Slashdot Mirror

You do have a good point. There has been talk about supporting additional image formats (JPEG 2000, TIFF, MNG) using imagelib extensions. They could do the same for different video codecs, as well.

I notice that Firefox is an open source project, so all it takes is someone to come forward to do the work. I also notice that the Google Summer of Code will be starting over the next several months. Are there any students out the that want to make some extra $$$, get great software development experience, and add a significant new capability to one of the most popular browsers?

As usual with Firefox features, if you don't like it, you can probably fix it. Try the oldbar extension. There is probably a way to disable it without an extension, ISTR there is a setting in about:config for 3.0 at least, but you can google that yourself. Personally I love the awesome bar, although I don't think I will flip out about the new version for a whole 6 months, but each to their own.

All the Idiots who scream here that the OS is doing something worng: no, it's not.

This is called "hiding behind the standard" (a disease very common among kernel developers). Just because the standard doesn't specify behaviour in a certain situation doesn't mean that any behaviour is equally okay. In this case, ext4's behaviour very much hurts the robustness of the system, which is rather important in unreliable environments like laptops.

In this case, what KDE does is certainly not unreasonable (and its developers are certainly not "idiots"). It doesn't overwrite configuration files in place, which would be bad even in the absence of system crashes, as doing it that way is not atomic. Instead it creates a new temporary file, writes the new contents, then renames the temporary file to the old one. This is an atomic operation on Unix: you either see the old contents or the new contents, but nothing in between. Now, the problem is that in case of a crash, ext4 gives you the worst possible outcome by reordering the operations: it will "recover" the rename for you, but not the actual write of the new data. So you end up with a 0-byte file - far from atomic. POSIX of course allows this, but POSIX allows just about anything: that doesn't mean its reasonable. The only guaranteed solution - use an fsync/fdatasync - is something that almost nobody does because the performance is horrible (ext3 in fact will write the entire journal, IIRC, when doing an fsync() on a single file - this really hurt Firefox 3 performance). So the KDE developers can be excused for not doing that.

It's the job of a modern filesystem to ensure robustness and performance. If you don't use an fsync, you should expect that there is a time window during which transactions might become undone (not the end of the world for configuration files), but they should never be reordered. For instance, this is how Berkeley DB works if you disable fsync: it guarantees ACI but not ACID. For many desktop applications, that's good enough. Destroying every file that has been updated since the last fsync isn't. And your users aren't going to be impressed by the argument that POSIX allows it.

The remaining question for users is: Has someone yet developed a plugin to block google ads entirely? And if not, how long will it take now?

Ad-Block Plus. Assuming, of course, that you are using Firefox.

CustomizeGoogle

https://addons.mozilla.org/en-US/firefox/addon/743

A lot of people seem to have little-to-no understanding as to what ActiveX is. It is a plug-in infrastructure based on COM, nothing more, nothing less. It allows for a library to provide a visual component that can be loaded by another application to display content. That plug-in infrastructure was used in Internet Explorer to load browser plug-ins. Those plug-ins run within the browser process under the current user security context. There is absolutely no functional difference between ActiveX in Internet Explorer on Windows or an XPCOM plug-in for Firefox on Linux.

The problem is that in both cases those plug-ins have to have a fairly wide amount of functionality. If that plug-in is intended to display video then it has to be able to work with the video API of the platform in question. As such these plug-ins generally cannot be sandboxed too tightly otherwise they would no longer be able to function and their usefulness of being able to extend the functionality of the browser is lost.

https://addons.mozilla.org/en-US/firefox/browse/type:7

This website lists the XPCOM plug-ins available for Firefox. There are quite a few more if you follow the link to the bottom. If a vulnerability is identified in ANY of those plug-ins a successful exploit will be fully capable of trashing the profile of the current user and there is nothing that Firefox can do to stop it, even on Linux.

Yeah because those extensions just instantly came out when the browser did.

From wikipedia:

Mozilla Firefox 3 was released on June 17, 2008

Wikipedia's source:
https://developer.mozilla.org/devnews/index.php/2008/06/17/firefox-3-available-today-at-1700-utc-10am-pdt/

Let's look at the oldbar download and usage statistics:
Zoom out all the way on this graph:
https://addons.mozilla.org/en-US/statistics/addon/7637

Mouse over the extreme left of the graph. Notice the date displayed: 06/17/2008. Notice that there were 97 downloads that day.

Do the same for hideunvisited:
https://addons.mozilla.org/en-US/statistics/addon/7429

Notice that on 06/17/2008 there were 2,215 downloads.

Oldbar and hideunvisited *were* available on the day that Firefox 3.0 shipped.

You are an idiot.

I respectfully request that you retract this statement.

I'm pointing out that some really bad decisions and horrendous attitudes are turning one incredible browser into a troublesome piece of shit.

You're not making your point. All you've done in this thread is rail endlessly about how much the AwesomeBar sucks and how it's the death of Firefox. *I* *don't* *agree* *with* *you*. If you have other examples, you need to start bringing them out. This one example is very weak, as it's based on matters of taste.

Yeah because those extensions just instantly came out when the browser did.

From wikipedia:

Mozilla Firefox 3 was released on June 17, 2008

Wikipedia's source:
https://developer.mozilla.org/devnews/index.php/2008/06/17/firefox-3-available-today-at-1700-utc-10am-pdt/

Let's look at the oldbar download and usage statistics:
Zoom out all the way on this graph:
https://addons.mozilla.org/en-US/statistics/addon/7637

Mouse over the extreme left of the graph. Notice the date displayed: 06/17/2008. Notice that there were 97 downloads that day.

Do the same for hideunvisited:
https://addons.mozilla.org/en-US/statistics/addon/7429

Notice that on 06/17/2008 there were 2,215 downloads.

Oldbar and hideunvisited *were* available on the day that Firefox 3.0 shipped.

You are an idiot.

I respectfully request that you retract this statement.

I'm pointing out that some really bad decisions and horrendous attitudes are turning one incredible browser into a troublesome piece of shit.

You're not making your point. All you've done in this thread is rail endlessly about how much the AwesomeBar sucks and how it's the death of Firefox. *I* *don't* *agree* *with* *you*. If you have other examples, you need to start bringing them out. This one example is very weak, as it's based on matters of taste.

Yeah because those extensions just instantly came out when the browser did.

From wikipedia:

Mozilla Firefox 3 was released on June 17, 2008

Wikipedia's source:
https://developer.mozilla.org/devnews/index.php/2008/06/17/firefox-3-available-today-at-1700-utc-10am-pdt/

Let's look at the oldbar download and usage statistics:
Zoom out all the way on this graph:
https://addons.mozilla.org/en-US/statistics/addon/7637

Mouse over the extreme left of the graph. Notice the date displayed: 06/17/2008. Notice that there were 97 downloads that day.

Do the same for hideunvisited:
https://addons.mozilla.org/en-US/statistics/addon/7429

Notice that on 06/17/2008 there were 2,215 downloads.

Oldbar and hideunvisited *were* available on the day that Firefox 3.0 shipped.

You are an idiot.

I respectfully request that you retract this statement.

I'm pointing out that some really bad decisions and horrendous attitudes are turning one incredible browser into a troublesome piece of shit.

You're not making your point. All you've done in this thread is rail endlessly about how much the AwesomeBar sucks and how it's the death of Firefox. *I* *don't* *agree* *with* *you*. If you have other examples, you need to start bringing them out. This one example is very weak, as it's based on matters of taste.

I don't like having to use extensions for things that use to be configurable but that developers decided to push by disabling that option for the sake of their own agenda.

IDK if you're a software developer. If you are, then I'm sure that you're acutely aware of the significant testing and maintenance burden that additional configurable behaviours add.
The core Firefox team cannot possibly maintain and test *everyone's* pet behaviour. It is for this reason that the Extension and Add-on systems were built into Firefox. These systems allow the Firefox team to indirectly support *everyone's* pet behaviour by offloading the testing and maintenance burden to the third-party devs who are willing to take up the cause of marginalized users such as yourself.

I suppose you're suggesting you've never ever made a mistake visiting a site you shouldn't have, or installing something you shouldn't have?

IDK about the PP, but I daresay that I fit this description. Safe computing is not hard, when you put some thought into it.

I also don't accept that "most people" like awesomebar. I've seen no evidence of that whatsoever. Take a look at the complaints all over the net...

Consider us the silent majority who use Awesomebar to get our browsing done faster and get back to work.

You haven't provided one concrete solution to address any of my concerns.

That's 'cause you've already provided the concrete solutions that address your concerns. You have two extensions that -when used in concert- give you the Firefox behaviour that you desire. There's nothing more to be said.

There ... things about [Firefox that have] ... left me feeling quite bitter about it. I've only reported these things here.

Did you discuss these things in the Firefox IRC channel or mailing lists? If deemed appropriate, did you file bugs against the bad behaviour? If not, then why not?

What's worse is version 1.0 of this tool was great...

Why aren't you still using Firefox 1.0? Do you not know where to go to download a copy?
Look here:
ftp://archive.mozilla.org/pub/mozilla.org/firefox/releases/
But before you *use* any of those browsers, be sure to read this:
http://releases.mozilla.org/pub/mozilla.org/firefox/releases/README

Is there anything more that we can do for you tonight, sir?

I don't like having to use extensions for things that use to be configurable but that developers decided to push by disabling that option for the sake of their own agenda.

IDK if you're a software developer. If you are, then I'm sure that you're acutely aware of the significant testing and maintenance burden that additional configurable behaviours add.
The core Firefox team cannot possibly maintain and test *everyone's* pet behaviour. It is for this reason that the Extension and Add-on systems were built into Firefox. These systems allow the Firefox team to indirectly support *everyone's* pet behaviour by offloading the testing and maintenance burden to the third-party devs who are willing to take up the cause of marginalized users such as yourself.

I suppose you're suggesting you've never ever made a mistake visiting a site you shouldn't have, or installing something you shouldn't have?

IDK about the PP, but I daresay that I fit this description. Safe computing is not hard, when you put some thought into it.

I also don't accept that "most people" like awesomebar. I've seen no evidence of that whatsoever. Take a look at the complaints all over the net...

Consider us the silent majority who use Awesomebar to get our browsing done faster and get back to work.

You haven't provided one concrete solution to address any of my concerns.

That's 'cause you've already provided the concrete solutions that address your concerns. You have two extensions that -when used in concert- give you the Firefox behaviour that you desire. There's nothing more to be said.

There ... things about [Firefox that have] ... left me feeling quite bitter about it. I've only reported these things here.

Did you discuss these things in the Firefox IRC channel or mailing lists? If deemed appropriate, did you file bugs against the bad behaviour? If not, then why not?

What's worse is version 1.0 of this tool was great...

Why aren't you still using Firefox 1.0? Do you not know where to go to download a copy?
Look here:
ftp://archive.mozilla.org/pub/mozilla.org/firefox/releases/
But before you *use* any of those browsers, be sure to read this:
http://releases.mozilla.org/pub/mozilla.org/firefox/releases/README

Is there anything more that we can do for you tonight, sir?

Ah-ha! If that's the case, this slowness on Linux is a symptom of Firefox foolishly doing an fsync() every time it writes to the filesystem (see this bugzilla entry). Lots of discussion there, but I don't think it's been resolved.

The short version: If fsync() is expensive on your kernel/filesystem/etc. you're going to see terrible performance.

Sometimes it's the site server that's slow, not the connection. And I download images large enough to give me time to hit the Stop button :P ... Which, incidentally, doesn't work right in Firefox (Bug 58880). Try viewing an image directly and Stopping it when it's half-loaded; it disappears and is replaced by an error message, instead of staying on the screen half-loaded. Every other browser does this right, including Classic Netscape, and it's Firefox's code ancestor.

I lied again! There is an addon that lets you remove the Menus as well. https://addons.mozilla.org/en-US/firefox/addon/1455

Firefox 3 also can't open the html5 spec page for some reason. Neither can Opera, Seamonkey, etc., without spiking to 99% CPU.

See bug 475606 and associated dependencies.

The behavior is nice IMO, but the UI change is horrible. ~341,000 people agree.

Besides the fact that you tested a trunk build when you should have tested this one: ftp://ftp.mozilla.org/pub/firefox/nightly/3.1b3-candidates/build1/win32/en-US/, you do realize that these javascript benchmarks are far from useful for determining real-world web browsing experience? Furthermore, the V8 benchmark in particular is so tailored to show Chrome's performance advantage that it can not be considered useful for anything other than showing Chrome is excellent at performing a handful of cherry-picked functions. Better yet, go play with google maps or some other heavy ajax app and then tell us that FFx 3.1 is not up to snuff. I have and I'm telling you that Ffx 3.1 is fast.

The memory bloat bugs that I see get dismissed are generic complaint bugs with little or no information with how to reproduce it. Here's a hint: "Just browse around for awhile and you'll see the problem" is not helpful. You don't have to look beyond this very thread to see plenty of people who can't reproduce those problems with their regular browsing habits. If you can't give useful steps to reproduce or a testcase that clearly shows the problem, how can you expect it get any traction with anyone? On the other hand, look at even a recent bug like Browser not responsive while loading HTML5 spec to see how much attention a bug can get when useful information is provided with it.

And I don't recall anyone saying anything about it being a feature, not a bug, other than Ben's oft-cited and ancient blog post that's not even relevant anymore given how much has changed with Gecko's memory management since then.

What's more, you shouldn't have to dig around in about:config to change a setting that doesn't actually do what you want.

The max rich results setting just means it won't display any search results. That's not even remotely the same as going back to an old-school auto-complete functionality.

Exactly. (Mod parent up.) There is no way to disable the Awesome Bar in the sense that d3ac0n means, i.e. returning to a sensible autocomplete dropdown rather than the search-based algorithm it uses now. And there apparently won't be, given that this bug is "RESOLVED WONTFIX".

To be fair, I hated it at first (and at times I still do) but while it sometimes has completely random matches, there are a number of sites that I can now get to much more easily, even without having bookmarked and tagged them. About the only thing that I do always do is use the oldbar extension as a basis for my CSS to get a slightly more sensible appearance (i.e. something that doesn't go half way down your screen with half a dozen results).

I don't hate it as much as I used to, and I recognize that 95% of users love it, but I'd still switch back if I had the option. I have miscellaneous usage problems I could rant in detail about (and yes, I have "trained" it--I've been using FF3 since Download Day), but my biggest problem is philosophical: it breaks expectations. The location bar is for typing locations. If I start typing a location, if it employs any kind of "smart" searching technology, then I can't predict what will be in the dropdown--whereas a bar that simply autocompletes rather than searches is predictable and useful.

In the WONTFIXed bug, the developers encourage feedback about how to make the awesome bar customizable, how to change the weightings applied to the search function, etc. They completely miss the point that no amount of tweaking and preference-weighting will make an algorithm that can exactly predict what I want 100% of the time. The entire premise of "search" in the location bar is flawed.

Admittedly, that's my opinion. And as I mentioned above, I recognize that the vast majority of people like it. I don't ask for it to be removed, or for it to not be the default. All I ask is for the option to revert to the old behavior.

And you know, as I am sitting here at my home PC in the family room with my kids running around behind me, it might be possible that I don't want them seeing the names and URLs some of the more Adult oriented sites that Mommy and Daddy surf together after they go to bed.

This might help:
https://addons.mozilla.org/en-US/firefox/addon/1306

And I like the bar because I frequently click the star on pages I want to be prioritized in the search, but don't need to be bookmarking.

Dear Adobe, please fix your flash plugin. Seems that once a day if I go to a page with considerable flash (which is most pages these days), the browser will crash [...]

There's a third-party fix available: Flashblock

Not a fix, I know, but have you considered Flashblock?

Indeed. If it were possible to completely remove IE, including the rendering engine, it's incredible just how much it would break. Steam and IETab come to mind immediately.

There is a very tenuous connection.

First note that ECMA script is the core language. Add a few fewtures and you have the cross browser script base. The (cross browser script base)+DOM is what is portable between browsers.

I use the term cross browser script base, because unfortunately there really is no name for this language.

Everybody implements something that is basically an extended version of the cross-browser base script.

Microsoft calls their implementation JScript.

Netscape (and now Mozilla) call their implementation Javascript.

Pretty much nobody else named their implementations, so they are generally also called javascript, although strictly speaking that is incorrect.

One of the features of Mozilla's ECMAscript implementation which is what can rightfully be called Javascript is a specific binding for java objects, allowing them to be manipulated by javascript must as though they were native Javascript objects.

See https://developer.mozilla.org/en/JavaScript for more information. Details of the Javascript Java binding can also be found on that wiki.

Every time I find myself on a Windows box using any other browser I wish I could expand text boxes (like the one I'm typing in now) to be able to see my whole comment. It's been years now.

Chrome does this, and Firefox can too.

I think you mean Fennec for the mobile browser. The name follows a theme, the Fennec is a small African fox.

It's easier to install the Chromifox addon for Firefox. Same GUI, with plugins, and it works on any OS that supports firefox ;)

https://addons.mozilla.org/en-US/firefox/addon/8782

wget http://download.mozilla.org/?product=firefox-3.0.6&os=win&lang=en-US

of course

It's difficult to get away from Windows and IE, because of their anticompetitive behavior.

It's not difficult at all - you go to http://mozilla.org/ and click on "Install Firefox". You just have to care enough first, and most people don't. And, trust me, they won't care any more if Windows installer will, at some point, ask them "Do you want IE or some other browser?". In fact, they won't see it at all, because most installations are done by OEMs, anyway. And, while it's good if the OEMs get full freedom to choose which browser to bundle, do you seriously think that more than a few will include anything but IE, and that any at all will actually remove IE completely, even if the option is there?

Bah, I can do it too !

Java [is] actually integrated much less than, for example, Silverlight ... it can actually interact with the Javascript on the page.

Untrue. LiveConnect allows communication between Java applets and Javascript, and it works just fine.

or mozilla can fix the fucking bug.

https://bugzilla.mozilla.org/show_bug.cgi?id=446139

Just FYI: the RFE to remove those addons was marked WONTFIX by mozilla, because "they should be removed by the Installer that put the files there".

IMO it has to be possible to remove them from the Add-On manager.

Fuck, messed up the quote, damn boards with their crappy syntax :D

If you use Firefox you can install the Slashdotter extension. In addition to several handy options for tweaking Slashdot, you get a nice "Reply to selected text"-option when right-clicking. This inserts a blockquote for you. Recommended!

Firefox 3 can work on Windows 98 as long as you're prepared to follow the instructions to hack 98 a bit.

Unless they whitelist programs that you can execute, you can just go to the Mozilla FTP server, find the nightly that corresponds to the latest release version of Firefox, and download the zip file. No installation necessary.

Unless they whitelist programs that you can execute, you can just go to the Mozilla FTP server, find the nightly that corresponds to the latest release version of Firefox, and download the zip file. No installation necessary.

Here is the bug to which the attachments are attached:
https://bugzilla.mozilla.org/show_bug.cgi?id=478721

If you have a Mozilla account then please comment on the bug requesting it to be deleted, as I have done.

exactly like DC8s, without propellers

They then proceed to show how allowing unicode in the hostname continues to confuse and confound people. Register a cert for *.foo.com, then set up a hostname of www.google.com[unicodeslashlike]login[unicodeslashlike]blah[unicodeslashlike]blah[unicodeslashlike]blah.foo.com and presto, you have a valid certificate for a site that looks more or less like https://www.google.com/login/blah/blah/blah.foo.com [google.com], except that it's not hosted by google.

Fixed in 2005?

Exactly. Of course, if you use the Petname Toolbar, none of these attacks work.

Maybe they should check for homeless on the planets of Polaris? using Firefox 3.1 beta to view them?

PGO not yet turned on for linux releases, not even for trunk hourlies/nightlies yet.
See https://bugzilla.mozilla.org/show_bug.cgi?id=418866
Some work now happening (at a guess the back and forth between BZ and I above has got some prompting done - BZ has links/is inside mozilla somewhere.

PGO is when you run the instrumented program first, take the results and optimize againt the used code paths

http://mxr.mozilla.org/mozilla-central/source/build/pgo relevant bit of the tree

http://forums.mozillazine.org/viewforum.php?f=29
You can find some 64 bit builds (search for Autofox) and a pgo build or two maybe - (search for Ted's builds). Note it does say 3rd party - not official mozilla releases.

Due to some ancient history I am one of the ones that test latest builds and so I know some of the history and a bit of may way around. I am also tqft at forums.mozillazine.org if you ask a question over there and I answer.

PGO not yet turned on for linux releases, not even for trunk hourlies/nightlies yet.
See https://bugzilla.mozilla.org/show_bug.cgi?id=418866
Some work now happening (at a guess the back and forth between BZ and I above has got some prompting done - BZ has links/is inside mozilla somewhere.

PGO is when you run the instrumented program first, take the results and optimize againt the used code paths

http://mxr.mozilla.org/mozilla-central/source/build/pgo relevant bit of the tree

http://forums.mozillazine.org/viewforum.php?f=29
You can find some 64 bit builds (search for Autofox) and a pgo build or two maybe - (search for Ted's builds). Note it does say 3rd party - not official mozilla releases.

Due to some ancient history I am one of the ones that test latest builds and so I know some of the history and a bit of may way around. I am also tqft at forums.mozillazine.org if you ask a question over there and I answer.

And no, I cannot possibly consult a dictionary or wikipedia each time I encounter a new word.

Not even those long ones?
And why not?
Does google charge for every search you make where you live?

It sure as hell takes less time to do that, than to make a slashdot comment.
An uninformed slashdot comment might I add.
You CAN add search engines (including dictionaries) to Firefox, you know?

Now, I'm not one to do this normally, but seriously - what you need is Adblock Plus.

If you like to support sites you like by leaving the ads on, you can white-list them.

On Facebook, I see no ads. I also don't add many applications, though.

I'm freeloading, but I use the site so infrequently that it doesn't matter. You could say that I add value by making other people more likely to spend more time looking at their ads, because they have one more friend's profile to browse. That would imply that I have lots of friends that look at my facebook profile all the time, but still, even though I make no money for them directly, they still get something out of me :)

Facebook is making plenty from others with ads and data-mining. If you want to keep up with your friends while not "paying" for Facebook with your advertising eyes or data, block their ads and don't give them much personal data. Simple.

For reference
turn on profile-guided optimization on fx-linux-tbox
https://bugzilla.mozilla.org/show_bug.cgi?id=418866
Last comment May 2008.

I tried to find a better bug - meta or tracking bug - without luck yet.

PGO is enabled in the trunk versions - not yet in the release versions, it does compile and run. Finished compiling a short while ago (an hour or so) as I type.

Ted M gave up pushing for pgo on linux after a while - trying to get it done without much help I think was the problem.

"On the other hand, the Windows build was faster even without PGO enabled on Windows."
I believe you. You seem to have done the research and it wouldn't be the first time I was wrong.
I have seen a reason for it somewhere but can't remember what or where though.

PGO on linux will help.
People also not writing sucky js for websites (slashdot comes to mind) would perhaps help more.

And static linking will not work on firefox - the UI dfepends on XUL which is diabled if you enable static (but I think you know this already).

https://developer.mozilla.org/en/Configuring_Build_Options
ac_add_options --enable-libxul (default)
        Builds the core gecko components as a single library called libxul. This improves startup and runtime performance by reducing the number of relocations performed.

ac_add_options --enable-static --disable-libxul
        These options build a larger single executable, which has components linked statically. --enable-static requires --disable-libxul. If you use --enable-static, it is recommended that you also --disable-tests. This option is not recommended for Firefox. It still exists only for Thunderbird and SeaMonkey which cannot yet build in the libxul configuration.

[v2+ of seamonkey uses XUL so enabling static seamonkey will no longer be an option for new versions].

Most of the current performance focus seems to be on js (TraceMonkey). Though Gecko 1.9+ did include significant speed ups. I am don't know if this is the best - but that is what is happening. However, performance regressions when noticed do get attention. But I only watch the ff trunk build threads and seamonkey v2+ stuff closely - so anything with current ff is beyond my knowledge.

Happy
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b3pre) Gecko/20090216 SeaMonkey/2.0a3pre ID:20090216123246
user.

PS: session restore just hit seamonkey trunk so I can go crash happy and not lose my place too much.

For reference
turn on profile-guided optimization on fx-linux-tbox
https://bugzilla.mozilla.org/show_bug.cgi?id=418866
Last comment May 2008.

I tried to find a better bug - meta or tracking bug - without luck yet.

PGO is enabled in the trunk versions - not yet in the release versions, it does compile and run. Finished compiling a short while ago (an hour or so) as I type.

Ted M gave up pushing for pgo on linux after a while - trying to get it done without much help I think was the problem.

"On the other hand, the Windows build was faster even without PGO enabled on Windows."
I believe you. You seem to have done the research and it wouldn't be the first time I was wrong.
I have seen a reason for it somewhere but can't remember what or where though.

PGO on linux will help.
People also not writing sucky js for websites (slashdot comes to mind) would perhaps help more.

And static linking will not work on firefox - the UI dfepends on XUL which is diabled if you enable static (but I think you know this already).

https://developer.mozilla.org/en/Configuring_Build_Options
ac_add_options --enable-libxul (default)
        Builds the core gecko components as a single library called libxul. This improves startup and runtime performance by reducing the number of relocations performed.

ac_add_options --enable-static --disable-libxul
        These options build a larger single executable, which has components linked statically. --enable-static requires --disable-libxul. If you use --enable-static, it is recommended that you also --disable-tests. This option is not recommended for Firefox. It still exists only for Thunderbird and SeaMonkey which cannot yet build in the libxul configuration.

[v2+ of seamonkey uses XUL so enabling static seamonkey will no longer be an option for new versions].

Most of the current performance focus seems to be on js (TraceMonkey). Though Gecko 1.9+ did include significant speed ups. I am don't know if this is the best - but that is what is happening. However, performance regressions when noticed do get attention. But I only watch the ff trunk build threads and seamonkey v2+ stuff closely - so anything with current ff is beyond my knowledge.

Happy
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b3pre) Gecko/20090216 SeaMonkey/2.0a3pre ID:20090216123246
user.

PS: session restore just hit seamonkey trunk so I can go crash happy and not lose my place too much.

Here's a Stylish style I made that makes Slashdot not look like shit:
http://pastebin.com/f39655fa6

yes I've heard of gprof that's profiling not a profiling optimizer they are different things. Turns out the mozilla build system does support PGO with gcc though: https://developer.mozilla.org/en/Building_with_Profile-Guided_Optimization . I'm not sure how it compares with the PGO in MSCC. I don't even know if the distros tend to build it with PGO. In any case gcc tends to lag behind commercial compilers in terms of performance (that is after all the selling point of commercial compilers). GCC tends to be ahead in terms of standards compliance.