Microsoft Has It All—Almost (1985)
nytimes.com
nytimes.com
The really young techies won't remember this since it was over 30 years ago but what Microsoft did during the 1980s in the software space was astounding.
The 1970s Microsoft was originally a "programming languages" company. They sold MS BASIC and MS COBOL.
In the 1980s, they added productivity apps like word processors and spreadsheets (Word and Excel). What's amazing is that they didn't acquire those from a competitor (like the later acquisitions of Spyglass for IE or Vermeer for MS Frontpage). Word & Excel were homegrown at Microsoft and Bill Gates hired Charles Simonyi to lead their development.
They added operating systems like MS-DOS and later MS Windows & Window NT. MS-DOS was an acquisition but Windows NT was homegrown by hiring Dave Cutler.
Nobody else replicated Microsoft's successful diversity in software. That included Lotus, Ashton-Tate, Computer Associates, Adobe, etc.
That's 3 very different areas of software : programming + MS Office + operating systems. All 3 were huge profit machines.
In the 1990s, they continued success with server-side software like SQL Server and MS Exchange.
To give a modern analogy, imagine Jetbrains just selling programming IDEs like IntelliJ and PHPStorm. And then Samsung comes along and says, "we need an operating system for the millions of smartphones we're going to sell" so Jetbrains hires Andy Ruben[1] to write the Android os for Samsung. And then Jetbrains decides that Android also needs a picture sharing app and hires Kevin Systrom[2] to develop Instagram. That absurd combination from one company is what Microsoft did in the 1980s and in that alternate universe, the Jetbrains owners would be multi-billionaires.
Compare that to Google where it's almost 20 years old and its major revenue is still AdWords. Diversifying revenue is not easy.
A comparable (and possibly more astounding diversification) in modern times would be Amazon's origins from selling books & CDs to selling AWS computation resources.
[1] https://en.wikipedia.org/wiki/Android_(operating_system)#His...
I'm impressed with he way the succeeded with different approaches: plenty of homegrown applications, some acquisitions, some joint ventures. Sure, there were plenty of projects that didn't work out, but the ones that did work out did really, really well.
MS-DOS was before MS Word and Excel. It took a long time for Word or Excel to really gain ground against the market leaders, and arguably they weren't really dominant forces until they were packaged together as Office in the 90s.
Windows NT may have started development in the late 80s but it certainly wasn't released until the 90s and took a while to find success.
You also miss out on the main reason Microsoft was able to become a dominant force in so many fields; MS-DOS. DOS was the PC operating system from the 80s into the 90s, until Windows 95 finally took over, and gave Microsoft the money and position to do whatever it wanted.
None of the rest of its software suite would have become dominant without being built on top of DOS and later Windows.
Way back in my angrier days when I would have spelled Microsoft as Micro$oft The Evil Empire, I would have totally agreed with you.
I've since revised my opinion. The reason is counterexamples of other wildly software that didn't need operating systems to be successful like Intuit Quicken, Photoshop, Duke Nukem games, etc. If others can sell millions of copies without an os to run on MS-DOS, Microsoft itself can too. Also MS-DOS & Windows dominance doesn't guarantee success because MS Money lost to Intuit's Quicken.
For example, with MS Word, was it successful because of Windows or was it better than WordPerfect? Sure, Windows is a part of it but MS Word was actually easier to use and crashed less often the WordPerfect-for-Windows. Wordperfect-DOS wasn't WYSIWIG. Wordperfect-DOS required memorization of function keys F01 to F12 to bold and italicize. MS Word early versions[1] had obvious buttons for bold and italics.
Now that Microsoft is weakened by iPhone/Android and Amazon AWS, I think I'm able to look back at MS in a more detached perspective. I now think they did amazing work in the 1980s and 1990s which got later overshadowed by the ugly "browser wars" and antitrust lawsuit.
[1] https://tmft.files.wordpress.com/2014/09/krnl386_004.png
I'm not saying that the Office products were bad, nobody was going to buy software that totally didn't work, and I'm not trying to argue the relative merits of Wordstar or WordPerfect or Lotus 123 or whatever, but if you don't think the "home field advantage" of DOS and Windows is a big part of the success of Office, then I think you're in denial.
I remember that MS kicked off an office suite price war and that Borland and Lotus matched the low prices with their suites. MS Office still outsold Borland and Lotus when prices were the same. Borland Quattro & Approach and Lotus Ami Pro word processors were not enough of a differentiator.
MS Word and Excel were also the biggest sellers on Apple Mac and obviously, there was no Windows os there.
For laypeople, that may be fine. MS Word was atrociously bad for high-volume typing. Had family in transcription, and they could fly with WordPerfect, doing > 100 WPM. This was a combination of having external tools ("plugins", effectively, to do word expansion with macros), WP being DOS-based (faster screen UI/etc), and, to some measure, experience.
One family member got switched to MS Word on Windows. Well.. the whole department did. They suffered through an immediate drop to about 60% of previous productivity, and it took a couple of years and severe hardware upgrades to get most of the people back to around 80% of previous productivity levels.
Why switch? MS was bundling the whole offering and tempting the company to upgrade. The management of an 'integrated stack' was seen as attractive, and the massive drop off in dept-level productivity was accepted. Management still bitched that stuff was late, and quality dropped, but... hey, network admins have a bit less work to do, and some other dept saved a few bucks.
The bundling/integration of stuff on to the OS was a big selling point, even when it led to demonstrably poorer results for some parties involved. Cool! - there's a menu with a button to do some funky text manipulation. That only takes 8 seconds instead of "having" to memorize some arcane keystrokes. When the keystrokes are memorized, though, it takes .4 seconds to accomplish the same thing.
Totally agree. That's one of the reasons why law firms and their typists hung on to Wordperfect-DOS for a decade even after MS Word was mainstream. Very long legal documents that would choke MS Word worked fine in WordPerfect DOS.
When Windows 95 first came out, it crashed daily on my computer and friends of mine had the same experience. Sometimes it crashed multiple times a day.
I was young then and thought nothing of it. I just rebooted and played whatever game had been interrupted. But now I wonder how they got away with it.
Microsoft was excellent at marketing and when they achieved their monopoly position, they did an excellent job of shutting out competitors. By the late nineties the phrase "No one has ever been fired for buying Microsoft products" was a very popular phrase. But I wouldn't say I was wowed by their work. Windows 95, COM, MFC, etc... I don't see (or read) anyone thinking back on those days fondly.
It was perfectly possible to run Windows 95 without it crashing. I ran it for around a year before the launch (the Chicago beta) and it was very stable by the end IF you had decent hardware and IF you set it up correctly.
But the PC hardware industry wasn't very well developed in 1995, and there was a huge variation in products that took a while to shake out.
Bear in mind that Windows 95 was running on tens of millions of PCs that had never been designed to run it, and that had never been tested with it. A lot of the ones in the field were almost unique (because of variations in processors, RAM, graphics cards etc), and Microsoft had never seen many of them.
And they got away with it because people could get a PC with a good GUI and masses of compatible software for a very low price.
This is the second time recently I've seen someone on HN refer to this expression with a Microsoft branding.
The late nineties are a little too early for me to have had any awareness of technology politics. I never heard the expression you describe.
I did hear, many times, "nobody ever got fired for buying IBM". It seems odd that that would have been a later form of the expression than "nobody ever got fired for buying Microsoft".
The saying was based on the fact that IBM, like the 7th Cavalry, would ride in to rescue large accounts if they screwed up.
I never heard it said of Microsoft in the 90s because Microsoft was never a big proportion of anybody's spending. It sold to hundreds of millions of consumers and small companies, rather than Fortune 500 companies.
I did once discuss this with one of the Microsoft salesmen who handled giant corporations, and who had been hired from IBM. He said IBM tried to get 33% of a company's IT sending whereas Microsoft tried to get 3%.
Microsoft's original DOS spreadsheet was called Multiplan.
I think that's just that the market has changed. Google has programming languages, operating systems and office suites, and they're all more complicated than the offerings of 30 years ago, but given away for free (with advertising).
Here's one way to squint: they'd have to pay a lot more if they didn't have Android.
There's no direct value in any of those. Google gets very little direct revenue from the things you list nor could they ever hope to transform any of them a significant revenue source in the event their primary revenue stream (ads) went into a downturn.
In my view, Chrome and Android are part of a defensive strategy. Imagine a Google without Chrome or Android. Microsoft and Apple (or someone else) would own the devices that users use to access Google services. They could easily choke Google off if they wanted to. Remember how Apple decided Google Voice should be a web app? As much as people say Google is vulnerable, it would be much more vulnerable. Chrome and Android are worth every penny Google spends on them even if they make zero revenue. Plus, with Chrome and Android, Google not only gets a seat in the table for discussions on how to steer the web forward, Google pretty much drives the discussions on how to steer the web forward. For all the talk about Microsoft et al. using sponsorship money to influence FOSS projects like Maria DB, it is far more effective to do things in house (Google Chrome) as opposite to paying an outsider (Mozilla Firefox).
They don’t have to do that for customers and platforms that use chrome by default.
I found Robert Gaskin's fascinating book on the history of PowerPoint and its acquisition by Microsoft in an HN comment a few weeks back and it's a fascinating read of that fascinating paradigm shift in productivity apps that happened in the mid-to-late 1980s (that Microsoft was able to play a hot hand in):
http://www.robertgaskins.com/powerpoint-history/sweating-bul...
That, with the exception that Instagram is more of a "cool idea that caught on with a run-of-the-mill implementation" and had the luck of getting acquired else they'd still be looking for a revenue source, whereas technically what MS did was far more impressive (and with actual paying customers).
Google is poorly diversified and even Amazon is not as well diversified as one might think.
I wonder what falls in the "other" bucket. Visual Studio / MSDN come to mind. Perhaps Enterprise Services is split out from Azure & Windows Server. It looks like Dynamics might be reported separately from Office although I'd normally expect it to be grouped with it instead.
- SQL Server
- Video games (Halo, Minecraft...)
- Hardware like mice, keyboards, hololens, ...
- Probably many others
The 18% other should be Windows Phone but it isn't.
I think Amazon's apparent lack of diversification is actually a lack of granularity in the data presented. The real question is how diversified Amazon Products is.
It's worth mentioning that Windows NT was their third try at a high-end OS, and it was the only one that was entirely homegrown.
First was Xenix, which was a collaboration with SCO and based on Unix code licensed from AT&T. Then, the government gave AT&T the go-ahead to commercialize Unix themselves, so Microsoft panicked and abandoned Xenix, eventually giving SCO full ownership.
To replace Xenix, Microsoft partnered with IBM to create OS/2. Ultimately, Microsoft and IBM fell out, and IBM ended up with ownership of OS/2, so that didn't work out.
But DEC was going through a bunch of turmoil at the time, so Microsoft took advantage of it and poached VMS creator Dave Cutler and had him create NT.
It's amazing how things worked out.
(just something I heard a couple decades ago, don't sue me)
They sent Lotus 1-2-3 and Visicalc packing. They destroyed WordStar and WordPerfect. Lotus Notes, Groupwise? Gone. They destroyed Novell. IBM couldn't even sell software for DOS/Windows. Borland? History. Netscape? See ya.
The only people safe from Microsoft's wrath were companies like Adobe.
Tower Records destroyed a lot of cool record stores.
Wall Mart destroyed a lot of neighborhood grocery stores. Lidl and Aldi did the same in Europe.
Barnes & Noble did the same for charming bookstores.
My point: it is not the winner that is ruthless, it is the game.
But in the case at hand it was a bit more than ‘the game’. MS broke the law multiple times and were at minimum deeply unethical in several other cases, and were held to account for it. Not enough in my opinion, but it was there.
Ultimately though they also made the right moves at the right times and their competitors didn’t.
Could you expand on that?
I'm not a lawyer, but my understanding is that it's not against the law to be a monopoly, but it is against the law to abuse one's monopoly position.
What I'm hazy on is whether antitrust enforcement defendants are informed by the government that they're a monopoly prior to being charged with abusing said monopoly.
Is that the case, or no?
Did Microsoft break any laws that it wouldn't have broken had the government not decided that they were a monopoly?
It is how monopolies, fueled and protected by copyright and patent laws, who themselves are breaking the laws (anti competitive), work. You see, being a monopoly isn't against the law but abusing your overly dominant market position is.
> Lidl and Aldi did the same in Europe.
(There's no such thing as Aldi. There's Aldi Nord and Aldi Sud [0])
Hi, I'm from Europe. Aldi has 431 stores in The Netherlands, Albert Heijn 947. And that's probably nothing compared to what Superunie owns. [1] UK is a similar story. Lidl and Aldi are nowhere near the market share of Co-op or Tesco. [2] So... what did Lidl and Aldi do in Europe? Lidl or Aldi are nothing even close to a monopolist. Wall Mart is huge, but has healthy competition as well. [3]
Microsoft had virtually no competition in the desktop OS market from mid to late 90s until recent due to tablets and ChromeOS finally becoming a competitor. That actively stiffened the innovation on desktop in other related areas such as the browser market where everyone had to run dirty hacks in HTML and CSS to adhere to the defacto standard called MSIE. It was a horrible cesspit where the only browser you could reliably use only ran on one OS (Windows) and was also terribly insecure. And hey, it used to run on different OSes but Microsoft decided to quit those ports. The port they maintained the longest was for that 1% market share company "we're not a monopoly excuse cause Apple".
We didn't mention Intel yet. They're an entire different subject, another parasite who run in symbiosis with the monster of the 90s and 00s called Microsoft (or Mammon ;) [4]). Microsoft lies dormant now, as the market it had previously in complete control is marginalized by iOS and Android. Thank you, Apple and Google. And we shouldn't forget the wonderful work by KHTML/Konqueror team (which spawned Safari and Chrome), or Mozilla (and formerly Netscape) who initially was the dominant competitor of Microsoft in the browser space.
[0] https://en.wikipedia.org/wiki/Aldi#Geographic_distribution
[1] https://en.wikipedia.org/wiki/List_of_supermarket_chains_in_...
[2] https://en.wikipedia.org/wiki/List_of_supermarket_chains_in_...
[3] https://en.wikipedia.org/wiki/List_of_supermarket_chains_in_...
If someone from the 80s or 90s got a time machine and skipped to today, one of their biggest surprises would be that Bill Gates is considered a beloved philanthropist rather than the personification of corporate evil.
Furthermore WordPerfect people were touch typists, users of function keys. In previous times these people were the typing pool. Now we don't have typing pools, managers and others that would have used secretaries worked out they could do okay by themselves in MS Word with its toolbar buttons to embolden text etc.
Sadly the touch typing WordPerfect crew that had ensured the DOS years didn't really make it to the new GUI ways of working. Their muscle memory and need to keep the fingers on the keyboard kept them at the WordPerfect level of working, much like how you have programmers using 'vim' instead of a modern GUI like 'Notepad', there was no perceived improvement.
I remember how lame the Wordperfect for Windows was, it was not performant and the last thing you need with a word processor is that feeling your computer could BSOD any moment. It was all wrong for the Windows world but, as mentioned, Microsoft had the jump on that, they could even build in features to Wondows to support what would become Office if need be, Wordperfect didn't have that under the hood access.
> If someone from the 80s or 90s got a time machine and skipped to today, one of their biggest surprises would be that Bill Gates is considered a beloved philanthropist rather than the personification of corporate evil.
The problem rather is that people have a short memory. I for myself have not forgotten.
Microsoft used astroturfing campaigns in newspapers and magazines talking up their products and talking down their competitors. These tactics are much more widely known today, primarily because MS.
Microsoft created all kinds of undocumented apis that enabled their products to function much better than the competition. This is illegal for monopolists to do. IBM was forced to document all of their OS APIs because they were a monopoly, so that application developers would be on the same level and able to compete with IBM.
Microsoft changed apis and functionality that somehow always seemed to break their competitors products, but not their own.
Microsoft dominated to such an extent, that they essentially were forced to bail out Apple so that they could have a fig leaf excuse that they weren't actually a monopoly.
The game Microsoft played was one of enclosures, monopoly-building, corporate warfare, bordering on corporate fraticide, and engrossment.
Mind, that's a fairly standard game across a whole slew of economic activities, within which I'm looking for (and seem to be finding) some common grounds. Software, operating systems, hardware platforms (chipsets), communications networks, broadcasting, print media, copyright cartels, transportation (road, rail, sea, canal, air), and the finance, insurance, and real estate sectors all seem to share that certain je ne sais quois.
It seems fundamentally dendritic.
Oh wait, that's not what happened. Maybe some selection bias is going on?
I think the selection bias is 1980s companies because that was the context. (Linux is 1991, Java 1995, Google 1998, etc)
From the 1980s, the major software companies that's still alive today, you have:
Intuit (Quicken/TurboTax),
Adobe,
Computer Associates,
Autodesk
Adobe and Autodesk survived because Microsoft never entered the professional graphics editing market of Photoshop and CAD drafting. If Microsoft did try to compete, they still might have lost to Adobe and Autodesk but it sure helps when Microsoft ignores that market. (Microsoft sort of had a competing edocument standard of XPS to Adobe's PDF but that initiative didn't seem to have all of MS's attention behind it.)
Computer Associates (now CA Technologies) was this weird corporation that basically did a bunch of acquisitions of forgettable companies. I can't think of an instance where CA and Microsoft competed head-to-head. CA bought a database programming 4GL (Nantucket Clipper) in 1992 but the DOS based programming languages were already starting to die by then so MS Visual Basic for Windows didn't really compete fiercely with it.
Most of 1980s companies didn't survive, whether they competed with Microsoft or not. Microsoft did, but that doesn't mean it killed every other company that didn't.
But Microsoft competed with Apple, and Apple didn't die. Xenix - Microsoft's UNIX offering - went nowhere (though it wasn't even MS product, technically), and other Unixes still alive. In DOS times, there were pretty successful competitors to MS-DOS (of course, those died when DOS computing died, but I'm not sure this qualifies as direct competition).
It was a scavenger that extracted the remaining value from failing and failed companies.
I'm not saying it was bad for the ecosystem ;-)
>> MSIE? Dominates the world.
IE did dominate the world. Remember Netscape? IE was at like 90% in the early 2000's until Firefox showed up.
>> Java? Completely unknown
Never made much of an impact on the desktop. What's the most widely used Java desktop app? Was there ever one? Eclipse/Netbeans comes to mind.
Ironic, because these are just Java IDEs.
For a brief period. So did many other short-term phenomena, from BASIC to MS Access to Clipper to Pacman. And now they are stuff for the museums. Except for Pacman, of course.
> Never made much of an impact on the desktop
So? It still survived competition with Microsoft - and, arguably, won on every platform except Windows, non-Windows .net use is still basically zero and Linux rules server world, so...
> What's the most widely used Java desktop app?
I honestly have no idea. Well, Eclipse of course, IntelliJ suite of tools, probably, but it could be some other apps are Java too and I just don't know it. The point is Java is alive and well despite competition from mighty invincible Microsoft.
Lotus notes: some good ideas but horrible outdated app
wordperfect/wordstar: not even close to matching MS Word's feature or ease of use
Novell: what did they make other than networking that ended up being a standard part of all machines and all OSes, not just windows
Lotus 123/Visicalc: did Visicalc ever make it out of text mode? Excel beat those be being a better product.
Borland: also got out competed. Their only product that didn't suck was they're dev products but they couldn't keep up with Visual Studio. They had one database product that took them year to try to update to be competitive and lost their market by taking too long.
None of that has anything to do with "wrath" of MS.
The 'wrath' of any company is one juggernaut leader at the top. Apple's 'wrath' was Steve Jobs and Microsoft's 'wrath' was/is Bill Gates.
As much as we like to state it otherwise, Bill Gates was ruthless at execution, and he was workaholic with no parallel. People aren't exactly replaceable, and by that definition what they bring to the table.
Totally:
"As it happened, Gates's mother sat on the same charity board as IBM's chairman. She got IBM to talk to Gates with the promise of an OS for their underwhelming PC (which was nothing two kids in a garage couldn't have come up with – in fact, it perpetrated many of the flaws of the Apple II, which for its flaws was still very innovative for the time). Gates didn't have an OS, so sent IBM to talk to Gary Kildall of CP/M fame, but Kildall was not prepared to talk to IBM, so IBM went back to Gates. Gates was friends with the chairman of Seattle Computer Systems, who he convinced to sell him the OS for $50,000 – that bit is accurate. Gates did bluff IBM, but he then bought QDOS for Quick and Dirty Operating System from SCS." [1]
Rather, what people should realize is that back in 80s it was IBM who was the monopolist, and it was Microsoft who became the monopolist IBM once was. The smartphone and mobile markets [2] have smashed up the tech sector. Nokia, once a giant, is no more, and Microsoft is far less relevant because they missed the boat.
Regarding Windows NT, its interesting to note who Dave Cutler [3] was (one of the designers of DEC's OpenVMS, and a UNIX hater). Windows NT put a dent in the UNIX server market because it was a cheap alternative running on commodity hardware (the happy Wintel marriage), but in the end it was *BSD and Linux (or GNU/Linux) which replaced UNIX and partly Windows NT as Linux dominates the server market.
> [...] so Jetbrains hires Andy Ruben [...]
Andy Rubin.
[1] http://homepages.ihug.com.au/~ijoyner/Ian_Joyner/Pirates.htm...
[2] No source here, I'd just like to note its Apple iPod which created a cash cow for Apple. Its what caused the momentum (and cash) for Apple to start the iPhone.
They didn't miss the boat, they got on a much earlier boat that wasn't quite going in the right direction. Some better touch screen controls and MS would have had the iPhone a decade earlier.
> Rather, what people should realize is that back in 80s it was IBM who was the monopolist, and it was Microsoft who became the monopolist IBM once was.
This is true, but Microsoft only became the IBM of PC software. Intel became the IBM of processors. Cisco became the IBM of routers. And so on....
IBM was the IBM of everything, and at one point, was twice as big as every other IT company put together. Today's youngsters have absolutely no idea how powerful it was. (And after decades of failure, IBM is still almost a $90 billion company.) Microsoft never had that degree of power.
In the end, IBM's monopoly was split between a lot of different but less powerful companies, each with its own niche. Today, I think, only IBM's mainframe monopoly is left....
I seem to recall that MS Word at least drew on an existing wordprocessing system, though I can't pin that down.
I believe there may have been a word processor out of PARC itself that went on to become a product elsewhere. I don't know if that became or directly fed into MS Word or not.
And I'm really hazy on this, so don't take me too seriously.
That's like saying Apple's major revenue is still people paying cash in exchange for products — ads isn't a product, but a revenue model, an exchange currency that apparently people are happy to pay instead of money.
Btw, Google has 2 really successful operating systems, Android and Chrome OS. That they don't have much direct revenue from them is inconsequential for the topic at hand, because these operating systems ensure their continued dominance of the marketplace.
Microsoft genius was to control the market with its operating systems. Even if Office and Internet Explorer were good from a technical perspective, having their own merits, I'll always suspect that the tight integration with Windows, along with the deals Microsoft did with computer makers (possible because of MS DOS and Windows) is what killed their competition.
And now Google has taken that to the next level. To not see it is ignorant and unhealthy for the ecosystem. Because yes, Chrome is the new Internet Explorer.
The "ads" (uncapitalized generic word) is not a product but AdWords™ is: https://www.google.com/search?q=adwords&source=lnms&tbm=isch
AdWords is the system that advertisers log into to see their keywords campaigns, dashboards, analytics, pay with their credit-cards, etc. That's been Google/Alphabet's major cash cow.
Google wants other multi-billion profit sources besides AdWords. Google Cloud Platform is growing in revenue but I don't think they've publicly disclosed whether they are making significant profits from it yet.
Their X divisionp[1] is arguably tasked with finding more multi-billion dollar ideas: https://en.wikipedia.org/wiki/X_(company)
Just scroll through that list and it's going to get you amazed and marveled.
To be fair, Google has released number of products in many of the same areas as Microsoft but usually with different spin including Office, Sharepoint, SQL Server, Outlook and so on. With about 1/3rd of the size and 1/4th of the age, they have been immensely productive as well.
Between the fact that it was in '85 (long before the days of windows 95/98 and microsoft really "taking over the world"), the revenue total that wouldn't even register as a margin error for today's dominating tech companies, the fact that a deal with IBM meant they had "won it", or that their main competitor is a company so that they utterly destroyed a decade later ... What a blast from the past.
The IBM deal meant that DOS was going to be the Operating System for IBM's Personal Computer and became the de facto OS for all IBM PC clones. That was the deal that made Microsoft and shaped the tech industry for a generation. Indeed they won it.
I could see in today's world how if Watson was picked as Windows' assistant instead of Cortana, news articles would talk about how IBM had "won it" with that project.
Man OS/2 Warp was going to be AWESOME and than the infighting killed the whole thing. :(
In 1984 we also got IMB PCjr and that was IBM's home push and boy that lasted 13 months. IBM was just horrible in consumer computers.
> Man OS/2 Warp was going to be AWESOME and than the infighting killed the whole thing. :(
OS/2 Warp was actually the IBM monopoly's last concerted attempt to kill little upstart Microsoft. One of IBM's senior managers at the time told me "We're going to burn Bill's butt."
By that stage, it wasn't infighting, it was war. (The infighting was Microsoft's attempt to get IBM to accept Windows NT as OS/2 version 3.)
As it happened, Microsoft had a very strong ally in Kirkland called ... IBM. Big Blue's PC division was interested in selling PCs and it was selling lots of PCs running Windows. PCs running OS/2 were virtually unsaleable.
Eventually, Gerstner arrived and told the OS/2 guys they'd lost and they should just stop.
Microsoft and IBM started the development of OS/2 because IBM wouldn't support Unix. (Microsoft had the most widely used Unix of the day, in Xenix.) IBM required a proprietary OS and had a long-running war with AT&T.
Microsoft backed OS/2 strongly but sales were, in Bill Gates's word, "dismal". It was a flop. However, Windows was a hit.
Microsoft wanted to go to a 32-bit OS/2 v3, which is what became Windows NT.
IBM had sold tons of 16-bit IBM PC/AT machines and insisted on a 16-bit operating system.
Hence the divorce, where Microsoft kept Windows and got a bundle of cash ($25 million, if I remember correctly). IBM got OS/2 and an ultra-sweet deal on Windows ($9 a copy, if memory serves).
However, the Unix business was a huge mess at the time. First, the all vendors were at war with one another, then they divided into two warring camps (SVR4 vs OSF).
While they were busing fighting one another, Microsoft nicked their cheese.....
Update: Of course, Windows NT was designed by the same guy as DEC's VAX VMS, so it's got that in its DNA. Maybe he was thinking of servers as well...
Someone should write a book, "100 Decisive Tech Battles: From Ancient Times to the Present," a tongue in cheek reference to https://www.amazon.com/100-Decisive-Battles-Ancient-Present/...
The book is excellent, and analyzes battles with a clinical detachment not really found in most history texts.
But each battle is decisive: it shaped the world. There are many skirmishes that would be interesting to analyze but out of scope for the book. That's what makes it a fascinating collection.
If we had to think of 100 technology "battles" that reshaped the world, I wonder what they would be? There is so much freedom in the criteria that it's hard to know where to constrain it: Electricity, plumbing, grocery stores, etc have all shaped the world. Many had a "decisive" effect in that it was technology vs technology, and one tech came out the winner.
I think computing alone could fill a book of 100 tech battles, and it would be interesting to try. Which stories were decisive? It would take months to decide, but it would be enjoyable work.
For example, VHS won out over Betamax but Sony remained a successful consumer electronics company and I'm not sure things would have played out in a materially different way had Betamax won. (Though maybe the industry would have been a bit more concentrated.)
AC electrical power over DC is probably one with implications. MS-DOS and Windows are certainly examples. As is Linux. x86? Maybe although I'm not sure how different the world would be if you swap in Motorola instead.
> Microsoft is now unleashing a barrage of programs, including ... Access, a communications package for the I.B.M. computer.
This is the 'other' MS Access - the one that predates the database.
Really amazing what he had done by 29. As someone older than that, I'm really starting to feel like I'm hitting the "older tier" of tech ages.
I made a large compilation list of famous founders a while back. You might be surprised at how not particularly young most of them are. I won't repost it here, but here's a link to that thread:
http://www.nytimes.com/1994/06/11/obituaries/mary-gates-64-h...
She had been a regent at the University of Washington since 1975, the same year she became the first woman to serve as a director of First Interstate Bank and the first to serve as the president of the King County's United Way. She was later appointed to the board of the United Way of America; in 1983, she became the first woman to lead it. Right Time, Right Place
Her tenure on the national board's executive committee is believed to have helped Microsoft, based in Seattle, at a crucial time. In 1980, she discussed with John R. Opel, a fellow committee member who was the chairman of the International Business Machines Corporation, the business that I.B.M. was doing with Microsoft.
Mr. Opel, by some accounts, mentioned Mrs. Gates to other I.B.M. executives. A few weeks later, I.B.M. took a chance by hiring Microsoft, then a small software firm, to develop an operating system for its first personal computer.
And this...
http://prospectingprofessor.blogs.com/prospecting_professor/...
Why did IBM even go to Redmond, WA in the first place?
At the time, IBM president John Opel served on United Way's national board...as did Mrs. Gates. In 1980, when someone mentioned Microsoft to John Opel, he responded, "Oh, that's run by Bill Gates, Mary Gates' son." Behind the scenes, Mary Gates had spoken with John Opel about the new breed of small companies in the computer industry, which she felt were under-appreciated competitors of the larger firms with which IBM traditionally partnered.
That's not a mystery: it went to license Microsoft Basic, which was the most widely recognized Basic of its day. It dated back to the MITS Altair.
So IBM would have gone to Redmond regardless of Mary Gates.
However, it probably helped with the DOS deal, whereby Microsoft offered to save IBM's operating system bacon. Basic was a known quantity but MS-DOS didn't even exist.
Very interesting
1980 | $8m sales (40 employees)
1981 | $17m sales
1982 | $24.4m sales (220 employees)
1983 | $55m sales
1984 | $97.4m sales
1985 | $140m sales
1986 | $197m sales
1987 | $345m sales
1988 | $590m sales
1989 | $804m sales
1990 | $1.18b sales | 5,635 employees
1991 | $1.84b sales
1992 | $2.76b sales
1993 | $3.78b sales | $953m profit
1994 | $4.7b sales | $1.1b profit
1995 | $6.1b sales | $1.45b profit
1996 | $8.67b sales | $2.19b profit | 20,561 employees
1997 | $11.4b sales | $3.45b profit
1998 | $14.5b sales | $4.5b profit
1999 | $19.7b sales | $7.78b profit
2000 | $22.9b sales | $9.4b profit
Before microsoft, software was expensive. Many people were programming their own games or software with the language of their computer (a mix of proprietary basic and assembly).
After, everybody had pirated copies of Win3.1 and Office and were coding excel macros :-(
For me, there was an explosion of creativity between 1983 and 1992 where people were empowered to program. After, programming was for "specialists". I have hated a lot Microsoft at that time.
That's an important point that I forgot about until you mentioned it: I bought Microsoft software because it was easy to copy, or more accurately, lacked copy protection. This was in the days of having to shove the original disk in the drive even if you had it copied to the hard drive. It'd be one thing if floppies were more reliable, but $DEITY help you if that master disk went bad, because Lotus would be...less than helpful. So I rewarded the "good guy" by buying their unprotected software. Hey, if they trust me enough to not copy-protect it, the least I can do is pay for a copy.
And when Windows NT went out, I remember big debate about whether Microsoft could do enterprise software, given that all they'd really done to that point was put out applications and dev tools, and a cute little graphical shell that went on top of their cute little toy operating system.
It's writing and editing certainly leave a lot to be desired but from a historical standpoint about Microsoft, it's a fantastic read.
(Paul Andrews worked for the Seattle Times and Microsoft was his beat. I guess he just had more and better contacts.)
Hard Drive did well because it came out first and filled a big hole in the market. However, Gates was sufficiently better to become a best seller.
I guess some things never change.
There are probably more bugs in most non-Microsoft applications, it's just that they don't have enough users to find them....
(And they did have multiple text-mode PC versions, VMS, Unix, Improv for NeXTStep...)
(...and it all worked on a 4.77MHz 8088.)
I loved WP and 123. Begrudgingly switched to MS Office in 97.
I would posit that it was due to WP screwing the pooch at every turn. They held out until they couldn't on releasing a Windows version, and it was as buggy and unreliable as you would expect from a rushed product designed to check a marketing checkbox. WP also bet on OS/2, and they did it by porting what was already a poor product, so the OS/2 version ended up being a poor, buggy product that had the added bonus (due to an interpretation layer) of being dog slow.
So maybe Microsoft's success can be attributed somewhat to bundling, but much of it can also be laid at the feet of competitors like WP and Lotus.
As it turned out, IBM refused to effectively market OS/2 and Microsoft was busy cutting bundling deals with manufacturers while releasing their "munchkins" (look it up, it's well documented) on Compuserve and Usenet. And then Windows NT came out, giving Microsoft a much less laughable operating system in their quiver, and that was pretty much it for OS/2.
You had to vape your PC to run it.
By contrast, you could pick up a copy of Windows for $40 and run it on top of DOS without losing your DOS software. If you didn't like it, it hadn't cost you much time or money.
And there was a boom in Windows software with lots of cheap programs to try.
Seriously, how would anyone expect OS/2 to win that battle?
I remember our Pentium 1 75 MHz being shipped with OS/2 Warp. It was dogshit slow (and on top of that, I was an impatient kid). I was used to DOS (which could be fired up from OS/2 but then it didn't work well, and DOS 6.x was known to work better than that version). I was sitting more in DOS 6.x (booted from a floppy) than OS/2 Warp.
Eventually, we bought a Windows 95 pirated copy for about 50 Dutch guilders (not sure how that translates to current currency due to Euro and inflation). It was a low risk deal indeed for the reason you suggested (cheap) however the real deal was expensive and we already paid for an OS, and we had a contact who would press thousands of these (in hindsight, probably related to Twilight CDs etc as both sourced from the infamous TUe network). So we went with that.
We ended up not using it much either because we were used to DOS and didn't find Windows 95 stable. Not that DOS was super stable, it (6.x) was just more stable.
The first Windows version I've ever used which was reasonably stable, was Windows 2000 (yes, I used NT 3.5x and NT 4.x and Windows 2.0 and Windows 3.1). Arguably, it was even more stable than XP which is probably the first version the general consumer used which was reasonably stable.
Now, how much did a copy of Windows 95 cost back in the days? Well, I'd love to read some other sources but the fact we bought a pirated one for 50 Dutch guilders plus the source I found which says Microsoft suggested a retail price of 210 USD suggest a different price than 40 USD for Windows 95 [1]. Which leads me asking which Windows version you're referring to. No offense, btw.
[1] http://community.seattletimes.nwsource.com/archive/?date=199...
Unfortunately I can't find a list of Windows retail prices, but originally Windows was sold much like an application. That is, it assumed you already had DOS installed.
From what I recall, the official retail was something like $99 and you could pick up a copy for around $80 or so. (Some programs came bundled with a run-time copy of Windows, though you couldn't used that copy with a different application.)
Windows 95 upped the price because Microsoft was charging for both DOS and Windows at the same time. On that basis, something like $200 doesn't sound wrong, and that's around the price Windows has sold at ever since. (Unless, of course, you buy a pre-launch offer.)
Software and hardware prices were generally much higher in the old days, of course.
On stability, I found both NT4 and Windows 2000 to be very stable. With other versions, they were stable if they were correctly set up on decent hardware, you only used Microsoft drivers, you were pretty careful, and you restarted every day.
I could usably run Win98-XP for about week. I would never find out how long I could run NT4 because I was dual-booting it ;-)
Eggs are fragile, but they generally don't break if you handle them carefully. It was much the same with Windows...
You get 2 OS with Windows 9x? Really? Or is it that Windows 9x is just a GUI on top of DOS? At the very least its product tying.
One thing which stood out was how you excused the terrible stability of Windows 9x. For me the fragility of Windows 9x is unacceptable. What's worse is that a lot of people thought this was normal, acceptable behavior for computers. Millions of work hours have been wasted thanks to this instability. Back then Linux desktop had its fair share of issues which were also unacceptable, but instability wasn't one of them.
This is an interesting question...
Windows started out (early-mid 1980's) as a graphical layer on top of DOS (Sold separately). To DOS, Windows added memory management, device independent graphics, windowing, multiple processes, and a cooperative multitasking scheme that let the multiple processes run at the same time.
The next 10 years of history (through to Windows 95, 98, and Me) are then largely about Windows taking over more and more functionality from DOS.... but never quite supplanting it. If a Windows 3.1 system launched Windows at the end of DOS's autoexec.bat, Windows 95 did essentially the same thing, but internally and not in a script. (In fact, Andrew Schulman got Windows 95 running as DOS 7.0.)
So the way to view it is that this lineage of Windows is really a hybrid design, with elements of both DOS and Windows.
The Windows NT kernel and OS/2 >=2.0 are both totally different... to the extent they run DOS, they run it as a child process under a native 32-bit OS. (Not 64-bit because x86-64 doesn't support the V86 mode necessary to make virtual DOS machines run.)
SAA tied together all IBM's enterprise systems (except the Unix ones) and it mandated IBM's version of OS/2 on the client.
Some people felt that IBM had sufficient market power to impose it. I assume Lotus did, too. It turned out that IBM was wrong, and Lotus's strategy sank along with IBM's.
OS/2 was part of a strategy to take back control of the PC industry by shifting the whole thing to an IBM-controlled base with PS/2 systems, MCA buses, and the OS/2 operating system.
The original PC was released in 1981, followed in 1984 by the PC AT. One of the things the PC AT was supposed to do is bring multitasking and larger (>640K) RAM - mostly thanks to the 80286 processor. The trouble that there wasn't a mainstream OS that used all of the 80286's capabilities until OS/2 came out in 87/88. At that point in time, OS/2 was both the solution to some long-term problems in the PC space and the annointed successor to DOS by the IBM/Microsoft pair that led the PC market at the time.
For a whole range of reasons, OS/2 turned out to be expensive to buy/run and incompatible with most of the software people actually ran. The market therefore kept its momentum mostly behind DOS until August 1990's release of Microsoft Windows 3.0. Microsoft had an internal skunk works project that brought 80286-specific features to Windows, but in a way that was more compatible with DOS and cheaper than OS/2. (Despite some technical limitations.) That's when Microsoft took a risk, and pivoted away from IBM and OS/2 and towards a Microsoft Windows specific strategy. It paid off for them and the rest is history.
Plus it all ran together, through OLE and COM, so that spreadsheets could be embedded into documents or presentations. If the recipient didn't have the suite, they'd need to buy it. Viral marketing, 1990s style.
WP did get caught flat-footed on the Windows transition, but everyone, including Microsoft, had been pushing OS/2 up until that point.
True up to a point. From at least the early 90s, Microsoft was trying like crazy to get companies to write Windows software instead ;-)
And it's worth noting that Windows itself didn't really show up big until Windows 3.11 for Workgroups. Win 3.1 Workgroup was released in October, 1992, and the really stable version, 3.11, in November of 1993. That was still being rolled out up to several years later (enterprises are conservative). There was app development for it, yes, but Microsoft seemed to have already established its Office strategy. (Just checked: the initial announcement was in 1988.)
This is somewhat ancient history, and I rather dimly remember it (my primary platforms then were already largely server: Unix and VMS, plus others), but desktop was how I accessed the "real computer".
Borland, Ashton-Tate and Lotus refused. In late 1987, Philippe Kahn said "People are afraid Microsoft is going to take control of the operating system." [1]
Microsoft was afraid that IBM going to take control of the operating system.
Windows 3.0 was enough of a success for Microsoft to intensify it efforts....
Their solution was to go multi-platform, with an all-singing cross-platform super-product that would run everywhere. Except it didn't actually run anywhere; it broke under its own weight.
So they were very much behind when they finally buckled down to porting the 1-2-3 codebase to Windows in 1991.
Source: I was on the team that shipped the Motif and OpenLook versions of 1-2-3, separate but concurrent to the above.
The other thing I should've mentioned is that at the time, OS/2 did look like the surer bet than Windows. The success of Windows surprised everybody (including, I think, Microsoft.)
(I did like MS OS/2 2.0 though...)
This fabulous kludge was driven by the testing regime: if we changed one byte of the spreadsheet engine it would trigger person-years of manual testing.