

I left PLS to work at IBM as a Unix developer for a short time before my hobby, running the first modern ISP, evolved from nerd-net to a consumer commodity. I have no doubt that a PC based Intelliphone would have become the first Internet browser appliance had Gary not died when he did. That is one of the many directions this project was heading. Gary had more than a passing interest in the Internet, and was intermittently discussing it with me prior to his death. He was trying to create something like the Internet, prior to the Internet being opened to the general public by way of initiatives sponsored by senator Al Gore. --Georgedotcom 16:48, 21 March 2007 (UTC)
 * Is there any connection of Kildall's Prometheus Light and Sounds or their so called Intelliphone with US Order's & Colonial Data Technologies' Intelliphone also named "Telesmart 4000" and mentioned in a 1996 article of Popular Mechanics (that is, just two years after Kildall's death)?
 * --Matthiaspaul (talk) 13:05, 22 August 2021 (UTC)
 * --Matthiaspaul (talk) 13:05, 22 August 2021 (UTC)

Comments on the main article
As for the assertion that Gary had "occasional private outbursts of bitterness" over the Dos issue, it is true he was somewhat bitter, but I would never call these events "outbursts". The most memoral event in this regard was when we upgraded from windows 3.0 to windows 3.1 on a system running DrDos. DrDos had the reputation for running windows better than MsDos, due to the integrated memory manager that was superior even to Quarterdeck's QEMM. In an effort to kill DrDos, Microsoft implemented a fake blue screen that would appear if DrDos was detected. When this was discovered there was no outburst, as one would expect... Gary just made some brief comments, and decided the windows test machine (a 16mhz Compaq 386 system) was to be our new file server.

Regarding the comment "Kildall's interest was primarily in inventing and writing programs that mattered to him, and not in building an industry or a large company", this is silly, as his main focus was building an industry AND a large company based on the Intelliphone. This effort was what kept him from dwelling on bitterness from the past.

His home in Westlake was not a ranch, and the garage had 6 stalls, not 14. The garage was to display his collection of famous race cars that he aquired, each having been restored to ready-to-race condition by the original car builders. While I never saw his Pebble Beach house, it is my understanding that there he kept a collection of Lamborghini Countachs... THAT garage was large, and may very well have held 14 cars. He drove around Austin in a Lamborgini LM4, and had a beat-up black Chevy truck for hauling things. He very briefly owned a Lear Jet, but the fuel consumption was so high that he sold it and purchased a Gulfstream wide-body jet to replace it.

The reason he never sued over MsDos is that there was no legal basis early on, and by the time Lotus invented the look and feel concept, the statute of limitations had run out and the purportedly stolen code was no longer in use... A number of CP/M clones existed, many having been written by Gary's college students. Gary told his students that the best way to learn programming is to write your own operating system, and CP/M was the standard his students cloned. So he wasn't in the habbit of suing over cloned versions of his OS. --Georgedotcom 15:59, 21 March 2007 (UTC)


 * 23-March-2007: Thank you for your comments. I have swapped "outbursts" to become "statements" as less POV-slanted.  In general, claims of extreme behavior require multiple sources for Wikipedia: for example, saying he was married at his local church might go unsourced, but saying he exchanged vows sailing in a Caribbean lagoon would need multiple, independent sources.  I will revise and re-source the other issues in the near future.  Thanks, again. -Wikid77 15:22, 23 March 2007 (UTC)

26-March-2007: After finding contrary evidence, I have removed the speculative phrase that Kildall's interest was "not in building an industry or a large company" and substituted a fact ("after gaining US$ millions from selling DRI, he started another company, Prometheus Light and Sound"). The prior speculation did seem like nonsense, considering Gary Kildall had named DRI "Intergalactic Digital Research" -- not the name of a quiet, little hackers venture. See more rationale at: Talk:Gary_Kildall, below.

References list
23-March-2007: A Wikipedia "Notes" section contains footnotes, such as defined from ref-tags. The footnotes (displayed by " " ) can cite sources or just explain details as an aside comment. A Wikipedia "References" section is a bibliography, typically in alphabetical order by name of author or organization. To condense citation footnotes, reuse the same ref-name with a trailing slash "/" (" "), then for books, list all relevant page numbers in the one reused ref-tag (example: ). Read more at: Guide_to_layout. -Wikid77 15:01, 23 March 2007 (UTC)

Motives sources (2007)
26-March-2007: Articles on Wikipedia tend to avoid descriptions of a person's private motives. A statement of motives or priorities would require detailed evidence, particularly written quotations from the person directly, and not just another person's opinion. Although motives are certainly valuable information, a description of motives is very difficult, due to the high standard of evidence required to define a person's motives. I am removing the speculative notion that Gary Kildall did not want to lead a large corporation; that notion was based on one guy's opinion, while other information contradicts that notion: the DRI company was originally called "Intergalactic Digital Research" rather than some small hackers, temporary venture. Also, Gary Kildall's final company, called Prometheus Light and Sound, was developing PBX telephone systems to compete with high cell-phone prices, not the domain of a small operation. I think any other statements of motives should be backed by direct quotations from Gary Kildall, rather than from hearsay. -Wikid77 00:03, 26 March 2007 (UTC)

Cause of Death
The official Digital Research site states "Gary Kildall passed away following a blow to his head at the Franklin Street Bar & Grill in Monterey, California" Do you think the Death section of the article should be updated from "sustained an injury at a Monterey restaurant" to "sustained a blow to the head at the Franklin Street Bar & Gril in Monterey"?

The site also contains his eulogy however I don't know if it would make sense to link to it. Suggestions please? 64.93.163.34 09:07, 10 November 2007 (UTC)

He had a serious alcohol problem. He fell off a bar stool and died from the resulting head injury. After his death, his wife sued his doctors for malpractice because she claimed that they only treated the head injury, rather than treating the underlying problem of alcohol abuse.--76.93.42.50 (talk) 06:48, 22 February 2008 (UTC)


 * I heard that story too, but it is one of multiple conflicting stories. Nobody knows what happened because Gary didn't remember himself. 67.168.160.218 (talk) 00:08, 8 March 2008 (UTC)

It's a silly legend though; how could the doctors treat the underlying alcohol problem of alcohol for a man who comes with a severe blow in his head and dies three days later? A little logic is needed here. --Robert Abitbol —Preceding unsigned comment added by 172.163.244.97 (talk) 07:02, 30 March 2008 (UTC)

Hope this explains more: On July 6, 1994 Kildall, 52, walked into a Monterey bar. He was wearing motorcycle leathers with Harley-Davidson patches;a would-be biker. There were some real bikers in the bar. Something was said. There was pushing and shoving, and Kildall died from injuries sustained to his head. An inquest called the death "suspicious," but no one was charged. Calling Gary an alcoholic is the same as calling Ballmer a monkey. It was not an accident. — Preceding unsigned comment added by 188.89.247.48 (talk) 11:20, 26 July 2012 (UTC)

Gary's book
Known is that Gary wrote a book shortly before his tragic -but convenient for Microsoft- death about Ballmer and Gates. Where can this book be obtained? From what I read, a family member has the book but is scared for reprecussions by Gates' lawyers. The book should be opened for publication. — Preceding unsigned comment added by 188.89.247.48 (talk) 11:29, 26 July 2012 (UTC)

Questionable analysis
I removed the following text excerpt from the article added on 2012-08-07T19:10:00‎ by user Wdhowellsr:


 * "A complete forensic analysis performed and published in the July 2012 issue of IEEE Spectrum concluded that, "QDOS was absolutely not copied from CP/M, and MS-DOS showed no signs of copying either. Kildall’s accusations about Bill Gates were totally groundless."

While Bob Zeidman's article is certainly interesting (and even sometimes entertaining) to read, I removed it because I think his "forensic analysis" is fundamentally invalid in several ways and the conclusions drawn from it are superficial and highly misleading (comparing apples with oranges and then drawing conclusions on bananas). I could not help but to get the impression that this was written by someone trying to help the sales of the mentioned tools. If you think this pseudo-scientific analysis should really be mentioned in the article, we should at least find some text framework putting it into perspective. --Matthiaspaul (talk) 23:55, 7 August 2012 (UTC)


 * CP/M and QDOS / MSDOS were clearly different code bases and as such would have survived any legal challenges according to current case law. While I have no connection to Zeidman's company and for that matter Microsoft or CP/M, I will see by your response if in fact you are simply biased against Microsoft or related to Kildall.  If you do replace this again, I expect a very clear explanation as to the legal reason for QDOS / MICROSOFT being in clear violation of copyright.  I look forward to your response. — Preceding unsigned comment added by Wdhowellsr (talk • contribs)


 * Wdhowellsr restored the text and I too have again removed it. Not only is the "forensic analysis" itself flawed and cannot be used to draw the conclusions that the author attempted to make, but the specific text Wdhowellsr added is borderline inflammatory and most certainly not the kind of thing we want in articles. --Tothwolf (talk) 19:45, 10 August 2012 (UTC)


 * Comment. My sense is the WP article already covers the issue: comments about MS/DOS being a "clone"; Kildall believed it infringed; "Davis told him that intellectual property law for software was not clear enough to sue"; Kildall compromised. Some of the reverted text is not supported in the ref and could be original research. There is not a claim about what sort of infringement Kildall alleges.
 * Consequently, I do not have a problem with reverting the proffered text.
 * The Spectrum article is poor, sounds like an advertisement, and has some silly claims. Nevertheless, I believe the article could be used for some narrow attributed statements in the article. Zeidman may qualify as a forensic expert with a significant viewpoint.
 * Disclosure: I met Kildall in 1972, and he gave me a copy of his thesis. One of the claims in the Spectrum article sounds like a corrupted comment about the TOPS-10 operating system.
 * Glrx (talk) 20:40, 13 August 2012 (UTC)

I guess the editors who researched this article should be glad that it has finally brought some coherence to the CP/M – DOS controversy. Zeidman clearly relied on it. Unfortunately, in researching whether "Kildall's accusations were groundless," he forgot to check what Kildall's accusations were. Computer Connections, as represented in They Made America, made that very clear: DOS copied CP/M's interface. This has never been a secret or up for debunking. If Kildall had more than that, can anyone imagine why he would leave it out of a book written specifically from bitterness against Gates? 24.22.217.162 (talk) 03:27, 18 August 2012 (UTC)
 * The Court of Appeals for the Federal Circuit has ruled that APIs are copyrightable and the Supreme Court sided with them.
 * I think that's a bad ruling in general, but it does mean that legally speaking the IBM-Microsoft camp was wrong. Of course legally, they became right again when Kildall signed his right to sue away.
 * But the more important question would be whether MS-DOS would be copy from a moral perspective, and whether Kildall should have gotten greater recognition, and all talk about legal issues and forensic analyses detracts from the question that really matters. — Preceding unsigned comment added by 82.139.82.82 (talk) 17:41, 4 August 2015 (UTC)

John Torode
I'm quite surprised that an article on Gary Kildall has no mention of John and Patsy Torode who, with Gary Kildall, started Digital Systems. John Torode designed most of the hardware on which Gary's earlier software ran. Danensis (talk) 14:58, 10 December 2012 (UTC)
 * In fact, they might be notable. If you can, please write the Digital Systems (Seattle) (currently only a redirect) and/or John Torode (physicist) articles. --Matthiaspaul (talk) 12:08, 4 April 2015 (UTC)

My first edit today
Apologies. I'm not entirely sure what happened, it looks like I accidentally copied a much older version of the page into place. I'll try the edit again later (which is not the one I just did.) --98.254.202.225 (talk) 17:17, 14 August 2013 (UTC)

Jack Sams
FTA: IBM lead negotiator Jack Sams insisted that he never met Gary

Verbal agreements are legally binding, so he cannot say anything else really. — Preceding unsigned comment added by 82.139.82.82 (talk) 17:27, 4 August 2015 (UTC)

Memoires Released to public under strict individual license
http://www.computerhistory.org/_static/atchm/in-his-own-words-gary-kildall/ Part of the confidential book written by Gary is released under a license not allowing to do citation, but it is allowed to link to this distribution page. Only the first chapter is released. — Preceding unsigned comment added by 91.214.169.69 (talk) 02:03, 5 August 2016 (UTC)
 * Now added as an External Link. This should help improve visibility and accessibility, even though this had already been included in the article as a reference.--Concord gioz (talk) 23:01, 3 June 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 2 external links on Gary Kildall. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20090524153356/http://norran.se/familj/article233923.ece to http://norran.se/familj/article233923.ece
 * Added archive https://web.archive.org/web/20161119154226/http://www.cadigital.com/kildall.htm to http://www.cadigital.com/kildall.htm

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 11:35, 11 October 2017 (UTC)

Recent changes
A lot of referenced material has recently been removed from this topic, replaced with unreferenced material that's often inaccurate and poorly-written. Has there been a conversation about these changes? I've removed them once already, since they were completely unreferenced. That removal was reverted, and I think the quality is suffering as a result. -- Mikeblas (talk) 15:53, 10 August 2021 (UTC)
 * I took a run at cleaning up after those changes, and I convinced myself that it was better just to revert. Too much of the new material has problems, and too little references. -- Mikeblas (talk) 03:00, 16 August 2021 (UTC)
 * Thanks, Mike, I started to work myself through these recent changes as well but without edit summaries or sources I found them rather difficult to check, therefore, to play it safe, I was about to revert to the last stable revision as well.
 * Alpha Lion, if there really are aspects we haven't covered properly in the article yet, please let us know, so that we can discuss them here before incorporating them into the article.
 * --Matthiaspaul (talk) 20:56, 16 August 2021 (UTC)

Homicide?
I was reading this earlier and noticed an edit by user 187.58.157.239 basically saying that Stewart Cheifet was the prime suspect for Kildall's murder. This was completely uncited, and there are no readily accessible sources for the content which said that Cheifet was evasive about his whereabouts, struck a reporter and Kildall and Chiefet didn't get on and would 'set deadly traps' for one another.

If anyone has sources, feel free to add what I removed back in, but I just see borderline libel. 37.60.108.144 (talk) 04:07, 20 December 2023 (UTC)