Firefox 2.0 into release?

Jacek Poplawski · 2006-10-24 09:51:00

Will FireFox 2.0 be included in 0.8 Voodoo release?

Jacek Poplawski · 2006-10-24 13:12:23

Lots of people will install 0.8 release as their first Arch. It is not important for current users, but for new users. We have new xorg, new KDE/GNOME, new kernel and new gcc. It would be nice to have also new FireFox.

dtw · 2006-10-24 13:44:07

I agree entirely and it does matter a great deal to new users! Let's remember that without new users Arch will die with the current users (albeit in 70 odd years)

cactus · 2006-10-24 20:40:08

Jacek Poplawski wrote:

Will FireFox 2.0 be included in 0.8 Voodoo release?

No idea. I would hypothesize and say that within a week or two after mozilla.org releases it, it should appear in the repositories.
If voodoo 'presses' before then, it would be a 'no'.
If it 'presses' after that, then I think it is highly likely.

Being that firefox doesn't effect that many other things, I don't think it would stay overly long in 'testing' before hitting the main repos.

hope that helps.

phrakture · 2006-10-24 21:44:11

I split this. Lets not do the flame-war thing.

Suffice to say: yes, hopefully. 0.8 may be some time away, it may not - we don't have a tight schedule for these things.
firefox 2.0 should be publically consumable by then.

fax8 · 2006-10-25 11:57:36

firefox 2.0 released yesterday.

ralvez · 2006-10-25 12:49:16

I installed and tested FF2 (got a copy from the mozilla.og site).
I'm rather disappointed with the restuls, though.
It's definitely sluggish, particularly if compared to FF1.5, and in the two hours I played with it it crashed at least 3 times.
I also noticed that some audio/video streams that play fine in FF 1.5 would not play in 2.0
I have not taken the time to examine the reasons for the problem, but it most certainly does not inspire a feeling of "getting better" with a new version, which to me is is disappointing, because FF is a very visible OS project and may "color" things for a lot of people out there.

:cry:

superstoned · 2006-10-25 14:09:31

ralvez wrote:

I installed and tested FF2 (got a copy from the mozilla.og site).
I'm rather disappointed with the restuls, though.
It's definitely sluggish, particularly if compared to FF1.5, and in the two hours I played with it it crashed at least 3 times.
I also noticed that some audio/video streams that play fine in FF 1.5 would not play in 2.0
I have not taken the time to examine the reasons for the problem, but it most certainly does not inspire a feeling of "getting better" with a new version, which to me is is disappointing, because FF is a very visible OS project and may "color" things for a lot of people out there.
:cry:

considering how slow FF 1.x is, it's a bad thing 2.0 got even slower. I hope 2.0.1 will fix this...

imho it really sucks that the two most visible FOSS projects, OO.o and FF, are two which once where proprietary - as it really shows in quality (both suck in terms of clean code, performance and duplication). I'd rather see more efforts and marketing go to 'real' FOSS projects like KDE and Gnome and their respective software (Koffice, Gnomeoffice and Khtml).

Jacek Poplawski · 2006-10-25 14:44:12

Why exactly do you think that Firefox 2.0 is slower? I've noticed that it's faster. Do you have any benchmarks?

JGC · 2006-10-25 15:57:01

I want to have a firefox in current that complies to the mozilla licensing/trademark stuff before 0.8 gets released. This can either be Firefox 2.0 without branding, or IceWeazel 1.5.0.7, it depends on what the restrictions are for the firefox "community edition" (what community? mozilla is almost non-free software, the only difference is that you get the source).

ralvez · 2006-10-25 16:11:12

Jacek Poplawski wrote:

Why exactly do you think that Firefox 2.0 is slower? I've noticed that it's faster. Do you have any benchmarks?

Because I have both (the old and the new) and when I click on a link it takes a while (anywhere from 5 to 9 sec.) for it to go, while with the old it just goes there.

Update:

I've manually made the following changes to about:config in FF2.0 which greatly improved the performance and responsiveness of the browser, much like my old 1.5:

    user_pref("network.http.pipelining", true);
    user_pref("network.http.proxy.pipelining", true);
    user_pref("network.http.pipelining.firstrequest", true);
    user_pref("network.http.pipelining.maxrequests", 8);

    user_pref("network.http.max-connections", 96);
    user_pref("network.http.max-connections-per-server", 32);
    user_pref("network.http.max-persistent-connections-per-proxy", 24);
    user_pref("network.http.max-persistent-connections-per-server", 12);

The question is,though, why do the user has to do this "by hand"? It should be there by default ... I mean: don't they want to have the browser as fast as possible?

cougar · 2006-10-25 16:29:38

ralvez wrote:

The question is,though, why do the user has to do this "by hand"? It should be there by default ... I mean: don't they want to have the browser as fast as possible?

Because it's not very server respectful

Snarkout · 2006-10-25 16:52:09

Because some sites purposely throttle/reject connections that greedy. Also aggresive pipelining breaks a lot of stuff IME.

dolby · 2006-10-25 17:26:40

swiftox 2.0 in aur seems to work as 1.5.0.7 did. thx to eXire for the fast supply of the pkgbuild

stonecrest · 2006-10-26 00:00:55

cougar wrote:

ralvez wrote:
The question is,though, why do the user has to do this "by hand"? It should be there by default ... I mean: don't they want to have the browser as fast as possible?
Because it's not very server respectful

Yeah, seriously. Why don't you also just enable the rape-server option while you're at it?

Firefox 2 renders faster than 1.5 here, without any omfg-optimize tweaking. I realize it's more popular to criticize projects for every little thing we don't like, rather than point out a project's improvements and benefits, but having used firefox 2 and the rc's for weeks, I think it's a much appreciated improvement.

And 5 to 9 seconds from clicking on a link? That doesn't sound like a firefox issue...

Global Havok · 2006-10-26 02:22:20

stonecrest wrote:

cougar wrote:
ralvez wrote:
The question is,though, why do the user has to do this "by hand"? It should be there by default ... I mean: don't they want to have the browser as fast as possible?
Because it's not very server respectful
Yeah, seriously. Why don't you also just enable the rape-server option while you're at it?
Firefox 2 renders faster than 1.5 here, without any omfg-optimize tweaking. I realize it's more popular to criticize projects for every little thing we don't like, rather than point out a project's improvements and benefits, but having used firefox 2 and the rc's for weeks, I think it's a much appreciated improvement.
And 5 to 9 seconds from clicking on a link? That doesn't sound like a firefox issue...

I've always refrained from pumping up pipelining, but wonder about the actual effects of pipelining on servers. How much of an effect do you think people enabling this has on sites dominated by Firefox (mainly the Slashdot/digg effect)?

Does actually effect bandwidth? Or is it more of a using up too many ports issue? (Sorry I'm clueless when it comes to most networking issues.)

ralvez · 2006-10-26 02:38:21

stonecrest wrote:

cougar wrote:
ralvez wrote:
The question is,though, why do the user has to do this "by hand"? It should be there by default ... I mean: don't they want to have the browser as fast as possible?
Because it's not very server respectful
Yeah, seriously. Why don't you also just enable the rape-server option while you're at it?
Firefox 2 renders faster than 1.5 here, without any omfg-optimize tweaking. I realize it's more popular to criticize projects for every little thing we don't like, rather than point out a project's improvements and benefits, but having used firefox 2 and the rc's for weeks, I think it's a much appreciated improvement.
And 5 to 9 seconds from clicking on a link? That doesn't sound like a firefox issue...

Take it easy champ ... you are too sensitive.
I just made a timed comparison of my installed FF 1.5 and 2.0 and the first has no delay the second does ... that is not my system since both are in the same system.
The tweaks were given by mozilla and I found them Googling, so ... perhaps they should enable the rape-server option themselves
All I'm saying is that I noticed a clear difference in the response of the new browser and if that is the way things are then other people will notice that too and may feel that is not as "powerful" as it used to be.
And just in case you did not notice it I'm not criticizing the project, FF IS my favorite browser I'm just surprised because it was my understanding that it should be faster than FF 1.5

shadowhand · 2006-10-26 05:54:49

This operates on the same principle as a "download accelerator", such as aria2c or axel (search AUR). It opens several connections to a server, and gets different parts of a file at once. Or in the case of Firefox, starts opening different parts of the requested page, one thread for an image, another for a CSS file, etc until it maxes out the number of connections.

cougar wrote:

Because it's not very server respectful

You are making Firefox open a larger number of connections than a standard user (there is an RFC about this) would make. It does however, use your full potential bandwidth. What's more important to you, using all the bandwidth you have, or having to wait a couple of seconds and be nice to the servers you visit? I'd be upset if I tried to come to archlinux.org and couldn't access it because the connections were maxed out. (Maybe /. someday?)

JGC · 2006-10-26 06:30:43

Another thing is that microsoft servers don't even like it when you tune firefox using Fasterfox set to the optimized setting that's still inside RFC specifications. MS IIS will start corrupting images on big pages or just screw up completely and only serve half of the images. Since a big piece of websites is running on MS IIS, I think it's shooting yourself in the foot when you over-optimize the HTTP pipelining/concurrency settings.

iphitus · 2006-10-26 06:34:51

JGC wrote:

Another thing is that microsoft servers don't even like it when you tune firefox using Fasterfox set to the optimized setting that's still inside RFC specifications. MS IIS will start corrupting images on big pages or just screw up completely and only serve half of the images. Since a big piece of websites is running on MS IIS, I think it's shooting yourself in the foot when you over-optimize the HTTP pipelining/concurrency settings.

Always enabled pipelining here, never had such a problem.

ralvez · 2006-10-26 11:38:19

shadowhand
I see your point... and it's fair enough.
But do you see mine? I would expect that the newer version would be as responsive (if not more) than the old version. That, which in itself is just a minor thing (when you compare that to all the added improvements and the fact that is a "good looking" browser and the add-ons an so forth) is where those adverse to FOSS will pick.
I'm sure that FF 1.5 had some tweaks that make it faster or more responsive and I feel it would be nice if the mozilla devs. would have added them "by default".
I want my browser to be #1 !!

shadowhand · 2006-10-26 22:39:01

By the way, I was reading Lifehacker's Firefox 2.0 config tweaks and noticed the bit about "prefetch". I have a feeling that what you are experiencing with the extra load times is probably due to prefetch. I just installed swiftfox from the AUR to test this theory, and noticed a very significant difference when I turned it off. Might be worth looking into.

Also, if you hate the FF 2.0 theme (I do), try this: https://addons.mozilla.org/firefox/3479/

ralvez · 2006-10-26 22:51:53

I already changed my theme
The default theme is not that bad, mind you, but I found it a bit too faded
Thanks for the idea of the pre-fetch, I'll look into it.

benplaut · 2006-10-26 23:41:04

.

Last edited by benplaut (2021-06-25 12:36:25)

ralvez · 2006-10-27 00:09:58

I noticed that most of the themes and add-ons (at least the ones I like ) have been ported to the new FF2
I'm sure you will not be waiting too long.

Arch Linux

#1 2006-10-24 09:51:00

Firefox 2.0 into release?

#2 2006-10-24 13:12:23

Re: Firefox 2.0 into release?

#3 2006-10-24 13:44:07

Re: Firefox 2.0 into release?

#4 2006-10-24 20:40:08

Re: Firefox 2.0 into release?

#5 2006-10-24 21:44:11

Re: Firefox 2.0 into release?

#6 2006-10-25 11:57:36

Re: Firefox 2.0 into release?

#7 2006-10-25 12:49:16

Re: Firefox 2.0 into release?

#8 2006-10-25 14:09:31

Re: Firefox 2.0 into release?

#9 2006-10-25 14:44:12

Re: Firefox 2.0 into release?

#10 2006-10-25 15:57:01

Re: Firefox 2.0 into release?

#11 2006-10-25 16:11:12

Re: Firefox 2.0 into release?

#12 2006-10-25 16:29:38

Re: Firefox 2.0 into release?

#13 2006-10-25 16:52:09

Re: Firefox 2.0 into release?

#14 2006-10-25 17:26:40

Re: Firefox 2.0 into release?

#15 2006-10-26 00:00:55

Re: Firefox 2.0 into release?

#16 2006-10-26 02:22:20

Re: Firefox 2.0 into release?

#17 2006-10-26 02:38:21

Re: Firefox 2.0 into release?

#18 2006-10-26 05:54:49

Re: Firefox 2.0 into release?

#19 2006-10-26 06:30:43

Re: Firefox 2.0 into release?

#20 2006-10-26 06:34:51

Re: Firefox 2.0 into release?

#21 2006-10-26 11:38:19

Re: Firefox 2.0 into release?

#22 2006-10-26 22:39:01

Re: Firefox 2.0 into release?

#23 2006-10-26 22:51:53

Re: Firefox 2.0 into release?

#24 2006-10-26 23:41:04

Re: Firefox 2.0 into release?

#25 2006-10-27 00:09:58

Re: Firefox 2.0 into release?

Board footer