Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Firefox Mozilla News

Firefox Is Lagging Behind, Its Co-Founder Says 646

sopssa writes "Firefox's co-founder Blake Ross is skeptical about the future of Firefox. He says that 'the Mozilla Organization has gradually reverted back to its old ways of being too timid, passive, and consensus-driven to release breakthrough products quickly.' Within the past year Chrome has been steadily increasing its market share, along with the other WebKit-based browsers like Safari. Meanwhile Mozilla's (outgoing) CEO says that while Firefox is more competitive than ever, they're looking forward to their mobile version of Firefox. 'Clearly, both are annoyed at what has happened to their former renegade web browser. But, by many accounts, Firefox is no longer considered to be the light, open alternative it once was.'"
This discussion has been archived. No new comments can be posted.

Firefox Is Lagging Behind, Its Co-Founder Says

Comments Filter:
  • Certainly not light (Score:5, Informative)

    by AnonGCB ( 1398517 ) <7spams@gma i l . com> on Tuesday May 18, 2010 @09:18PM (#32259924)
    In order of resource usage, from a consumer's standpoint I'd rank them: Chrome Opera FireFox Internet Explorer This is not based on any tests but simply my experience using them all. Personally, Chrome is good but Opera has more features I use and is more customizable, so Opera wins out overall - and now Opera is nearly as good as Chrome in benchmarks.
  • And in other news (Score:3, Informative)

    by joeflies ( 529536 ) on Tuesday May 18, 2010 @09:26PM (#32259986)

    Mozilla official cite that the innovation of new features in other browsers suspiciously correlate to the sudden appearance of black duck eggs at restaurants near the Mozilla office.

  • Re:Things Mature (Score:5, Informative)

    by sopssa ( 1498795 ) * <sopssa@email.com> on Tuesday May 18, 2010 @09:30PM (#32260000) Journal

    Opera has been really successful with providing a browser that feels light to use but is still powerful and full of native features. That's probably the reason why Opera feels so constant and fast - all the features are build-in and have the same level of quality. While a better addon system would be good, besides ad blocker (which I use Ad Muncher for), there's not really any features that are missing. And the whole GUI and usage feels a lot more robust than Firefox's XUL-based interface.

  • Re:Firefox 4... (Score:2, Informative)

    by t0y ( 700664 ) on Tuesday May 18, 2010 @09:31PM (#32260010)
    Out of process plugins are coming out soon with 3.6.4 for select plugins.
    In the dev release (minefield) this is already enabled for all plugins and it's fairly stable, save from some corner cases like java's modal security popups. This is seen working perfectly when the plugins hang/crash, which is fairly often with certain builds.

    And I really, really hope they won't go for the process per tab like planned. It takes a lot of extra memory for little benefit (IMO).
  • by AnonGCB ( 1398517 ) <7spams@gma i l . com> on Tuesday May 18, 2010 @09:33PM (#32260016)
    You may not like Opera's default interface, but it's cleaned up in 10.52 and no matter what version it's incredibly customizable. As for an old computer, if internet speed is an issue Opera has a function called 'Turbo', it sends all your requests to Opera's servers and has the pages compressed, speeds up load times immensely.
  • by Ekuryua ( 940558 ) on Tuesday May 18, 2010 @09:42PM (#32260064) Homepage
    Also, it is true that jemalloc(the memory allocator) of firefox is quite broken under linux. It is however quite good under windows, and there are very few leaks from firefox itself nowadays. Most leaks come from non-updated extensions/a couple known culprits(see firebug/video download helper/menu editor, and a couple others)
  • Re:Yes... (Score:5, Informative)

    by ducomputergeek ( 595742 ) on Tuesday May 18, 2010 @09:46PM (#32260088)

    Funny, because originally that goal was to create a stripped down version of Mozilla/Netscape that was lightweight and fast. They seem to have forgotten that it wasn't supposed to be a wholesale replacement for Netscape/Mozilla with all the bells and whistles.

  • by amicusNYCL ( 1538833 ) on Tuesday May 18, 2010 @10:01PM (#32260186)

    I'm an Opera fan myself, but I would not rank IE last in resource usage, certainly not below Firefox. IE deserves a lot of scorn, but I think that one of Microsoft's goals with IE is to use resources as efficiently as possible (considering their huge customer base), and I think they've accomplished that goal. I would be pretty shocked to see any real-world benchmarks where Firefox beats IE in terms of memory use. Granted, IE is going to execute Javascript an order of magnitude slower than anything else, but it's going to do it while using less memory. Although, if Mozilla doesn't get its act together then it's going to soon find itself lagging behind IE9 in Javascript performance. That would be embarrassing. Of the top 5 browsers, Firefox is currently ranked 4th in Javascript performance. The IE9 preview already beats Firefox 3.7, but the IE9 preview isn't an actual browser yet.

    Like I said, IE deserves a lot of scorn for the bugs and differences between everything else, but I think it's safe to say that resource usage might be IE's strongest point (fighting for first place with an easy-to-use UI).

    That being said, Opera still rocks everything else.

  • by Anonymous Coward on Tuesday May 18, 2010 @10:03PM (#32260194)

    Really? Because I just back-navigated to a page on 5.0.375.38, and was taken to the same position.

    But Chrome still has session problems, I think. Several times I have tried to "go back" to restore form data, and the data was lost. I don't know if it still affects Chrome, as it forced me to change my browsing habits to be more careful.

  • by Anonymous Coward on Tuesday May 18, 2010 @10:17PM (#32260264)

    ftp://ftp.mozilla.org/pub/firefox/releases/

    So hard to find. Oh god oh god.

    Most open source stuff is on FTP. Get used to it.

  • Misses the point (Score:3, Informative)

    by Runaway1956 ( 1322357 ) on Tuesday May 18, 2010 @10:22PM (#32260312) Homepage Journal

    Ever since KHTML was branched into WebKit, I've been mildly excited by the possibilities. There are only so many underlying engines on the market. IE, Opera, Webkit, and Firefox each uses a different engine. IMHO, Wekit is the way forward. IE's engine is closed source, and no one can do anything with it - ditto for Opera's. That leaves Webkit versus Gheko. Gheko has been a good, reliable engine, which has been bent and stretched, folded, and mutilated time and again to perform as various coders and/or coding teams have seen fit. But, I think it is nearing the end of it's life.

    From the wikipedia:

    WebKit-based browsers

            * Arora
            * Web Browser for Android (mobile device platform)
            * BOLT browser
            * Google Chrome
            * Epiphany (web browser)
            * iCab (version 4 uses WebKit; earlier versions used its own rendering engine)
            * Iris Browser
            * Konqueror (version 4 can use WebKit as an alternative to its native KHTML[18])
            * Midori
            * OmniWeb
            * OWB
            * Safari
            * Shiira
            * Sputnik for MorphOS (based on S60 WebCore)
            * SRWare Iron
            * Stainless
            * TeaShark
            * Uzbl
            * Web Browser for S60, used in all Nokia Symbian smartphones.
            * WebOS, used in the Palm Pre mobile
            * WebPositive, browser in Haiku

    Grab a couple of them, and test drive them. That should satisfy the Google bashers who might want to experiment with Webkit.

  • Re:Firefox plugins (Score:5, Informative)

    by Peach Rings ( 1782482 ) on Tuesday May 18, 2010 @10:33PM (#32260380) Homepage

    Chrome's noscript is really bad though. You can only allow scripts for the current domain, so if the page uses scripts from a different domain then you have to visit that separately and allow it. And there are no temporary allows, only permanent. And wildcards don't work, so you have to unblock news.slashdot.org separately from yro.slashdot.org.

    Still, Firefox is frequently infuriating and I only use it because Midori isn't mature yet.

  • by IRoll11!s ( 1609859 ) on Tuesday May 18, 2010 @10:35PM (#32260404)
    Sounds like you are hitting the limits of either your ram or disk cache. (in FF) Install the Cache Status addon. Sure it's a bloated interface to what is essentially a few preference settings, but it gives you a visual status in your um, status bar.
  • Bad moderation (Score:3, Informative)

    by imtheguru ( 625011 ) on Tuesday May 18, 2010 @10:37PM (#32260424)

    Both pieces of information in the parent are irrelevant to the problems highlighted by grand parent poster.

    Compression is great for high-latency networks but that isn't even close to the problem expressed above.

    Further, Opera's "cleaned up" default UI is in a version which is yet to be released. It's 2010 and Opera is just getting around to sorting out the default UI. I relent that the previous versions have all been greatly customisable, but then what excuse does Opera have for not starting simple and allowing the users to expose features to meet their needs?

  • Re:Yes... (Score:5, Informative)

    by amicusNYCL ( 1538833 ) on Tuesday May 18, 2010 @10:37PM (#32260428)

    Of course development has slowed - it has achieved the goal most users/developers have wanted for it: To be a stable, fairly secure platform that allows a decent plugin model, and works consistently between platforms.

    What? Where did you get that from?

    From Wikipedia [wikipedia.org]:

    The Firefox project began as an experimental branch of the Mozilla project by Dave Hyatt, Joe Hewitt and Blake Ross. They believed the commercial requirements of Netscape's sponsorship and developer-driven feature creep compromised the utility of the Mozilla browser. To combat what they saw as the Mozilla Suite's software bloat, they created a stand-alone browser, with which they intended to replace the Mozilla Suite.

    Intended to combat feature creep. It was designed to be a lightweight standalone browser. See any mention in there about a decent extension model (plugins aren't the same as extensions BTW; Flash is a plugin, Adblock is an extension)?

    From Computer World [computerworld.com] in Sept. 2002, the week Phoenix 0.1 was released:

    The Mozilla development project, Mozilla.org, this week released Phoenix 0.1, a speedier version of its open-source Web browser.
    The Phoenix browser is designed to improve upon Mozilla 1.1, released in August, with additional features such as a new design, customizable tool bar and improved bookmark manager...

    The Phoenix browser, which uses a large amount of the Mozilla code, is "a lean and fast browser" that loads in about half the time of Mozilla 1.1, Mozilla.org said.

    Again, emphasis is on performance. The line in that article talking about the plugin management for version 0.2 is referring to classical plugins, not Firefox extensions. Extensions were not added to Firefox until version 2.0. Extensions were never an original design goal. I don't have a source for this, but I actually remember downloading Phoenix 0.1. It was distributed as a single zip file without an installer, you just unzipped it and ran the executable. What people were impressed with for that release were the disk size of the files, the startup speed, and the memory footprint. All performance metrics.

    It's fine if you want to defend Firefox, but there's no reason to try to rewrite history by saying the design goals for Firefox were different than what they actually were. It's a fact that the current version of Firefox does not live up to many of the ideals that the designers of Phoenix started with. It's also a fact that the current version includes several useful things that were not part of the original goals. Again, there's no reason to rewrite history. People like to defend Firefox because of its extensions, but the fact is that extensions were never part of the plan, speed and performance were the goals. The extension model was added because the core browser lacked many features that could not be included and still meet the performance goals. So, now we have an extension model and worse performance. That's the way it goes.

    And yes, I remember this happening. I remember downloading and using Phoenix, I remember the name change to Firebird and then to Firefox, and the initial release in 2004, 2 years after Phoenix started. The release of Firefox 1.0 was a major event in the tech world, they even ran full-page ads for it in the New York Times funded by donations (you got your name listed in the ad, I was there). I remember using the 1.x line, I remember when the extension system was announced for 2.0 and how much it excited everyone, 2 years after the release of 1.0. I remember continually seeing the performance of the browser decline. That hasn't really stopped, even the IE9 preview is now faster at Javascript than Firefox 3.7. So the conclusion holds, the original design goals of Firefox have been neglected or ignored in part, and some of them have

  • by Ilgaz ( 86384 ) on Tuesday May 18, 2010 @10:53PM (#32260510) Homepage

    Using Opera for my needs, such as 720P HDTV TV (try other browsers!), Nokia E71 Symbian S60 which Google didn't even bother etc.

    When you use it in such conditions, you figure the sad fact. Even if Opera switched to open source, it would be some developer PR nightmare since most of the commits would be rejected. Even a single line in HTML renderer must be coded with a Symbian OS, some plane video terminal, some car dashbard, some SD gaming console "web channel" in mind. What amazes me is, they still manage to keep up with the trends and actually implement on impossible to count platforms equally. For example, their Android beta has JIT compiler for Javascript while Google couldn't manage/care enough to ship Chrome to PPC/OS X.

    If Firefox really wants a truly mobile version, they should be ready to shave a lot of the code and reject a lot of commits. For example, if someone's super cool JIT patch doesn't work on ARM, the should reject it.

    They should have started experimenting with Symbian right after "S60 V3" handsets started to ship. Calling Nokia evil or joking with others RAM (their so called supporters) didn't help of course.

  • Re:Things Mature (Score:5, Informative)

    by geekboy642 ( 799087 ) on Tuesday May 18, 2010 @10:54PM (#32260520) Journal

    Or, perhaps more significantly in the day of netbooks, try Chrome, Firefox, and Opera with a slow or very small main drive. Opera is the only one I've found that doesn't lock up loading a page with the hdd-led solid red, and that is why Opera has evicted every other browser from my EeePC.

  • by sznupi ( 719324 ) on Tuesday May 18, 2010 @11:17PM (#32260652) Homepage

    That model isn't bad on desktops / laptops, either. The browser is becoming / has become(?) the main app many people use on their PCs. And possibly one of only two (the second being video editing) which can still greatly benefit from improvements in processing power or...attention to its performance during development.

    I would hope we can prefer the second, less wasteful, solution.

    Some places already seem to do so; places where a PC has typically much longer lifetime. Ukraine [ranking.com.ua], where Opera is handily the #1 browser; Russia [rankingru.com], where it's the #1 "alternative" one.

  • Re:Firefox plugins (Score:5, Informative)

    by eosp ( 885380 ) on Wednesday May 19, 2010 @12:01AM (#32260896) Homepage
    Chrome's AdBlock still downloads the offending components; it just hides them from the user.
  • Re:Firefox plugins (Score:3, Informative)

    by Aczlan ( 636310 ) on Wednesday May 19, 2010 @12:05AM (#32260922)
    Does AdBlock in Chrome block ads BEFORE they are downloaded? Or does still it hide them AFTER they are downloaded? Last time I tried it all that it did was hide ads, which doesn't help when ad servers are being slow.

    Aaron Z
  • Re:Things Mature (Score:4, Informative)

    by scdeimos ( 632778 ) on Wednesday May 19, 2010 @12:11AM (#32260960)
    Having worked on legacy codebases I can say that "add more features, expect more bloat" doesn't hold as a rule. Adding a new feature often causes a refactoring of code because of the awful way it was written originally. This often leads to a reduction in code complexity and size. Firefox is about as legacy as it gets, with code going back beyond Phoenix into the hallowed halls of Communicator and Navigator. Firefox gets jacked-up occasionally to have entire subsystems replaced, as was the case in FF3.0 when Mork was (thankfully) wheeled out and replaced with SQLite for storage and the Gecko layout engine was upgraded to 1.9.
  • by BZ ( 40346 ) on Wednesday May 19, 2010 @12:29AM (#32261046)

    There are several things at play here:

    1) Webkit provides an HTML/XML/DOM/CSS renderer, period. Gecko provides that plus a
            networking library, SSL implementation, and so forth, on most platforms. To create a
            usable browser on top of webkit you have to provide all those components. But if you
            have to custom-write them anyway for your crazy hardware or OS, then the existing Gecko
            implementations don't do you much good. Also, if you want to do something very
            different from what the existing infrastructure in Gecko is set up to do in terms of
            document navigation, etc, then the existing functionality might get in your way
            instead of helping.
    2) Webkit is perceived as being simpler and easier to hack than Gecko. It's not clear to
            me how much truth there is to this perception nowadays; back when Apple picked khtml I
            think it was more true.
    3) Webkit has better PR in some ways. It's been actively marketed to developers more
            than Gecko has.
    4) People seem to have a double standard on embedding the two (e.g. demanding binary
            compatibility out of Gecko across releases but not making any such demands on Webkit
            for some reason).
    5) There are existing Gecko-based browsers on mobile devices (e.g. the n810 and n900
            default browser is Gecko-based).

    For apple's original decision to use khtml, I believe it was a combination of #2 above and wanting something they would have more of a chance of controlling (hence the forking that happened).

  • by surveyork ( 1505897 ) on Wednesday May 19, 2010 @01:10AM (#32261224) Journal
    Finally, Mozilla developers have acknowledged there really is a problem: https://bugzilla.mozilla.org/show_bug.cgi?id=490122 [mozilla.org] Apparently, these micro-pauses are related to garbage collection, cycle collection and other synchronous I/O activity. The folks at Mozilla are working on garbage collection and I/O improvements. This should help with the micro-pauses.
  • Re:Misses the point (Score:3, Informative)

    by Randle_Revar ( 229304 ) <kelly.clowers@gmail.com> on Wednesday May 19, 2010 @01:39AM (#32261346) Homepage Journal

    I think Presto is Opera's engine

  • by AlgorithMan ( 937244 ) on Wednesday May 19, 2010 @03:23AM (#32261854) Homepage
    I was sick of FF's performance, so tried to improve it. Here are my conclusions:
    When you delete (or move) your firefox settings directory, then firefox is fast as hell again, so FF is not inherently slow. It becomes slower over time, so I looked for things that changed over time and I found that the performance stays good (except for flash!) even if you navigate dozens of sites on 4 year old hardware, if you do the following things:
    • restrict yourself to the indispensable add-ons (although comparing browsers with add-ons to browsers without add-ons is quite pointless, imho)
    • use f*cking adblock! ads eat up so much performance! adblock is the one add-on that will improve your performance
    • clear the history and see how fast firefox suddenly becomes again... I have set firefox to delete the history and search history on closing the prog. This can become a pain sometimes, but the performance is woth the sacrifice

A failure will not appear until a unit has passed final inspection.

Working...