Ask HN: How did Microsoft hold back the Internet for 6-7 years? | Hacker News Reader

Ask HN: How did Microsoft hold back the Internet for 6-7 years?

44 pointsbiznerd69 comments

I binged (jk googled) and couldn't find anything.

This was in a comment located here: https://news.ycombinator.com/item?id=8673760

And it piqued my interest.

How did this happen? Or did it happen?

The very short, oversimplified version:

Microsoft leveraged their dominance of desktop computing to gain dominance of browser installed base with IE. Having won the battle, they stopped improving the browser as an application platform.

During the first period, Microsoft developed and deployed a variety of browser technologies such as Dynamic HTML (aka the DOM) and XMLHTTPRequest that moved the browser toward being a viable application platform. During the second period, IE stagnated and it was left to Firefox and, later, Chrome to pick up the baton. But they had to fight for market share for several years before having enough influence to make significant progress. IE has recently caught up, but that still leaves several years before enterprise customers will deploy the improvements.

Thus in 2014 there are still major gaps in the browser platform that really should have been solved some years back, and a large portion of installed base still using the transitional IE 8 and 9.

Microsoft tied browser development to OS development. After Windows XP (Oct 2001) they didn't release an OS until Vista (Jan 2007). Things far, far more important to Microsoft's overall strategy than HTML parsing stalled during that gap.

Dynamic HTML is not "aka the DOM", nor is XHMLHttpRequest an application platform.

Microsoft developed function calls and API calls which were only available internally, and for preferred enterprise customers, that would fail on Netscape.

In addition, "IE has recently caught up" is hardly the truth and you can visit HTML5test.com or CSS3test.com to test your browser; or visit caniuse.com to browser around to find large gaps and inconsistencies between IE and any other browser (and not just HTML but the web APIs that work everywhere but IE).

Additionally, many enterprise web applications developed in the early 2000s solely targeted IE and often used ActiveX controls, which made many companies standardize on IE6, and in turn made IE6 support (and dealing with its limitations and bugs) a requirement for most web applications. The result was that even as other browsers started to innovate, it wasn't practical to leverage those advancements in many web applications.

"a large portion of installed base still using the transitional IE 8 and 9."

Looks to be about 3.4%, per http://www.w3schools.com/browsers/browsers_explorer.asp , vs 5.8% for IE10/11

The truth is more shaded than saying they held back the web. What I remember is that IE 5 came out in 1999, and was a significantly better browser than netscape 4, which was my favorite browser at the time. So, netscape had an inferior product. Meanwhile microsoft also decided to bundle IE with windows, for free, on all new PC's. So netscape had a worse product and worse user acquisition. If you're a startup in that position, what is the logical thing to do? Well, netscape decided the most logical thing was to start a multi-year rewrite of their entire product, with the goal of having an identical UI, but a fresh codebase. IMHO, it is fair to say microsoft helped netscape into the grave, but it is also fair to say netscape did much of the walking.

So, from about 2000 until about 2006 IE was the only game in town because there just weren't any viable competitors (well, ok, there was opera, but...). Looking around and noticing they didn't have competition, microsoft figured they didn't need to iterate their product, so they didn't.

Now, this will sound strange to say now, but IE 6 had the best standards support, in 2001. However, it also had a lot of proprietary features which made things easy to do that were hard to do using W3C standards, which as standards tend to be weren't as developer-friendly as they could have been (I still think CSS's layout model is a big mistake). Web developers being web developers they couldn't resist those features to build stuff quicker, and they ended up building a lot of IE-only sites, which created the legacy which we are still battling today. And that made it very hard for upstart browsers like firefox to gain marketshare.

Now, again IMHO, it is fair to say microsoft did nothing to discourage people from using those proprietary features and getting locked into a dead-end platform. However, it is also fair to say you could and can build a standards-compliant codebase which is IE 6 compatible so developers were helping the jailer put on the chains.

I think blaming it all on MS is easy but inaccurate. It was a shared blame across netscape, microsoft and the web development community of the early 2000's, which ended up in a stagnated browser market from 2000 to 2005/2006.

NS betting the farm on JavaScript Style Sheets and a strictly-hierarchical element access tree certainly didn't help matters (and the CSS-to-JSSS translation layer never really worked). Your CSS had to allow for every possible hierarchy in which the target element(s) could be found. Developing for it was largely a matter of figuring out why the hell that one paragraph/bullet point/header/table cell didn't look anything at all like the ones around it, and knowing you weren't going to like the answer when you found it. ("Oh, I see... changing the font colour here overrides the font size I set back there. Makes sense...???")

No, it's pretty accurate. When you gain a monopoly and people stop testing on other browsers, you create a big problem. I never stopped using Netscape/Mozilla. It doesn't matter if IE was better for a period of time. Once it became dominant and incompatible, all other browsers fought a huge uphill battle. IE is still the most widely used desktop browser.

This is unfortunately starting to happen once again with Chrome.

The irony is that many sites now have poor compatibility with IE due to using proprietary webkit-only CSS styles. On mobile it's even worse, with most "mobile optimized" sites being broken even on firefox and opera. Even big sites perform poorly. Engadget's mobile version is broken on anything that's not webkit. It is frustrating to see that people never learn. They're quite happy to change jailers, but quite unwilling to free themselves.

Yeah, it's gotten to the point that some alternative mobile browsers are starting to support -webkit- prefixes.

Up until IE5 I remember being a bit of leapfrogging from both sides. I'd use IE version X, then Netscape version X+1, then IE etc etc. So Netscape definitely faced situations previously where they were no longer the best browser. As you note, what really started killing them was IE finding its way automatically into nearly every single home computer on the market, and the fact that for many people "the internet" and "the blue E" were synonymous (Stories were legion of people who thought their internet was broken because the IE shortcut was removed from their screen).

As someone who hasn't used Windows since right around that time (I was Linux/FreeBSD only for several years until transitioning to OSX) that ushered in a dark area - there were many websites that I simply couldn't use. Their loss, not mine.

Another aspect that is often overlooked is Microsoft's decision to closely couple IE with Windows. On Windows, the installed IE version was used within third-party applications as part of the standard application SDKs provided by Microsoft. For example, any QuickBooks user can tell you stories about IE security settings affecting their accounting software in a negative fashion. This made it difficult for Microsoft to iterate IE, because so many applications relied on proprietary aspects of IE's rendering engine.

The very short answer is that they of course didn't. It's utter nonsense. I was around in those days, worked on Mozilla and Nutscrape etc. Back in the day there were many browsers, all spun off the Spyglass original code, the internet was pre javascript and it was a very different place. Then the browser wars started, there was Netscape and IE and others, back then it was viable for individuals to actually make a competitive browser, but over time they got to be big complex pieces of code, basically a VM. Security became a big deal and the open source movement built the best browsers, Mozilla, Phoenix and Firefox ten years ago. Microsoft tried to get the Internet to come to it rather than the other way round, just as companies like AOL did. They tried to add proprietary technology to lock in the internet to their Windows platform. Can't really blame them for that, but meanwhile every user on the planet was free to install any other browser they wanted. Just because they didn't isn't Microsoft's fault. They didn't prevent you doing so.

These days the internet has shifted from the desktop to laptops to mobile phones and tablets where Apple and Google have the lock in as Microsoft did. Apple allow other browsers provided they don't want fast javascript. Yet nobody's accusing Apple of holding up the internet.

The main things that really did hold back the internet was bandwidth, it was/is the phone and cable companies because they really do have a stranglehold on their customers.

Others here have suggested microsoft tried to stop others from innovating. Again, total bullshit, they forced exactly nobody to use their software.

I am no fan of MSFT, I haven't used their products in years, but someone who was around then really needs to set the story straight.

Microsoft was declared a monopoly and ordered broken up as punishment largely because Gates held firm that a web browser was an integral part of an operating system. (It was only the switch in US Presidential administrations that prevented the "remedy" from taking effect.)

Seems so fucking silly in 2014.

The laws around antitrust are complex and this is a huge oversimplification -- but I have a hard time believing anyone here thinks that shipping a computer with a browser pre-installed, and in no way preventing access to downloading another browser with it, deserves government intervention into the structure of your company.

Remember, this was an era when software came on many, many floppy disks and a 28.8 modem card cost $299 and required a screwdriver to install. Off by one CSS bugs and allowing both VBScript and JScript and OLE controls to run in the browser wasn't exactly the roadblock to an open information nirvana.

Microsoft was convicted of using their monopoly in operating systems to prevent other browsers from being installed on Windows computers by OEM manufacturers. The browser/OS integration was a sideshow to the main legal issue.

There was no "conviction" (it's not a criminal case) - the judge's findings of fact centered on 1) monopoly & monopolistic behavior and 2) "tying" - specifically saying that a browser was a part of the operating system.

how come people don't upvote this answer? Yes, it was in MSFT own interest to curb the development of the web, but there were also other factors involved.

this is a very narrow minded view. microsoft began creating plugins for utilities that did not follow web standards such as activex that would integrate the browser with the operating system. on top of them being horribly flawed with security bugs, these plugins would then only work for ie. microsoft then pushed people creating web content to use these non-standard plugins in their site because they would not modify ie to include w3c standards that did the same thing. so now many people and companies were forced to use ie because some site they needed to use required one of those plugins.

this continued on until so many major security flaws were found in ie that it drove people who didnt even know what a web browser was to firefox.

This too is a narrow view. Another way to look at it is that people were building apps in the browser and they needed to do real stuff. Browser apps needed to talk to devices, legacy systems, card readers, EEPROM programmers, medical imagers... whatever the hell it is that people jam into Windows. And those devices weren't magically going to sprout a REST interface. There had to be a transitional period.

Microsoft had been making developers happy for a decade by giving that exact sort of functionality in local file Explorer, in Word, in Excel, on the desktop, via COM/OLE/VBScript and god knows what else. So they tried it.

The blood/brain barrier between OS and Browser remains up for debate a decade later.

The W3C remains woefully understaffed and even today hasn't solved basic problems that were solved in the dumb terminal era of the 1970s.

There's a lot of shit that got shoved into the browser when we were excited about browsers that shouldn't be there. Likewise there's a LOT of OS-level functionality missing from the browser that may or may not belong there, but which I don't see appearing in the next 10 years either.

> microsoft began creating plugins for utilities that did not follow web standards such as activex that would integrate the browser with the operating system.

It should be noted that the technology behind ActiveX actually started as a Netscape plugin developed by a 3rd party. I remember trying it out way before ActiveX became a thing in IE. It was a crazy idea but back then everything was a crazy idea. There were hundreds of different plugins for all the browsers. Nobody knew what was going to stick. ActiveX stuck because it was Microsoft and because COM was a well established technology.

I hate to give just a wikipedia link but I really have nothing more to add than what is there:

http://en.wikipedia.org/wiki/Embrace,_extend_and_extinguish

While this might not explain "6-7 years of holding back the internet" it definitely was not beneficial.

Short summary: there was a time when people paid for browsers. Microsoft started bundling IE for free along with the OS. This killed the paid browser market completely (bringing down companies like Netscape) as everyone started using IE since everyone was using windows. Once their position was established, Microsoft stopped developing IE. The IE team was mostly disbanded and there was no updated in IE for many many years.

Not progressing the internet was in MS' best interest. They wanted a world where desktop apps running in their OS was the future.

It was a particularly shortsighted gambit, too, since native applications still beat the pants off of any JS web-based implementation for all but the most trivial of applications. Even very popular web apps these days are less efficient than the (native) applications and servers that back them, where the majority of real work by the computer is done, and arguably are less efficient and more clunky (to both use and maintain) than corresponding native front-ends would be. Until/unless browsers actually have capabilities that rival VMs in some sense and are used as such, it will remain that way.

It wasn't short sighted. It very effectively hobbled a competing ecosystem. They did hold the web back by 6 or 7 years and if they hadn't done their best to kill it in its crib then web apps would be 6 or 7 years better than they are now. (Look at the progress since Mozilla broke their stranglehold, it led directly to the iPhone amongst other things)

Its very hard for people to contemplate hypotheticals though (Taleb discusses this in Fooled by Randomness) so they got way with it.

> bringing down companies like Netscape

To be fair, Netscape Navigator could be a real POS, and IE 5/6 were actually leaps forward at the time. The real problem was Microsoft killing off all competition and then, having won the Browser Wars, disbanding the IE development team, leaving IE to stagnate.

Other issues:

- Killing off competition made it so that the browser landscape on other platforms was barren.

- MS having a separate version of IE for Mac that was a completely separate codebase with it's own set of bugs/quirks also didn't help with the browser as an OS agnostic platform.

- IE's lack of standards compliance paired with it's "still work even with HTML horribly broken" also had a hand in things. E.g. for years a good portion of the Internet was delivered as broken HTML because it "worked in IE" and other browsers were "broken" because they were more strict about the standards.

Yea, I remember it also affecting Opera.

Netscape was NOT a POS except in the eyes of those trying to get web sites to work in it that worked in IE. Microsoft created IE with various APIs that only worked in IE but would fail in Netscape.

I strongly doubt you ever tried to make a site work in Netscape 3. If you'd nest layers (yes, the <layer> tag) more than 2 levels, random stuff would break.

You talk about APIs but really, if JS compat was your only problem you were a happy developer. It was the markup engine itself that was crap. IE's was simply also crap, just in a different way.

In 1998 or so, Scottrade, my broker, advised me to quit using IE and only use Netscape because of all horrible issues they had making their code work with it and that they advised all their clients to do the same. That's almost a quote.

What on earth are you talking about.

I worked at a company that built a web app. We recorded 50+ separate crashers for various version of netscape 3-4. There was a running internal contest about the shortest html that could crash netscape. And the funniest. And the most interesting. It was utter shit software.

I think OP might have only remembered Gecko based Netscape. As a side note, I wonder how many of the crashes would be considered exploitable nowadays.

No, Netscape was a total POS. The definition of bloated buggy software. A huge number of features that web developers take for granted originated in IE.

As Marc Andreesen told Charlie Rose on his PBS show last year, the code he wrote that was used in IE was "horrible" and they re-wrote everything for Netscape.

try not to look at Marc Andreessen's comments in the context of 2014. Remember that at the time NOBODY was making a reliable browser -- not Netscape and certainly not Microsoft. Ecommerce was in it's infancy. Web sites were hand-tooled, un-ready to handle, let alone scale, secure internet transactions. Every browser had severe programmatic flaws. By today's standards, yes, the browsers of the time were POS, as were the web apps being constructed. No kidding... I'm surprised no one has mention "the main thing" (Netscape CEO Jim Barksdale's favorite phrase) here: Netscape leadership team's panicky decision to sell out to AOL in November of 1998. That decision, not any technological shortcomings, is what ruined the Netscape browser, and doomed the company. I remember pulling into the Netscape parking lot on Ellis St in Mtn View and hearing the news that AOL had purchased us. My heart sank to the floor of the car. To their eternal disgrace, Netscape's leadership sold the company to Steve Case's glorified sales marketing firm at a time when Netscape's browser still had 40% market share and an incredibly loyal customer base. They took the money and they ran, making pathetic reassurances of how it was the best thing for the company and the only realistic option, given Microsoft's competitive market advantage. It was an act of cowardice -- there's no other honest way of framing it. The disappointment inside Netscape headquarters following the sale was tangible. Despite Andreessen's and Jim Barksdale's backpedaling, we knew that AOL was buying Netscape to remove it, so they could go toe-to-toe with IE in the nascent ecommerce market without dealing with a trouble-making, nimble firm that had the technological chops to survive, and perhaps prevail. The technologically illiterate AOL chumps from Virginia who descended on Netscape (wearing ties, being douches) quickly confirmed our suspicions. To overcome the disillusion and resentment brewing inside Netscape, AOL paid us two months salary to stick around. Most of the technical talent I worked with took the money and started looking for the next gig -- but when it's over, it's over... trust me, Netscape alumni still contemplate the "what if" scenario -- what if Netscape's leadership team had shown some backbone, hadn't succumbed to their own greed (AOL purchased Netscape for $4.2B and they got most of it). At the time of the AOL purchase, the Netscape browser had an extremely loyal user base, a dedicated and talented technical staff, and a very good fighting chance against the MS behemoth that has rarely been anything more than a mediocre innovator.

Based on your down-thread comment, I'm guessing that you never actually developed for Netscape 4.x (or 3.x for that matter) and are basing this entirely on a single comment from a non-technical person at a single company. Keep in mind that it's quite possible that Scottstrade's tech team was crap, and that's why they were having issues. It's also quite possible that you're talking about "Netscape 6.x vs IE 6.x" instead of "Netscape 3.x/4.x vs IE 5.x".

I do remember hearing that the IE for Mac was actually much better than the Windows version. Not being a Mac user back then though, I have no idea if that was really true or just the sort of smug superiority that Mac users always took to their variants of Windows software back then (e.g. Marathon vs. Doom/Quake).

If you ever get the time read "Competing on Internet Time"

It really highlights what a mess Netscape was internally, with multiple codebases for different browser versions, constantly having to fix bugs in multiple places, it's no wonder it was buggy.

IE5 was significantly better than Navigator, and in particular crashed way, way less often.

Microsoft had no problem with progressing the Internet. They made billions in investments from 1996 to 2002 trying to do just that. They were a big spur in pushing broadband forward at that time, putting large sums of money into numerous access providers; Gates & Co. did a lot of speeches and articles touting how important broadband was. This was at a time when dial-up was overwhelmingly dominant.

What they wanted, was to prevent the Web from eating Windows. That's a very different context than what you're painting, by claiming they were trying to prevent progress on the Internet.

Case in point: an ethernet jack on the Xbox in 2001. People thought they were NUTS. They wound up owning online gaming for a decade.

"Microsoft cast a shadow over the software world for almost 20 years starting in the late 80s." It was not only the internet. In some way it was an evil monopoly that did everything to prevent others from innovate and forced usage of it's own crappy software. The citation come's from PG's essay from 2007 "Microsoft is dead" [0].

[0] http://www.paulgraham.com/microsoft.html

How did they do it though? PG's opinion wasn't very detailed in that respect.

He writes about the dangers Microsoft saw in Javascript and generally an open web that escapes Microsoft's lock-in, and that they kept Javascript broken to prevent the escape.

I posted the link mainly because I think that a large portion of Hackernews audience is young and doesn't have a feel of times when Microsoft was casting the shadow on everything... And I think this "Ask HN" is an example of this.

- Using secret OEM agreements to prevent preinstallation of alternative operating systems such as BeOS

- Making 3.x versions of Windows crash when run on DR-DOS instead of MS-DOS

- Using undocumented Windows APIs in other MS software that let them do things third parties (like WordPerfect) couldn't do as easily.

- Running FUD campaigns against Linux

- Funding SCO to run FUD campaigns against Linux

- As others mentioned, bundling and integrating IE in a way other browser vendors could not match (e.g. as explorer.exe), then stopping development

- Railroading useless standards like OOXML instead of using and improving the OpenDocument standards.

- Running for years on WMA instead of MP3

- etc.

Some references:

http://www.groklaw.net/staticpages/index.php?page=2007021720...

https://en.m.wikipedia.org/wiki/Halloween_documents

My favorite is the OS/2 2.0 fiasco.

Seems like a far cry from "cast[ing] a shadow over the software world for almost 20 years..." Is there anything else?

That's the thing about shadows; they can be subtle, and introduced in boring ways, little by little. See John Oliver's explanation of the boringness of the net neutrality debates for an example of how boring, seemingly trivial minutiae can have lasting serious effects. The anti-dual-boot provisions in MS's OEM agreements were a particularly egregious example. Killing Internet progress for years is another.

Yeah, I just don't see it. It takes more than one party to make an OEM agreement and I don't see a problem with Microsoft doing just as they please with their own OS, just as Apple and Google do with their own properties.

I get that people like to blame one party, but let's look at the bigger picture here: every company plays hard-ball with their own properties if they can. In my opinion, Apple and Google have been far more damaging to the software world than Microsoft had ever been. Apple because they popularized the walled-garden approach to everything and Google because they evangelize taking all control away from users by making every app web or cloud driven with zero choice.

Microsoft didn't just "play hardball"; they abused a monopoly position to extinguish competition, and they lost multiple antitrust suits because of it. The fact that other companies are misbehaving now doesn't negate the damage MS did in the 1990s. OEMs couldn't refuse MS agreements because of MS's monopoly. To an extent the computing world has recovered, so if you weren't old enough or not yet into the computing industry you might not have noticed, but the market would be way more interesting without MS's anticompetitive bullying.

http://en.wikipedia.org/wiki/Microsoft_litigation#Government...

Just like today's "mean people" essay, PG was likely wrong in that essay as well. Firstly, regarding the "shadow that was cast", let's look at some data rather than rely on what "everyone knows", shall we? Here's a relevant paper that looks at some data (caveat - it only covers events up to 2000, but then that's the period PG talks about as well):

"Did Microsoft Deter Software Innovation?", Josh Lerner, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=269498

Spoiler: It follows Betteridge's Law.

And of course Microsoft is not "dead" or "irrelevant" and does not produce "crappy software", as much as the Silicon Valley hive-mind would like to think so. This is trivially disproved by looking at 1) their revenues and 2) their ranking in the top brands worldwide over the last decade and a half.

You should read Microsoft's reply to the essay [0] which contains your arguments and PG's reply to it [1].

It's interesting to review claims from years ago and I think PG was largely right. Microsoft is not completely irrelevant (yet?) but they definitely lost the cultural war and moves like open sourcing .NET show that they want to join the old opponent's positions.

I'll add that I'm not sure if Microsoft's historical influence was evil. They prevented fragmentation and created a unified environment and probably enabled PC revolution.

[0] http://dondodge.typepad.com/the_next_big_thing/2007/04/since... [1] http://www.paulgraham.com/cliffsnotes.html

> disproved by looking at 1) their revenues and 2) their ranking in the top brands.

Enjoy chowing down on McDonalds, because apparently, by your own measure, it's not crappy food.

Heh, I actually happen to like McDonalds! I think it has the best fast food you can get out there, especially for the price. Imagine that, people have opinions contrary to yours!

Now imagine something harder to believe: the vast majority of people have opinions contrary to yours.

They didn't hold back the Internet 6-7 years. IE was far, far beyond whatever competitors offered, it might be more fair to say that Microsoft released IE6 years too early. AJAX, CSS, JS, all the cool technologies that make the web today were part of IE 6 (but not quite at the level they are today of course), and in addition to that there were some DirectX accelerated graphics, an advanced plugin system while other browsers only supported NSAPI, etc.

After that there was little innovation from Microsoft, but there was little need for innovation either, most developers looking for advanced capabilities used Flash instead of addressing the web browser natively, because that was the trend back then and Flash works on other brands of browsers as well, and for a while because a lot of people still used older browsers. For those reasons there was no developer demand for more advanced features, the features that were offered were hardly used for a long time

For example AJAX was publicly introduced in IE5 in 1999, while other non-beta versions of competitors appeared from 2002 to 2005. Websites using AJAX thus were rare until about 2004-2005. There was no need for Microsoft to add more technology until the competitors caught up. Unfortunately for Microsoft, the competitors didn't just catch up, they overtook them and implemented some features differently than in IE, those missing and different features in IE have caused plenty of grief for web developers ever since.

I think it is more than a fair statement to say Microsoft held back the web[1] for 6-7 years, but we settled on the web browser as the delivery for internet services and are still trying to shoehorn everything into it.

Microsoft had a monopoly in the OS market, gave IE away free, and provided tools and incentives to develop for IE and nothing else.

From a company's point of view, it makes a lot of sense. Cost and 90% of customers don't have any problems. The company has a standard development and testing platform. Cost is an amazing motivator and having that OS monopoly was an easy leverage point.

1) there are still issues with using educational / testing sites in any browser but an outdated version of IE. The college textbook with integrated websites could really use some disruption

Just venturing a guess, I'd say it's a reference to early versions of Internet Explorer (particularly IE6, but also later) which combined a dominating position (thanks to Windows) with (1) an insistance on doing things their own way instead of following standards and (2) a very slow pace of innovation. This forced coding to a very low common denominator and/or duplication of effort for advanced features and different browsers. The impasse ended when other browsers became popular enough to make IE's dragging its feet counterproductive for MS, too.

Very slow pace of innovation? There was no innovation for a period of roughly 5 years until growing firefox adoption forced their hand. I seem to remember the internal team was disbanded and had to be reconstituted as well. Microsoft intentionally retarded the development of the browser as an application development target to protect their OS monopoly. And, indeed, exactly what microsoft feared came to pass: once apps targeted the browser, you have chromebooks, tablets, and phones displacing windows. It's not like you need a full desktop to browse facebook, twitter, instagram, pinterest, youtube, or gmail.

[1] http://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Net...

This is oft-repeated but wrong. Sure, there was no innovation in IE, but innovation elsewhere on the web never slowed down. Even when IE dominated the browser market, browsers like Opera kept developing new features. Opera in the late 90s had features that I still don't see in most browsers today.

Now, those other browsers were less popular for a while, but let's not conflate "widespread use" with "innovation".

No.

I worked on web apps in the early and mid 2000s. What we could do was defined by the capabilities of ie 5-6. Everything else, with it's whopping up to 10% market share, was irrelevant.

Plus, innovation on its own is worthless. If I can't use feature X in the browser that the vast majority of my customers use, it may as well not exist.

http://www.hanselman.com/blog/microsoftkilledmypappy.aspx

I think it interesting that no one here has brought up the fallout from the dot.com bust and 9/11. Both of these had far more effects on the pace of innovation than anything Microsoft did or didn't do.

It terms of how (from memory + fact checks on Wikipedia): Microsoft and Netscape battled over browsers throughout the late 90s with Netscape starting in the position of the dominant/only browser and IE was seen as a joke, that quickly, by the time IE5 (1999) was released Netscape seemed completely in technical debt with it's product when they couldn't support even the most basic CSS support in Netscape 4.x. IE5 was also the release that added support for what is now called AJAX.

IE6 was released August 2001, at which point it had most of the market, IE also existed for the Mac and most people I knew at the time thought of Mozilla/Netscape as completely irrelevant as a development target. Opera has basically always been irrelevant in my view. This started an era of IE-only sites which further damaged the competition.

Microsoft disbanded it's IE development team and it wasn't until a few years later that people realized that this happened (It wasn't announced till 2003) - people seemed to assume Microsoft was working on new version of IE, which was natural since it was pretty much the only browser in town.

WHATWG was formed in 2004 so everyone else (except Microsoft) could work on web standard because this had basically stopped at that point.

Firefox wasn't released till November 2004, which was the first time it looked like there would be a credible threat to IE (though it had been pretty good for a year before with Mozilla, but still unknown to most).

Acid2 test was created 2005 which further highlighted the problems with IE6's rendering: http://en.wikipedia.org/wiki/Acid2.

IE7 was released in October 2006, by which point web developers who had been trying to more and more with the web were thoroughly frustrated with IE and it's rendering bugs. IE7 was a big disappointment because it whilst it fixed some long standing problems like it's box model, it was still a long way off the standards that had been produced since IE6 and it didn't pass that Acid2 test.

In terms of why: I've wondered why for a long time now. Mostly, I think IE6 was already too good at being web application platform and Microsoft was worried (as they had been with Java) that this would make Windows irrelevant. Given that IE was effectively free the probably assumed there would be no viable competition due to the lack of business model. Microsoft stopping work on IE they could allow websites to work, but continue to make web apps that were too clunky to use so people would write native Win32 apps.

Reasons for "why" not often discussed: 1) Vista consumed the top systems devs at Microsoft for seven years 2) the IE team wasn't "disbanded" so much as loyal to management at Microsoft that was discarded (Brad Silverberg, David Cole) 3) enterprise customers were plenty happy with IE6 4) the dotcom bust quieted the indie developer ecosystem 5) Microsoft honestly thought they could get NT kernels + .NET on small devices and leverage massive developer support and existing tooling in the late 90s.

Now it seems ISPs have been holding back the Internet (in the US). There's almost no competition on price, bandwidth, latency, etc. Nobody is pushing the envelope.

Nobody? In the last few years, Charter has gone from a few Mb/s downloads to 100Mb and my monthly price has been cut in half and there are no data caps.

If it is actually referring to IE6, in some ways IE's stagnation was helpful. Prior to IE6 was an arms race-like flurry of different browser vendors haphazardly slapping on features to their products, and the IE6 code freeze we saw thereafter gave standards-writers like the W3C and WHATWG time to catch up writing specifications for the web.

It was a good thing. Not just for the standardization bodies, for everyone. The "Fierce Idiocy of 'New!'"[1] died down for a little while and briefly it was possible to rely on a stable development environment. Features are great if they're the correct ones and they're done well, but stability and reliability are even nicer.

1. http://www.constitutionaldaily.com/index.php?id=737%3Athe-fi...

Reply on news.ycombinator.com