Thoughts on the fading away of the age of IE | Hacker News Reader

Thoughts on the fading away of the age of IE

medium.com

37 pointssocrateslee106 comments

First, why the market share of IE is so resilient to die? Second, why did IE lose to Chrome from the position of market domination while Windows has always been dominating the desktop OS market? Now I realized that the answers to the two questions are the same — the mechanism of the software update.

This can't be right though, because IE always had updates too, via Windows Update. Although they weren't as frequent as Chrome's

The answer of why IE hung around for so long probably has to do with corporate intranets and Microsofts huge enterprise customers needing to preserve backwards compatibility. Backwards compatibility is one of Microsoft's strengths but with IE it became a weakness for them.

The answer of why it lost to Chrome is that Chrome was a very strong entrant in the battle from day 1 of its release. Remember the Chrome launch comic? https://www.google.com/googlebooks/chrome/index.html

As of the article, major version updates only with new versions of windows. And "why it lost to Chrome is that Chrome was a very strong entrant" begs the question.

Windows updates could be turned off back then. And often was

> Backwards compatibility is one of Microsoft's strengths but with IE it became a weakness for them.

I know that Microsoft in general puts a lot of thought and resources into backwards compatibility.

But the thing is, with IE, I never felt it.

Back in the day, whenever a new version of IE came out, so much stuff would break, and the whole company would scramble to get everything fixed and compatible again - and getting things to work on multiple versions of IE simultaneously (and flawlessly) was such a pain.

We never had any similar problems with Firefox or Chrome... those updates never hurt as much. IE was the only browser, where new versions would cause all-hands-on-deck meetings...

>This can't be right though, because IE always had updates too, via Windows Update. Although they weren't as frequent as Chrome's

It was typically the case that Internet Explorer major version updates were optional, so you could use IE6 on Windows XP SP3, despite IE8 being out at that time.

IE6 was actually supported until 2016 on embedded versions of Windows XP!

When I was a younger (and more-foolish) web dev few years ago, already well in the decline of IE, I met someone much older who said he was one of the developers of IE. He must have seen it coming in my face from my grin, or he had experience on the topic, so he quickly reminded me of something that was absolutely true; back then when IE was released it WAS at the top of web tech development, the problem came with the trail of users still active years/decades later, since the tech itself was pretty good when it was released. Sure in some releases they might not have been literally the best, and they did employ shady techniques, but it's true that they were always technically advanced and competent releases.

I guess IE’s reputation was also shaded by the reputation of Microsoft. The Microsoft back in the 90s and early 00s was IMHO, not a very nice company. But how they have evolved! For someone starting out with Linux in the 90s I could never dream of Microsoft being a part of the open source ecosystem. In my book, Satya Nadella has been a very good steward of his company.

Personally, I preferred the Microsoft of the 90s and 00s who were concerned about getting my money at the cash register.

These days, Microsoft are concerned about getting my money at my god damn desk every second of the god damn day.

Announcing Microsoft Nickel & Dime 365! No fiddling with “purchases” or “subscriptions”. Everything you earn is transferred directly to us!

> But how they have evolved!

I'de disagree with that statement as some of their moves (such as purchasing GitHub) could be equally viewed as part of their typical EEE strategy that permeated the entire business.

WSL could also be, and indeed is by many, viewed the same way ("Microsoft is just trying to pseudo-own the linux ecosystem by embedding it within Windows").

Personally, seeing how EEE permeated the entire business in the infamous 90's & 00's era Microsoft and all of the dirty laundry that ended up being aired, I don't doubt for a second that the current-era Microsoft is just as bad, but better at hiding it.

Apples don't fall far from the tree.

Fool me once...

I'll disagree with your disagreement. They carry with them the reputation from the 00s, which most of us seem to be incapable of letting go of. There a fundamental assumption being made here that they are incapable of change.

To me their actions speak louder than slogans and have proven themselves good citizens in many areas. Scrutiny is good. They should constantly be scrutinized.

Yet we have Google and Apple who regularly escape the same scrutiny that they ought to be getting, and instead are given a free pass for more nefarious actions than ms have committed.

If we're going to take a stance if distrust, it needs to be extended to everyone. Otherwise it's just politics and brand loyalty.

Okay, but they still haven't changed.

They haven't done jack shit for open source. Buying Github doesn't mean anything.

They also haven't done jack shit for open platforms either.

They can't ship working cross-platform software that matters to save their lives. (VSCode doesn't matter. There are a million other text editors)

Things that DO matter, that they have locked into their ecosystem are things like Office or Teams.

Teams is based on electron, and they still fucked up the cross-platform capabilities. Teams on Linux is absolute ass.

Why not port Office to Linux? I thought Microsoft <3 Linux. They don't. They haven't changed.

They lose a ton of devs to Linux and Mac, since doing any kind of dev work is a pain.

Their solution? Ship an entire Linux OS (WSL2) alongside their own OS, since their dev tools are dogshit. They didn't do this out of goodwill, they are trying to stop the bleeding.

So please tell me, in the last 20 years, is there really anything microsoft has _actually_ done to foster any kind of goodwill from Linux people?

> their dev tools are dogshit

I agree with you except for this. Microsoft has, since the '90s, always had top-tier dev tools. They're a different ecosystem than unixy development, and hence incompatible and unfamiliar with a different user experience. Linux's dominance in the server environment has caused it to gather developer mindshare in the last 20 years and so there are many developers who prefer the Linux tooling ecosystem. And yes, Microsoft is trying to (and succeeding at) keep them in Windows with WSL. I expect to see them use the 'extend' playbook to trap them on the platform (if they haven't already, I'm not following very closely).

> given a free pass for more nefarious actions than ms have committed.

Not to defend Apple or Google, as they are both toxic companies, but Microsoft literally supported SCO.

How is SCO relevant in 2023? That company has been dead for two decades.

Hands up if you've ever heard of Xenix!!

SCO isn't relevant as a company.

The grandparent comment said that Google and Apple had done much more nefarious than Microsoft. Microsoft who funded SCOs baseless lawsuit to completely destroy Linux.

If you can think of "more evil" things that Google and Apple have done then I'm all ears!

They just cultivated the same howling mob of die-hard fans to shield themselves from valid criticism. Its way cheaper to run a hype-cult then produce good products or perform a clean up after a PR-Disaster.

Meanwhile they absurdly managed to make people beg for technological innovations fromt he main group at fault for technologic stagnation aka huge undefeatable monpolies.

To be emotional towards a company, means to become a p3wn.

> but it's true that they were always technically advanced and competent releases.

Wow. Historical revisionism if ever I saw it. Netscape imploded like the fall of Rome, rent by incompetent business leaders and crumbling from unassailable technical debt. IE defeated the competition by anti-competitive practice, then had no competition as their opponents scored all of the own goals. Mozilla was a slow rising Phoenix from the ashes of that catastrophe.

From there, Microsoft did what big companies do best, having destroyed their competition. Absolutely… sweet… stuff all. Years of nothing. Technically advanced? Competent? Ha I say! If miserly underinvestment is now “technical brilliance” I weep for the youth. Thank goodness that competition eventually returned; but there are reasons every single human who could jumped on chrome when it was released.

> but there are reasons every single human who could jumped on chrome when it was released.

...Advertising? Antitrust action?

Because, let's be honest, Firefox had been around for a while, and Opera was there even longer. The people who waited until Chrome for to jump ship were, mostly, people who hadn't heard of an alternative.

Chrome was fast and minimal when first released while Firefox famously ate all of your memory. No matter how much you had, it would eat it. I remember being mocked for the number of tabs I would always have open, but secretly, they were just jealous that it was actually possible.

This was in the days of “don’t be evil” and people even liking google. Firefox had its advocates, but it wasn’t the simple “just way better than IE” that chrome was. I shamefully moved from Firefox to chrome. For the tabs man, I needed those tabs.

> No matter how much you had, it would eat it. I remember being mocked for the number of tabs I would always have open, but secretly, they were just jealous that it was actually possible.

Confused about this comment. At all points during the history of even single-process Firefox, as far as I know, it could handle far more tabs than Chrome, as in not even worth comparing. Chrome would start actively falling apart at 100 tabs. Firefox could eat up all of your memory with 20 tabs, but would only be marginally worse at 1000 tabs.

Firefox died because it systematically started eliminating all of its advantages over Chrome and imitating it in even superficial ways, and became openly user-hostile as it became completely financially dependent on Google. Imo it had only a little bit to do with Chrome's performance, which was often based on visually cheating and assuming that nobody needed more than 20 tabs, but also had tab isolation so the browser wasn't always hard crashing. Still, plenty of people are impressed by a quickly appearing unresponsive UI, plenty don't need more than 20 tabs, and hard crashes suck.

> For the tabs man, I needed those tabs.

I needed tree-style vertical tabs, so I was locked in because Chrome wasn't flexible enough to build them.

You are almost certainly right. I blamed all the memory eating on the tabs, but it probably had nothing to do with it.

And the thing was, in those days when Firefox was a memory hog, it was a 32-bit process. So it was limited to a max of 4GB anyway. It made efficient use of that 4GB and worked surprisingly well with 1000 tabs, even though there were inconveniences now and again. By this time, buying an extra 4GB of RAM for Firefox was pretty cheap.

> I shamefully moved from Firefox to chrome. For the tabs man, I needed those tabs.

Opera had tabs before Chrome even got released. So i'm assuming the key difference was the ability for Chrome to advertise its features to you.

I am the first one who started using Firefox as soon as I had any notion of anything and haven't used windows for 15+ years (Linux+Mac now), but it's also true that when a new Internet Explorer was released I (and all) web devs were relieved because soon most people would be using it, and the web compatibility was good with those (pity you if you had to support older IE though). They even went ahead and invented some features of the modern web. Would they be faster with more competition? Undoubtedly. But the hate from webdevs was mostly from the trailing users still stuck in IE, not for whenever a new release was launched, that was a good event.

Note: I don't dispute they were slow, anti-competitive, etc., just that now from a tech point of view IE seems garbage, but that's a false believe and when a new IE launched it was decently competent.

> just that now from a tech point of view IE seems garbage, but that's a false believe and when a new IE launched it was decently competent.

I think, perhaps, we are talking about different eras. When IE came on the scene in the mid 90s it was garbage. Anyone tech-savvy was using Netscape and laughing at the pitiful entrance from Microsoft. Netscape was vastly superior and had a huge runway (so it seemed). But the thing is, MS bundled the browser with the OS and called it the “internet”. Click here for the “internet”. How do you compete with a free app installed on the desktop of every computer? Netscape had an existential crisis on their hands and legal wrangling over anti-competitive behaviour was never going to be fast enough to save them.

Behind the scenes things were already chaos and the code was an irredeemable pile of technical debt (or so the story goes, I’ve never looked at the code). New Netscape versions came out, but the rapid improvements of the early years were gone. It was a buggy unstable mess that seemed to get worse with every release while IE got better (not great, but the trajectory was clear). Soon even nerds like me were using IE because Netscape had jumped all of the sharks. By the time we were partying like it was 1999, IE was the only game in town. And then MS called the internet “done” and went back to throwing chairs and yelling “developers” a bunch.

I’ve never been a web dev, but I expect that the pain of those lost years where MS cared little for standards due to lack of any meaningful competition, could have been dramatically shortened. I’ve always hated www, ever since it devoured my beloved gopher and brought on the eternal September, but it’s a real testament to the power of standards that this inefficient, ugly, slow but ubiquitous method for deploying content has been so successful. Once Microsoft and Adobe finally got out of the way.

I mean, Netscape did implode, and Microsoft was anti-competitive, but...

IE4 introduced DHTML, i.e., the paradigm that the entire HTML document is on the DOM and you can modify it with JavaScript (and style it with CSS, which is also on the DOM).

IE5 introduced XMLHttpRequest, i.e., the foundation of the modern SPA.

Both of these were major advances in web development, and websites started requiring (or preferring) IE because they wanted to use that stuff.

From there, yeah, IE6 was mostly bugfixes and webdev languished until Chrome came around (with Firefox being great, but not enough to topple IE.)

Sure, IE4 was the turning point where you couldn’t laugh at MS any more and Netscape was only referred to in terms of “what the hell is up with Netscape?”. Still, that “4” in the title means it had taken a bunch of “much worse than Netscape, but we know how to cheat real good” versions of anti-competitive behaviour before we got to this point. Might Netscape have survived had MS not been the amoral business demons that they were?

Of course, MS were still competing at that point and shear kinetic energy of the behemoth got you through the next couple of releases. Before how many years of nothing? And internet years are like dog years, it was basically infinity time before chrome burst onto the scene to actually move things forward again. Where might we be right now, if MS hadn’t have been able to leverage their OS install base and complete lack of integrity to kill Netscape dead before it had a chance to threaten their business model?

We probably wouldn't have Mozilla, for one.

So the open-source energy that went into Firefox would've gone into... KHTML? Maybe?

It’s a common misunderstanding: because something is the dominant technology it must be good in some way. The history of technology, not just computers but all technology, repeatedly and consistently shows otherwise.

Kubernetes :-)

I don't know which timeline you are from, but even at v3.0 IE was noticeably ahead of Netscape Navigator, both in terms of speed and stability. Trust me, I browsed the internet back then on a 486 with 16mb. Every little speedup was noticeable.

At 4.0 the gap widen. While IE's rendering speed got even faster, its integration with Active Desktop etc. slowed down the start up in Windows 98. Netscape got even slower across the board. It's funny that even without the fancy features Netscape 4 was slower than IE4.

By 5 the dust was largely settled. Netscape was on rapid decline and IE5 optimized/got rid of all the IE4 crap in Windows 98SE. My desktop (Pentium III 800Mhz) was faster with 98SE' IE5 than IE4, which was much better than Netscape 4.0.

I’m from the timeline that is slightly older than you by the looks of it. The first couple of releases of IE were garbage, and I would argue it was IE4 where the writing was clearly on the wall for Netscape. Mind you, if Netscape had continued apace, then it should have been fine. 1995 Netscape Navigator 2 was full of promise and life. A little bundle of hopes and dreams. IE2 around the same time was the Cuckoo in the nest. Ugly, dishonestly planted, and soon taking all the nutrients from the Netscape hatchling until it withered away.

Ahh Microsoft; ever were you the great teacher on the reality of power.

IE4 is the first what I would call "recognizably modern" browser, because it's the first browser that has a concept of "reflow" accessible through Javascript. I recall writing some Javascript attached to a button that would add another row to a table, and that new row would have a button in it with attached Javascript that would add a new row. Yeah, it's not how you'd do it today, and I crashed the browser more than once trying to do this sort of thing, but it worked like you expect a browser to work today; a row was added to the table and the page reflowed around it.

Prior to that, nothing worked that way. What limited page modification there was after formatting couldn't ever cause a reflow, and you couldn't even do all the things that couldn't cause a reflow, either. Reflows were simply not a thing through Javascript. (Deferred image loading could cause reflows, and a few other special cases the browser would handle, but you didn't have generalized access to it.)

Netscape tried to compete but they clearly couldn't. They tried to build a "layer" concept, which was ultimate just several chunks of page that could be completely rewritten, but couldn't be reflowed as we'd use the term today. It was a weak concept that was very difficult to use for anything.

IIRC, Netscape qua Netscape never solved this. It was only after Firefox that it could compete. There was a long time where IE was really the only recognizably-modern browser.

It, too, fell behind, its internal architecture becoming simply too limiting. But there was a period of time where it was legitimately the best browser, no matter how anyone may have felt about it.

> rent by incompetent business leaders and crumbling from unassailable technical debt.

I think you've mistaken this for a brand-off between IE and Netscape. Netscape got very, very bad, then died. That allowed IE to stop working at all on the browser, and even to use their bugs as a moat.

If memory serves, Opera was always more performant, they just got hosed by dominant web sites like hotmail, msn, etc. intentionally breaking the css/markup being served and other websites that used explicit browser sniffing and would refuse to load real content for anything but IE.

> Wow. Historical revisionism if ever I saw it. Netscape imploded like the fall of Rome

It looks you concede the fact that IE was clearly better in its 4.0-6.0 releases than its competition (chiefly Netscape), but hate this fact so much, that you then still somehow dispute it.

MS had its usual anticompetitive practices, but that does not disqualify the fact that IE was at the time also technical state of the art.

(My teenager MS hating me at the time reacted similarly and stubbornly used/advocated Netscape/Opera anyway).

> It looks you concede the fact that IE was clearly better in its 4.0-6.0 releases than its competition

Yes, but only because they cheated. They poisoned their opponents and claimed victory after they fell over. Just because the new releases showed steady improvements while Netscape was thrashing around in its death throes doesn’t mean they were “technically excellent” in any meaningful sense. And once it was clear that Netscape wasn’t going to be getting back up they completely stopped trying. That’s the bit I’m most annoyed about.

Microsoft's cheating had no influence on the fact that Netscape's engineering (and/or its management) messed up bad. The decision to do a complete rewrite (Mozilla) was the final nail in the coffin.

> Microsoft's cheating had no influence on the fact that Netscape's engineering (and/or its management) messed up bad.

I completely disagree with this. The unfair competitive advantage that MS created was directly causal in Netscape acting out of desperation. Desperate companies do desperate things. Often really stupid desperate things.

If they had continued with the dominant market share and comfortable revenue stream, they would have made slow comfortable decisions. Instead they made rash, desperate decisions and imploded.

Microsoft had no valid justification to take that much market share that quickly (Netscape was much better for the first couple of versions). They could only do it because they could leverage their OS.

If it had been a fair playing field, the outcome might have been very different.

Ironically, Microsoft wanted a web based Outlook, and thats what essentially gave birth to XMLHttpRequest, which was a huge factor in web apps becoming a thing, and launching the web 2.0 era that saw IE get unseated by competition.

I wish I could explain to people how GMail seemed damn close to magic when it came out.

Why?

Page updates without reloading (AJAX) were a pretty novel thing at the time. Threaded conversations too. And 1 GB of storage! Not 10 MB like others! For free! This was back during peak "Google is actually amazing" era too.

Correct.

As I recall, Microsoft actually stopped development of IE, saying the job was done, the web browser complete.

Am I correct? Can anyone verify this?

The job was done in terms of market share achieved, and thus they divested their resources elsewhere. The mistake was to not invest more seriously in further development when Chrome gained popularity.

>Microsoft actually stopped development of IE, saying the job was done

Despite being a technology enthusiast who was paying attention at the time and who heard this when it happened in various discussions and articles, now some twenty years later this is hard to source!

This is one account: https://slashdot.org/comments.pl?sid=171546&cid=14288661

Mozilla lists it as such on their timeline (May 31, 2003):

https://wiki.mozilla.org/Timeline

Microsoft Drops the Ball with Internet Explorer: https://web.archive.org/web/20050506233847/http://www.window...

So it wasn't that users refused to upgrade their installations of IE, it's that Microsoft insisted for a long while there would be no IE releases past version six, since it was now a part of the operating system and the team they had had had been disbanded to work on other parts of Windows and Office by that point. It was only when enough bugs were seriously impacting Windows and the emergence of Firefox and Safari that they started taking serious aim at fixing those bugs and releasing later versions of IE, seven and so on..

No, you're wrong. IE was a superior product, and a lot of Microsoft haters tend to forget this. Netscape was buggy, slow, and bloated compared to contemporaneous versions of IE, and IE had more features like better CSS support, IFRAME, XmlHttpRequest and the fastest Java runtime out there. Yes, Microsoft made a better JVM than Sun, and Sun's response was to sue.

IE won the browser wars -- fair and square.

> IE was a superior product, and a lot of Microsoft haters tend to forget this. Netscape was buggy, slow, and bloated compared to contemporaneous versions of IE

Now explain Opera. I remember seeing the lengthy blog post by the CEO of Opera demonstrating how Microsoft was deliberately messing with the browser when users tried to access Hotmail and other Microsoft controlled sites. It's a matter of historical irony that Google used some of those same tactics on Microsoft later when they were trying to make Windows Phone a thing.

>IE won the browser wars -- fair and square.

There was nothing fair about it. Microsoft went so far as to tie the browser to the system in order to compete. It messed with compatibility to screw with users of competing browsers.

You can snow people who didn't live through it, but you cannot snow those of us who were there and did.

I'm glad someone pointed this out. Netscape was incredibly buggy and unstable. Anyone remember the Netscape 4.x days where it crashed hourly? In the early 2000's, there was only one viable browser: IE 6. It was like this for a long time.

back then when IE was released it WAS at the top of web tech development

I remember when IE4 came out. It was a fascinating piece of software (and extremely crashy too in beta) packed with all sorts of forward looking technologies like Active Channel, Active Desktop, a chat app, Outlook Express, and more. I'm a bit sad, but I could literally watch a 2 hour documentary on IE4, there must be so many stories behind its development and eventual demise.

It was always implementing proprietary stuff and breaking w3c standards following their predatory business model. It only gained traction because it was bundled with windows and MS lost that case in federal court. Maybe a MS dev believes it was good tech, but nobody I knew shared that opinion.

Makes you wonder what's stopping Chrom(ium-based) browsers from suffering the same fate. Maybe the simple fact that the browser exists to attract users so they can be packaged and sold to advertisers, vs the browser existing to trap you in a desktop OS ecosystem.

It's pretty remarkable that Firefox and Chrome were able to succeed over a web browser that came installed on everyone's devices.

A person had to use IE to go and find another software that does the same thing. And for a browser, there is relatively little UI to begin with - it's mostly an empty box to view websites. The technical folks are used to that, but the majority of market share is non technical people. The "be 10x better" paradigm is especially hard when it's your empty white box versus mine.

Thinking back, I suspect adblocking and pop-up blocking was a big part. The IE era web was extremely user hostile, with pop-up windows and pop-behind windows everywhere. And Firefox* offered a solution to it by blocking those, which I think was hard for Microsoft to follow due to their partnerships with major websites. Tabs were another, but I think that was less important.

* Then Mozilla, and Opera did it first

OS bundling is way overrated, historically. People choose the bundled item when they are literally forced to by the vendor, in the way Apple does.

Netscape won the original Browser wars because it was better and faster and smarter. Over time it became slow and bloated as Netscape literally tried to bundle an OS into its browser.

The original IE was faster and smarter than Netscape Navigator, people literally made conspiracy theories about why it was faster.

Microsoft stopped developing IE effectively, then Open-Sourced Firefox came along to challenge, but it was inherited from the same Netscape codebase and made minimal market progress at the time.

Google suddenly came out with Chrome, it was faster, smarter, and cleaner than IE and others. And worked best with google.

Microsoft controlled the OS, but it did not control the web, and in various ways the web said use "Chrome" because Chrome was better.

>>OS bundling is way overrated >>the web said use "Chrome" because Chrome was better I think you may be ignoring that google bundled chrome on android and "bundled" it with every web platform they owned. Youtube/gmail/maps etc. have suffered firefox incompatabilities and shown large pop ups suggesting users move to chrome.

If bundling is overrated why do you think google created chrome in the first place? and continued to fund it and push it on their platforms for years?

> Google suddenly came out with Chrome, it was faster, smarter, and cleaner than IE and others. And worked best with google.

... and it got bundled by default with the installers of many free software

> then Open-Sourced Firefox came along to challenge, but it was inherited from the same Netscape codebase and made minimal market progress at the time

No, this is not correct: https://gs.statcounter.com/browser-market-share/all/worldwid...

Firefox very much made market progress. Chrome didn't catch Firefox until 2011: https://gs.statcounter.com/browser-market-share/all/worldwid...

Android was already on version 4 by that time, so of course they were catching up as they were becoming dominant in that space, and Chrome was the default browser installed on every device.

Microsoft stopped developing IE because they were under intense anti-trust scrutiny having been found guilty of anti-trust violations and bound by a tough consent decree that made them timid for the first time ever.

That minimal Firefox market progress was about a quarter of the entire Web, and enough to see the content of the Web so dramatically transformed over a decade (that even Safari claimed to be Gecko) to a standards based platform with minimal reverse engineering all the IE-isms and documenting those for others to follow, and in so doing opening the playing field for standards compliant browsers.

Your summary, presented as fact, is quite poor and I'm gonna assume you have no particular expertise here and were instead sharing some opinions you picked up rather than sharing any actual history from experience, right?

I was there, a part of almost all of this for 25 years now, so there's my authority. What's yours?

You might have been part of the same crowd generating conspiracy theories! I don't know. But those ideas were a very important part of developing the conviction based communities which promoted it (not a criticism). Firefox usage peaked at 32% apparently 2010 market share against the IE which only had minimal development since early 2000s. Chrome proper>2010 effectively defeated Firefox in two years of issue and became dominant after 3 years. Why so quick? for the conviction based community, Chrome won because of advertising, bundling, and variations of force not because it was fundamentally a nicer experience for end users.

An interesting retrospective and nice to read, though I had to chuckle at "soft" discs. Is 'floppy' now considered offensive?

I started my web dev career when MSIE was accelerating and achieved crazy levels of dominance. Developing for IE was shit. I mean, I was never out of work, but the wasted time figuring out which quirk you were coding around was, frankly, ridiculous when you look back on it. But you had an arsenal of workarounds and techniques which that generation of developer fantastic problem-solvers and have lots of initiative.

Despite how they did it, it is essential that nobody in or around the industry forgets how good MSIE was for mainstream adoption of the internet/www. It literally made it so that everyone was online. This was when the beige box in the corner was your window to the www and very few people had that kind of connectivity in their pocket.

Reflecting on how MS achieved dominance and then essentially backburnered MSIE development really befuddles me to this day. Some of the thoughts in the comments make sense; Was it arrogance and MS seriously thought Win32/.NET would remain dominant? Or simply that they'd dropped the ball in other departments and had to redeploy resources to a profit centre?

>Or simply that they'd dropped the ball in other departments and had to redeploy resources to a profit centre?

You aren't entirely wrong; Microsoft tripped over themselves trying to get XP's successor out of development hell.

It was five years from Windows XP to Windows Vista, compared to one to two years between Windows releases up to XP. And then as we all know, Windows Vista completely flopped because it was too advanced for its time.

It took eight years from Windows XP to Windows 7 until Microsoft finally had a new Windows release that people would come out and buy. In a way, they really were too busy elsewhere to care about IE.

There's many company-intern intranet applications still in use out there, which rely on old IE features like ActiveX and whatnot. Afaik, older versions of .NET came with features that required IE.

Also Asia always was and to my knowledge still is using a lot of IE - not sure about the reason why.

A little history lesson:

* Netscape wanted to dominate the web, gain a monopoly and gain full control.

* IE came late to the party - long after Netscape was already established

* MS invested a *ton* of resources to catch up, rapid development pace, quick updates, constantly new features and improvements

* MS too wanted dominance and full control of the web

* MS gained marked share through "dirty tricks" leveraging their OS dominance

* MS got repeatedly sued for unfair competition, paid large sums, paid lip-service to the courts but never truly changed their strategy

* that hurt Netscape but didn't kill it

* what killed off Netscape was their own doing: a costly code rewrite, that took way to much time and delivered subpar results

* Microsoft won, they had what they wanted: full control of the web, and the power to shape it however they saw fit

* they did nothing. they just stuck with IE6, no new features or improvements for years, even bugfixes and security fixes became really slow. all momentum was lost.

* Firefox rose from the ashes of Netscape - MS still did nothing.

* Firefox gained significant market share - that finally kicked MS into gear

* MS re-started updating and improving IE - but they were technologically behind and had to play catch-up again

* Chrome entered the ring and boosted the pace of development

* IE never fully caught up

* Chrome gained full dominance of the market - but unlike IE did not slow down development

* MS tried all their dirty tricks again - but they still continued to loose market share

* they rebranded IE to Edge (still using the same render engine, but dropping backwards compatibility)

* they dropped their own render engine, and switched to using webkit

* despite even more dirty tricks, no breakthrough success in terms of market share

I believe the fading of IE pretty much started with Microsoft resting on their (un-earned) laurels, after Netscape died. And yes, that's even before IE peaked - while they still had upwards momentum.

That timeline all seems pretty accurate, but I'm unsure about EdgeHTML or Spartan or whatever it was called being "still the same render engine" - do you have more on that?

"still the *same* render engine" is poor wording on my part. Of course they overhauled/reworked things, dropped support for ActiveX and a lot of backwards compatibility stuff. So it probably would be more accurate to say it's a fork of the old engine.

That edgeHTML starts at version 12.0 is a hint for being a continuation of that lineage - but there's also a few bugs/idiosyncrasies edgeHTML has inherited from Trident.

I remember that edgeHTML did spoof the user agent so browsers would not detect it as IE, and would not apply IE specific hacks and polyfills. And that would cause problems with some web apps, that could be fixed by re-enabling a select few hacks/polyfills.

It's been a long time ago, so I don't remember what specifically it was... sorry.

All good, thanks for the follow-up! I've been curious about the (original) Edge rendering engine for some time.

This is a wonderful summary. The only thing I would add is that the fumbled rewrite was, IIRC, Netscape’s reaction to the rise of MS. So I would argue that IE really did kill off Netscape.

Also misses the Linux and Mac history - KHTML/Konquerer and WebKit/Safari, which of course was the basis for Chrome. Google didn’t need to start from scratch.

I know there's more history to webkit, but that didn't really seem relevant to the rise and decline of IE... I glossed over it for the sake of brevity. (And yes I know, I still failed to keep it brief, despite that).

As for the decision to do a rewrite - sure Microsoft put pressure on Netscape, and that did influence their decision, and sure Microsoft might have killed off Netscape anyway - but the rewrite still was Netscape's own decision, and it was also Netscape who failed to pull it off successfully.

Of course. I thought your summary was awesome - brought back a lot of memories - and was just trying to add colour.

I do remember reading Mozillazine to keep up to date with the rewrite… and at some point, it was a year at least, I just gave up.

I think the reason why MS did nothing with IE6 dominance was that they didn’t want the web to be a viable alternative to Windows/win32/.NET. Or, perhaps, they didn’t believe it could be and as such it wasn’t worth further investment.

And now, two decades later, MS is shipping web apps as a part of Office.

Trivia: at some point (around IE8, I believe) Google released a plug-in to Internet Explorer called IEFrame that displayed supported pages inside IE window, but using Chrome underneath. It was fairy popular.

> Google released a plug-in to Internet Explorer called IEFrame

It was called ChromeFrame.

There was also an ActiveX plug-in that allowed hosting XULRunner in IE, which allowed embedding Firefox inside it.

> Also Asia always was and to my knowledge still is using a lot of IE - not sure about the reason why.

ActiveX saw much heavier use in those markets than the US & Europe. Other parts to the east too. Partly this is because of language issues (terrible non-latin writing support and such in browser-native options), partly because of requirements set down in local government and/or bank policies (South Korea being the most significant example here, but not the only one) because of security concerns.

Early on trying to enforce the security requirements those governments/banks needed required running native code (remember that there was a US export ban on encryption supporting keys longer than 40 bits so relying on browser-only code communicating over HTTPS transport was dubious) and the options where ActiveX or Java. Largely because ActiveX was built-in to a browser, and partly because Java has a reputation for being heavier in terms of client system requirements, ActiveX often won over Java.

While most of those forces have ceased to be, or are at least greatly reduced, for a long time they created significant friction for moving away from IE in those areas. Once an option is as entrenched as IE became in some of those locations, it takes a long time for other options to be considered at all viable – there is a lot of momentum that they have to try arrest.

Just this morning I accidentally found an ancient (well, recently updated but never rebuilt), quoting tool I use from Intangi has an embedded web view when you click the help button. Being an old school 32 bit Windows app, I had a suspicion it was still IE based so I started browsing around their help site for a bit, trying to find a way out to a search engine where I could go to a page of my choosing. To make a long story short I used the path:

Intangi help -> Intangi search -> Intangi forum post with a GitHub link -> GitHub's legal notice pages -> a link to the Creative Commons license on the CC website -> CC main page -> CC search in the bottom right -> a search and CC blog post for a collaborating with Google from 2013 -> The link to the Google page about the collaboration -> some generic Google intermediary page with a Google products listing -> the Google search page.

In all of this most pages except the Intangi pages and the Google search page were broken somewhere between unrecognizable to "I'm surprised anything loaded at all" with a ton of JS script error messages along the way, which is why it wasn't as simple as GitHub search -> Google Search. This gave me the hint it wasn't the usual system IE 11 rather some kind of legacy embed. Searching for my user agent led me to see it was reporting itself as IE7 and running the ACID2 test did indeed show the failed rendering I'd expect from IE7.

This gave me quite the chuckle as it was as a kid it wasn't until the Internet Explorer 8 betas in early 2008 when the idea the browser was just a loadable program and I there might be others that are better hit me... and this was predating even that!

Props to whoever at Google insists on maintaining legacy versions of the search page. At this point it's just the main home page, search results are new and somewhat broken for old browsers, but it's still something. It even works when I load up an early IPv6 development stack into NT 4! And let me tell you, there are very few sites using IPv6 which work in browsers that old.

What we need now is the fading away of Windows Outlook and its Word based html renderer. Emails need to start using conditional rendering to show an “Outlook is no longer supported. Please upgrade to Thunderbird” message

Hard no, HTML in emails should be kept at bay as much as possible. If that means supporting a weird subset of Word HTML, so be it.

The new Outlook client is WebView2 (Edge, i.e. Chromium) based. It will be replacing the consumer focused Mail app in Windows as well, which I think was using the original Edge (non-Chromium) WebView.

Right. Well I'm just going to suggest to clients that we use <div>s from now on and see if anyone complains.

I'm too young to be any kind of oracle on IE and it's history, however I'm personally glad that IE was already fading away by the time I began my software engineering career. This is because, from the stories I've heard from older colleagues, IE rapidly turned into another loser EEE Microsoft boondoggle; trying to own the ecosystem and crush competition by:

1. Moving standards so fast, so randomly, and so opaquely, that nobody can keep up.

2. Abusing market share and WinOS (then sued for it)

3. Etc...

In my eyes, IE was never going to recover from all of that bad blood. It was ill-fated and doomed to die once greed took a hold.

I think the biggest pain of IE was there were few standards at all and the ones it did support it moved extremely slowly on. E.g. if you wanted to something as basic as add a drop shadow to an element you would do it with a DirectX filter. Sure, it was available before a standards process figured it out... but implemented in such a proprietary way Microsoft didn't even bother trying to put such functionality into the version of IE for Mac. In the meantime, anyone that wanted to move to the much simpler CSS way had to write two versions to do so as it took IE a long time to get the standards implementations and then a long time for the users to actually upgrade.

Microsoft itself should never recover from this bad blood either.

When you get a bully on the ground, you keep kicking and you don’t stop. Otherwise they’ll just come back and make you wish you’d never been born.

I have been thinking about two questions for a long time. First, why the market share of IE is so resilient to die? Second, why did IE lose to Chrome from the position of market domination while Windows has always been dominating the desktop OS market? Now I realized that the answers to the two questions are the same — the mechanism of the software update.

>First, why the market share of IE is so resilient to die?

Because it came bundled with Windows and because the internet was designed for IE. You can see that even today: Chrome is bundled with Android, and the internet is designed for Chrome.

>Second, why did IE lose to Chrome from the position of market domination while Windows has always been dominating the desktop OS market?

Because IE, and specifically IE6, was a product of 2001. When Firefox (NOT Chrome) finally usurped the throne, it was because almost five years had passed since IE6 had released and the technologies for the internet and computers in general were being held back by what was then an ancient decrepit relic that was IE6.

Firefox introduced (mostly) W3C compliant (X)HTML, advanced CSS, JavaScript, tabbed browsing (nonchalantly stolen from Opera), and browser extensions/plugins among other things. Firefox was plain better than IE6 which at that point was frozen ancient technology, and Firefox usurped the throne from IE6. Microsoft's kneejerk reactions that were IE7 onwards and Trident!Edge came far too slow far too late and they never recovered from the loss.

Now you're probably asking why Chrome is the dominant monopoly browser today, and that answer is also easy: Chrome incorporated what Firefox introduced but marketed themselves more aggressively, continued to improve where Firefox languished upon obtaining the throne, and for better or worse threw out a lot of conventional computer nerd traditions which ultimately made for a more performant browser in line with common user desires.

> Now you're probably asking why Chrome is the dominant monopoly browser today, and that answer is also easy: Chrome incorporated what Firefox introduced but marketed themselves more aggressively

That's one way of putting it. But I think the article indirectly explained it better:

> We know that binding IE with Windows is the key to IE’s domination of the browser market.

Google did a similar thing with Chrome. They bound it to Google's services. You got nagged every time you visited www.google.com. Understandably google's more advanced stuff didn't work with IE. But it also often didn't work with Firefox. This is what mozilla had to say about that https://www.zdnet.com/article/former-mozilla-exec-google-has... :

    A former high-ranking Mozilla executive has accused Google of intentionally and systematically sabotaging Firefox over the past decade in order to boost Chrome's adoption.

    He is not the first Firefox team member to come forward and make such accusations in the past eight months; however, his allegations span far beyond current events and accuse Google of carrying out a coordinated plan that involved introducing small bugs on its sites that would only manifest for Firefox users.

Large enterprises, medical organizations (think hospitals etc), governments (think smaller/local governments), and the elderly/tech un-savvy are very much on old Windows versions (e.g. XP) to this day. They don't even realize there are other browsers, or in the case of enterprise/medical, the cost and risk of updating is too great and requires vetting a bunch of stuff, which costs money and time.

Defaults are powerful, and IE was default.

If you’re a large bureaucracy in the early 2000s you went with whatever was default - and that was IE.

Defaults were seen as so powerful, that MS was anti-trusted for this. They did it anyways at the time because the juice was worth the squeeze.

I still do not understand how Apple gets away doing the same (or worse) with Safari on iOS.

Because Microsoft, Google and Apple are the fig leaf for each other. No Monopoly in sight.

Microsoft had to keep Apple alive to be their fig leaf, Google keeps Mozilla alive to be their fig leaf. Apple is not as dominant as Microsoft was.

> Second, why did IE lose to Chrome from the position of market domination while Windows has always been dominating the desktop OS market?

> Now I realized that the answers to the two questions are the same

Yes - Android and the iPhone.

The only real mobile browsers are Chrome and Safari, and Safari is only available within its walled garden, so to get a mobile-like development experience on desktop people used Chrome. It all followed from there.

(IE6 was Windows-only and tried to achieve lock-in with ActiveX, which let you run native COM controls in the browser. Hilariously inappropriate security model, but popular in the enterprise, and at one time mandated by the South Korean national ID system.)

It's a niche, but Firefox runs just fine on Android.

And it runs full featured ublock on Android giving me an ad-free web experience.

When Chrome launched, its rendering speed was in a league of its own, delivering immediate results without the need to repeatedly click a link for faster loading. The design was clean and intuitive compared to alternatives such as Firebug and Fiddler the HTTP debugging proxy.

Such things rarely have a single succinct answer.

In reality, there's other factors:

- Legacy applications where nobody wanted to rewrite them to work with non-IE

- Chrome gained market share because the entire user experience was better. That includes updates, but also it was just really snappy to use. Speed is probably the bigger factor; normal users don't care about updates.

- As some have pointed out, lots of less technical users just use the default

- Did I mention those couple decades worth of legacy apps? It bears repeating, since MSFT had successfully kept other browsers out of market share through their 1-2 punch of strategic browser incompatibility and shipping with the OS.

IE being the default also meant that malware and viruses targeted it, and VBA scripting in Office.

Chrome started out not being targeted by these, and in addition, Google pushed security patches pretty quickly.

Anyone got any numbers on the frequency of exploits, and the mean time-to-patch, for IE vs Chrome, over time?

Why did IE last so long and refuse to die? Simple:

    <meta http-equiv="X-UA-Compatible" content="IE=7”>

During the oughts a lot of large companies had a lot of applications written using Microsoft DHTML “behaviors” and other specific tech. At an insurance company I worked on an app in the late teens that required IE7 and moving that to Chrome/Firefox was a 4 developers for 6 months effort. Which was actually really good since no one thought it could be done.

Many of these companies “standardized” on ie11 and wouldn’t even install chrome or Firefox. That’s why IE lasted so long.

I think people forget how much query played a role in peeling the fingers of IE’s death grip away as well.

Better alternatives plus compatibility layers.

There are lots of antiquated techs that deserve to be relegated to the dustbin of history, but I feel like some people have done jquery dirty by pretending that we never needed it in the first place.

jQuery?

Looks like autocorrect got one but not the second, yes.

Browser in those days, is not a good business. IE6 sticks years no updates, watching Firefox or Chrome grab the market, because Microsoft do not gain 1 cent from browser software, they were waiting for Google's bad luck. Until a decade later, everybody saw browser's value.

Browser in any period is not a good business.

I don"t think any business has made money on building and supplying a browser.

So they have to get the money back from somewhere.

Google from getting you to use other google things and getting more advertising revenue.

MS by making you use Windows (yes IE was available for the Mac but I think that was to keep Apple afloat to stop anti-trust problems).

Apple first to get out of IE and then to get a browser on iOS and then not have to use Chrome and now get paid by Google to use its search engine. Possibly blocking Flash also was a reason.

Netscape - well that is the problem where do they get revenue - Google paying them to be the default search and probably to head off anti-trust.

Opera?

Vivaldi - from payments from search engines to use them and from clients why pay for bookmarks. https://vivaldi.com/blog/vivaldi-business-model/

I used OmniWeb on NeXT and OSX and did pay for that but they could not compete so not profitable.

I wonder if we will have to handle more outdated browsers soon, now that Chrome dropped support for Windows 7/8 and older Androids.

Windows 7 is already down to ~3% of Desktop browsers on Windows and Android 6 or older represent even less than that as a share of the Android version. 2-3 browsers out of 100 probably spells trouble for e.g. government sites but by the time these are outdated enough versions to cause real concern for normal sites it might even be down farther. When IE 8 was still 30% of the share it made sense to put the extra time into it, when legacy become so low the opportunity cost of holding the whole site years back or maintaining multiple versions is hard to justify.

Some companies simply stop advancing when they’ve “won”.

Reply on news.ycombinator.com