The last time Hackerfall tried to access this page, it returned a not found error. A cached version of the page is below, or clickhereto continue anyway

Lightspeed – a browser experiment | Verdi@Mozilla

I’ve been using Mozilla code since Mozilla was still the code name for an internal Netscape project to build the next generation browser suite.

I *tried* to use Netscape 6, and rapidly came to the conclusion it was only released because it had been so long since the effort started, and there was a feeling the project had to get *something* out the door, just to show there *was* development. It should never have been released.

I *did* happily use Netscape 7, and moved to the Mozilla Suite when AOL terminated the Mozilla project and it transitioned to an independent effort.

I somewhat reluctantly moved to Firefox when the decision was made to concentrate on it and the Suite would no longer get development.

For the most part, I’ve been pleased with Firefox, but my pleasure is beginning to fray. What I liked about Firefox was the architecture and configurability. It was built on Gecko. Gecko rendered HTML, CSS, JavaScript and XUL. What Gecko rendered didn’t have to be a browser, and you could theoretically implement a desktop in it. You could create themes to modify how Firefox looked, and extensions to modify and extend what Firefox *did*. Being *able* to do that is why I *use* Firefox instead of Chrome, Opera, or Safari.

But I’m getting the impression the Firefox devs are suffering the “echo chamber” effect. They all work together, and only talk to each other. “Hey! We can do *this!* Wouldn’t that be *cool*?” “Yeah! That would be cool! Let’s do it!”

NO. Cool is in the eye of the beholder. It doesn’t *matter* whether *you* think it’s cool. What *matters* is whether millions upon millions of Firefox *users* around the world think it’s cool, but you won’t know *that* unless you actually *talk* to some.

The current Horrible Example is Australis. I was able to get it to a point where I could use it, but it took doing. The biggest issue was removing the Addon Bar. I run a lot of addons – enough that imitating Chrome and putting their icons on the Nav bar doesn’t work. There simply isn’t enough room. I looked at the Australis design, and said “If I wanted something that looked like Chrome, I’d *use* Chrome!” Fortunately, I found an addon or two that restored the Addon Bar and let me tweak Australis in desirable ways, but it says bad things that I had to use third-party code to do it. Whether to have an Addon Bar and ability to tweak the interface should be part of Firefox. (And it only worked for me because I’m an experienced long-time Firefox user who has spent time popping the hood and fiddling, and knows how to do stuff like that. Heaven help a newer user who hasn’t developed that expertise.)

And Australis got shoved down *everybody’s* throat, like it or not. If anybody in Mozilla actually polled *users* to see what they thought of the changes, I never heard about it. Make all the changes under the hood that you like, but tread *very* carefully on UI changes, announce them well in advance, and where possible, make them *optional*. Unexpected UI changes are *unpleasant* surprises, and you want user surprises to be pleasant. “But it’s *better*!” Not to the user, it isn’t, and it’s the user’s opinion that matters.

In UIs, one size *doesn’t* fit all. Windows 8 should have demonstrated that to the world.

And Firefox has gotten steadily larger and slower. One of the machines I have it on is an ancient notebook with a <1ghz CPU,slow IDE4 HD, and a whopping 256MB of RAM. It runs Linux, but I don't even *try* to run Firefox. It takes 45 seconds simply to load and initialize, in a *very* minimal configuration with only a few addons, and is perceptibly slow once up, with noticeable delays trying to do things like type into a text box. A current version of Chrome will actually run sort of acceptably. Something is *very* wrong there. (Yes, it's an ancient very low end box, and the vast majority of users will have something more current, but why does Chrome perform so much better?)

I was a bit bemused when the Firefox OS came out. Mozilla has only so many development resources, and the more it tries to do, the more it risks taking its eye off important balls. As a practical example, where does Mozilla funding come from? What happens to its revenues and ability to do the work it does if users leave Firefox in droves because they don't like the directions it's taking? Nothing good, but I think Mozilla is at risk of that.

Firefox is Mozilla's crown jewel, and should be treated like it. Take a leaf from the Hippocratic Oath: "Primum non nocere." (First, do no harm.) Changes should be good ones. Surprises should be pleasant. Performance should *improve*. Any other way lies madness and failure.

I'm all in favor of research, and using things like this to develop new approaches. I'm all *opposed* to foisting it on people just because someone at Mozilla thinks it's a good idea. Develop this as a separate product if you must. Don't make Firefox into it. I am far from the only one who would drop Mozilla code period if you do.

Continue reading on blog.mozilla.org