Google was able to build Chrome in-house faster and at lower cost than it would have been to rewrite Firefox from scratch under Mozilla governance. They would have had to convince Mozilla to do all of this stuff. https://www.google.com/googlebooks/chrome/index.html
The real advantage of WebKit (or KHTML) was that it was way, way, way easier to embed than Gecko. That made experimenting much easier.
> They would have had to convince Mozilla to do all of this stuff. https://www.google.com/googlebooks/chrome/index.html
People tend to forget that the Chrome team was initially a Firefox team paid by Google to work on "the chrome" (which has always been the name of the back-end of Firefox).
Legend has that Google decided to roll out their own browser when they realized that they could not arbitrarily land any feature in Firefox without a review from someone out of Google. More specifically, legend has it that Google prepared a large refactor of Firefox called "Places", reviewed it internally, then attempted to land this as a single patch to Firefox, without any public review. It is my understanding that this is how Google open-source typically works (big code drops), but definitely not how true open-source works (incremental changes). This was blocked by Mozilla until it could be split into smaller changes that could be reviewed and discussed individually. Places eventually landed (and new versions of Places are still part of Firefox to this day) but, says the legend, Google was pissed off and decided that they would be better off with their own, in-house, browser.
Chrome when released on September 2008 had the faster Javascript v8 engine. (Built by Lars Bak team in Denmark.) The improved performance was very noticeable on js heavy websites like Google Maps when scrolling the map and zooming in and out.
A paper[1] showing Chrome winning on various performance benchmarks matches my recollection of performance differences.
Also, Chrome's per-process isolation for each tab also made it more stable when dealing with buggy websites (bad javascript code leaking memory). Firefox would crash more often. The tradeoff was that Chrome used a lot more RAM than Firefox.
[I didn't downvote your comment.]
The V8 javascript engine was the real deal however. It forced Mozilla to significantly invest in its own javascript engine.
That's not entirely true. If memory serves me correctly, at the time V8 came out, Mozilla already had the TraceMonkey JIT implemented in Spidermonkey. However, the first version of Google Chrome (with V8) was released before the first version of Firefox with the new JIT.
IE shipped activex and jscript (MS scripting host) exploits first, too. Which made the whole tab isolation stuff useless.
Unfortunately for them I not at a single point has the difference been big enough to justify leaving Firefox behind.
Even today as broken as Firefox is compared to what it used to be it still outshines Chrome for my workloads it seemsm
If Firefox was broken broken all other browsers would be beyond broken IMO as none of the mainstream browsers really hold a candle to it for my workflows even today.
But compared to itself a few years ago the advantage over other browsers has diminished in every aspect I care about except maybe security.