I know there was a big fuss when Firefox moved over to the rapid release model, but it really seems to have energised the developers. We're seeing a lot of improvements in areas that never seemed to get the love before, and I suspect that knowing the users will get their hands on your changes in 6 weeks is responsible for a lot of that.
Most of the fuss around the rapid release model is emotional. There were a few legit criticisms (e.g. extension compatibility), but they were solvable. 95% of the fuss is just "OMG they'll reach Firefox 400 by the end of the year, how idiotic!!!"
We're seeing a lot of improvements in areas that never seemed to get the love before
Mozilla started increasing its paid staff about 1.5 year ago and it almost doubled. Most of those people are now well worked in. It would have been sad if there was no user visible improvement. (Of course, not all new hires went to desktop Firefox with critical projects like B2G starting etc)
Increasing the release schedule coincided pretty well with that in getting the new work out faster, too.
Noticed the 'tabs-on-demand' feature a few weeks back in Nightly. That's a great feature, that should have been in there earlier.
You can turn this behaviour off as well. I'd like to see perhaps the option to turn it off on an individual site basis. Or perhaps set a default value of off for pinned tabs (I haven't tested, the behaviour of pinned tabs.)
I have the feeling that in the beginning, the idea of preserving your tabs was precisely because you wanted them to be prerendered. Then it became a way to have some sort of "front line bookmarking"
Exactly, I think they caught on as they are a more intuitive UI than delving into your history or using bookmarks. When rendered repeatedly until you read them (perhaps weeks after,) they make for very expensive bookmarks.
Tabs/history/bookmarks I believe are the most pressing area for UI development in the browser.
A lot of this management could be pushed into the OS shell. If you could manage Windows easier - you wouldn't need tabs.
Tab behaviour is so inconsistent across browsers, apps and different OSs it hurts.
I don't know about the Opera version of tabs but in Mozilla 0.9.3 I used tabs because the new window delay was ~2s due to excessive XUL overlays and I installed the save tabs extension because the browser crashed every 2 hours or so and I wanted to keep my tabs.
I feel like browsers have failed to innovate on bookmarks. The problem with bookmarks is that once you have too many they become unweildy. There should be a separate between "stuff I visit a lot" and "this great article about Early Modern Spain that I want to read later". I'm glad that Safari is delving into the latter a bit now.
The feature actually was there since Firefox 4, but wasn't default. In Firefox 4 it was an about:config setting, sometime later a checkbox was added to Firefox preferences (General, Don't load tabs until selected). I've been using this feature for as long as it has existed, and with the way I use tabs it has been a great help to me.
I love that FF is focusing on performance. FF startup time used to be really annoying but is it that important anymore? Do people actually close their browser?
Unless I encounter memory leaks or similar problems the only time I open my browser is when I restart my computer, which is at most once a month.
I'm curious if this data exists. Is there data on how often people close / open their browser?
Anecdotally I close my browser 5-10 times a day. It opens immediately so there's no reason to keep it open if you're done with your tasks. And having fewer programs open makes it easier for me to focus on my task at hand.
I leave my browser open always, only closing it to clear out the memory leaks. So start-up isn't terribly important to me.
My FF opens pretty quickly by my standards. What seems to take forever is closing it. It seems to me that there ought to be some performance optimizations to be found in that area.
I've seen that my battery life suffers when I have too many programs running that I don't currently use, so I make it a habit to close a program when I notice I haven't used it, or know I won't be for a few minutes.
There was a push a couple years back to collect that data. Mozilla Test Pilot. They kept track of how long people kept browsers open, how many tabs they had and how they were used, opened, closed... All of this was anonymized and collected in all sorts of pretty graphs.
The site seems down, now, but that data was taken and helped shape the changes we're seeing now.
From a user experience point of view, I find that Chrome's pre-rendering of pages when you start typing them is a big deal for me now. Sometimes it makes a big difference on how fast you perceive the page to load - it loads instantly. Mozilla might want to copy that feature or do something similar.
It requests data from a website before you actually press enter to make the request? Or it fetches whatever locally cached data it may have for that website? Big difference. I assume the latter. I haven't used Chrome in a year or so.
I am heavy user of browsers and I have never noticed any difference in speed between Firefox, Chrome and for that matter IE in terms of my normal browsing. I usually have several browser windows open and sometimes over 100 tabs or more. Firefox wins hands down for its vastly superior tab handling and addon infrastructure.
I'm a die-hard Firefox user; can't live without tab mix plus but Chrome is significantly faster rendering my web application than Firefox or IE. We recommend it to all our customers for that reason -- the difference is striking.
Our application is not JavaScript heavy but it does contain a lot of HTML.
I switched from Chrome to Firefox a few weeks ago. Yes, Chrome is noticeably faster for every rendering task. In particular, it's nice that single tabs don't slow down the whole browser.
In the end though, both work fine and I don't mind the difference in speed too much.
Why the switch? I don't imagine there are too many people switching from Chrome to Firefox. I simply can't live without the features I've gotten used to, but if I didn't know any better I'd probably just stick with Chrome.
For some reason, my main Chrome installation broke (kept resetting itself every few days). Also, I became fed up with sending all my settings, bookmarks and history to Google.
So I tried Opera. It had a few quirks which ultimately threw me off too much. Hence, Firefox. It does not annoy me (much). That is all I ask from good software.
Also, Firefox syncing is quite amazing.
Divorcing myself from Google is still in progress though, with maps and calendar being the last holdouts.
I work on the cycle collector. The basic idea of a cycle collector is to trace through suspicious objects to find ones that can be freed. If you can tell that an object is alive, then you don't need to trace through it. There was already some support for this (if a DOM object is in a document being displayed, it must be alive), but we've added support for many other common cases (for instance, if a JS object is reachable from a DOM object that is in a document that is currently being displayed, it and all JS objects reachable from it must be alive).
This is mostly a process of logging what the cycle collector is looking at, then trying to figure out if there are reasons why particular objects are always alive that we could take advantage of. The most common cases have been taken care of in Firefox 12 and 13, so we're left with a variety of little cases. This results in exciting bugs with names like "unmark gray objects held alive by nsXBLDocGlobalObject of skippable nsXBLDocumentInfo" ( https://bugzil.la/735550 ).
The bug grayrest linked to has a good list of them, as well as one of its subbugs ( https://bugzil.la/716598 ).
I don't agree with the browser-on-demand solution, and while I'm glad I can turn that option off, I don't think it goes far enough. What I want is for background tabs to be loaded after the current tab, and done in such a way that doesn't disrupt my current browsing activity. Maybe load them one at a time after the current one is done rendering.
The reason I prefer using chrome than firefox is that I get freeze when viewing a video. I don't see them with chrome. But I prefer firefox's management of memorized passwords.
It's great that FF is advancing in performance terms and slowly closing the gap with Chrome.
However, the big pain is the memory leaks caused by the addons (namely Firebug) which make the memory climb to 2 gigs after 4-5 hours of development. I guess they should be focusing on these issues too, since some of them are so popular, that users prefer FF thanks to them (like the aforementioned one).
35 comments
[ 3.1 ms ] story [ 86.1 ms ] threadMozilla started increasing its paid staff about 1.5 year ago and it almost doubled. Most of those people are now well worked in. It would have been sad if there was no user visible improvement. (Of course, not all new hires went to desktop Firefox with critical projects like B2G starting etc)
Increasing the release schedule coincided pretty well with that in getting the new work out faster, too.
You can turn this behaviour off as well. I'd like to see perhaps the option to turn it off on an individual site basis. Or perhaps set a default value of off for pinned tabs (I haven't tested, the behaviour of pinned tabs.)
Tabs/history/bookmarks I believe are the most pressing area for UI development in the browser.
A lot of this management could be pushed into the OS shell. If you could manage Windows easier - you wouldn't need tabs.
Tab behaviour is so inconsistent across browsers, apps and different OSs it hurts.
Unless I encounter memory leaks or similar problems the only time I open my browser is when I restart my computer, which is at most once a month.
I'm curious if this data exists. Is there data on how often people close / open their browser?
My FF opens pretty quickly by my standards. What seems to take forever is closing it. It seems to me that there ought to be some performance optimizations to be found in that area.
https://hg.mozilla.org/mozilla-central/rev/139e8e845aa5
The site seems down, now, but that data was taken and helped shape the changes we're seeing now.
Old link: https://testpilot.mozillalabs.com/testcases/tab-open-close/a...
Speed, for me, just has never been an issue.
Our application is not JavaScript heavy but it does contain a lot of HTML.
In the end though, both work fine and I don't mind the difference in speed too much.
So I tried Opera. It had a few quirks which ultimately threw me off too much. Hence, Firefox. It does not annoy me (much). That is all I ask from good software.
Also, Firefox syncing is quite amazing.
Divorcing myself from Google is still in progress though, with maps and calendar being the last holdouts.
The reason I switched was some niggling privacy concerns and then discovering that google-chrome won't allow Ghostery to block doubleclick.
Firefox with:
addresses all the privacy concerns. Then: give me pretty much the same UX as I had in google-chrome.This is mostly a process of logging what the cycle collector is looking at, then trying to figure out if there are reasons why particular objects are always alive that we could take advantage of. The most common cases have been taken care of in Firefox 12 and 13, so we're left with a variety of little cases. This results in exciting bugs with names like "unmark gray objects held alive by nsXBLDocGlobalObject of skippable nsXBLDocumentInfo" ( https://bugzil.la/735550 ).
The bug grayrest linked to has a good list of them, as well as one of its subbugs ( https://bugzil.la/716598 ).
However, the big pain is the memory leaks caused by the addons (namely Firebug) which make the memory climb to 2 gigs after 4-5 hours of development. I guess they should be focusing on these issues too, since some of them are so popular, that users prefer FF thanks to them (like the aforementioned one).
Really? Except for OSX users that is. Still, improving startup time is a good thing.