Memoirs from the Browser Wars (2003) | Hacker News Reader

Memoirs from the Browser Wars (2003)

ericsink.com

92 pointsluu26 comments

This is a fascinating read, but this one line stuck out to me:

> Scott told me that the IE team had over 1,000 people

What on earth could they possibly have had those thousand people doing?

It is indeed absolutely mind-boggling when you consider that MS later did next to nothing on the browser for 5 years.

Hell, Apple forked KHTML in 2001, a few months before the release of IE6, and had 5 whole years to develop Safari (and WebKit) with essentially no significant (Sorry Opera) commercial competition. In fact, between 2001 and late 2004 (FF 1.0) no Netscape progeny had significant share, so Microsoft had no real competition of any sort. If MS had kept their foot on the gas we may never have seen an opportunity for Chrome to dominate.

They were basically trying to figure out how to "integrate" the browser into Windows. But there were vast amounts of technology based around or that used ActiveX - ADO, ASP, ASF, you name it - Microsoft created it. Besides that, there were a lot of standards they were integrating.

At some points, Microsoft had two teams working on different versions of IE, in order to leapfrog Netscape. The extra team developed the "componentized" version where the browser was built from re-usable objects, where Netscape's code was still a rambling mess. (This helped Microsoft win the AOL business: it used IE even after it bought Netscape.)

The US DoJ's Janet Reno was a major factor. She forced Microsoft to sign a Consent Decree (1995?) that prevented Microsoft from tying separate products to Windows but allowed it to provide integrated improvements. The DoJ followed up with an anti-trust suit (on Netscape's behalf) over browser tying.

Microsoft won the browser case by 2-1 on appeal. However, all this bloodletting may have been a factor in Microsoft putting off the next browser, which eventually appeared in Vista. Which was late, following the Longhorn debacle, which prolonged XP's life, which enabled IE6 to get entrenched, which caused web designers so much pain.

The world might be a better place if the DoJ hadn't interfered, but we'll never know....

If you haven't already, you should definitely read "How Microsoft Lost The API War" (http://www.joelonsoftware.com/articles/APIWar.html) by Joel Spolsky. One of my favorite articles of all time -- it speaks to some of the things you raised... specifically how Microsoft intentionally let IE6 development come to a halt because it was no longer strategically beneficial to them (not because the DoJ interference made them stop). Also hints at the idea that the stability of IE6 for a while actually created a good environment for innovation in the web app space to take place because things we're changing all the time.

Great point, I'd missed that. Even though I'm a huge Spolsky fan, read the whole website, bought the book, interviewed the man etc ;-)

Otherwise, I'd have picked his first Bill Gates review as one of my favorite articles of all time...

IE6 had a lot of features, many of which are obscure now, but which took a long time to be replicated by the standards. Consider "zoom" (like a subset of CSS transforms), "filter" (a combination of CSS transforms, text-shadow, box-shadow, and CSS filters), page transitions (like CSS transitions, sorta), colored scrollbars, and so on. There were also random features like smart tags that are best forgotten…

Not to mention Dynamic HTML Behavior Components, that let you implement COM components in JavaScript.

Querying each other's interfaces and aggregating with the controlling IUnknown.

I expect 990 of them were supervising the 10 that did all the work.

But Netscape was running at a much faster pace. They got ahead of us on features and they began to give their browser away at no cost to end users. This made Netscape the standard by which all other browsers were judged. If our browser didn't render something exactly like Netscape, it was considered a bug. I hated fixing our browser to make it bug-compatible with Netscape even though we had already coded it to "the standard". Life's not fair sometimes. :-)

Everyone who's not the current marketshare leader goes through this feeling. And it's changed quite a few times.

I was stunned. That was 50 times the size of the Spyglass browser team. It was almost as many people as Netscape had in their whole company. I could have written the rest of the history of web browsers on that day -- no other outcomes were possible....Looking back on the browser wars, Tim Krauskopf remarked that we had beaten everybody who didn't outspend us by a factor of five. :-)

Pretty much describes what Chrome did to the competition.

My view is that Chrome won because they provided the best developer tools, so developers began building with Chrome in mind.

and good timing -- chrome took off right as it became apparent how shitty firefox had become (slow, crashy, memory pig -- and before you say anything, that's w/ no plugins but flash and no flash running). I tried chrome, was entranced by a browser that didn't suck, and quickly switched.

One thing I find curious is how in the early days right after NCSA Mosaic, "Mosaic" became a generic term for browsers - We got Netscape Mosaic and Spyglass Mosaic. NCSA quickly asserted their name, but it would be interesting how we could have had an alternative universe where we were using Mosaic Safari and Google Mosaic :)

Internet Explorer up to version 6 contained a copyright note about containing Spyglass Mosaic code in the about dialog. Obscure layout glitches and behaviours from Mosaic were definitely in IE up to version 8 (e.g. mouse pointer related code). Many things (UI, seperation of code into libraries, API) of IE 11 trace back to IE 3.

There's a commonly repeated story that Microsoft's licensing of Spyglass Mosiac required them to pay Spyglass for every copy of IE sold, which of course ended up being nothing. Not sure how true that is.

Oh, here's a wikipedia link with a cite: "By including it free of charge on their operating system, they did not have to pay royalties to Spyglass Inc, resulting in a lawsuit and a US$8 million settlement on January 22, 1997." [1]

[1] https://en.wikipedia.org/wiki/Internet_Explorer_versions#Ear...

It's very bittersweet. Not foreseeing that the browser would ever be given away for free screwed them twice. In their IE licensing, and up against Netscape.

> I could have written the rest of the history of web browsers on that day -- no other outcomes were possible.

It's really amazing how much things can change. Eric wrote this in 2003, when IE was dominant by a ridiculous margin and everything else was a tiny niche. It was written from the perspective that IE had won the browser wars forever.

Now, of course, IE has fallen from grace hard, and it's all about browsers based on WebKit, Blink, and Gecko.

There are a number of things you can thank for that. One is MS sitting on their browser for 5 years and doing nothing with it while everyone else kept innovating. Another, probably more important one is the rise of mobile. Mobile IE only ran on WinMo, when the early history of smartphones was dominated by Symbian and its WebKit-based browser, and then iOS and Android took over with their own WebKit-based browsers...

sitting on their browser for 5 years and doing nothing with it

This was a consequence of the backfiring of the "embrace, extend, extinguish" strategy and the enforcement of the consent decree. IE could no longer be a product distributed free with Windows; it had to be part of the OS, because that's what they'd argued in court. So nothing could change until the release of Vista.

Remember, IE was reactive (in the strategic sense). Microsoft didn't really want rich portable web applications to become a thing, they much preferred businesses to tie their internal processes to VB client-server apps or implement their functionality as ActiveX controls. A lot of them did and ended up stuck to IE6 for a decade.

http://en.wikipedia.org/wiki/History_of_Internet_Explorer

Well, I remember at some point late in Vista's development, a bunch of news articles appeared about how Microsoft was putting the long-ago-disbanded IE team back together in response to Firefox's growing market share.

It's not like they were developing IE7 alongside Vista all those years but couldn't release it until the rest of Vista was ready.

I'm curious about whether Eric and JWZ have ever sat down over a few beers and compared notes...

"I was asked to be the primary technical contact for Microsoft and their effort to integrate our browser into Windows 95. I went to Redmond and worked there for a couple of weeks as part of the "Chicago" team. It was fun, but weird. They gave me my own office. At dinner time, everyone went to the cafeteria for food and then went back to work. On my first night, I went back to my hotel at 11:30pm. I was one of the first to leave."

Wow. Being old as I am now, I'd certainly avoid to work for such a company without some immense compensation.

spyglass mosaic was one of the first browsers i used; even back then, i remember thinking spyglass was a peculiar name..

The original business for Spyglass was scientific visualization tools. When we did our "pivot" into web browsers, the company name stayed the same.