Heh, I remember finding shreds of these news groups threads and news letters posted to Geocities and such in the mid-90's when I was coming of age. "The MUF List" - Microsoft's Undocumented Features, etc. I still have my DOS programming books but I never really did much myself, other than some assembly programs written in debug. The list of dirty things MS did to other companies is extensive. The DriveSpace debacle was the one I remember most. If memory serves, they tricked the company that made a compressed filesystem driver for DOS/Windows to "license" the feature to MS, then MS just forgot to ever pay them. The company was so cash starved they couldn't outlast a lengthy lawsuit to recoup what they were legally owed. MS did every dirty trick they could. Bill Gates is one seriously dirty MFer.
> The company was so cash starved they couldn't outlast a lengthy lawsuit to recoup what they were legally owed.
They were actually paid $120M by Microsoft [1]:
> A Los Angeles jury Wednesday ordered Microsoft Corp. to pay $120 million for infringing a software patent held by Stac Electronics Co., a much smaller software firm based in Carlsbad, Calif.
Stac sued MS and won. But, Microsoft counter-sued Stac for reverse engineering MS software-- Microsoft's claim was that the features used to implement Stacker were undocumented by Microsoft, and only used internally by Microsoft. Microsoft won this suit. And, ultimately "invested in" Stac and paid some royalties as the ultimate settlement.
The "funny" thing was that Microsoft was claiming that it documented everything and that competitors could make equivalent competing products to their own on their platforms to the court in the anti-trust case that was around the same time.
MS management and their company were and are sleazy and ruthless to a fault.
Somehow it seems like "paid" is the wrong word when you're forced by a court to do it. Almost like stealing a car and then claiming you "donated" it to the true owner upon arrest.
Winning a settlement doesn't automatically get you the money. It just gives you legal rights when trying to collect -- you still have to actually collect, which can be difficult and time-consuming.
Do we know if MS actually paid this settlement, and if so, on what schedule?
The way I understood the story back then was that they wanted a license deal with MSFT just like MSFT had with the hardware companies. MSFT wanted to just buy them out. Lawsuits, court, and so on finally forced the sale of the company to MSFT for the amount in the settlement.
Microsoft entered into negotiations with Stac Electronics to licence Stacker, but the deal fell though (allegedly, Stac claim that Microsoft didn't offer any money, just the right to sell enhancement products).
So Microsoft instead licensed a competing product called DoubleDisk from Vertisoft. So of course MS didn't pay anything to Stac Electronics, but presumably Vertisoft was compensated (hopefully fairly, we haven't heard otherwise... But this is Microsoft)
Stac later successfully sued MS for patient infringement, which is where the story about Microsoft alleged "no cash licensing" offer came from.
Is it? How so? I'm not saying they're angels, but I don't know if I'd consider the MSFT of today "worse" than in the 90s and early 2000s. It's certainly a more complicated beast than it was then.
Someday this is going to be in an engineering ethics class.
Microsoft literally broke every anti trust law, but the DOJ was too incompetent to prosecute them properly due to the lack of education.
A lot of these decisions remind me of SV companies: if it’s not specifically, illegal it’s fine; Ex: Goog/Fb invasive Tracking users across the web, even if the user requests not to be. I hope every person that works for these companies is someday ashamed of their time there.
“It Is Difficult to Get a Man to Understand Something When His Salary Depends Upon His Not Understanding It.”
Also see current AI doom debates. See Yann LeCun tweets. He has some the silliest metaphors, like comparing AI development to aviation development. As if the two are similar in order of magnitude of downside risk.
Salary and income are one of the most powerful decision making and ethics shaping factors. The flow of money literally defines everything for 95% of people, including their ideological and political preferences.
Yes, incompetent, and not due to as of yet unknown payoffs to key officials. With that much money at stake, in a far less transparent world than we live in today, briefcases full of cash worked a lot better. How would you even deposit that today without some sort of money laundering scheme?
Briefcase cash is so gauche - you just hire them when they “retire” from public service. Much more proper and you don’t even need to explicitly offer - everyone knows how the bread is buttered.
Tell your parents to sell their house, but the contract includes lifetime tenancy and guaranteed maintenance and renovations. Surely that's not bribery.
It's not bribery if you just happen to visit a new friend's mansion in Hawaii any time you want a vacation.
Perhaps your spouse would be a good choice to be VP of Something Boring at this non-political non-profit charity. Isn't it nice that the employees of TotallyNotCorruptCo always make sure to donate so much money each year? Did we mention that the VP job comes with a car and driver? And can be done from home except for the fund-raiser galas. Totally not a bribe.
Microsoft didn't need to bribe anyone in the 80s and 90s because our country was still snorting that "woo hoo free markets fix everything" pixie dust. We weren't allowed to interfere with big business anymore. Clearly they knew what was best for us.
IIRC the ruling was under Clinton, but the 'punishment' under Bush was that Microsoft gave free copies of Windows to schools (which were a Apple bastion at the time).
I hope the people who make these kind of decisions will get prison time. It's much easier saying sorry than asking for permission and some people exploit it on the back of the society, no less
> DOJ was too incompetent to prosecute them properly due to the lack of education.
It was not the lack of education. It was a political policy choice that tied DOJ's hands.
Reagan era changed US antitrust into so called Borkism. Robert Bork's "The Antitrust Paradox" (1978) criticized the US United States antitrust law and solidified the thinking of University of Chicago school of economists.
Borkism has very narrow view into monopolies. It has created monopolies in ever sector of the economy.
That microtracking made a lot of tiny and very cool niche consumer packaged goods brands a lot of money in the early 2010s. They can't afford to use untargeted TV advertising like the mega brans unfortunately, but Facebook did a great job helping them connect with their niche. It's much harder for new brands to get started these days.
Keep in mind that this isn't an isolated case: Microsoft used this as a tactic with Quicktime amongst others. Their tactics were numerous, either bringing spurious error messages, stealing associations to bring error messages/dodge the invocation of the competitor software or simply altering windows code to produce legitimate errors in competitor software.
That's on top of plain old threatening other companies if they dared use technology that they were trying to suffocate.
True, this is one case that isn't mentioned much, but this gave Excel and Word its massive market share.
When Windows 95 was being developed, MS kept telling Lotus (and maybe Word Perfect) that OS/2 was the real future, so Lotus was focusing on a GUI version of their main product for OS/2. And Notes was originally developed on OS/2.
So Lotus had a nice version of 123 for OS/2 about to go, but the had to shift gears to Windows when W95 was released. Also there were rumors that MS was keeping some APIs secret to make it appear their products were faster than competitors.
When W95 came out, by magic, MS had a GUI version of Excel and Word, thus eventually killing those companies. Lotus had to shift gears and may have lost a couple of years to MS.
I used to work on Wine and don't recall encountering any APIs that made things magically faster. That was a popular rumour on Slashdot in the early 2000s because at the time, most devs were unaware of how IE could start so much faster than Mozilla, and how MS Office could start so much faster than OpenOffice. Cheating via pre-loading was assumed. It was eventually proven that this couldn't be the case when these apps started faster than their competitors even on Wine on Linux. The actual cause was that MS cared about and heavily optimized startup time. No magic.
The OS/2 history isn't quite how I remember it. The fact there'd be a new version of Windows surprised nobody. Chicago development wasn't exactly a secret that caught anyone unawares. Also OS/2 Warp came out before Windows 95 did.
> When Windows 95 was being developed, MS kept telling Lotus (and maybe Word Perfect) that OS/2 was the real future, so Lotus was focusing on a GUI version of their main product for OS/2. And Notes was originally developed on OS/2.
This story doesn’t make sense to me. IBM and Microsoft’s falling out was public knowledge by July 1991 - there was an article about it in the New York Times. [0] Windows 95 wasn’t released until July/August 1995 - 4 years later. Everyone who was paying attention knew by mid-1991 that Microsoft’s future was Windows not OS/2.
> Lotus was focusing on a GUI version of their main product for OS/2.
Lotus and Wordperfect both spent a lot developing for almost every platform other than Windows. It always surprised me that they never had even a fallback for Windows for their main products, given how dependent they each were on a single product.
GUI Excel 2.2 was released for Windows 2.0 in 1989, 3 major releases before Office for Windows 95. Essentially same for Word.
WordPerfect had a perfectly working GUI version for Windows by 1994, it just couldn't compete with Word. Same for Lotus123, they were out of fashion long before Windows 95.
Excel and Word both ran on Windows 2.0 (released Dec 1987). I supported Excel on Windows 2 in production.
In the late 1980s, yes, Microsoft and IBM told the industry that OS/2 was the future, because they both thought it was.
OS/2 1.0 was released the same month as Win 2.0) had no GUI, because it wasn't ready.
OS/2 1.1 had in, released nearly a year later in Nov '88. This was what Lotus and others needed. 1.1 was not good and barely worked. For example, OS/2 1.1 still had the Windows 2 MS-DOS executive style user interface.
OS/2 1.2 was nearly another year: Oct '89.
Now, it worked, it included advanced features such as the HPFS filesystem, meaning long filename support, it had a revamped and improved user interface, and the performance was OK... but also the 80386 had been out for 2 years and 386 PCs were becoming mainstream. OS/2 1.X was limited to the 286. That meant a maximum of 16 MB of RAM, that meant no ability to multitask DOS applications, and other serious limitations. The high end PC market was visibly accelerating away from the new operating system.
It also had formidable hardware requirements. For example it wanted 8 MB of RAM minimum.
Somewhere around the timeframe of OS/2 1.1 or OS/2 1.2 is when Microsoft realised that this project was running late, it had high hardware requirements, and yet it still couldn't fully exploit the hardware of new PCs shipping in the late 1980s, and it was becoming apparent that the future was not OS/2 after all.
So it's somewhere around the 1988 to 1989 that Microsoft realised that's unless it was willing to sacrifice the PC operating system market, it needed a fallback plan, it needed some kind of alternative: something that could run on high-end mainstream commodity PCs, which means a 386 with 2 to 4 MB of RAM, that offers more compatibility with DOS, DOS device drivers, and DOS applications than OS/2.
That's roughly the time that the school choir project within Microsoft started to turn into Windows 3. Windows 3.0 was released in 1990, it had the new user interface from OS/2 1.2, it could support and use the new facilities of 386 chips… but it could also run in just 1 MB RAM, ran quite well in standard mode in 2 MB of RAM, and ran really well with 4 MB of RAM in 386 enhanced mode with multitasking of DOS applications. In other words, in half the RAM requirement of OS/2 1.2, they could actually do more than that operating system could do, in terms of important things that MS-DOS users actually wanted and needed.
So yes in the very late 1980s, Microsoft did lie to large software vendors, promising them that OS/2 was the future, while it was working on something that would ultimately destroy OS/2. But even now I am no admirer of Microsoft, I have to admit that it was very far from a certainty that Windows 3 was going to be a hit. In the time frame that Microsoft was developing Windows 3, Word for Windows and Excel were already shipping products.
Yes Lotus and WordPerfect believed the lies. Yes it's true, Lotus and WordPerfect could've targeted Windows 2, but Windows 2 was pretty terrible and was a commercial flop. So if they had targeted Windows 2 it would cost a lot of money and the products would never have made it back. Yes they made a mistake, and yes later on Microsoft encouraged them to continue investing in that mistake, knowing that there was a gamble that might make it all a huge waste of money. But what they did probably looked like the sensible prudent decision the time.
And more to the point, relevant to your comment, this all occurred in the late 1980s, long before the industry went 32-bit and long before anybody even dreamed of Windows 4 which would eventually become Windows 95. That stuff happened in the 1990s, when frankly the OS/2 war was already long over.
Lotus was at a crossroads. They could continue with DOS, port to Windows, or port to OS/2. Which path should a cash-rich company with one product pick?
All three.
I have little sympathy for Lotus. They need to own their decisions. Blaming their competitor for misdirection is eye rolling.
I think they still introduce small problems for people who run software they don't like. For example, if you run Linux on your desktop, and want to use a Windows VM for work, you can't run WSL2 or Hyper-V in the KVM/Qemu guest, even if nested virtualization is enabled and working with other operating systems. It used to work, but now it doesn't, and no know knows exactly why.
These situations are murky. I don't know that they broke it deliberately. Maybe it broke on its own and the problem affects so few people that they just don't care about it.
And you can't really demand that they spend their developer resources on things that they think won't help them. But on one hand, they wrote their own Wayland server for Windows 11, but on the other hand, they say, we can't make Teams work on Wayland.
They do lots of odd stuff. It's pretty easy, and pleasant, to run a headless linux server on Hyper-V, but setting up a proper desktop system on your own is hard. They don't actually create a wall you can't get around, but they create obstacles that make it easier to do the things they want you to do.
It's not fair or reasonable to get mad at them about this stuff. It's more that we should be clear about where they're coming from.
A while back, my Ubuntu LTS system kept badgering me to upgrade to a new Ubuntu. I finally relented, and installed the latest. It installed without any errors, but the display was all messed up.
After much fiddling and googling about, I deduced that apparently didn't support the graphics card anymore. It never said "don't install me, cuz it won't work with the card I used to work with." It just died on it. I wound up reinstalling the older Ubuntu.
I don't believe there was any conspiracy by the Ubuntu developers. No conspiracy is necessary - it's really hard to not break things. I knew some people who worked in appcompat at MS, and they had a lot of stories to tell about how hard they had to work to not break things. A lot of software misused Windows APIs making it hard for MS to improve things.
> I think they still introduce small problems for people who run software they don't like.
Linux, FreeBSD and Apple OSX operating systems routinely break without warning older software. I've had the fewest breakages with Windows operating systems.
Just in the last few months, I had to buy a new iphone because Waze, Twitter, and my banking app simply ceased working on it. No warning, no nothing. Those apps upgraded themselves, and just stopped. No check for compatibility with an older iphone.
Bought a new iphone, transferred the apps over, and they magically started working again.
I remember somebody at Microsoft arguing that AARD had a sort of legitimate purpose -- Windows reached deep into DOS in ways that in modern times would be anathema, and did stuff that one couldn't expect to necessarily work properly on a DOS clone, since it wasn't part of the public API.
> As a simple example, when Windows started up, it increased the size of MS-DOS’s internal file table (the SFT, that’s the table that was created by the FILES= line in config.sys). It did that to allow more than 20 files to be opened on the windows system (a highly desirable goal for a multi-tasking operating system). But it did that by using an undocumented API call, which returned a pointer to a set of “interesting” pointers in MS-DOS.
[...]
> But the problem was that the MS-DOS 4.0 SFT was 2 bytes larger than the MS-DOS 3.1 SFT. In order to get Windows to work, I had to change the DOS loader to detect when win.com was being loaded, and if it was being loaded, I looked at the code at an offset relative to the base code segment, and if it was a “MOV” instruction, and the amount being moved was the old size of the SFT, I patched the instruction in memory to reflect the new size of the SFT! Yup, MS-DOS 4.0 patched the running windows binary to make sure Windows would still continue to work.
So yeah, by modern standards the way old DOS software worked was all kinds of screwy.
It was later revealed exactly what the purpose was, when internal mail discussions came out in court.
"What the guy is supposed to do is feel uncomfortable, and when he has bugs, suspect that the problem is dr-dos and then go out to buy ms-dos. or decide to not take the risk for the other machines he has to buy for in the office."
Would it be legal to create similar incompatibilities if there is both a legal business justification (that particular solution can be implemented in X developer time and would cost us Y money which wins against a proper implementation) and an engineering justification (that particular solution would use X less bytes of disk/RAM and would run better on Y% of our customer's machines) for an implementation which heavily depends on undocumented internals of another product?
Microsoft has done many questionable things but the AARD code was never enabled on a shipping version of DOS so it doesn't seem quite fair to blame them for that.
I also find the whole argument silly because it was obvious even during the Windows 3.0 days that Windows was going to be the whole OS and no one would be buying two operating systems, one to use as a bootloader and compatibility layer for legacy apps and the other to be the GUI OS.
The whole idea never made any sense and all the sturm & drang about Dr DOS proposes an impossible future where a niche for a paid text-mode PC DOS OS was going to continue existing period. It wasn't and it didn't. It's like complaining about Microsoft hijacking the market for TCP/IP drivers. Even if the conspiracy theory was true it was never going to matter.
> Microsoft has done many questionable things but the AARD code was never enabled on a shipping version of DOS so it doesn't seem quite fair to blame them for that.
Here’s how a Microsoft executive described it at the time: “What the guy is supposed to do is feel uncomfortable, and when he has bugs, suspect that the problem is DR-DOS and then go out to buy MS-DOS"
It’s certainly true that DOS’s time was limited but people were still buying it for years after he said that in 1992. Far more importantly, however, you have to think about the situation with OEMs. Microsoft was absolutely cutthroat on licensing with PC manufacturers to prevent alternative operating systems – entire product lines had to be Microsoft-only, they demanded DOS & later Windows licenses be purchased even if the buyer didn’t want them, volume discounts depended on exclusivity or bundling other apps, etc. The hardware companies put up with that because they didn’t have much choice: there were too many business apps and games which their customers wanted to refuse Microsoft’s terms. DR-DOS offered a way out of that: everything ran, often better due to its superior memory management, and the last thing Microsoft wanted was someone doing something like selling systems with, say, OS/2 or GeoWorks Ensemble (a quite decent desktop app suite at the time) and relying on DR-DOS as the operating system choice for customers who wanted DOS compatibility to avoid Microsoft’s exclusivity requirements for that particular model. I don’t know that this would have fared better than, say, System 76 is doing now but they were keenly aware of just how much leverage they’d lose if an alternative got enough marketshare to let OEMs do more than acquiesce to Microsoft’s standard terms.
I understand the quote and 100% believe that some people within MS wanted it purely to prevent competition. But clearly someone with a cooler head looked at it and made the call not to enable it. That's the only thing that actually matters.
I get what you are saying about Dr DOS and the OEMs but it was a battle already lost - Windows 3.x was a massive blockbuster for MS. I suspect that was part of the internal argument for not shipping the AARD code: it was a liability for a battle already won. Windows 95 started development in 1992 and was intended from the start to be a single OS without requiring a separate DOS.
Not to say Microsoft didn't do bad things. The OEM licensing terms were abusive in my opinion... why not focus on that instead of some AARD code that was never enabled and wouldn't have mattered even if it was?
To Microsoft, by my reckoning, DR DOS was an imitator of MS-DOS. In assessing Microsoft’s reaction to DR DOS, I must think what would be own reaction if I had designed and implemented something, and someone copied my design into a distinct implementation of their own. I would have to accept that copyright law gives me no protection and I might console myself with the proverb about imitation being flattery, but I would think it outrageously unjust if some law would somehow have me owe anything to the imitator.
I'm sorry I have to strongly disagree with this view. MS-DOS was developed in 6 weeks and was a clone of another popular OS of the era called CP/M created in 1974 by Digital Research. The same company that created DR-DOS. If there were laws regulating common and fundamental designs and implementations as we do copyrighted works Microsoft wouldn't exist. Digital Research would have sued Microsoft in '81 and that would have been the end of Microsoft.
The section "Common Ancestry" acknowledges the prior history. Though it feels a bit odd. "MS-DOS is there said to be variously a copy or clone of an operating system named CP/M" seems to gloss over the details by sticking specifically to Caldera's Consolidated Statement of Facts, and not mentioning QDOS/86-DOS.
But the author really assert that "DOS" is a very specific term:
> To me, DOS meant MS-DOS and its licensed variants such as PC DOS. DR DOS was no more a brand of these than juices pressed from soy beans, almonds or cashews are brands of milk. This is not to say they have no merit—some of those juices are more enjoyable than is most American milk—but they’re not different brands of the same thing. To me, DR DOS could not even be a candidate for consideration as another brand if it would always, as seemed likely, need special accommodation in anyone’s DOS programming.
so the idea of a "DOS market" doesn't make sense.
Further, the author probably places you into this camp:
> The preceding question seems to have got conflicted in the early 1990s—and may be still in the apparently growing community of retro-computing enthusiasts, some of whom are vigorously revising history, at least as recorded by Wikipedia, to favour DR DOS. The conflict I mean is that celebration of DR DOS as a compatible alternative to MS-DOS for running MS-DOS programs often coincided with requests that MS-DOS programmers should keep in mind what different writing may be needed for DR DOS.
I remember testing a protocol called DDE back in the Windows 3.1 days, and we couldn't figure out what Excel behaved so much faster than Lotus 123. Then we dug into the code and found that Windows 3.1 would free up all available memory as it launched Excel. A sneaky trick which I guess meant that users would find Excel much more responsive that 123 and ultimately help Microsoft get Office domination that it has today.
Someone else in this thread who worked on Wine said there was no evidence for such a thing since the faster launch times also occurred in Wine. What forensics did you use to determine this?
I don't believe the context of the Wine related comment was Win 3.1. OpenOffice wasn't around to compare launch times. Internet Explorer was around, a bit, but was competing with Netscape, not Mozilla.
The idea of introducing bugs to users that doesn’t conform 100% to intended use of your software is not good idea, what will happen is that your software will be perceived as buggy in general.
Not your software, a competitor's software. The intent here was to inject bugs to make DRDOS look bad. And it worked.
Later Microsoft would have to spend many long years dealing with how many extension points they'd put in Windows which were opportunities for third parties to inject crashes accidentally into Microsoft executables. That's why there's an entire subsystem to allow video drivers to crash and restart.
Depends on whether it's common or not. If a smaller competitor has bugs, the default is something else, and folks will go yo the default and look askance at the ones sticking with the odd duck software. Source: Linux user since the 90s, LibreOffice/OpenOffice/StarOffice since early '00s.
When they reach that unusability limit, most people will call their IT department and have the offending buggy software reinstalled as a fix. IT won’t remove your buggy copy of Microsoft Word and Teams and replace it with WordPerfect and Slack.
In case you're wondering what AARD means (I did): "The name was derived from Microsoft programmer Aaron R. Reynolds (1955–2008), who used "AARD" to sign his work; "AARD" was found in the machine code of the installer" (from Wikipedia).
If the QDOS/86DOS angle interests you check out a talk that Tim Paterson, of Seattle Computer Products and later Microsoft, gave at VCF West 2019. He describes the history of 86-DOS (licensed to become PC-DOS and MS-DOS). API compatibility with CP/M, was the intention but no code was taken from CP/M.
CP/M was an Intel 8080-based OS. The 8086 was mostly source compatible (but not binary compatible) with the 8080 (i.e. 8080 assembler programs could be assembled targeting the 8086 with only minor changes) and QDOS/86DOS supporting CP/M APIs was to help it fit into this re-assembly workflow.
Memories of seeing my Dad’s boxed copy of “Doctor DOS” sitting on the bookshelf next to his 286 PC are forever burned into my brain. Using DR DOS and XTree on that computer was my first taste of PC usage. Prior to that I’d only been allowed to use the family Atari 800. Years later, when Wolf3D hit the scene, I finally managed to convince him to let me upgrade it to MS-DOS 6, install a VGA card, and upgrade to 4MB of RAM.
Has anybody implemented this in the open source world ?
It's far enough in the future, it would be good to see a reproducable demo of Win95 on DrDOS (or even freedos) - it seems like something someone on the virtuallyfun site might do eventually.
Actually it was shipped. “Microsoft disabled the AARD code for the final release of Windows 3.1, but did not remove it, so that it could have become reactivated later by the change of a single byte in an installed system.” [1]
[1] Schulman, Andrew; Brown, Ralf D.; Maxey, David; Michels, Raymond J.; Kyle, Jim (1994) [November 1993]. Undocumented DOS: A programmer's guide to reserved MS-DOS functions and data structures - expanded to include MS-DOS 6, Novell DOS and Windows 3.1 (2 ed.). Addison Wesley. ISBN 0-201-63287-X.
As documented in the excellent book "Undocumented DOS", whose red cover I can still picture. (Noted in the article, but it was a good book that I still recall fondly.)
IIRC, the author of "Undocumented DOS" discovered and disclosed this feature during the beta, before release.
It seems at least plausible, if not probable, that Microsoft disabled the xor encrypted bogus error message generating code in the release version as a result of this disclosure.
> In assessing Microsoft’s reaction to DR DOS, I must think what would be own reaction if I had designed and implemented something, and someone copied my design into a distinct implementation of their own.
I mean yeah, but it’s not like they just copied it from him.
> Microsoft, which needed an operating system for the IBM Personal Computer,[8][9] hired Tim Paterson in May 1981 and bought 86-DOS 1.10 for US$75,000 in July of the same year. Microsoft kept the version number, but renamed it MS-DOS.
Heh, I remember finding shreds of these news groups threads and news letters posted to Geocities and such in the mid-90's when I was coming of age. "The MUF List" - Microsoft's Undocumented Features, etc. I still have my DOS programming books but I never really did much myself, other than some assembly programs written in debug. The list of dirty things MS did to other companies is extensive. The DriveSpace debacle was the one I remember most. If memory serves, they tricked the company that made a compressed filesystem driver for DOS/Windows to "license" the feature to MS, then MS just forgot to ever pay them. The company was so cash starved they couldn't outlast a lengthy lawsuit to recoup what they were legally owed. MS did every dirty trick they could. Bill Gates is one seriously dirty MFer.
> The company was so cash starved they couldn't outlast a lengthy lawsuit to recoup what they were legally owed.
They were actually paid $120M by Microsoft [1]:
> A Los Angeles jury Wednesday ordered Microsoft Corp. to pay $120 million for infringing a software patent held by Stac Electronics Co., a much smaller software firm based in Carlsbad, Calif.
[1] https://www.latimes.com/archives/la-xpm-1994-02-24-fi-26671-...
Stac sued MS and won. But, Microsoft counter-sued Stac for reverse engineering MS software-- Microsoft's claim was that the features used to implement Stacker were undocumented by Microsoft, and only used internally by Microsoft. Microsoft won this suit. And, ultimately "invested in" Stac and paid some royalties as the ultimate settlement.
The "funny" thing was that Microsoft was claiming that it documented everything and that competitors could make equivalent competing products to their own on their platforms to the court in the anti-trust case that was around the same time.
MS management and their company were and are sleazy and ruthless to a fault.
2 replies →
Justice delayed is justice denied.
I wonder if we can say the same for timely payments especially when maintaining proper cashflow is critical for small businesses.
Somehow it seems like "paid" is the wrong word when you're forced by a court to do it. Almost like stealing a car and then claiming you "donated" it to the true owner upon arrest.
Winning a settlement doesn't automatically get you the money. It just gives you legal rights when trying to collect -- you still have to actually collect, which can be difficult and time-consuming.
Do we know if MS actually paid this settlement, and if so, on what schedule?
The way I understood the story back then was that they wanted a license deal with MSFT just like MSFT had with the hardware companies. MSFT wanted to just buy them out. Lawsuits, court, and so on finally forced the sale of the company to MSFT for the amount in the settlement.
That's not the story on Wikipedia.
Microsoft entered into negotiations with Stac Electronics to licence Stacker, but the deal fell though (allegedly, Stac claim that Microsoft didn't offer any money, just the right to sell enhancement products).
So Microsoft instead licensed a competing product called DoubleDisk from Vertisoft. So of course MS didn't pay anything to Stac Electronics, but presumably Vertisoft was compensated (hopefully fairly, we haven't heard otherwise... But this is Microsoft)
Stac later successfully sued MS for patient infringement, which is where the story about Microsoft alleged "no cash licensing" offer came from.
>Bill Gates is one seriously dirty MFer. Agreed. Completely agreed. Just watched the death of OS/2 video, and it confirms it.
Unfortunately, MS is even worse these days.
Is it? How so? I'm not saying they're angels, but I don't know if I'd consider the MSFT of today "worse" than in the 90s and early 2000s. It's certainly a more complicated beast than it was then.
2 replies →
True. They are waiting for the days of subscription only. as soon as they can.
[flagged]
Someday this is going to be in an engineering ethics class.
Microsoft literally broke every anti trust law, but the DOJ was too incompetent to prosecute them properly due to the lack of education.
A lot of these decisions remind me of SV companies: if it’s not specifically, illegal it’s fine; Ex: Goog/Fb invasive Tracking users across the web, even if the user requests not to be. I hope every person that works for these companies is someday ashamed of their time there.
Ashamed? The opposite, usually they’re proud.
“It Is Difficult to Get a Man to Understand Something When His Salary Depends Upon His Not Understanding It.”
Also see current AI doom debates. See Yann LeCun tweets. He has some the silliest metaphors, like comparing AI development to aviation development. As if the two are similar in order of magnitude of downside risk.
Salary and income are one of the most powerful decision making and ethics shaping factors. The flow of money literally defines everything for 95% of people, including their ideological and political preferences.
Sad but true.
2 replies →
The auto complete black box is just soooo dangerous haha you better regulate all the risky other ones, mine's cool though d/w chief
> “It Is Difficult to Get a Man to Understand Something When His Salary Depends Upon His Not Understanding It.”
see also: cloud
Yes, incompetent, and not due to as of yet unknown payoffs to key officials. With that much money at stake, in a far less transparent world than we live in today, briefcases full of cash worked a lot better. How would you even deposit that today without some sort of money laundering scheme?
Briefcase cash is so gauche - you just hire them when they “retire” from public service. Much more proper and you don’t even need to explicitly offer - everyone knows how the bread is buttered.
5 replies →
Tell your parents to sell their house, but the contract includes lifetime tenancy and guaranteed maintenance and renovations. Surely that's not bribery.
It's not bribery if you just happen to visit a new friend's mansion in Hawaii any time you want a vacation.
Perhaps your spouse would be a good choice to be VP of Something Boring at this non-political non-profit charity. Isn't it nice that the employees of TotallyNotCorruptCo always make sure to donate so much money each year? Did we mention that the VP job comes with a car and driver? And can be done from home except for the fund-raiser galas. Totally not a bribe.
Microsoft didn't need to bribe anyone in the 80s and 90s because our country was still snorting that "woo hoo free markets fix everything" pixie dust. We weren't allowed to interfere with big business anymore. Clearly they knew what was best for us.
Plus when Bush took over, looks like he had the DOJ drop the case. At least under Clinton they were trying, though incompetently as you said.
IIRC the ruling was under Clinton, but the 'punishment' under Bush was that Microsoft gave free copies of Windows to schools (which were a Apple bastion at the time).
The two primary parties any country generally votes for:
1. Corrupt over there and incompetent everywhere else
2. Incompetent over there and corrupt everywhere else
3 replies →
The DOJ competently prosecuted Microsoft for its illegal monopolistic practices: https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor....
But then Bush took over as president and ordered DOJ to effectively drop the case. Microsoft got away with a slap on the wrist. https://www.latimes.com/archives/la-xpm-2001-sep-07-mn-43010...
I hope the people who make these kind of decisions will get prison time. It's much easier saying sorry than asking for permission and some people exploit it on the back of the society, no less
> DOJ was too incompetent to prosecute them properly due to the lack of education.
It was not the lack of education. It was a political policy choice that tied DOJ's hands.
Reagan era changed US antitrust into so called Borkism. Robert Bork's "The Antitrust Paradox" (1978) criticized the US United States antitrust law and solidified the thinking of University of Chicago school of economists.
Borkism has very narrow view into monopolies. It has created monopolies in ever sector of the economy.
That microtracking made a lot of tiny and very cool niche consumer packaged goods brands a lot of money in the early 2010s. They can't afford to use untargeted TV advertising like the mega brans unfortunately, but Facebook did a great job helping them connect with their niche. It's much harder for new brands to get started these days.
Which anti-trust law says you harm consumers by giving them free stuff?
Keep in mind that this isn't an isolated case: Microsoft used this as a tactic with Quicktime amongst others. Their tactics were numerous, either bringing spurious error messages, stealing associations to bring error messages/dodge the invocation of the competitor software or simply altering windows code to produce legitimate errors in competitor software.
That's on top of plain old threatening other companies if they dared use technology that they were trying to suffocate.
>Keep in mind that this isn't an isolated case
True, this is one case that isn't mentioned much, but this gave Excel and Word its massive market share.
When Windows 95 was being developed, MS kept telling Lotus (and maybe Word Perfect) that OS/2 was the real future, so Lotus was focusing on a GUI version of their main product for OS/2. And Notes was originally developed on OS/2.
So Lotus had a nice version of 123 for OS/2 about to go, but the had to shift gears to Windows when W95 was released. Also there were rumors that MS was keeping some APIs secret to make it appear their products were faster than competitors.
When W95 came out, by magic, MS had a GUI version of Excel and Word, thus eventually killing those companies. Lotus had to shift gears and may have lost a couple of years to MS.
I used to work on Wine and don't recall encountering any APIs that made things magically faster. That was a popular rumour on Slashdot in the early 2000s because at the time, most devs were unaware of how IE could start so much faster than Mozilla, and how MS Office could start so much faster than OpenOffice. Cheating via pre-loading was assumed. It was eventually proven that this couldn't be the case when these apps started faster than their competitors even on Wine on Linux. The actual cause was that MS cared about and heavily optimized startup time. No magic.
The OS/2 history isn't quite how I remember it. The fact there'd be a new version of Windows surprised nobody. Chicago development wasn't exactly a secret that caught anyone unawares. Also OS/2 Warp came out before Windows 95 did.
4 replies →
> When Windows 95 was being developed, MS kept telling Lotus (and maybe Word Perfect) that OS/2 was the real future, so Lotus was focusing on a GUI version of their main product for OS/2. And Notes was originally developed on OS/2.
This story doesn’t make sense to me. IBM and Microsoft’s falling out was public knowledge by July 1991 - there was an article about it in the New York Times. [0] Windows 95 wasn’t released until July/August 1995 - 4 years later. Everyone who was paying attention knew by mid-1991 that Microsoft’s future was Windows not OS/2.
[0] https://www.nytimes.com/1991/07/27/business/microsoft-widens...
> When Windows 95 was being developed, MS kept telling Lotus (and maybe Word Perfect) that OS/2 was the real future, so
Microsoft's committment to OS/2 ended with Windows 3.0. But until 3.0, they themselves believed OS/2 was the (imperfect) future.
Windows 3.0's ability to use protected mode on a 286 changed it all.
https://www.landley.net/history/mirror/ms/davidweise.html
> Lotus was focusing on a GUI version of their main product for OS/2.
Lotus and Wordperfect both spent a lot developing for almost every platform other than Windows. It always surprised me that they never had even a fallback for Windows for their main products, given how dependent they each were on a single product.
GUI Excel 2.2 was released for Windows 2.0 in 1989, 3 major releases before Office for Windows 95. Essentially same for Word.
WordPerfect had a perfectly working GUI version for Windows by 1994, it just couldn't compete with Word. Same for Lotus123, they were out of fashion long before Windows 95.
You have the timeline all messed up.
Excel and Word both ran on Windows 2.0 (released Dec 1987). I supported Excel on Windows 2 in production.
In the late 1980s, yes, Microsoft and IBM told the industry that OS/2 was the future, because they both thought it was.
OS/2 1.0 was released the same month as Win 2.0) had no GUI, because it wasn't ready.
OS/2 1.1 had in, released nearly a year later in Nov '88. This was what Lotus and others needed. 1.1 was not good and barely worked. For example, OS/2 1.1 still had the Windows 2 MS-DOS executive style user interface.
OS/2 1.2 was nearly another year: Oct '89.
Now, it worked, it included advanced features such as the HPFS filesystem, meaning long filename support, it had a revamped and improved user interface, and the performance was OK... but also the 80386 had been out for 2 years and 386 PCs were becoming mainstream. OS/2 1.X was limited to the 286. That meant a maximum of 16 MB of RAM, that meant no ability to multitask DOS applications, and other serious limitations. The high end PC market was visibly accelerating away from the new operating system.
It also had formidable hardware requirements. For example it wanted 8 MB of RAM minimum.
Somewhere around the timeframe of OS/2 1.1 or OS/2 1.2 is when Microsoft realised that this project was running late, it had high hardware requirements, and yet it still couldn't fully exploit the hardware of new PCs shipping in the late 1980s, and it was becoming apparent that the future was not OS/2 after all.
So it's somewhere around the 1988 to 1989 that Microsoft realised that's unless it was willing to sacrifice the PC operating system market, it needed a fallback plan, it needed some kind of alternative: something that could run on high-end mainstream commodity PCs, which means a 386 with 2 to 4 MB of RAM, that offers more compatibility with DOS, DOS device drivers, and DOS applications than OS/2.
That's roughly the time that the school choir project within Microsoft started to turn into Windows 3. Windows 3.0 was released in 1990, it had the new user interface from OS/2 1.2, it could support and use the new facilities of 386 chips… but it could also run in just 1 MB RAM, ran quite well in standard mode in 2 MB of RAM, and ran really well with 4 MB of RAM in 386 enhanced mode with multitasking of DOS applications. In other words, in half the RAM requirement of OS/2 1.2, they could actually do more than that operating system could do, in terms of important things that MS-DOS users actually wanted and needed.
So yes in the very late 1980s, Microsoft did lie to large software vendors, promising them that OS/2 was the future, while it was working on something that would ultimately destroy OS/2. But even now I am no admirer of Microsoft, I have to admit that it was very far from a certainty that Windows 3 was going to be a hit. In the time frame that Microsoft was developing Windows 3, Word for Windows and Excel were already shipping products.
Yes Lotus and WordPerfect believed the lies. Yes it's true, Lotus and WordPerfect could've targeted Windows 2, but Windows 2 was pretty terrible and was a commercial flop. So if they had targeted Windows 2 it would cost a lot of money and the products would never have made it back. Yes they made a mistake, and yes later on Microsoft encouraged them to continue investing in that mistake, knowing that there was a gamble that might make it all a huge waste of money. But what they did probably looked like the sensible prudent decision the time.
And more to the point, relevant to your comment, this all occurred in the late 1980s, long before the industry went 32-bit and long before anybody even dreamed of Windows 4 which would eventually become Windows 95. That stuff happened in the 1990s, when frankly the OS/2 war was already long over.
12 replies →
Lotus was at a crossroads. They could continue with DOS, port to Windows, or port to OS/2. Which path should a cash-rich company with one product pick?
All three.
I have little sympathy for Lotus. They need to own their decisions. Blaming their competitor for misdirection is eye rolling.
You don't have to spread old lies to make microsoft look bad, they do just fine at being terrible on their own.
> stealing associations to bring error messages/dodge the invocation of the competitor software
So... just like what they're doing with browsers these days then
I think they still introduce small problems for people who run software they don't like. For example, if you run Linux on your desktop, and want to use a Windows VM for work, you can't run WSL2 or Hyper-V in the KVM/Qemu guest, even if nested virtualization is enabled and working with other operating systems. It used to work, but now it doesn't, and no know knows exactly why.
These situations are murky. I don't know that they broke it deliberately. Maybe it broke on its own and the problem affects so few people that they just don't care about it.
And you can't really demand that they spend their developer resources on things that they think won't help them. But on one hand, they wrote their own Wayland server for Windows 11, but on the other hand, they say, we can't make Teams work on Wayland.
They do lots of odd stuff. It's pretty easy, and pleasant, to run a headless linux server on Hyper-V, but setting up a proper desktop system on your own is hard. They don't actually create a wall you can't get around, but they create obstacles that make it easier to do the things they want you to do.
It's not fair or reasonable to get mad at them about this stuff. It's more that we should be clear about where they're coming from.
If it helps, I'm currently running a Windows kvm guest that runs WSL2, but I had to change my processor type.
Ubuntu 22.04 Host, virt-manager/ qemu/ kvm hypervisor Win10x64 Guest, CPU model is set to "Broadwell-noTSX-IBRS"
It's definitely not as performant with this setting, but it gets the job done so I can run WSL2 and Docker Desktop in my Windows vm.
YMMV.
Thanks, I'll definitely give that a try.
A while back, my Ubuntu LTS system kept badgering me to upgrade to a new Ubuntu. I finally relented, and installed the latest. It installed without any errors, but the display was all messed up.
After much fiddling and googling about, I deduced that apparently didn't support the graphics card anymore. It never said "don't install me, cuz it won't work with the card I used to work with." It just died on it. I wound up reinstalling the older Ubuntu.
I don't believe there was any conspiracy by the Ubuntu developers. No conspiracy is necessary - it's really hard to not break things. I knew some people who worked in appcompat at MS, and they had a lot of stories to tell about how hard they had to work to not break things. A lot of software misused Windows APIs making it hard for MS to improve things.
> I think they still introduce small problems for people who run software they don't like.
This is completely unfounded speculation.
How old is the graphics card? My work machine has a GPU that's quite niche and about 14 years old, still rock solid at 4k etc etc
> It's not fair or reasonable to get mad at them about this stuff. It's more that we should be clear about where they're coming from.
You're describing a wild animal: "It's not fair to get mad at the wolf for biting you - that is its nature"
Perhaps then we should also treat them as the wild, hostile animal they are.
Linux, FreeBSD and Apple OSX operating systems routinely break without warning older software. I've had the fewest breakages with Windows operating systems.
Just in the last few months, I had to buy a new iphone because Waze, Twitter, and my banking app simply ceased working on it. No warning, no nothing. Those apps upgraded themselves, and just stopped. No check for compatibility with an older iphone.
Bought a new iphone, transferred the apps over, and they magically started working again.
1 reply →
I remember somebody at Microsoft arguing that AARD had a sort of legitimate purpose -- Windows reached deep into DOS in ways that in modern times would be anathema, and did stuff that one couldn't expect to necessarily work properly on a DOS clone, since it wasn't part of the public API.
Still, that idea could have been better executed.
Aha, here:
https://learn.microsoft.com/en-us/archive/blogs/larryosterma...
> As a simple example, when Windows started up, it increased the size of MS-DOS’s internal file table (the SFT, that’s the table that was created by the FILES= line in config.sys). It did that to allow more than 20 files to be opened on the windows system (a highly desirable goal for a multi-tasking operating system). But it did that by using an undocumented API call, which returned a pointer to a set of “interesting” pointers in MS-DOS.
[...]
> But the problem was that the MS-DOS 4.0 SFT was 2 bytes larger than the MS-DOS 3.1 SFT. In order to get Windows to work, I had to change the DOS loader to detect when win.com was being loaded, and if it was being loaded, I looked at the code at an offset relative to the base code segment, and if it was a “MOV” instruction, and the amount being moved was the old size of the SFT, I patched the instruction in memory to reflect the new size of the SFT! Yup, MS-DOS 4.0 patched the running windows binary to make sure Windows would still continue to work.
So yeah, by modern standards the way old DOS software worked was all kinds of screwy.
It was later revealed exactly what the purpose was, when internal mail discussions came out in court.
"What the guy is supposed to do is feel uncomfortable, and when he has bugs, suspect that the problem is dr-dos and then go out to buy ms-dos. or decide to not take the risk for the other machines he has to buy for in the office."
-- Brad Silverberg, Microsoft, 1991, as quoted at https://www.theregister.com/1999/11/05/how_ms_played_the_inc... and in many other reports of the court case that one can still find.
That was the purpose.
Would it be legal to create similar incompatibilities if there is both a legal business justification (that particular solution can be implemented in X developer time and would cost us Y money which wins against a proper implementation) and an engineering justification (that particular solution would use X less bytes of disk/RAM and would run better on Y% of our customer's machines) for an implementation which heavily depends on undocumented internals of another product?
Or "Gaslighting" in modern idiom.
3 replies →
Interesting! I stand corrected, though I think the MS explanation does sound plausible for the times.
Can this "Windows poked around in the guts of DOS" and "DOS 4.0 patched Windows" sort of mechanic be verified elsewhere?
6 replies →
Microsoft has done many questionable things but the AARD code was never enabled on a shipping version of DOS so it doesn't seem quite fair to blame them for that.
I also find the whole argument silly because it was obvious even during the Windows 3.0 days that Windows was going to be the whole OS and no one would be buying two operating systems, one to use as a bootloader and compatibility layer for legacy apps and the other to be the GUI OS.
The whole idea never made any sense and all the sturm & drang about Dr DOS proposes an impossible future where a niche for a paid text-mode PC DOS OS was going to continue existing period. It wasn't and it didn't. It's like complaining about Microsoft hijacking the market for TCP/IP drivers. Even if the conspiracy theory was true it was never going to matter.
> Microsoft has done many questionable things but the AARD code was never enabled on a shipping version of DOS so it doesn't seem quite fair to blame them for that.
Here’s how a Microsoft executive described it at the time: “What the guy is supposed to do is feel uncomfortable, and when he has bugs, suspect that the problem is DR-DOS and then go out to buy MS-DOS"
https://www.wired.com/1999/05/caldera-ms-cheated-in-dos-war/
It’s certainly true that DOS’s time was limited but people were still buying it for years after he said that in 1992. Far more importantly, however, you have to think about the situation with OEMs. Microsoft was absolutely cutthroat on licensing with PC manufacturers to prevent alternative operating systems – entire product lines had to be Microsoft-only, they demanded DOS & later Windows licenses be purchased even if the buyer didn’t want them, volume discounts depended on exclusivity or bundling other apps, etc. The hardware companies put up with that because they didn’t have much choice: there were too many business apps and games which their customers wanted to refuse Microsoft’s terms. DR-DOS offered a way out of that: everything ran, often better due to its superior memory management, and the last thing Microsoft wanted was someone doing something like selling systems with, say, OS/2 or GeoWorks Ensemble (a quite decent desktop app suite at the time) and relying on DR-DOS as the operating system choice for customers who wanted DOS compatibility to avoid Microsoft’s exclusivity requirements for that particular model. I don’t know that this would have fared better than, say, System 76 is doing now but they were keenly aware of just how much leverage they’d lose if an alternative got enough marketshare to let OEMs do more than acquiesce to Microsoft’s standard terms.
I understand the quote and 100% believe that some people within MS wanted it purely to prevent competition. But clearly someone with a cooler head looked at it and made the call not to enable it. That's the only thing that actually matters.
I get what you are saying about Dr DOS and the OEMs but it was a battle already lost - Windows 3.x was a massive blockbuster for MS. I suspect that was part of the internal argument for not shipping the AARD code: it was a liability for a battle already won. Windows 95 started development in 1992 and was intended from the start to be a single OS without requiring a separate DOS.
Not to say Microsoft didn't do bad things. The OEM licensing terms were abusive in my opinion... why not focus on that instead of some AARD code that was never enabled and wouldn't have mattered even if it was?
"DOS isn't done until Lotus won't run" ... not the first time for Microsoft
Provenance of this reviewed at https://news.ycombinator.com/item?id=10432608 )
To Microsoft, by my reckoning, DR DOS was an imitator of MS-DOS. In assessing Microsoft’s reaction to DR DOS, I must think what would be own reaction if I had designed and implemented something, and someone copied my design into a distinct implementation of their own. I would have to accept that copyright law gives me no protection and I might console myself with the proverb about imitation being flattery, but I would think it outrageously unjust if some law would somehow have me owe anything to the imitator.
I'm sorry I have to strongly disagree with this view. MS-DOS was developed in 6 weeks and was a clone of another popular OS of the era called CP/M created in 1974 by Digital Research. The same company that created DR-DOS. If there were laws regulating common and fundamental designs and implementations as we do copyrighted works Microsoft wouldn't exist. Digital Research would have sued Microsoft in '81 and that would have been the end of Microsoft.
I believe you're correct about QDOS (which was bought to become MS-DOS) being a quick knockoff of DR's CP/M.
So I didn't understand the author's comment about DR-DOS being the imitator, at least not without acknowledging the prior history.
The section "Common Ancestry" acknowledges the prior history. Though it feels a bit odd. "MS-DOS is there said to be variously a copy or clone of an operating system named CP/M" seems to gloss over the details by sticking specifically to Caldera's Consolidated Statement of Facts, and not mentioning QDOS/86-DOS.
But the author really assert that "DOS" is a very specific term:
> To me, DOS meant MS-DOS and its licensed variants such as PC DOS. DR DOS was no more a brand of these than juices pressed from soy beans, almonds or cashews are brands of milk. This is not to say they have no merit—some of those juices are more enjoyable than is most American milk—but they’re not different brands of the same thing. To me, DR DOS could not even be a candidate for consideration as another brand if it would always, as seemed likely, need special accommodation in anyone’s DOS programming.
so the idea of a "DOS market" doesn't make sense.
Further, the author probably places you into this camp:
> The preceding question seems to have got conflicted in the early 1990s—and may be still in the apparently growing community of retro-computing enthusiasts, some of whom are vigorously revising history, at least as recorded by Wikipedia, to favour DR DOS. The conflict I mean is that celebration of DR DOS as a compatible alternative to MS-DOS for running MS-DOS programs often coincided with requests that MS-DOS programmers should keep in mind what different writing may be needed for DR DOS.
2 replies →
I remember testing a protocol called DDE back in the Windows 3.1 days, and we couldn't figure out what Excel behaved so much faster than Lotus 123. Then we dug into the code and found that Windows 3.1 would free up all available memory as it launched Excel. A sneaky trick which I guess meant that users would find Excel much more responsive that 123 and ultimately help Microsoft get Office domination that it has today.
Someone else in this thread who worked on Wine said there was no evidence for such a thing since the faster launch times also occurred in Wine. What forensics did you use to determine this?
I don't believe the context of the Wine related comment was Win 3.1. OpenOffice wasn't around to compare launch times. Internet Explorer was around, a bit, but was competing with Netscape, not Mozilla.
The idea of introducing bugs to users that doesn’t conform 100% to intended use of your software is not good idea, what will happen is that your software will be perceived as buggy in general.
> what will happen is that your software
Not your software, a competitor's software. The intent here was to inject bugs to make DRDOS look bad. And it worked.
Later Microsoft would have to spend many long years dealing with how many extension points they'd put in Windows which were opportunities for third parties to inject crashes accidentally into Microsoft executables. That's why there's an entire subsystem to allow video drivers to crash and restart.
You are literally commenting under an article of a company having a 2billion market cap, built also from such underhanded tactics.
While it's immoral, it doesn't seem to produce the result that you hope it's be producing.
Microsoft's software is definitely perceived as buggy (And I never said anything about revenue).
Those users would have been running DR-DOS, meaning they would assume THAT software is buggy.
Which was the entire point and intention, as we saw in an email chain that came out during court.
Most people will tolerate the bugs until they absolutely cannot use whatever software they're running though.
Depends on whether it's common or not. If a smaller competitor has bugs, the default is something else, and folks will go yo the default and look askance at the ones sticking with the odd duck software. Source: Linux user since the 90s, LibreOffice/OpenOffice/StarOffice since early '00s.
When they reach that unusability limit, most people will call their IT department and have the offending buggy software reinstalled as a fix. IT won’t remove your buggy copy of Microsoft Word and Teams and replace it with WordPerfect and Slack.
3 replies →
[dead]
In case you're wondering what AARD means (I did): "The name was derived from Microsoft programmer Aaron R. Reynolds (1955–2008), who used "AARD" to sign his work; "AARD" was found in the machine code of the installer" (from Wikipedia).
If the QDOS/86DOS angle interests you check out a talk that Tim Paterson, of Seattle Computer Products and later Microsoft, gave at VCF West 2019. He describes the history of 86-DOS (licensed to become PC-DOS and MS-DOS). API compatibility with CP/M, was the intention but no code was taken from CP/M.
CP/M was an Intel 8080-based OS. The 8086 was mostly source compatible (but not binary compatible) with the 8080 (i.e. 8080 assembler programs could be assembled targeting the 8086 with only minor changes) and QDOS/86DOS supporting CP/M APIs was to help it fit into this re-assembly workflow.
https://www.youtube.com/watch?v=R2Qh0O3Dt10
Memories of seeing my Dad’s boxed copy of “Doctor DOS” sitting on the bookshelf next to his 286 PC are forever burned into my brain. Using DR DOS and XTree on that computer was my first taste of PC usage. Prior to that I’d only been allowed to use the family Atari 800. Years later, when Wolf3D hit the scene, I finally managed to convince him to let me upgrade it to MS-DOS 6, install a VGA card, and upgrade to 4MB of RAM.
Has anybody implemented this in the open source world ?
It's far enough in the future, it would be good to see a reproducable demo of Win95 on DrDOS (or even freedos) - it seems like something someone on the virtuallyfun site might do eventually.
Apparently there was some work inside Caldera to do exactly that.
https://news.ycombinator.com/item?id=28358828
Wow, DR DOS. That's a memory I forgot I had.
MEMMAX. XDIR. XDEL. DISKMAX.
Getting flashbacks yet? (-:
I don't remember the first and last but the middle two is vaguely familiar, reminded me of XCOPY.
Remember ARJ?
1 reply →
This article is about the AARD code, which was present in a beta release of Windows but was never shipped.
Actually it was shipped. “Microsoft disabled the AARD code for the final release of Windows 3.1, but did not remove it, so that it could have become reactivated later by the change of a single byte in an installed system.” [1]
[1] Schulman, Andrew; Brown, Ralf D.; Maxey, David; Michels, Raymond J.; Kyle, Jim (1994) [November 1993]. Undocumented DOS: A programmer's guide to reserved MS-DOS functions and data structures - expanded to include MS-DOS 6, Novell DOS and Windows 3.1 (2 ed.). Addison Wesley. ISBN 0-201-63287-X.
Is it shipped if it cannot execute under any conditions? Philosophical question I'm afraid.
6 replies →
As documented in the excellent book "Undocumented DOS", whose red cover I can still picture. (Noted in the article, but it was a good book that I still recall fondly.)
I can touch my copy without moving from the chair that I am sitting in. (-:
2 replies →
Thanks for the recommendation! Just bought a copy off eBay for $4 for my collection.
However it was present in the release shipped to reviewers, who wrote their reviews which including saying how windows worked better on ms-dos
IIRC, the author of "Undocumented DOS" discovered and disclosed this feature during the beta, before release.
It seems at least plausible, if not probable, that Microsoft disabled the xor encrypted bogus error message generating code in the release version as a result of this disclosure.
> In assessing Microsoft’s reaction to DR DOS, I must think what would be own reaction if I had designed and implemented something, and someone copied my design into a distinct implementation of their own.
"MS-DOS was a renamed form of 86-DOS – owned by Seattle Computer Products, written by Tim Paterson." - https://en.wikipedia.org/wiki/MS-DOS?useskin=vector
I mean yeah, but it’s not like they just copied it from him.
> Microsoft, which needed an operating system for the IBM Personal Computer,[8][9] hired Tim Paterson in May 1981 and bought 86-DOS 1.10 for US$75,000 in July of the same year. Microsoft kept the version number, but renamed it MS-DOS.