Gran Paradiso Alpha 3

kbrosnan writes "Gran Paradiso Alpha 3 is a release of the Gecko rendering engine for testing purposes only. Here are the release notes. While this release uses the interface of Firefox, no significant interface changes have been made. These alpha releases focus on making improvements to the core elements: graphics, JavaScript, page rendering, etc."

Changes.

[Seumas]

no significant interface changes have been made. These alpha releases focus on making improvements to the core elements: graphics, JavaScript, page rendering, etc."

In other words, there have been no changes to anything that MSIE users care about, like 3d buttons, blocky, chunky, weird, chunky tabs and nifty "click" noises every time you click something, load something, go back a page, go forward a page, scroll down a page, select something, delete something, type something, submit something or refresh something?

Re:

[NekoXP]

Aww but I love the click in IE. Sometimes when I click in Firefox I wonder if it's actually DOING anything.. it does seem to sit there and churn a lot in the background. When I click in IE and it's locked up because of some dumb flash anim and not responding to my button press, it doesn't make any sound.

The click makes it very clear when the browser is sucking ass, and when it is not :)

Re:

[gerrysteele]

Perhaps someday soon they might invent some innovative technology to update you of these events.

I would call it "The Status Bar".

Re:

[644bd346996]

So the throbber and status bar are not enough? Submit a request for enhancement to bugzilla. And call it an accessibility option. It shouldn't be hard to add an optional click noise.

Re:

[BibelBiber]

IE users don't go for Alpha Software, do they? When you use MS Windows you expect fully functional software that always works as expected. I'll recommend to wait for Firefox 3 Final.

Release notes and comments

[FishWithAHammer]

* Animated PNG (APNG) images are now supported.
* The DOM clientLeft and clientTop attributes are now supported.
* Introduced support for , which puts resources into the browser's offline cache. This allows a web application to ensure that its resources are available in the cache when the browser goes into offline mode. See * * * Marking Resources for Offline Use for further details on offline support.
* Improved precision of layout and scaling across a wide range of screen and printer resolutions.
* Implemented cycle collection in XPCOM, which detects cases where two released objects hold one another, but neither is held by anyone else. In this scenario, both objects can safely be purged. Previously, the holds each has on the other would have prevented them from being purged.
* Added support for the HttpOnly cookie attribute, which marks a cookie as readable only by the server and not by client-side scripts.
* Added a new preference, "Warn me when web sites try to redirect or reload the page", which notifies the user when the page specifies HTTP-EQUIV=refresh.
* Windows 95, Windows NT 4, Windows 98, and Windows ME are not supported for Gecko 1.9.
* OS X 10.2 is no longer supported, and OS X 10.3.9 or better is required.
* The non-standard JavaScript Script object is no longer supported.
* Moving DOM nodes between documents now requires a call to importNode or adoptNode as per the DOM specification.

It's kind of sketchy that they're not supporting older Windows or OS X versions, but I don't think that's a huge deal. I wish they'd reintroduced MNG instead of APNG (purely a personal preference; APNG is probably actually a better way of doing it), and any fixes to JavaScript are nice to have.

Re:

[Frosty Piss]

It's kind of sketchy that they're not supporting older Windows or OS X versions

At a certain point, this sort of decision has to be made: Move forward, or live in the past. The technical issues that surround supporting old systems verses moving forward with more elegant solutions for modern systems.

But what I don't understand is why they continue to insist that there are no memory usage issues when there is a lot of practical evidence that there are?

Re:

[bunratty]

they continue to insist that there are no memory usage issues when there is a lot of practical evidence that there are?

Where do "they continue to insist that there are no memory usage issues" and where do you find "a lot of practical evidence that there are"? I find that Firefox does have some memory leaks, and they are being fully acknowledged by the developers. Do you think there are some memory leak reports that are being ignored? If so, you should be far more specific about what these are.

Re:

[Frosty Piss]

As for the memory usage, it seems that the answer is always "it's not a memory problem - that's just the result of how Firefox uses it's page caching features to provide fast recall!". I'm not sure that makes a lot of sense to me, because it strikes me as odd that any kind of browser "feature" would possibly consume 500mb or more memory on a regular basis.

But if a large number of people us FF in a way that causes problems, it is a design flaw. Either that, or they just don't care what their users want.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[Frosty Piss]

Don't you think users want the lesser of two evils?

I think users want Mozilla to build a browser that takes into account common usage patterns, and not respond with flip comments that it's being used wrong.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[Tim C]

But now that we are all used to tabs, and can open links in new tabs just as easily and quickly as in the current tab, does Joe Average really use the back button that much? (Honest question, as I have no idea)

Certainly I almost exclusively open links in new tabs unless I'm sure I'm done with the page I'm currently on, and with the number of comments I read here from people who describe having dozens of tabs open simultaneously, it seems I'm not the only one. I freely admit to not being Joe Average though,

Re:

[bunratty]

it strikes me as odd that any kind of browser "feature" would possibly consume 500mb or more memory on a regular basis.

Firefox using 500 MB of RAM sounds like a serious problem to me. Tell us how we can reproduce the problem so someone can write up a bug report and a developer can fix the problem.

Re:

[LoveMe2Times]

Open a bunch of tabs that use plugins. I haven't pinned it down exactly, but one time a couple weeks ago I had Firefox using 1.5 GB of RAM. Yes. GIGABYTES. I was stunned, frankly. I had a dozen tabs open, tops. I was surfing Adobe's website, Verisign's website, googled a few things, so there were a couple of Flash movies playing and a PDF loaded. One and a half freakin gigabytes. The amount of swapping caused by this resulted in much anomalous behavior, and I was forced to use task manager to termin

Re:

[Cid Highwind]

When you say "a couple of flash movies" do you mean "five or six 300MB TV episodes from Youtube"? If so, that would be most of your 1.5GB memory usage right there...

Re:

[LoveMe2Times]

Actually, the size of the flash file (.SWF or .FLV) doesn't matter. Those are cached to disk, not in memory. But no, as I mentioned I was on Adobe's website, which does part of their site in Flash, not surprisingly. I believe (without anything to back me up at the moment) that the problem comes from the Flash/ActionScript bridge, where no JS objects ever get released or somesuch. Tellingly, on the machine that reached the 1.5 GB barrier, I do not have FlashBlock installed, which exacerbates the problem.

Re:

[bunratty]

I think Acrobat Reader also leaks memory, at least when used as a plugin in a Gecko based browser, very badly, perhaps for the same reason.

An Introduction to R [r-project.org] (PDF file)
Opened in Firefox 2.0.0.3 with the Acrobat plugin, VM Size 14 MB
Opened in Opera 9.10 with the Acrobat plugin, VM Size 22 MB
After closing the tab with the PDF file, VM Size did not drop significantly in either browser. Still don't see any problem with memory in Firefox. Remember to open only that one page just after you start the browser

Re:

[bunratty]

Sorry, but "Open a bunch of tabs that use plugins." isn't exactly a reproducible set of steps. I open a bunch of tabs that use plugins all the time, and I have seen memory usage go up to 400 MB, but memory usage almost always goes back down to below 200 MB after I close the tabs. Don't all browsers use lots of memory when you open a bunch of tabs that use plugins, and then release the memory when you close the tabs?

If you still think that Firefox has memory problems that other browsers don't have, please

Re:Release notes and comments

[bunratty]

Just to make my vague, general statements concrete, I picked three sites at random, each of which uses a different plugin:
The official US time clock [time.gov] (Java)
weatcher.com interactive map [weather.com] (Flash)
Panda Pang [zeronews-fr.com] (Shockwave for Director)

With these three pages open Firefox 2.0.0.3 on Windows XP has a VM Size of 175 MB. Huge memory problem in Firefox? No, Opera 9.10 on Windows XP has a VM Size of 171 MB. After closing the tabs in Firefox, VM Size goes down to 46 MB. Doing the same in Opera, VM Size goes down to 59 MB. If anything, it looks like Opera may have a problem releasing unused memory. Keep in mind for a fair comparison that you must open only those sites after starting the browser, otherwise, you could see the built-up memory usage form hours or days of use in a browser that you've been visiting other pages in.

If you can come up with a series of steps that causes high memory usage in Firefox, and not high memory usage in other browsers, maybe you're on to something.

Re:

[bunratty]

To me that's a huge amount of memory for displaying a few kbs of data.

Perhaps, but all browsers use that much memory for displaying pages. Is there some problem specific to Firefox you're seeing?

Re:

[bunratty]

I don't understand. All browsers use about the same amount of memory, but only Firefox has the memory problem? You're going to have to explain that one pretty carefully.

Re:

[bunratty]

The point of Firefox used to be that it was lightweight. Now it isn't. This is a clear regression.

Did Firefox at some point use significantly less memory than other browsers? I suppose you can clearly demonstrate this regression? If so, I suppose you have a point. If not, I suggest that you're speculating and are possibly incorrect.

Re:

[bunratty]

So if another browser sucks it's ok for your browser to suck too?

No, of course not! But if all browsers suck, why say "Firefox sucks"? That implies other browsers don't suck. Why not say "all browsers suck," which conveys what you're trying to say much more clearly?

about:config

[SavvyPlayer]

browser.cache.memory.enable

Re:

[vidarh]

On my machine you just run it for a couple of days with lots of tabs. After a while closing the tabs will not make any difference at all in memory consumption.

I have a Linux box and a Macbook Pro running OSX, and it happens consistently on both of them, and have since 1.0.x at least.

Re:

[bunratty]

After a while closing the tabs will not make any difference at all in memory consumption.

According to the operating system, no it doesn't make any difference. That's because, due to memory fragmentation, memory is being released somewhere in Firefox's memory pool so that it cannot be returned to the operating system. Memory is being released and reused internally, but not back to the OS. It's a well-known effect that isn't a problem specific to Firefox. All browsers exhibit this behavior.

Generally speaki

Re:

[Nutria]

Open 10 tabs. Wait.

How long?

I've had 40 tabs open, and (in the v1.5 & 2.0 series) see the RES mem get up to 300m, but no higher.

Re:

[Nutria]

Nope. Debian Sid.

Re:

[bunratty]

Ok. I'll try this one, too. I've opened up Firefox 2.0.0.3 on Windows XP with just this page open, and it has a VM Size of 16 MB. After one reload, it's up to 22 MB. I notice that the ad changed, so memory cache and the Flash plugin are probably responsible for the extra memory, as is memory fragmentation. After another reload, it's down to 20 MB. After the third reload, still at 20 MB. Fourth reload, 24 MB. Fifth, 23 MB. After the 22nd reload, it's using 31 MB. So, the memory is going up over time, but the

Re:

[mboverload]

Uh, brainfart. 2000 still supported.

Re:

[Jeffrey Baker]

If you're happy with win2k you'll be happy with Firefox 2. Simple concept.

Re:

[kv9]

If you're happy with win2k you'll be happy with Firefox 2. Simple concept.

uh, no I will not be. FF3 is progress. anything after W2K is bloat. and FF3 will support W2K (RTFGP) so your point doesn't stand.

Re:

[neongrau]

i remember ppl saying the same about windows 98SE when w2k was released.
ppl like you prefer the eye-candyless w2k.
and i now hear ppl saying it about XP since vista is out (including me).
and i'm pretty sure ppl will say the same about vista once the successor is released.

so like someone earlier posted:
it's ok to live in the past

(for a while at least)
we don't want to come to a total halt in technology

Re:

[shawn443]

I don't remember 2K bothering me quite as much as XP. Everytime I see that balloon telling me I have unused desktop icons I get mad. Eye candy is useless to me unless it enhances my work. Otherwise, give me TWM [wikipedia.org]. That said, I am sitting down with a Vista Home Super Duper Bee's Knees Fucking Ultimate cd coupled with a Barnes and Noble style training session sometime this week. I am sure I am going to be mad.

Re:

[oddfox]

You'd be much less angry at Windows if you'd simply use Google or another resource to find out how to tell it to not do what you don't want it to.

Re:Release notes and comments

[mhall119]

It doesn't support older Windows because it uses Cairo [cairographics.org] for faster rendering, which I've read doesn't support older Windows versions. I'm not sure if it's the same reason older versions of OS X are not supported.

Re:

[Dh2000]

funny thing is, cairo is dog slow

Re:

[atamido]

Cairo is not a speed demon, but it has been getting noticeably faster with ever release that they've made. It may be slower than what Firefox is currently using, but it's also a lot more useful. Their current Gecko rendering engine doesn't have the capability to do the things that they want to do, without a significant rewrite. So in this sense Cairo will add a ton of ability, fix a lot of memory leaks, and probably be at least as fast in a few releases. They will also gain time from all of the developer

Re:

[mhall119]

Right, I think Cairo is being used specifically because it would allow Firefox to finally add features like full page zoom (no just text resizing) that has been available in other browsers for a while now. It should also improve page printing.

Re:

[dvice_null]

> It's kind of sketchy that they're not supporting older Windows or OS X versions

There is a reason for this. The changes they have made to the core won't work with older versions of Windows as easily as they do with more modern versions. This means that a lot of work would be need to add support for them. But it has been said that if anyone is interested doing the job, he/she is quite welcome to do it.

Bug!

[Zarel]

I had a feeling the rendering engine improvements would break something. The Quick Contacts list of GMail with Chat has a huge space on the bottom that increases each time you hover over a user. I wonder if it's a rendering engine bug or a GMail bug.

Re:

[Nasarius]

So file it [mozilla.org] if it's not there already. That's what the alpha is for.

Re:

[kbrosnan]

It is https://bugzilla.mozilla.org/show_bug.cgi?id=36349 6 [mozilla.org] - "gmail talk, onclick textarea creates new space in contacts list"

Re:

[G-funk]

Does it support inline-block yet FFS?

Re:

[daeg]

Yes, as of Alpha 2: http://www.mozilla.org/projects/firefox/3.0a2/rele asenotes/ [mozilla.org]

The inline-block and inline-table values of CSS 2.1's display property are now implemented.

Re:

[G-funk]

Beeeeeeeeeee-yoooo-diful

Re:

[Ant P.]

I couldn't care less about having another format for obnoxious animated ads, TBQH. What really pisses me off is that an actual useful format [wikipedia.org] got killed off for BS political reasons.

Not a gecko release

[El_Muerte_TDS]

It's an alpha release of Firefox 3, it uses the Gecko 1.9 engine.

Re:

[jesser]

Since it contains a lot of changes to Gecko but very few changes to the Firefox UI, I think it makes more sense to call it a "Gecko 1.9 alpha" than a "Firefox 3 alpha".

Re:

[Tim C]

Surely you've just agreed with the OP? It's an alpha release of Firefox 3.0 (which uses Gecko 1.9), not a release of Gecko 1.9 itself.

ACID 2 Compliance

[nahdude812]

Yes, it does completely pass the Acid 2 [webstandards.org] CSS compliance test.

Re:

[jez9999]

Aaaaactally, the nose on GPA3 appears 1px too far to the right and to the bottom. ;-P

Re:ACID 2 Compliance

[jesser]

It's ok if the nose is a pixel offset from the reference-rendering position. IIRC, it depends on the order in which borders are drawn, which isn't specified by CSS. It's even ok, and considered ideal by some, if the nose is anti-aliased to be "half a pixel offset" from the reference-rendering position.

Re:

[nahdude812]

Normally don't nibble at AC trolls, but just in case you do actually come back to read responses to your own, or in case someone else reads your drivel, it will make me feel better =)

Whoopee fuckin doo! Somebody call the press! Lets have a party!

Now that there is a browser that passes the all-important-almighty-mission-critical-life-or-de ath ACID 2 test, I guess we'll have to come up with a new cool sounding buzzword browser compliance test in order to perpetuate the browser snobbery that so many have com

What?

[mboverload]

WTF? Where did the cars go?

Re:

[drinkypoo]

They were destroyed by Cammy [wikipedia.org].

redraws involve headache-inducing white flashes

[dankelley]

On OSX, Gran Paradiso Alpha 3 had an annoying habit of flashing a white screen before redrawing a page. To test this, just go to http://www.mozilla.org/products/ [mozilla.org] and click from tab to tab.

One can only hope that this won't occur in the release versions, because it is really quite annoying.

Re:redraws involve headache-inducing white flashes

[Seumas]

That's a feature, dude. The white screen in between page loads is where the government flashes the subliminal commands, instructing you to consume, worship and be content.

Re:

[jesser]

I think this is bug 361600. I agree that it's annoying, and it looks like it will be fixed in time for Firefox 3.

It's even more annoying in debug builds ;)

Re:

[jesser]

Yeah. For a solid percentage of the 6-digit numbers, Google will give you a Debian bug report, a Gnome bug report, or a Mozilla bug report.

Re:

[Anc]

The new theme for FF3 will look even less like OS X apps, and will continue to have ugly Windows 95-ish form controls. Actually, they'll be switching to Windows 3.1-ish controls.

Quite the opposite. One of the already implemented changes that will make it to Fx3 is enabling native Cocoa widgets.

Re:

[dankelley]

I like Camino, Safari, and Opera, but I have come to rely upon a variety of FireFox plugins, so the aforementioned aren't really options.

Re:

[Goaway]

...Opera? Did you just complain about Firefox not looking like an OS X app, and then recommending Opera?

Release schedule?

[Tester]

I was looking at the mozilla wiki and I couldn't find a release schedule for Gecko 1.9 / Firefox 3. Does it exist somewhere? I read allusion to like March and then to the end of the year. I guess if we are still in Alpha, its more like Q3/Q4 ?

I tried recent nightly builds, and I really liked what I saw on the painting front. I hope we can get a stabilized release soon.

Re:

[dvice_null]

Firefox 3 is planned to be released on November: http://wiki.mozilla.org/ReleaseRoadmap [mozilla.org]
Wikipedia quotes the same source.

Run both firefox and Gran Paradiso ?

[mallumax]

Is there a way to run both firefox and Paradiso without affeting firefox settings and extensions etc ?

Re:

[teh kurisu]

I keep a separate user account for this sort of thing. Not exactly convenient though.

Re:

[obender]

Is there a way to run both firefox and Paradiso without affecting firefox settings and extensions etc ?

Unzip firefox in a separate directory, for example c:\beta\firefox, make an extra directory for profiles and start firefox with the -profile option:

set MOZ_NO_REMOTE=1
c:\beta\firefox\firefox.exe -profile c:\beta\profile

The MOZ_NO_REMOTE variable will prevent it from connecting to another running instance of Firefox. All the settings are stored in the profile directory so it will leave the regular installation alone.

Ask your admin to grant you write access

[tepples]

If you don't have write-access to c:\beta\firefox, then you are not the administrator. If you are not the administrator of a machine, and the administrator refuses to grant you this write access, then you probably shouldn't be running alpha software on the machine.

Re:

[obender]

What if you don't have write-access to c:\beta\firefox ?

Then use a directory where you have write access. c:\beta was just an example.
I thought you would easily figure that out given that you are running Debian.

Re:

[Nutria]

Then use a directory where you have write access. c:\beta was just an example.
I thought you would easily figure that out given that you are running Debian.

The point is: why should you need write access to a non-$HOME directory, no matter where it is?

Re:

[obender]

The point is: why should you need write access to a non-$HOME directory, no matter where it is?

You don't. Make subdirectories under $HOME and you're OK. The whole point of using -profile is to allow you full control on where you place you settings.

And last, for some of us c:\beta is home :)

Re:

[jginspace]

"Is there a way to run both firefox and Paradiso without affeting firefox settings and extensions etc ?"

First of all I recommend you create a new profile for V3 (Gran Paradiso). After that it's possible to run new instances of Firefox from a batch file with the contents:

set MOZ_NO_REMOTE=1
firefox -P

Mac users, give it a try!

[chrysalis]

For OSX users, Gran Paradiso is a huge improvement over previous Firefox versions. It's way faster, and it feels as fast as Safari. While there are still some bugs especially with forms, this is definitely something OSX users should try.

Re:

[dr.badass]

For OSX users, Gran Paradiso is a huge improvement over previous Firefox versions. It's way faster, and it feels as fast as Safari.

If it looks like Firefox, acts like Firefox, but is as fast as Safari -- it's still Firefox. Fuck that.

Re:

[jrockway]

Yeah, I hate visiting sites and seeing them render properly! That really sucks! Fuck firefox!

Re:

[dr.badass]

Yeah, I hate visiting sites and seeing them render properly!

Where did I say anything about rendering? As far as I can tell, Safari/WebKit and Firefox/Gecko are both more or less accurate. Safari passed Acid2 back in 2005, while Firefox is just now getting there. Besides, if I wanted to run a Gecko browser, I'd run Camino over Firefox, because it doesn't feel like a bad port from Windows.

Re:

[hritcu]

Once it's stable sure. Most people feel absolutely no need to be on the bleeding edge.

Even Mozilla guys ignore non-x86 Linux

[cyba]

There's only single "Linux" download link that refers to Linux/x86 binary. If leading Free Software project doesn't treat non-x86 platforms seriously, how can we expect something different from e.g. hardware manufacturers?

Re:

[644bd346996]

Anybody running Linux on a non-x86 box will know how to install from a distro's repositories. The same can no longer be said for Linux on x86, because of people installing ubuntu on their grnadmother's computer, etc. However, in this case it really doesn't matter. This is alpha software. If you are messing with it, you should know how to compile it yourself. It is not really worth Mozilla's time to acquire obsolete non-x86 desktop hardware for building nightlies and alpha releases.

I find your comment about

Re:

[eraserewind]

Free Software in the FSF's terms doesn't ship with any binaries at all. You got the source code, compile it yourself. Personally I think it's a bit much to expect builds for every platform out there from an alpha release.

how can we expect something different from e.g. hardware manufacturers?

Nobody wants binaries from hardware manufacturers. I think most Linux devs would be over the moon if the manufacturers released only source code for non-x86 (or even x86).

Re:That felt weird

[Seumas]

Just to be ironic

I do not think that word means what you think that it means.

Re:

[ProfessionalCookie]

Don't worry...he's just been hanging out with Alanis Morsette. On second thought, worry.

Alanis uses irony correctly

[gvc]

"Saying the opposite of what you mean" is but one sense of irony. Here's another:

OED 2 fig. A condition of affairs or events of a character opposite to what was,
or might naturally be, expected; a contradictory outcome of events as if in
mockery of the promise and fitness of things.

http://photo.net/bboard/q-and-a-fetch-msg?msg_id=0 07HIc [photo.net]

Ironic you should think so

[alienmole]

Ironically, it's exactly this sense of the word that most people are claiming Alanis failed at. For example, neither "Rain on your wedding day" or "A traffic jam when you're already late" are outside of what might naturally be expected -- as many people have observed, it's simply bad luck. One could argue that both qualify as an outcome "as if in mockery of the promise and fitness of things", but the problem with this is that they don't meet the test of being opposite to what might naturally be expected.

Um... no.

[Millennium]

The only ironic thing in her song is that despite being about irony, there is in fact no irony in it. Maybe that was part of the intent of the song in the first place, but it doesn't change the fact that there isn't a single correct use of irony in the piece, just a lot of depressing coincidences.

Re:

[icepick72]

I was actually going for more of a sense of situational irony [wikipedia.org]; however I admit not much thought went to the wording at the time. The best part has been the enjoyment of reading through all the comments about my statement. Isn't that just a little bit ... ironic?

Why should it?

[Anonymous Coward]

I consider it a feature if a browser prevents people from visiting MySpace.

Re:Why should it?

[pipatron]

I knew the MySpace-crowd were young, but 5 months, wtf?

Re:So..

[Jeffrey Baker]

If they "fixed" this "leak", the other bunch of fucknozzles would come back asking "Why are back and forward so slow?!??". I think for the time being you people are slightly less annoying, so the "leak" stays.

Re:

[jesser]

It's possible to turn off the back-forward cache by setting browser.sessionhistory.max_viewers to 0 in about:config. That said, for some users, Firefox uses a lot of memory due to actual leaks rather than this kind of caching.

New word!

[Anonymous Coward]

Fucknozzles! That's a new one. I suppose if you've already got an asshat, you need some matching fucknozzles to go with it.

Re:

[Nutria]

Fucknozzles! That's a new one. I suppose if you've already got an asshat, you need some matching fucknozzles to go with it.

Just wait for the metric assload of fucknozzles.

Re:

[Man of E]

Other browsers have shown (Opera, Safari) that it's possible to have speedy back and forward buttons without taking up a gig of memory. You can claim that the Firefox back-forward code is so poorly written that it would have to be redone from the ground up, and that the developers consider a new spellchecker to be higher-priority than a time-consuming rewrite of this memory-hogging component. But please don't pretend that this is an intrinsic trade-off in browser design.

Re:So..

[bunratty]

Nice rant, but Firefox does not seem to use more memory than other browsers. See my above posts and the following links:
Radically New IE 7 or Updated Mozilla Firefox 2--Which Browser Is Better? [pcworld.com]
IE 7 vs IE 6 [zimbra.com]
Firefox 2 - the lean, mean browser [mozillazine.org]

If you can give a set of steps that causes Firefox to use "up to a gig of memory" and does not cause other browsers to use nearly as much memory, let's have it. Then whatever problem you're seeing can be reported and fixed.

Re:

[bunratty]

It's not programmer logic. The "fastback" feature was one of the most requested enhancements before it was added. It was at the request of users that the feature was added, not because of "crappy programmer logic." Anyway, the feature seems to use only tens of megabytes at most (at least on the vast majority of web sites), and can be turned off entirely if it's using too much memory for your liking.

Re:

[Bazzargh]

*Holds up hand*

my fault, assuming you are using an alpha or nightly build. There are still a bunch of font bugs in cairo, in particular when the text is scaled. Most of those were fixed in cairo 1.4.2, they should land in the mozilla soonish, and hopefully the next release will look a lot better. Mac fonts on cairo trunk are now pretty much up to par with the other platforms, and Robert O'Callahan, Vlad, etc have done great stuff making it all perform well too.

-Baz (maintaining mac font stuff in cairo)

Re:

[Breakfast Pants]

Me too. This feature is long awaited.

Re:

[stu42j]

Actually, the page zoom at least is being worked on. That's what the Cairo backend is all about.