Q&A with Firefox's Blake Ross

dotlin writes to tell us the Seattle PI is running a lengthy and interesting interview with Firefox's Blake Ross. In the interview Ross addresses many of the issues surrounding the future of Firefox including their attempt to streamline Firefox in 2.0, the feature comparison between Firefox and IE, different ways of measuring browser market share, and many more.

Comments

Re:old news

[LiquidCoooled]

Considering how slashdot is designed to cross post, I don't see how this can be avoided.
After all submissions are made based upon what users of websites find, so its inevitable that some of those sites are on your bookmarks list.

As it happens I read 2/3 of the sites you listed, but hadn't read this interview so slash is doing its job.

Pesky users

[PinkyDead]

Fortunately we've become kind of accustomed to complaints from the blogosphere and from geeks, which is generally where this line of pressure comes from. We're pretty good at picking out the points that are important to us and really just letting the others go.

Having wombled around the Firefox support site for awhile looking for answers to memory issues, I came to the conclusion that there was a certain level of disinterest in problems that were less than exicting to fix; more so, than other OSS projects. (I fully accept the subjectiveness)

This snippet sort of ties in with this feeling.

Sure, OSS developers can do what they like - I'm not paying them so I don't have much right to complain, fair enough.

But if you want to compete against MS, who are too customer focused then maybe a balance needs to be found which doesn't involve letting so many go.

Re:Pesky users

[bluebox_rob]

Well the paragraph before the one you quoted seems to say the exact opposite, i.e. less emphasis on whizz-bang new features and more tuning under the bonnet:

It looks like the 1.0 release because most of the work that has been going on has been to make it more stable, how do we fix the memory problems that people are complaining about, how do we make everyday tasks easier

I'd say they're heading in the right direction...

Re:Pesky users

[Anonymous Coward]

Sometimes the user is wrong. One of the strengths of Open Source is that the developers can do the right thing because they don't have to answer to their users. Of course, sometimes the developers are wrong and the users are right, so it's always wise to at least listen to your users' complaints, but if you know that a certain complaint is a misconception, it is better to clear up that misconception than to try and fix something which isn't broken. Some people will not listen when you try and clear up the misconception. As a developer you simply have to let these people go.

Re:Pesky users

[CastrTroy]

It's because the memory issues don't actually exist with Firefox. First there's 2 kinds of memory issues. 1 is the problem with the feature that has it store the pages in memory so that the history feature works better. This isn't a bug, and isn't a memory leak. The second is users complaining that Firefox takes up 700+ MB of RAM. I don't know what kind of funky extension these people are running, but I've never seen that happen. I've had firefox running for days at a time without seeing anywhere over 100 MB. I rarely ever see it go over 75 MB. Then again, I haven't kept it open for months at a time. Maybe if I did, then I may see problems. Then again, its a web browser. You can turn it off once in a while. Get one of those session saving extensions if you don't want to lose all the tabs you have open. They have more important things to fix, like trying to make sure it's CSS/Other standards compatible, and ensuring that there are no security holes.

Re:Pesky users

[FireFury03]

This isn't a bug, and isn't a memory leak.

Well, it might not be a memory leak, but I'd argue that it is a bug. If I leave my FireFox pointing at a auto-refreshing page for a couple of days it *will* OOM my machine. Whether or not that's a memory leak, I'd argue that causing the OOM killer to come out and start blowing away applications is a bug. Now I understand that this memory is supposidly used to cache content to speed up the browsing experience but I'd counter that argument by pointing out that if FireFox is so deep into swap space that it causes my machine to go on holiday for 5 minutes every time I do something because it's thrashing the swap then this isn't speeding up anything.

I've had firefox running for days at a time without seeing anywhere over 100 MB. I rarely ever see it go over 75 MB. Then again, I haven't kept it open for months at a time. Maybe if I did, then I may see problems.

I never close my FireFox unless I absolutely have to. Currently it's using about 281MB:
    PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
  1934 steve 15 0 281m 94m 8624 S 0.0 12.5 191:18.84 firefox-bin
(Yes, I know this includes mmap()ed resources, but I doubt FireFox is mmap()ing much huge stuff).

Then again, its a web browser. You can turn it off once in a while.

That's not really an excuse though is it... Hey, no need to fix memory leaks in Windows, it's only an OS, you can reboot it every so often... :) Shutting stuff down in order to work around a bug is a horrible and very annoying kludge.

Re:Pesky users

[aaronl]

There are two times that *I* am not running a browser, and most people I know are in the same boat. The two times are: rebooting the machine or Firefox crashed.

Also, if you leave FF open for a while without using it, it takes a couple of minutes (really, *minutes*) before you can use it. The UI doesn't update, and the machine is thrashing. This has gotten better over the last few releases, but it still happens, particularly under Windows.

Crashes occur most commonly for me because of Java, Flash, Acrobat, or the Flashblock extension. Java pretty consistently screws up the browser whenever it's started, as does Acrobat Reader. Flash is a random thing, so I'm not certain of how to reproduce it, but sometimes it will trash FF. Flashblock has a memory leak in its JS code. The biggest memory leaks seem to happen by just using the browser, though. Right now, on my Ubuntu box, it's using 256MB total virtual, and 154MB actual RAM.

As far as why you might do this... well, often you are looking at something, and have to leave. People that go and do thing that aren't sitting at the computer 24/7 often have this happen. I doubt even the most hardcore anti-social type on this site never leaves their computer. It's silly to clog up your bookmarks with something you only wanted to finish looking at, too. So, as you're going through things, you pop up a window, minimize it, and come back later. Later might be in a day or two.

Another reason might be that you're reading documentation, and haven't finished what you're doing. This is also rather common. You might have a few sets of documentation that have to be gone through to finish something, so you may even have *several* windows open, that you need to go through.

There are tons of reasons why you might leave the browser running. People leave their IM client open, a lot leave their email client open, you leave your UI open, and your OS running. Why should we have to restart our web browser when we don't restart any of those?

Re:Pesky users

[Sloppy]

Sure, if you leave your program on for days it will be using a lot of memory and you'll eventually have to shut it down and restart, but why leave it on for a couple of days?

Who knows? Who cares? A user can leave non-buggy apps running for weeks without them misbahaving, whether or not I know the user's reason for doing that. Users are unhappy that Firefox doesn't behave like a non-buggy app.

Re:Pesky users

[aaronl]

Considering that all of my machines have 1GB, it isn't that. I have free memory, and I disabled the previous page cache. It's a well known flaw in Firefox.

Re:Pesky users

[killjoe]

"But if you want to compete against MS, who are too customer focused then maybe a balance needs to be found which doesn't involve letting so many go."

If you read TFA you would have read the part about where MS abandoned IE for years. Where it didn't give a flying fuck about what was happening to their customers in terms of security and the features their customers wanted.

Do you really think MS track record with IE is better then firefox? If so you need to get educated. MS abandoned this project while their customers were suffering from a relentless assault of security problems.

Know thy enemy.

[Volanin]

Quoting from the article:

The truth is that [Internet Explorer 7] actually looks pretty good. People don't expect me to say that, they expect me to say that it's terrible [...] I think that it's a solid product, but I think that by the time it comes out, we're going to be another world ahead of them again, so I think it's kind of a step or two behind us.

And quoting The Art of War from Sun Tsu:

So it is said that if you know your enemies and know yourself, you will not be imperiled in a hundred battles; if you do not know your enemies but do know yourself, you will win one and lose one; if you do not know your enemies nor yourself, you will be imperiled in every single battle.

I, for one, have pleasure being in the Firefox side of this "war".
And it's relieving to know that Blake seems to have a very clear sight while leading this.

at least

[RockModeNick]

I'm personally more than greatful for firefox, because back in the day, netscape sucked so bad I actually really LIKED IE.

Re:Seamonkey vs. Firefox/Thunderbird

[Oxide]

What's funny about what you mentioned is that when firefox orignally started, it was supposed to be the light weight version of Mozilla. Now you're saying it is a bigger memory hog than Mozilla or seamonkey!!

Maybe now we need a light weight version of the light weight version of Mozilla :)

Re:Forms, Textboxes and the Cache

[NickFitz]

I would also like to know when firefox is going to allow its users to turn off the cursor going to the first text box on a page by default. It is a security risk.

That's not Firefox; at least, I never see this behaviour on any of the various versions I run on Mac , Windows or Knoppix. It sounds like the owner of the relevant page has stuck a bit of JavaScript in there to do this; that's the only way I've ever seen this behaviour implemented on any browser. Complain to the owner of the site(s) where you see this (as you say) risky behaviour.

I think people see Google doing this and think "Oh, it must be OK, Google do it". They are morons, because behaviour that enhances usability on Google's home page (where one wants to type in a search query, otherwise one wouldn't be there) can, as in the case you cite, actually detract from usability in other circumstances.

(I suppose the culprit might also be an extension: people have been known to dump irrelevant and unnecessary "cool" features like this in them, too.)

Re:Seamonkey vs. Firefox/Thunderbird

[CTho9305]

See, reports like this are why developers get jaded. Hundreds of people say, "use the browser for 30 minutes". Developers use the browser for days and don't experience the problem. Now what? The user is generally either unable or unwilling to get into the nitty gritty of real leak hunting, so nothing can be done. The developer gets frustrated, wondering where this problem is that he can't find.

Fortunately, David Baron wrote the Leak Monitor extension, that looks for a relatively common type of leak, which you can install, but it doesn't catch everything.

Re:Seamonkey vs. Firefox/Thunderbird

[knarf]

Been there, done that:

Firefox' development strategy being what it is there is not that much opportunity - other than by filing bugs - to influence priorities and design criteria.

The bug reports are there. They just happen to hang around a long time... Often the memory problems are blamed on:

• extensions (but as I already said I use a clean profile (without extensions) for testing purposes)
• 'it is not a bug but a feature to make your browser faster'. It doesn't make it faster if it thrashes the cache...
• 'I never see the problem on my $_box with $_memory and $_tabs open'. Good for you. Others are less lucky.
• 'use the leak detector extension'. I do sometimes just to see what it warns about. It obviously does not warn about the browser hogging memory when that is considered to be a feature.
• 'just use about:config to change the defaults'. If that is necessary the defaults should be changed, Firefox was intended as a browser for everyone - not just the about:configging /etc/sendmail.cf grokking crowd...

I really hope the memory problems (or features if you prefer) get sorted out as Firefox has quite some momentum behind it. It would be sad to see this momentum lost because of some (mis)feature eating PCs alive...