Dos Ain’t Done ’til Lotus Won’t Run

I was originally going to do a post on this, but Adam (who interviewed me on the topic before he came back to Microsoft) just posted this article and did a far better job of it than I could have ever done (he actually went out did research and stuff).

So go to Adam's blog and enjoy Adam's thorough debunking of a canard.


Comments (16)

  1. vince says:

    Hmmm, but do you deny that certain windows 3.1 betas intentionally would give weird error messages and refuse to run on DR-DOS? for example.

  2. Vince, that’s totally unrelated to whether or not there was a motto "Dos Ain’t done…".

    I wrote a bit on the AARD check here a year or so ago here:

    and here:

  3. vince says:

    > Vince, that’s totally unrelated to whether or not there was a motto

    Well, the entire tone of the article is "Haha, it’s completely silly to believe Microsoft would ever do anything to break backwards compatibility or do anything to make a competitot’s product not work". So I think it is relevant.

    The whole DR-DOS issue has some interesting similarities with current events. Back then DOS 3.3 had been stagnant for years, with the lousy 4.0 release not doing much to improve things. It was a really awful interface, but was "good enough". The DR-DOS came out with all kinds of neat features and started to steal away some customers. So then MS went crazy, in one hand starting a massive campaign about how DR-DOS was incompatible and in the other desperately copying all of the features it could (probably under the name of "innovation").

    The same is happening today with IE. IE stagnated for 5 years at IE6, and nothing at all happened until Firefox started gaining some market share. And now comes the rush to clone Firefox (also probably mis-labeled as "Innovation" by the marketing people).

    Yes, I don’t think there was some sort of company wide conspiracy to make Lotus not run, but I also wouldn’t put it past your company to do the bare minimum for competing products while at the same time playing every trick in the book to get your own products working better. The best thing for the stock-holders… yes. best thing for the users… no, and I wish the company wouldn’t pretend otherwise.

  4. Vince, this is so wrong on so many levels.

    DOS stagnated in the late 1980’s because Microsoft had no interest in improving it – Microsoft was 100% focused on OS/2. In fact, the reason I left the DOS team was because I felt that it had no future (biggest mistake in my career – I was later told I was about to be offered the dev manager position on MS-DOS).

    The reason MS-DOS 5.0 took off was two-fold. First, Microsoft had the rights to the whole operating system (we didn’t have rights to distribute the IBM authored pieces until DOS 3.3 and the Joint Development Agreement). And most importantly, Windows 3.0 shipped in 1990. Even though the wikipedia says that MS-DOS 5.0 shipped in response to DR-DOS, the real driving factor towards shipping DOS 5 was Windows 3.0. Windows 3.0 was shipping 10 million copies a month – more than any other software product in history, and all of a sudden, MS-DOS was a viable and interesting platform.

    DOS 5.0 bundled a clone of IBM’s workplace shell, a text-only shell that IBM developed for PC-DOS 4.0, that’s where the motivation for that came from, not DR-DOS. It also put in a bunch of stuff to improve the experience of Windows (like reducing the DOS footprint to enable more memory for Windows apps)

    I’m not saying that Digital Research didn’t push the DOS team, obviously they did. But DR was moving in a totally different direction from where the MS-DOS team was going – DR was pushing running DOS apps in v86 mode with their app and OS managing the transition. Microsoft was working on improving MS-DOS and ensuring that Windows worked better.

    And IE7’s much more than a FF clone. Yes, it has some FF features, just like FF has some IE6 SP2 features (can you say security bar?). But it’s got a lot that FF doesn’t have. Absolutely, FF is pushing the IE team. And that’s a very good thing.

  5. msemack says:


    Regarding your comments about DR-DOS, you might find this interesting. This was posted in Adam’s weblog, but I feel it is worth repeating here.

    "Someone earlier commented that Microsoft put an error message in Windows just to make DR-DOS look bad. This isn’t true. Contrary to popular opinion early versions of DR-DOS were not fully compatible with MS-DOS. In fact, several internal data structures were very different which caused problems with networking software such as LANtastic (first hand knowledge as a developer.) The error detection in Windows was valid during parts of the development stage. During shipment, these issues were resolved and the detection was turned off BUT the message was not removed. (At the time I was working for the owner of DR-DOS and WE had a hard enough time writing work arounds for that piece-o-crap.)

    Posted by: Joe at August 2, 2005 02:56 PM"

  6. vince says:

    You can say all you want about the DR-DOS issue, but it went to court for a long time (,1283,20128,00.html)

    and eventually MS settled (

    I’m sure you all will claim that the whole Stack/Doublespace issue was some sort of horrible misunderstanding as well.

    The computer industry does have a short memory, but some of us were there and still remember.

  7. vince says:

    That last link should have been,aid,14739,00.asp

    And yes, Firefox is pushing the IE team. Though you must admit it’s a miracle that anyone is pushing the IE team at all, considering how thoroughly MS destroyed netscape. At least the whole bundle-the-browser-in-the-OS is going to bite you guys, when it turns out that to get the "new" features of ie7 in MS OS’s other than Vista people are going to have to install Firefox.

  8. Vince, ALL I’m saying is that "DOS Ain’t Done ’til Lotus won’t run" is a canard, and never existed. Nothing more, nothing less.

    The DR-DOS was a big deal, but there were valid technical reasons for blocking it, and those valid technical reasons WERE legitimate (read the court ruling denying the injunction).

    Stac/Doublespace WAS a horrible misunderstanding. As I recall it, the Stac patent covered one specific compression algorithm and Doublespace used a totally separate compression algorithm. Unfortunately for Microsoft, the jury felt that even though the compression mechanism was totally different, the Stac patent covered the entire concept of compression of the disk. That’s the thing about juries – you can’t predict how they’ll act.

  9. Tim Smith says:

    I was there too Vince and Microsoft didn’t destroy Netscape, Netscape did. The same is true for WordPerfect. I was an AVID Netscape and WordPerfect user and I switched to MS products because they were better. I still remember trying to use Netscape 4.2 and getting tired of all the problems. Sure, IE was icky (technical term), but it worked better than Netscape. The same was true for Borland. I used Borland C++ compilers for years before I switched to MS because MS was better.

    BTW. This is being posted from a Firefox browser because it is better than IE. When IE7 comes out, I’ll take a look at it.

  10. Tim Smith says:

    Oh, and just because someone settles out of court or loses a lawsuit doesn’t mean they are guilty of what the press or the 30 second summary states. Patent abuse is becoming very widespread. Things are a lot more complicated than it appears.

  11. PaulJBis says:

    vince, if you are an american, you should know better than anyone that lawsuits have little to do with "justice" and more to do with commercial tactics. Often it’s just easier to settle out of court than to spend time and money litigating, even if you have reason on your side.

    Here’s a suggestion: why don’t you go back to Slashdot? I’m sure that you’d find a much more receptive audience there.

  12. vince says:


    > if you are an american, you should know better than anyone that

    > lawsuits have little to do with "justice" and more to do with

    > commercial tactics.

    Well yes, otherwise it would be impossible to explain how MS escaped any sort of real punishment during the most recent anti-trust suit.

    For some interesting reading, check out Caldera’s statement of facts from the DR-DOS trial: It makes some fascinating reading.

    > Here’s a suggestion: why don’t you go back to Slashdot? I’m sure that

    > you’d find a much more receptive audience there.

    Why do people assume that if you don’t like MS you are somehow some sort of Slashdot user?

    I have been vocally critical of MS’s business practices for way longer than Slashdot has even existed.

    I will admit there are 3 Microsoft products that I really liked and used extensively.

  13. Simon Cooke says:


    Actually, they were the same compression algorithm – Stac had a patent on it. Microsoft also had a patent on a compression algorithm too – the same one. (And one that they spent a lot of time searching for, so they wouldn’t infringe Stac’s patent). Unfortunately, the people at fault here were the USPTO.

  14. Tim Smith says:


    Do you happen to have a link to a patent search system or the patents in question? I thought IBM had one (patent search site), but I can’t find it.

  15. Wally Bass says:

    Maybe there is a corollary that "the blog ain’t done ’til the red herring is sung," or something like that. Lotus seems to me to be entirely a red herring in this particular phase. The general question implied by the phrase in question is whether Microsoft ever wrote code to torpedo a competitor. I never heard of that even being a serious question with regard to Lotus 1-2-3, but it was a serious question on the front as to whether Windows was intentionally coded to not run on DR-DOS. Andrew Schulman’s investigation on that topic is still available at – I suggest that the interested reader visit that page for themselves and make their own judgement.

Skip to main content