13 ms·

I read the article hoping for something insightful, but when I read the paragraph ...a half-dozen tabs that auto-open on launch thanks to the dozens of extensio

by gvb 12y ago

I read the article hoping for something insightful, but when I read the paragraph ...a half-dozen tabs that auto-open on launch thanks to the dozens of extensions I've accumulated over the years[...] I had to stop and say "wait, what?" To leverage his analogy that it is like coming home to a new house to find that most of your stuff is already there, it is like not cleaning your house and not taking out the trash for three years and then moving into a new house. Yeah, it's clean and new, but why is it better and will it stay better?

rockdoe 12y ago
To be fair quite a few users left Firefox because they found it "bloated" or "it didn't work anymore" but never realized they were the ones that installed the bloat/adware/malware.
- moe 12y ago
  I left Firefox because it was measurably slower than Chrome at the time (and not by a small margin) and because crashes/hangs would usually take the whole Browser down.
  - dragonshed 12y ago
    This. Have the scales tipped the other way recently? Chrome still feels faster to me, but I only have one active extension (Ad Blocker).
    - zxcvcxz 12y ago
      I think most people on HN using FF are running nightly versions, I know I was. The nightly versions of FF have always had the edge on Chrome, but I think it's just recently that the stable version of FF has really gotten pretty much on par with Chrome, and I'd say definitely better on the Linux 64bit.
      - wooger 12y ago
        Nightly -> Stable every ~12 weeks or so, so should feel just as fast. I noticed this difference too, but I believe that's as I use it with a fresh profile each time.
    - Ar-Curunir 12y ago
      Chrome doesn't work well for me on Linux at all, whereas Firefox functions beautifully.
      - voidlogic 12y ago
        While "work well at all" might be an exaggeration, I have found Firefox to run much better than Chrome/Chromium on Ubuntu 14.04 LTS as well.
    - cpeterso 12y ago
      The Ad Block Plus extension is itself a notorious memory and CPU hog [1]. Chrome's slimmer uBlock extension has now been ported to Firefox [2]. [1] https://blog.mozilla.org/nnethercote/2014/05/14/adblock-pluss-effect-on-firefoxs-memory-usage/ https://blog.mozilla.org/nnethercote/2014/05/14/adblock-plus... [2] https://github.com/gorhill/uBlock/releases https://github.com/gorhill/uBlock/releases
      - wtallis 12y ago
        Please don't put the blame solely on AdBlock Plus. Most of the problem is due to the rulesets, not the extension itself. EasyList and friends have a truly frightening number of broad CSS element hiding rules, which are much less efficient than the URL blocking rules that were the original AdBlock mode and should have stayed the primary means of blocking.
        wooger 12y ago
        This is patently untrue, as I subscribe to the exact same filter rulesets in uBlock, and it makes a dramatic, incredible difference to the browser's performance.
        wtallis 12y ago
        I didn't say that there was no difference in performance between uBlock and ABP. I was explaining that the performance problems with ABP are a result of it being misused. uBlock tolerates the abuse much better, but URL based blocking should still be the first line of defense.
      - vandahm 12y ago
        Thanks for the link to uBlock. I've been growing frustrated with Ghostery and AdBlock Plus, but didn't know of a good alternative. I'm giving uBlock a try, and so far, so good.
        slasaus 12y ago
        Although µBlock might be a lot slimmer than AdBlock Plus, according to this article [1] it's comparable with Ghostery. Ofcours µBlock is open source where Ghostery is not and I personlly really like µBlock's philosophy [2] about tracking and privacy. [1] http://www.extremetech.com/computing/199188-%C2%B5block-aims-to-block-ads-without-draining-system-resources http://www.extremetech.com/computing/199188-%C2%B5block-aims... [2] https://github.com/gorhill/uBlock#philosophy https://github.com/gorhill/uBlock#philosophy
  - hackuser 12y ago
    > I left Firefox mainly because it was measurably slower than Chrome at the time (and not by a small margin) I have no idea what your experience was, of course. Many people said this for years while Firefox consistently won or tied in objective comparisons. Having dealt with end-user complaints for years, I've learned to doubt reports that rely on subjective perceptions of things like 'speed'. Sometimes the new, cool thing feels faster. Perhaps Firefox should pump in some fake engine noise.
    - moe 12y ago
      Those "objective comparisons" then probably compared the rendering engines or some other micro benchmarks. My main problem was the GUI, not the rendering (though the latter probably played its part in slowing the GUI down). Opening/closing/moving tabs was sluggish. Just opening Firefox with my ~20 tabs open would take many times as long as opening Chrome with the same number of tabs. You didn't need a stopwatch to see the difference.
      - scholia 12y ago
        Of course, things have changed since then....
    - sliverstorm 12y ago
      I was one of the small group suffering (I think back around Firefox 24). It indeed was winning reviews & praise- but it was also completely unresponsive on my high-end machine. Several seconds of unresponsive, as if it/I were having problems with interrupts. I realize most people didn't have that problem; it was probably a freak corner case on my machine.
    - jlarocco 12y ago
      I switched to Chrome primarily because of that doubting. For a while a pretty good chunk of people were complaining about Firefox's speed and memory usage, and the dev team stuck their head in the sand and blamed plugins, extensions, slow hardware, OS problems, malware, and everything under the sun except Firefox itself. At some point people just said, "Fuck it, I'm not even trying to convince them any more, I'm just switching to Chrome." Also note that Firefox was "winning" objective benchmarks on things that didn't matter. Rendering speed doesn't mean much to me if it takes 10x longer to open the tab before rendering begins.
- speeder 12y ago
  For me the bloatware was Firefox 3.0 that installed itself against my wishes. I was using then Firefox 2 or 2.5 (don't remember) quite happily. It was just great (compared to Explorer). Chrome showed up... But Firefox 2.5 was still much better. Then one day, Firefox started to INSIST to install 3.0, then 3.5, then one day it just did, no matter how hard I tried to keep my old version. And then behold, Firefox 3.5 had a bigger interface, that used more screen space (and I am in Brazil, bigger screens and bigger resolutions get expensive FAST, my current laptop, that is a ASUS gaming one, is still 14 inches and the maximum vertical res is 768, screens better than that were unaffordable to me), and it was SLOW, VERY, VERY slow, like RIDICULOUS slow. So I hopped into Chrome... and I wish I could say I never looked back, but hogging 600mb just to load reuters.com webpage is NOT nice. I am only still using chrome for google-ecosystem reasons. If I could get rid of relying on google stuff (several work-related things depend on google) I probably would get on Firefox or Opera (or see if I can put Konqueror on Windows? I never tried that...)
  - speeder 12y ago
    Can you people explain why you are downvoting me for my anecdote?
    - hackuser 12y ago
      Feedback as requested: It's the all caps hyperbole. It contributes little; it undermines interesting, valuable discussions; it's contagious, spreading from one poster to another; and it's also used so much in so many forums that long ago it became tiresome and dull. That wasn't your whole post, but when I see it I stop reading and move on. Anyway, don't worry too much about a few negative mods.
  - zanny 12y ago
    Apparently Rekonq has a Windows version, at least according to a Google search and resultant blog post and winkde repo. http://lpapp.blogspot.com/2012/09/randa-rekonq-installer-for-windows.html http://lpapp.blogspot.com/2012/09/randa-rekonq-installer-for... Generally Rekonq is considered the better maintained KDE browser nowadays.
- nkozyra 12y ago
  I'd hate for this to become the revisionist history of the migration from FF -> Chrome. When I finally made the move FF had no extensions, no malware or any add-ons. I ran the same (mostly) version on Ubuntu, OSX and Windows 7. It had become unusably slow on all platforms. Looking around on the Internet in those days saw enough validation of the problem being widespread in terms of speed and memory resources that I tried Chrome and found it significantly snappier. In fact, I still do, although I note the great strides FF has taken in that regard. My problem is I haven't seen the same degradation in Chrome that I saw in FF a few years back. If it's getting bloated, I've missed it (in terms of palpable heft with slowdowns, crashes).
  - logn 12y ago
    The historical FF record: https://areweslimyet.com/ https://areweslimyet.com/
    - nnethercote 12y ago
      That site is the historical record of memory usage on one automated benchmark on one particular test machine. It's useful for finding memory regressions but you really can't draw general conclusions from it. In particular, it's worst-case behaviour (esp. spikes) that hurts the most with memory usage, and Firefox is much better in that regard than it used to be.
skrowl 12y ago
Throw bartab heavy on to firefox - https://addons.mozilla.org/en-US/firefox/addon/bartab-heavy/?src=search https://addons.mozilla.org/en-US/firefox/addon/bartab-heavy/... Now you can have HUNDREDS of tabs (in firefox's nice tab groups, of course) and they'll only stay open / in-memory for a configurable amount of time after you stop using them. Keeps memory utilization tight. I'm currently south of 500MB utilization with hundreds of tabs open and lots of extensions: Disconnect Search, DownThemAll, FoxyProxy, Greasemonkey, HTTPS Everywhere, Imagus, Stylish and uBlock
omaranto 12y ago
Well, as one piece of anecdotal evidence: I only use one browser extension (Vimium on Chrome, VimFx on Firefox) and still find Firefox significantly faster than Chrome. Also, I like using the same browser on my Android tablet as on my laptop and websites that use MathJax render incorrectly on Android Chrome (these sites render perfectly well on Android Firefox and on both desktop Firefox and desktop Chrome). So I'd probably use Firefox even if it were a little slower than Chrome.
- perishabledave 12y ago
  There is a Vimium equivalent for Firefox?! I'm sold!
  - markatto 12y ago
    There are several; vimperator has been around since before chrome even existed.
  - omaranto 12y ago
    Chronologically it's the other way around I think: Vimium was inspired by the Firefox plugins Vimperator and Pentadactyl. (The one I mentioned, VimFx, is more recent and more minimalistic that either Vimperator or Pentadactyl.)
    - commandar 12y ago
      The minimalism is why I vastly prefer Vimium to e.g. Vimperator, personally.
  - wooger 12y ago
    You're kidding right? Vimperator and the much better fork Pentadactyl have been around for years on Firefox, and implement everything Vimium does and more, in a much more reliable way. Every single control interface is available, and scriptable in Pentadactyl. Vimium isn't even the best vimperator clone for Chrome - cVim has now taken that crown by a very large margin, and is worlds better in terms of features, configuration and reliability.
emehrkay 12y ago
I'm not a Chrome-by-default user. But when it is closed and I drag a link to its icon, it is very infuriating that it just doesn't open the link that I asked it to, but whatever else it felt like opening alongside it.
- gknoy 12y ago
  You can disable the "open all the tabs I had open last time" behavior in the settings, though. Is Firefox more clairvoyant about what you want (e.g., startup with arguments does not load prior tabs, but normal startup does)?
falcolas 12y ago
He mentioned importing all of his plugins and preferences from Chrome back into Firefox, so it stands to reason that his benchmark between the two is roughly the same. As for the question of will it change? Well, to be frank, it shouldn't. It's known, however, that it will: Chrome in particular is known to not handle the accumulated history well. I haven't had a similar experience with Firefox, but that's only one data point.
- thefreeman 12y ago
  plugins / extensions are not imported between chrome and firefox, which is the bulk that is being referred to.
  - bryanlarsen 12y ago
    Given that the OP installed "cloud to butt plus" into Firefox, I think it's fair to say that he's doing a fair comparison.
ademarre 12y ago
Right. That paragraph about the extensions makes me question the credibility of his claims. Disappointing. I haven't experienced the performance problems he describes, and I use the Chrome dev channel for my main browser. I've noticed Firefox has been getting faster over the last couple years, but when I use it I still get the feeling that it gets in my way more than Chrome does. Edit: fix typo
- smt88 12y ago
  He's part of the rule that tech writers aren't technical people and often write about things they don't understand. As some famous person recently said, there's a huge amount of "intellectual dishonesty" in tech journalism.
  - bmelton 12y ago
    To spin it more positively, he and his ilk more generally represent our users than more "in the know" tech-journalists. I think this is a rough article, but likely one that either many users will empathize with, so the overarching idea that browsers should be doing something to prevent bloat is probably a sound one.
    - smt88 12y ago
      I guess I see journalism the old-fashioned way: you're not an expert on everything, but you do research and distill the concepts for regular readers. This guy clearly didn't do research. Yes, Chrome uses a massive amount of memory, but that was originally a feature. Most people had way more memory than they needed at any given time, so Chrome heavily used memory to decrease processing time and therefore improve rendering speed. It also split into many processes to insulate each part of the browser from crashes and security holes. Now that the age of SoCs/tablets/etc is upon us, people don't have as much extra memory as they did when Chrome first arrived. That may have an impact on how usable Chrome is (don't know if it's smart about "dialing down" its memory usage). But this guy didn't mention any of that. Like a modern "journalist", he just wrote about his own anecdotal experiences without learning about the underlying concepts.
sbochins 12y ago
I still use pentadactyl, even though it isn't being supported AFAIK. These kind of customizations are not possible on chrome. This is why I never switched even when it was cool.
- wyclif 12y ago
  I love Pentadactyl too, that's why I'm amazed it's no longer supported. Seems to me that with all the vim and Firefox-using devs out there, a lot of people would want to support it. But I have no evidence of that.
FilterJoe 12y ago
The problem is using many open tabs with Chrome. If you use no more than 3 or 4 tabs, Chrome is generally fine. If you use more than that, Firefox has some built-in features to help with tabs such as delayed tab loading and tab grouping. It also has more and better extensions for helping with multiple tabs. Furthermore, it doesn't launch a new process for each tab. In a browser comparison I wrote last month, I included a section for users who like to keep a lot of tabs open. Even Opera, which uses the same code base as Chrome, does tabs better than Chrome: http://www.filterjoe.com/2015/01/23/best-browsers-2015-which-browser-is-best-for-you/#ManyTabs http://www.filterjoe.com/2015/01/23/best-browsers-2015-which...
- emodendroket 12y ago
  I imagine this has something to do with the sandboxing in Chrome that prevents one tab from causing the entire browser to crash, no?
  - FilterJoe 12y ago
    Yes. Sandboxing makes sense when you think of browser as operating system. Isolate each process to increase security and crash-resistance. Makes sense, but it takes more memory.
    - emodendroket 12y ago
      Frankly I have more memory than I need for most tasks so I don't mind this trade-off.
      - frozenport 12y ago
        On the other hand memory memory takes more time to read, potentially leading to latency.
  - vxNsr 12y ago
    I don't know, I've had one tab cause other tabs to crash all the time, which is something that confused me considering the supposed isolation that should be happening. edit: not sure why I've been dv, but I'd like to add that one thing I've noticed is that this will usually happen when I open links in a new tab, somehow all those tabs become linked and if one becomes unresponsive they all do. For those who claim I should stop opening links in new tabs, I say: welcome to 2004, this is the whole purpose of tabbed browsing. If you;d like to try to reproduce this: go to facebook and open the first ten links in your newsfeed in new tabs (either by middle clicking, ctrl+click, or context menu) I don't normally do more than one or two at a time but ten will force the issue. then go to another tab that isn't connect to those other 11 it'll work fine while those 11 will look like they're on pause. Note: I haven't tested this in other browsers.
jackmaney 12y ago
I had the exact same reaction. Why would an extension automatically open multiple tabs? That sounds pretty malware-ish.
- emodendroket 12y ago
  I remember one or two extensions I liked started injecting ads into pages with updates, which led me to uninstall them but would probably go unnoticed by a lot of users.
- Domenic_S 12y ago
  Happens all the time. "Ghostery is updated!" "Thanks for updating RES!" "More update tabs opening up!"
engendered 12y ago
I had to stop and say "wait, what?" To be fair, much of the huffing and puffing "I'm leaving Firefox...Chrome is so much better!" was based upon the same sort of rhetoric. Early on Chrome didn't even have extensions, much less bloated monstrosities like Ad Block Plus and Firebug (don't get me wrong -- they both are fantastic, but most don't realize how incredibly onerous they are). So those people kicked Firefox to the curb and talked about how grand everything was with Chrome. Now that Chrome has an ecosystem, they put themselves in the same situation again.
throwawaymsft 12y ago
It doesn't matter. The user gets a bad experience and blames Firefox. "Oh, it's those naughty extensions slowing everything down!" smirk the developers. Do you think the user cares? Firefox is the one who opened themselves up to get abused by. (Same things happened on Windows, you'll get a machine full of crapware if you install popular apps. OSX has a curated app store. Look at the difference on an average windows and OSX machine.) Some solutions instead of blaming the user: * Name & shame poorly-performing plugins in the interface. (OSX does this with "apps using significant power" in the battery meter) * Have a plugin speed rating in the store * Have performance requirements for plugins (users will report usage data, you have X seconds to startup, if you average higher you will be reviwed/yanked.) Asking users to not install extensions when extensibility is one of your selling points is not a good idea. Fix how they interact with the browser.
rnernento 12y ago
Despite that fact I've run both without add-ons and I'm very happy to be back using Firefox. Like the author there was a point when I switched to Chrome, but even without add-ons it has become bloated and slow. My breaking point was when it decided to auto-install a Google Now notification center that started up with my computer.
nailer 12y ago
You could have realised this was linkbait from the domain name - dang can we please lower the rank for gizmodo.com links?

by gvb 12y ago

rockdoe 12y ago

To be fair quite a few users left Firefox because they found it "bloated" or "it didn't work anymore" but never realized they were the ones that installed the bloat/adware/malware.

moe 12y ago
I left Firefox because it was measurably slower than Chrome at the time (and not by a small margin) and because crashes/hangs would usually take the whole Browser down.
- dragonshed 12y ago
  This. Have the scales tipped the other way recently? Chrome still feels faster to me, but I only have one active extension (Ad Blocker).
  - zxcvcxz 12y ago
    I think most people on HN using FF are running nightly versions, I know I was. The nightly versions of FF have always had the edge on Chrome, but I think it's just recently that the stable version of FF has really gotten pretty much on par with Chrome, and I'd say definitely better on the Linux 64bit.
    - wooger 12y ago
      Nightly -> Stable every ~12 weeks or so, so should feel just as fast. I noticed this difference too, but I believe that's as I use it with a fresh profile each time.
  - Ar-Curunir 12y ago
    Chrome doesn't work well for me on Linux at all, whereas Firefox functions beautifully.
    - voidlogic 12y ago
      While "work well at all" might be an exaggeration, I have found Firefox to run much better than Chrome/Chromium on Ubuntu 14.04 LTS as well.
  - cpeterso 12y ago
    The Ad Block Plus extension is itself a notorious memory and CPU hog [1]. Chrome's slimmer uBlock extension has now been ported to Firefox [2]. [1] https://blog.mozilla.org/nnethercote/2014/05/14/adblock-pluss-effect-on-firefoxs-memory-usage/ https://blog.mozilla.org/nnethercote/2014/05/14/adblock-plus... [2] https://github.com/gorhill/uBlock/releases https://github.com/gorhill/uBlock/releases
    - wtallis 12y ago
      Please don't put the blame solely on AdBlock Plus. Most of the problem is due to the rulesets, not the extension itself. EasyList and friends have a truly frightening number of broad CSS element hiding rules, which are much less efficient than the URL blocking rules that were the original AdBlock mode and should have stayed the primary means of blocking.
      - wooger 12y ago
        This is patently untrue, as I subscribe to the exact same filter rulesets in uBlock, and it makes a dramatic, incredible difference to the browser's performance.
        wtallis 12y ago
        I didn't say that there was no difference in performance between uBlock and ABP. I was explaining that the performance problems with ABP are a result of it being misused. uBlock tolerates the abuse much better, but URL based blocking should still be the first line of defense.
    - vandahm 12y ago
      Thanks for the link to uBlock. I've been growing frustrated with Ghostery and AdBlock Plus, but didn't know of a good alternative. I'm giving uBlock a try, and so far, so good.
      - slasaus 12y ago
        Although µBlock might be a lot slimmer than AdBlock Plus, according to this article [1] it's comparable with Ghostery. Ofcours µBlock is open source where Ghostery is not and I personlly really like µBlock's philosophy [2] about tracking and privacy. [1] http://www.extremetech.com/computing/199188-%C2%B5block-aims-to-block-ads-without-draining-system-resources http://www.extremetech.com/computing/199188-%C2%B5block-aims... [2] https://github.com/gorhill/uBlock#philosophy https://github.com/gorhill/uBlock#philosophy
- hackuser 12y ago
  > I left Firefox mainly because it was measurably slower than Chrome at the time (and not by a small margin) I have no idea what your experience was, of course. Many people said this for years while Firefox consistently won or tied in objective comparisons. Having dealt with end-user complaints for years, I've learned to doubt reports that rely on subjective perceptions of things like 'speed'. Sometimes the new, cool thing feels faster. Perhaps Firefox should pump in some fake engine noise.
  - moe 12y ago
    Those "objective comparisons" then probably compared the rendering engines or some other micro benchmarks. My main problem was the GUI, not the rendering (though the latter probably played its part in slowing the GUI down). Opening/closing/moving tabs was sluggish. Just opening Firefox with my ~20 tabs open would take many times as long as opening Chrome with the same number of tabs. You didn't need a stopwatch to see the difference.
    - scholia 12y ago
      Of course, things have changed since then....
  - sliverstorm 12y ago
    I was one of the small group suffering (I think back around Firefox 24). It indeed was winning reviews & praise- but it was also completely unresponsive on my high-end machine. Several seconds of unresponsive, as if it/I were having problems with interrupts. I realize most people didn't have that problem; it was probably a freak corner case on my machine.
  - jlarocco 12y ago
    I switched to Chrome primarily because of that doubting. For a while a pretty good chunk of people were complaining about Firefox's speed and memory usage, and the dev team stuck their head in the sand and blamed plugins, extensions, slow hardware, OS problems, malware, and everything under the sun except Firefox itself. At some point people just said, "Fuck it, I'm not even trying to convince them any more, I'm just switching to Chrome." Also note that Firefox was "winning" objective benchmarks on things that didn't matter. Rendering speed doesn't mean much to me if it takes 10x longer to open the tab before rendering begins.
speeder 12y ago
For me the bloatware was Firefox 3.0 that installed itself against my wishes. I was using then Firefox 2 or 2.5 (don't remember) quite happily. It was just great (compared to Explorer). Chrome showed up... But Firefox 2.5 was still much better. Then one day, Firefox started to INSIST to install 3.0, then 3.5, then one day it just did, no matter how hard I tried to keep my old version. And then behold, Firefox 3.5 had a bigger interface, that used more screen space (and I am in Brazil, bigger screens and bigger resolutions get expensive FAST, my current laptop, that is a ASUS gaming one, is still 14 inches and the maximum vertical res is 768, screens better than that were unaffordable to me), and it was SLOW, VERY, VERY slow, like RIDICULOUS slow. So I hopped into Chrome... and I wish I could say I never looked back, but hogging 600mb just to load reuters.com webpage is NOT nice. I am only still using chrome for google-ecosystem reasons. If I could get rid of relying on google stuff (several work-related things depend on google) I probably would get on Firefox or Opera (or see if I can put Konqueror on Windows? I never tried that...)
- speeder 12y ago
  Can you people explain why you are downvoting me for my anecdote?
  - hackuser 12y ago
    Feedback as requested: It's the all caps hyperbole. It contributes little; it undermines interesting, valuable discussions; it's contagious, spreading from one poster to another; and it's also used so much in so many forums that long ago it became tiresome and dull. That wasn't your whole post, but when I see it I stop reading and move on. Anyway, don't worry too much about a few negative mods.
- zanny 12y ago
  Apparently Rekonq has a Windows version, at least according to a Google search and resultant blog post and winkde repo. http://lpapp.blogspot.com/2012/09/randa-rekonq-installer-for-windows.html http://lpapp.blogspot.com/2012/09/randa-rekonq-installer-for... Generally Rekonq is considered the better maintained KDE browser nowadays.
nkozyra 12y ago
I'd hate for this to become the revisionist history of the migration from FF -> Chrome. When I finally made the move FF had no extensions, no malware or any add-ons. I ran the same (mostly) version on Ubuntu, OSX and Windows 7. It had become unusably slow on all platforms. Looking around on the Internet in those days saw enough validation of the problem being widespread in terms of speed and memory resources that I tried Chrome and found it significantly snappier. In fact, I still do, although I note the great strides FF has taken in that regard. My problem is I haven't seen the same degradation in Chrome that I saw in FF a few years back. If it's getting bloated, I've missed it (in terms of palpable heft with slowdowns, crashes).
- logn 12y ago
  The historical FF record: https://areweslimyet.com/ https://areweslimyet.com/
  - nnethercote 12y ago
    That site is the historical record of memory usage on one automated benchmark on one particular test machine. It's useful for finding memory regressions but you really can't draw general conclusions from it. In particular, it's worst-case behaviour (esp. spikes) that hurts the most with memory usage, and Firefox is much better in that regard than it used to be.

skrowl 12y ago

Throw bartab heavy on to firefox - https://addons.mozilla.org/en-US/firefox/addon/bartab-heavy/?src=search https://addons.mozilla.org/en-US/firefox/addon/bartab-heavy/... Now you can have HUNDREDS of tabs (in firefox's nice tab groups, of course) and they'll only stay open / in-memory for a configurable amount of time after you stop using them. Keeps memory utilization tight. I'm currently south of 500MB utilization with hundreds of tabs open and lots of extensions: Disconnect Search, DownThemAll, FoxyProxy, Greasemonkey, HTTPS Everywhere, Imagus, Stylish and uBlock

omaranto 12y ago

Well, as one piece of anecdotal evidence: I only use one browser extension (Vimium on Chrome, VimFx on Firefox) and still find Firefox significantly faster than Chrome. Also, I like using the same browser on my Android tablet as on my laptop and websites that use MathJax render incorrectly on Android Chrome (these sites render perfectly well on Android Firefox and on both desktop Firefox and desktop Chrome). So I'd probably use Firefox even if it were a little slower than Chrome.

perishabledave 12y ago
There is a Vimium equivalent for Firefox?! I'm sold!
- markatto 12y ago
  There are several; vimperator has been around since before chrome even existed.
- omaranto 12y ago
  Chronologically it's the other way around I think: Vimium was inspired by the Firefox plugins Vimperator and Pentadactyl. (The one I mentioned, VimFx, is more recent and more minimalistic that either Vimperator or Pentadactyl.)
  - commandar 12y ago
    The minimalism is why I vastly prefer Vimium to e.g. Vimperator, personally.
- wooger 12y ago
  You're kidding right? Vimperator and the much better fork Pentadactyl have been around for years on Firefox, and implement everything Vimium does and more, in a much more reliable way. Every single control interface is available, and scriptable in Pentadactyl. Vimium isn't even the best vimperator clone for Chrome - cVim has now taken that crown by a very large margin, and is worlds better in terms of features, configuration and reliability.

emehrkay 12y ago

I'm not a Chrome-by-default user. But when it is closed and I drag a link to its icon, it is very infuriating that it just doesn't open the link that I asked it to, but whatever else it felt like opening alongside it.

gknoy 12y ago
You can disable the "open all the tabs I had open last time" behavior in the settings, though. Is Firefox more clairvoyant about what you want (e.g., startup with arguments does not load prior tabs, but normal startup does)?

falcolas 12y ago

He mentioned importing all of his plugins and preferences from Chrome back into Firefox, so it stands to reason that his benchmark between the two is roughly the same. As for the question of will it change? Well, to be frank, it shouldn't. It's known, however, that it will: Chrome in particular is known to not handle the accumulated history well. I haven't had a similar experience with Firefox, but that's only one data point.

thefreeman 12y ago
plugins / extensions are not imported between chrome and firefox, which is the bulk that is being referred to.
- bryanlarsen 12y ago
  Given that the OP installed "cloud to butt plus" into Firefox, I think it's fair to say that he's doing a fair comparison.

ademarre 12y ago

Right. That paragraph about the extensions makes me question the credibility of his claims. Disappointing. I haven't experienced the performance problems he describes, and I use the Chrome dev channel for my main browser. I've noticed Firefox has been getting faster over the last couple years, but when I use it I still get the feeling that it gets in my way more than Chrome does. Edit: fix typo

smt88 12y ago
He's part of the rule that tech writers aren't technical people and often write about things they don't understand. As some famous person recently said, there's a huge amount of "intellectual dishonesty" in tech journalism.
- bmelton 12y ago
  To spin it more positively, he and his ilk more generally represent our users than more "in the know" tech-journalists. I think this is a rough article, but likely one that either many users will empathize with, so the overarching idea that browsers should be doing something to prevent bloat is probably a sound one.
  - smt88 12y ago
    I guess I see journalism the old-fashioned way: you're not an expert on everything, but you do research and distill the concepts for regular readers. This guy clearly didn't do research. Yes, Chrome uses a massive amount of memory, but that was originally a feature. Most people had way more memory than they needed at any given time, so Chrome heavily used memory to decrease processing time and therefore improve rendering speed. It also split into many processes to insulate each part of the browser from crashes and security holes. Now that the age of SoCs/tablets/etc is upon us, people don't have as much extra memory as they did when Chrome first arrived. That may have an impact on how usable Chrome is (don't know if it's smart about "dialing down" its memory usage). But this guy didn't mention any of that. Like a modern "journalist", he just wrote about his own anecdotal experiences without learning about the underlying concepts.

sbochins 12y ago

I still use pentadactyl, even though it isn't being supported AFAIK. These kind of customizations are not possible on chrome. This is why I never switched even when it was cool.

wyclif 12y ago
I love Pentadactyl too, that's why I'm amazed it's no longer supported. Seems to me that with all the vim and Firefox-using devs out there, a lot of people would want to support it. But I have no evidence of that.

FilterJoe 12y ago

The problem is using many open tabs with Chrome. If you use no more than 3 or 4 tabs, Chrome is generally fine. If you use more than that, Firefox has some built-in features to help with tabs such as delayed tab loading and tab grouping. It also has more and better extensions for helping with multiple tabs. Furthermore, it doesn't launch a new process for each tab. In a browser comparison I wrote last month, I included a section for users who like to keep a lot of tabs open. Even Opera, which uses the same code base as Chrome, does tabs better than Chrome: http://www.filterjoe.com/2015/01/23/best-browsers-2015-which-browser-is-best-for-you/#ManyTabs http://www.filterjoe.com/2015/01/23/best-browsers-2015-which...

emodendroket 12y ago
I imagine this has something to do with the sandboxing in Chrome that prevents one tab from causing the entire browser to crash, no?
- FilterJoe 12y ago
  Yes. Sandboxing makes sense when you think of browser as operating system. Isolate each process to increase security and crash-resistance. Makes sense, but it takes more memory.
  - emodendroket 12y ago
    Frankly I have more memory than I need for most tasks so I don't mind this trade-off.
    - frozenport 12y ago
      On the other hand memory memory takes more time to read, potentially leading to latency.
- vxNsr 12y ago
  I don't know, I've had one tab cause other tabs to crash all the time, which is something that confused me considering the supposed isolation that should be happening. edit: not sure why I've been dv, but I'd like to add that one thing I've noticed is that this will usually happen when I open links in a new tab, somehow all those tabs become linked and if one becomes unresponsive they all do. For those who claim I should stop opening links in new tabs, I say: welcome to 2004, this is the whole purpose of tabbed browsing. If you;d like to try to reproduce this: go to facebook and open the first ten links in your newsfeed in new tabs (either by middle clicking, ctrl+click, or context menu) I don't normally do more than one or two at a time but ten will force the issue. then go to another tab that isn't connect to those other 11 it'll work fine while those 11 will look like they're on pause. Note: I haven't tested this in other browsers.

jackmaney 12y ago

I had the exact same reaction. Why would an extension automatically open multiple tabs? That sounds pretty malware-ish.

emodendroket 12y ago
I remember one or two extensions I liked started injecting ads into pages with updates, which led me to uninstall them but would probably go unnoticed by a lot of users.
Domenic_S 12y ago
Happens all the time. "Ghostery is updated!" "Thanks for updating RES!" "More update tabs opening up!"

engendered 12y ago

I had to stop and say "wait, what?" To be fair, much of the huffing and puffing "I'm leaving Firefox...Chrome is so much better!" was based upon the same sort of rhetoric. Early on Chrome didn't even have extensions, much less bloated monstrosities like Ad Block Plus and Firebug (don't get me wrong -- they both are fantastic, but most don't realize how incredibly onerous they are). So those people kicked Firefox to the curb and talked about how grand everything was with Chrome. Now that Chrome has an ecosystem, they put themselves in the same situation again.

throwawaymsft 12y ago

It doesn't matter. The user gets a bad experience and blames Firefox. "Oh, it's those naughty extensions slowing everything down!" smirk the developers. Do you think the user cares? Firefox is the one who opened themselves up to get abused by. (Same things happened on Windows, you'll get a machine full of crapware if you install popular apps. OSX has a curated app store. Look at the difference on an average windows and OSX machine.) Some solutions instead of blaming the user: * Name & shame poorly-performing plugins in the interface. (OSX does this with "apps using significant power" in the battery meter) * Have a plugin speed rating in the store * Have performance requirements for plugins (users will report usage data, you have X seconds to startup, if you average higher you will be reviwed/yanked.) Asking users to not install extensions when extensibility is one of your selling points is not a good idea. Fix how they interact with the browser.

rnernento 12y ago

Despite that fact I've run both without add-ons and I'm very happy to be back using Firefox. Like the author there was a point when I switched to Chrome, but even without add-ons it has become bloated and slow. My breaking point was when it decided to auto-install a Google Now notification center that started up with my computer.

nailer 12y ago

You could have realised this was linkbait from the domain name - dang can we please lower the rank for gizmodo.com links?