Firefox 3.5 Reviewed; Draws Praise For HTML5, Speed

johndmartiniii writes "Farhad Manjoo has a review of Firefox 3.5 at Slate.com this week. From the article: 'Lately I've been worried about Firefox. Ever since its debut in 2004, the open-source Web browser has won acclaim for its speed, stability, and customizability. It eventually captured nearly a quarter of the market, an astonishing achievement for a project run by a nonprofit foundation. But recently Firefox seemed to go soft.' The worried tone in the beginning of the review gives way to excitement over the HTML5 features being implemented, saying that thus far Firefox 3.5 'offers the best implementation of the standard — and because it's the second-most-popular Web browser in the world, the new release is sure to prompt Web designers to create pages tailored to the Web's new language.'" The final version could be here at any time; Firefox 3.5 is still shown as a release candidate at Mozilla's home page. Update: 06/30 15:31 GMT by T : No longer marked as RC; the Firefox upgrade page now says 3.5 has arrived.

Comments

As usual with new Firefox releases...

[Daetrin]

The main thing i want to know is if they've (finally) fixed the memory issues yet. Namely, if i keep a lot of tabs open for awhile (yes, i know, bad habit) and then close those tabs, will Firefox free up the memory (frequently over a gig of it) without requiring me to shut it down and restart it?

Re:As usual with new Firefox releases...

[A12m0v]

Firefox really needs a multiprocess architecture.

Re:As usual with new Firefox releases...

[oldspewey]

if i keep a lot of tabs open for awhile (yes, i know, bad habit)

This is a bad habit? I've always just thought of it as a convenient way to browse.

Re:As usual with new Firefox releases...

[Jorkapp]

They fixed most of FF's memory issues with FF3. I've been using 3.5 since beta 1, and I've never had any issues with memory.

Re:As usual with new Firefox releases...

[Alphager]

Or about stopping the auto-update. I use yum to install firefox automatically, then about 4 hours later I get message telling me that "Congratulations, you have firefox 3.0.11 installed", which breaks Google Streetview - it just remains black and no options actually appear in the Preferences->Clear Private Data popup. Reinstall Firefox using yum install, Google Streetview works again, and the cycle repeats.

How is this a Firefox-Issue? open a Bug with your distro to set the updates off. And turn off automatic updates in the preferrences.

Re:As usual with new Firefox releases...

[Darkness404]

I'm not sure if multiprocess architectures are really the way to go. Yes, they stop memory leaks but still take up more memory in the short run. Firefox is used heavily on older computers where IE doesn't cut it and the max memory is 512 MB or less.

Re:As usual with new Firefox releases...

[Daetrin]

Are you including virtual memory in that figure? I can't seem to fun FF without at least 100MB of physical memory, but I never see the sum of physical and virtual go over 600MB (Jesus! I have really lowered my expectations thinking that isn't a lot!) with 15 tabs open for a week.

I'm using version 3.0.11. I currently have three windows open with about 120 tabs between them. Process Explorer reports that the firefox.exe process has 585,384k private bytes and 689,916k virtual bytes. Over the next couple days the amount of memory consumed will continue to grow, probably until it hits around 1.5 gigs of private bytes. I know that i really shouldn't have that many tabs open, but as someone else pointed out it's a convenient bad habit. (Perhaps a quarter of those tabs are sites that i check and refresh fairly often, at least once a day. The rest are sites links that i've checked or the results of google searches that i either haven't finished reading yet or think i'll need to reference back to in the near future. (For example, over 30 of those tabs relate to the myriad of issues i've run into trying to get Oracle working through ADO.Net, and i'll need to keep a lot of them open for reference until this project actually works correctly.)

It's not that i mind Firefox taking up a lot of memory when i have a lot of tabs open (although almost 5 megs a page already seems a little high, though not as bad as your 40 megs per tab!) but i do mind that when i notice my computer slowing down and see that Firefox has consumed somewhere between 1 and 1.5 gigs of physical memory that doing a pruning of the tabs gets me almost no memory back. I have in fact closed everything down except for one google tab left on one window of Firefox and seen it still consuming over a gig of memory.

Re:As usual with new Firefox releases...

[Da Fokka]

... Namely, if i keep a lot of tabs open for awhile (yes, i know, bad habit)

Why is it a bad habit? The browser should facilitate the user, it shouldn't be the other way around.

Re:As usual with new Firefox releases...

[koreaman]

Er... Multi-process architecture is "a pile of shit" because you can't tell which process is which in your task manager?

That doesn't seem like a difficult problem to fix, and is hardly a fundamental problem.

Re:As usual with new Firefox releases...

[Anonymous Coward]

1.5GB looks like much, but it's only 12.5MB per tab. Considering that the browser has to keep the state and source data of every page, it doesn't seem excessive. Are you sure that the pages aren't running scripts which accumulate stale data over the course of days and weeks, because the programmers never expected their scripts to run for that long and didn't include any cleanup code, because that's usually handled by the browser when you leave the page or close the tab (which you never do)?

Re:As usual with new Firefox releases...

[ckaminski]

Multi-process != slow. What it equals is process isolation. If you global/static memory footprint is kept to a minimum, and you have fork() style semantics, your total overall memory usage is not significantly greater than if you were using threading instead.

Re:As usual with new Firefox releases...

[ckaminski]

Not true if your platform properly supports fork() semantics. You only make copies of data in child processes that you change - unchanged code pages are shared amongst all children.

What about Slashdot?

[macbeth66]

the new release is sure to prompt Web designers to create pages tailored to the Web's new language

Although, I would be happy if Slashdot would work right with the existing standards.

Re:HTML 5 and Javascript

[Tom9729]

Without installing any kind of plugin JavaScript is supported by virtually every modern desktop browser and a growing number of mobile browsers. Yes some websites use JavaScript to do annoying things like resize/move windows, but most browsers let you limit what a website is allowed to do.

Umm Flash on the other hand requires you to install a 3rd party plugin that may not work well (or at all) depending on what platform/browser you use.

IIRC the HTML 5 spec doesn't even say that JavaScript is "required" to play videos, it's just used for the UI.

Re:Softpedia claims to have it already

[pdboddy]

Do not download FF from anywhere except the main FF site. :P And FF3.5 is up now for download. Pretty snappy, speed-wise. :)

Re:"A nonprofit foundation"

[christopherfinke]

"Nonprofit" doesn't imply a lack of revenue.

Competition

[koreaman]

Am I the only one who doesn't see the multiplicity of real competition as a threat, but rather as the greatest success of the Mozilla Foundation? Had it not been for Firefox, Opera would still cost money, Google Chrome wouldn't exist, a few people who paid way too much for their computers would be running Safari, and most (l)users would be stuck with the latest version of IE -- IE6. Thank you, Firefox, for reigniting the browser wars, and here's hoping that this time around the wars will be fought with functionality, stability, security, and speed, rather than with a new incompatible extension to JavaScript every week.

Isn't this a little overkill?

[Xaedalus]

Please excuse me for being a paltry light user of Firefox... but aren't you an outlier in this particular case? The most tabs I ever have open on Firefox is three, maybe four. IMHO, you're a power user and while your comments are insightful, I have to wonder whether or not your insights are of relevance to the average user of Firefox? I'm all for improvement, but if the improvement is only noticeable when you've got 30+tabs open a day and are burning through close to a gig of RAM to keep everything operating... then what good is the improvement to the average user?

Re:Isn't this a little overkill?

[bemymonkey]

I tend to agree with this assessment. I consider myself a power user when it comes to tabs, and I only rarely have more than 20 tabs open (and that's when I haven't checked Slashdot for 2 days and need to read every article/summary/comment I've missed), and then only for a short time. Do a lot of people really leave the browser running _all_ the time with dozens and dozens of tabs open? I can't really imagine that being the norm...

Re:Isn't this a little overkill?

[SpinyNorman]

That guys usage does seem a bit strange, but each to their own!

However, there are different use cases for tabs that result in different numbers of tabs open:

e.g.

1) User wants to keep a number of sites open all day for quick switching - e.g. mail, news, documentation, etc. This is probably what you are thinking of, and it most likely is a relatively small number of sites - less than a dozen, say.

2) User wants to open a whole bunch of links off the sam page at once, since that's more convenient than flipping back and fro to open one read it, open the next, etc. Examples of this usage would be, for example opening search results (search engine, eBay, etc) or reading new posts on an online forum.

I routinely (many times a day - searching for collectables - made a $1500 profit on one yesterday) open 40-50 tabs at a time for eBay search results since it's so much faster to quickly run down a list of links doing open-in-new-tab (in background) then reading/discarding them, as opposed to doing it one at a time.

Re:Non-profit?

[hedwards]

It's not technically true, it's completely true. Don't the Mozilla foundation have tax exempt status?

All non-profits have to make enough money in one way or another to fund operations, after all people can't generally afford to work for free and suppliers expect compensation for their supplies. Google would be in a world of hurt if it clamped down on Mozilla as that would definitely trigger a swift DoJ investigation into anti-trust violations.

Re:Non-profit?

[koreaman]

Microsoft may have billions of dollars, but I'd be interested to see how much of that went to the IE budget pre-Firefox. I tend to think not much, given the quality of IE6. Don't get me wrong, I love Firefox, and use it whenever I need to go to a site Elinks can't handle, but I don't think it's all that "amazing" that a project set up by Netscape and funded by Google could compete with a project that Microsoft had abandoned. Especially since the Opera devs probably worked in similar (or worse) conditions, and Opera has been better than Firefox for years. (TFA's author might claim that this is less amazing because Opera is for-profit, and I think that's silly.)

Re:Explain to me again why this is not Evil

[maxume]

Mozilla (and Netscape before it) have long implemented things that are not in standards. This isn't what causes problems. What causes problems is not supporting the standard after it is released.

Really, gathering real world information about how an idea works is a valuable input to the standards process.