Slashdot Mirror
Firefox 3.7 Dropped In Favor of Feature Updates
Barence sends in a report from pcpro.co.uk that says "Under its original plans, Mozilla would roll out Firefox 3.6 and 3.7 over the course of 2009, each bringing minor improvements to the browser. However, a steady stream of delays to Firefox 3.6 has rendered that goal unobtainable, forcing Mozilla to rethink its release. As a result, Firefox 3.7 has been dropped and will be replaced with feature updates for Firefox 3.6 that will be rolled out with security updates. This should free up the team to work on the next major release, Firefox 4, slated for the last quarter of 2010, which is expected to follow the same development process." Updated 20100116 00:54 GMT by timothy: Alexander Limi, from Firefox User Experience, says that the PC Pro article linked above misinterprets the situation, and that 3.7 is still on the roadmap before 4.0. The confusion stems from a schedule realignment: the out-of-process plugins feature, originally slated to land in 3.7, will instead ship as a minor update in Firefox's 3.6 series. According to Limi, CNET gets it right."
I wonder what effect this is going to have on the implementation of SVG animation, which is part of gecko 1.9.3, which was to be used in 3.7. Is it going to be slotted into 3.6 sometime or will it get pushed to 4?
upon the advice of my lawyer, i have no sig at this time
People have said Firefox is multithreaded, and I'm no coder, but I know when a piece of software is using all available resources.
Firefox never goes above 25% CPU usage when I open up a new window (which in turn loads about 15 tabs). Maybe the Gecko rendering engine can't render two pages at once. All I know is that the Firefox becomes unusable/unresponsive on my quad core for about 5 seconds while everything loads. Chrome hits much higher CPU usage-- but it doesn't have [true] adblock.
So now we have to wait until 2011 for Firefox 4 to get tab previews in the taskbar? Time to investigate ad-block addons for IE8.
Benford's Corollary to Clarke's Law: "Any technology distinguishable from magic is insufficiently advanced."
I'm using it already as my predominant web browser of choice. Works like a champ so far. I know it's not even pre-release blah blah. It works for me.
I've been running the 3.7alpha nightlies for a while now (codename: Minefield...which is now, possibly, ironic) and it's been quite good. Shame to see it dropped, but hopefully they can move the code into 3.6 quickly.
What purpose does it serve to skip version numbers, except for some political or media-relations reason? The Linux kernel and many other open source projects have release cycles of "it's done when it's done" -- and a predictable version numbering system. What next, Mozilla Firefox 2010 Professional Edition? Delays are inevitable in any software development project.
Also, Slashdot -- this news post was like saying "X replaced by Y. Z reported jealous, but A and B are looking forward to bringing C onboard soon." Numbers should not be used in place of content. $WITTY_COMMENT. $RETORT. $TROLL. $VAGUE_REFERENCE_TO_SEXUALITY.
#fuckbeta #iamslashdot #dicemustdie
This should free up the team to work on the next major release, Firefox 4, slated for the last quarter of 2010, which is expected to follow the same development process.
Firefox will be dead before it hits version 5.0.
will be replaced with feature updates for Firefox 3.6 that will be rolled out with security updates
This seems to be a horrible idea to me, unless I'm misinterpreting it. I can see this being implemented in two ways:
One, Mozilla withholds security updates until there is a feature ready to go, which is just stupid - don't leave a hole if you've got a fix ready. One of the arguments in favor Firefox over IE is the more rapid security updates.
Two, Mozilla withholds features until a security update is necessary. I can't see any advantage to doing this, but there's a few obvious downsides (like withholding a perfectly good feature until someone finds something we're supposed to be hoping is not there).
Unless I'm missing something?
"A witty saying proves nothing." - Voltaire
Security updates should never be combined with feature updates. Anyone who doesn't want the feature update is then in the unfortunate position to decide whether they'll get the unwanted features or keep the unwanted vulnerabilities. Bad Mozilla.
Unless I'm missing something?
Well, two things you are missing is evidence that the Mozilla foundation will (1) "withhold security updates until there is a feature ready to go", and (2) "withhold features until a security update is necessary".
Perhaps they intend to roll out new features to 3.6 in the same manner as they do security updates; one 3.6.x release might be a bug fix, another might be new features and another a combination of the two. You don't have to bring out new features on major releases, so this might even mean that we'll get features added to 3.6 sooner than we would have done waiting until 3.7 before releasing them all in one go.
UNIX? They're not even circumcised! Savages!
I don't think the Mozilla Foundation is dumb enough to wait for new features for 3.6.x version security updates! I do think the version number could go as high as 3.6.15 (my guess) as security updates and the new features are "slipstreamed" in.
Yes, you're missing option 3.
Three, Mozilla rolls out a patch that includes a feature when it's ready, and rolls out a different patch when a security update is ready, and combines them if/when possible. That would still be "with" security updates, after a fashion, and it would be the logical, intelligent way to do so.
Canada: The US's more awesome sibling.
Since 3.5 released, I've had a grand total of maybe two crashes (at least one of which was clearly caused by Flash). It does use a decent amount of memory (100-400 MB depending on how I'm using it), but nowhere near what IE8 is using (often 50 MB or more per tab), and on my machine, I've got more than enough memory to handle it. Maybe you really are using unstable plugins and add-ons?
$_ = "wftedskaebjgdpjgidbsmnjgcdwatb"; tr/a-z/oh, turtleneck Phrase Jar!/; print
Maybe I'm lucky (conversely, maybe you are unlucky), but 32-bit Firefox 3.5x is 100%* rock-solid stable on my PCs. I can't compare this to IE's stability, as I never, ever, use IE. Granted, I only have 4 add-ons installed (ColorfulTabs, Flashblock, ForecastFox, and Oldbar), but Firefox simply works.
*Actually, I can remember 1 time that Firefox locked up on me, months ago, so its stability is 100% minus one_event.
Continuing to support old versions is a heavy burden, and has to end at some point. It's not a question of if people will have to make that decision, but when.
"Mozilla would roll out Firefox 3.6 and 3.7 over the course of 2009, each bringing minor improvements to the browser. However, a steady stream of delays to Firefox 3.6 has rendered that goal unobtainable."
[jay@gobstopper ~]% date
Fri 15 Jan 2010 12:32:18 EST
I think you're missing two things:
1) The article's first paragraph is taking a proposal for a possible future plan of action
and claiming that it is the plan of action.
2) Right now (Firefox 3.0 and Firefox 3.5) there are no features shipped as minor updates;
all features are "withheld" as you put it until the next major version.
The only firm current plan here is that one particular feature, namely out-of-process plug-ins, is currently planned to be backported to Firefox 3.6 and shipped in some form in one of the minor updates. Once it's judged ready and so forth. Since minor updates are all about security and stability, this particular feature fits well in their scope (for example, a significant fraction of Firefox crashes are actually Flash crashes).
There is also talk of possibly backporting some other small features (mostly performance-related) to the stable branch as they become ready. This may or may not happen. There is also discussion about what and when the next Firefox major update will be, and discussion about what and when the next Gecko release will be. These may not happen at the same time. None of that is decided.
Why don't you turn off automatic updates then?
My other account has a 3-digit UID.
I have seen you post this ANY time Firefox has been mentioned for the past couple of weeks, cut and paste style. You are either a shill of some sort, or forced to do this because of one of your clients. Either way, you aren't wanted here.
Living With a Nerd
Small feature updates are not conducive to getting corporate support. With large updates, a company can say, "We support Firefox 3.5+", and they can be reasonably confident that they don't need to fully test every minor release of Firefox 3.5. With small updates they have to say, "We support Firefox 3.6.7", and can't be sure that they will actually be able to support 3.6.8 without fully testing it. If you want corporate support, you have to have feature freezes, or support stops being worth the testing time.
I keep HEARING about all these serious problems, but the five computes in my household using Firefox 3.5x (two of them Ubuntu 9.10, three of them Windows XP SP3) haven't SHOWN me any of these problems.
These posts keep talking about how there are major problems with Firefox, and they keep getting worse...yet I haven't experienced nor do I know anyone in my relatively large nerd circle who has experienced what is being described after the release of 3.5.
It sounds like paid shill bullshit to me.
Living With a Nerd
No operating system should crash due to a misbehaving application. If it does crash, the operating system sucks.
It's true that my Fx has crashed seven times in the last three months. However, I can trace two of them to a faulty extension. The rest may very well come from the Flash plugin, which isn't entirely stable on Snow Leopard and hasn't been fixed in ages. Offhand I can't remember a single crash not directly related to Flash (excepting the extension, of course).
I'm willing to bet that a fair part of the stability issues people have actually comes from badly-written extensions and plugins. Remember that most other applications don't execute code written by Adobe (and yes, I see that as an argument as to why they're more stable).
USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
You're missing this:
(3) Mozilla does individual security fixes and feature updates for 3.6 as they are completed (maybe grouping the two together in an update if they happen to be ready at the same time, but not holding either to wait for the other), but doesn't have one big list of featur updates that must be complete for a "v3.7" that are released all at once. The "feature updates that will be rolled out with security updates", in this case, would mean that the feature updates are rolled into the usual chain of flowing, as-completed security update point releases rather than bundled together into a minor version release, not that each individual feature update must accompany at least one security fix.
Huh. I typed in about:crashes, and it was completely empty.
Anecdote vs. anecdote. To continue this argument you need real data.
Have you seen $200 million worth of development in Firefox?
http://planet.mozilla.org/
Spend a little time reading this on a regular basis, and you'll soon discover how many projects Mozilla handles, and all the developers they're paying.
The big projects include:
Firefox, Bugzilla, Camino, Fennec, Lightning, Sunbird, Seamonkey, and Thunderbird.
These are major multi-platform projects.
Mozilla has several projects for first-party add-ons for all of the above such as Firebug, Chromebug, . Then they have tons of major projects that most people never hear about. At the moment they're working on:
Jetpack
Raindrop
Bespin
Concept
Personas
Prism
Snowl
Test Pilot
Ubiquity
Weave
Electrolysis
A tool recently said the KDE code based purely on lines of code should have cost $175 million to develop, and that wasn't counting Koffice, and anything outside the main KDE trunk.
Mozilla also doesn't just do code projects, they do tons of community management and outreach projects like Mozilla Education, which costs even more money.
They also help support outside developers using Mozilla and Xulrunner for other apps such as Kompozer, Songbird, etc.
I don't know where all their money goes, but Mozilla does *A LOT*. To suggest they're not doing much development is ignorance or lies.
Firefox experiences a LOT of crashes and memory hogging, and has for years.
Firefox does crash for me from time to time, on Windows and Linux. I tend to use a lot of extensions, and the most common thing I hear is that extensions are the largest source of memory and stability issues. Do I get daily crashes, or 10 crashes a day? No. And I run daily snapshot builds. I maybe get 1 crash a week, if that.
As a Systems Engineer, I troubleshoot and support some big money apps that crash fairly often. Large software projects are going to have bugs. However, I wager if you run without extensions, you'll find that Firefox is pretty damned stable for such a massive multi-platform app.
Memory issues are all but lies these days. Memory usage has improved so much over the past few years. Firefox is actually better with memory usage than Chrome in many ways. The core app doesn't take too much memory on first load. It doesn't have memory leaks.
There are some intentional features which cause Firefox to eat up some memory that you can turn off, such as Firefox keeping fully rendered pages in memory, so that when you hit the back button, they just display immediately without having to re-render. When you close a tab, it still keeps that full session in memory for some time, so that you can reopen the closed tab with full rendered pages and history if you want.
If you don't like these features, turn them off. Not to mention, these are set to use dynamic chunks of memory which is preportional to your total memory. If you have a desktop with 8 gigs of memory that you're not using, why get upset that Firefox is using 300-400 megs of memory?
Unused memory isn't doing you any good.
Stop with the FUD. Real geeks know better and see right through BS and lies.
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
I don't see any of these problems. Firefox has never crashed for me under reasonable circumstances.
Sibling posts are right - this seems way too long and too calm. If someone writes a lot, it's usually a rant. Also, the bolding, etc.? It's way too artificial.
Yet Another Tech Blog
(but so much more, including game and movie reviews)
http://yanteb.peasantoid.org
"about:crashes" shows no submitted crash reports on my computers.
I run two to three Firefox windows with dozens of tabs 24/7, with active browsing of a variety of content types (Flash, images, embedded video, text, heavy scripting, AJAX, et cetera) for many hours daily, and a wide variety of addons installed. This particular install of firefox has been running for a little over a year.
My about:crashes is blank.
The randomness of failures suggests that Firefox writes to a random location memory that is important in some systems and not others. That's crucial in an unstable, poorly designed OS like Windows XP. Linux merely throws Firefox off the system.
This is ridiculous. You're obviously talking about things you don't know anything about -- concerning programs or OS's to start with, let alone Firefox specifically or the function thereof.
Dude. There have been threading calls on "other OSs" long before Microsoft butchered the design. You're misinformed or shilling.
Also, ffs, take an english class. The way you write makes me think you might actually be an MS coder trying to turf a little.
And the issue is what? Seriously. So no major overhauls until FF4, continuous minor updates both feature & security wise. This is what Microsoft, Linux, Amarok, Opera nd others do. Why would Mozilla be different?
Mountain out of a molehill. Start worrying if Mozilla stops talking about FF4.
Note: I know Amarok isn't in the same league as the other three up there, but I was going over their changelog yesterday and there were some pretty big updates done on a minor point change. Finally looking like it's back to all the functions of 1.4.10. Now if it would now get stuck scanning my collection at 49% I'd be gold
I call it 'The Aristocrats'
Not to feed a troll, but...
If he and a number of other people, as stated in his report (and I'll throw my experience in there as well), don't have the problems, it calls into question the original report of overwhelming issues.
Personally, I'd have to say that I use firefox for an average of 12 hours a day. I use it quite a bit at work and again when I get home. If you add in the time that my friends and relatives use firefox without on going crashing issues (ESPECIALLY those that take down windows, I've NEVER even heard of Firefox itself bringing down properly patched XP, and I know I've not had it take down either my Ubuntu, Gentoo or Fedora systems). I'd have to say that daily useage of Firefox in my circle has to be aproaching 80 or 90 hours per day.
I'm also saying that the "Automatically Generated Crash Reports" "Didn't happen" because, well, they didn't. mostly because there was no need for firefox to automatically generate a report on an event that didn't take place.
Is Firefox perfect? No, far from it. But I have found as many other people, that crashes, when they do occur are almost never caused by firefox itself, but one of the extensions. In the times I've heard of someone with a crash or two, they uninstall the last extension they put on there and they're back to stable. It's that easy. Same goes for slow load times, large memory usage, high CPU usage, etc.
Also, that originating post says that 91% of the Mozilla Foundation's income is $68M, and complains that we haven't seen $200M in development... Well, did you ever think that calculation is a bit off? after all, that'd only mean we'd see $75M which, last I checked is less than $200 by quite a bit.
You really should think about these things more before you post. If you bought that info, you might want a refund...
-=JML=-
I just read mine - I have one in Nov 2009 & one in Dec 2009. I seem to recall that both of those were caused by some script on cnet.com; it was certainly one particular site in both cases. I start each morning with a fresh 12 tabs open and go through the day opening & closing tons of tabs. Maybe this is "lighter" browser use, but I also have a machine at home which keeps 50+ tabs open for weeks at a time & almost never crashes. This leads me to agree with the GP, claiming that Firefox has major problems with crashing sounds like shill bs.
The randomness of failures suggests that Firefox writes to a random location memory that is important in some systems and not others. That's crucial in an unstable, poorly designed OS like Windows XP. Linux merely throws Firefox off the system.
What that suggests to me is that your memory is bad. Try running memtest and see if it reports any errors. Even if it doesn't, it might be heat related.
I've had issues with Firefox crashing in the past (although mostly due to my playing around with XPCOM while writing an extension), but I've never seen it crash the OS. If it's crashing the OS, it seems highly likely to me that there's something physically wrong with your system.
After all, even Linux crashes when the CPU physically falls out of its socket. (Don't try that at home.)
The last time I was routinely crashing Windows XP was due to an overheating issue with my graphics card. It'd run fine until I tried to play a game. Start up a game and then after a while, boom: PAGE_FAULT_IN_NONPAGED_AREA.
You are in a maze of twisty little relative jumps, all alike.
So you think the firefox team should just quit and that the people currently using firefox should switch to another browser? Just like that?
Clearly, your opinion of what counts as a "fatal flaw" is not widely held or firefox's market share wouldn't continue to grow as it has. People will continue to use firefox until they find another browser that is more appealing to them and as long as there's enough users to justify further development, the firefox team will continue to work on the code-base.
Out of curiosity, what are the "fatal flaws" as you see them?
*sigh* back to work...
Interesting post ..
I run FF in Linux and can't recall the last time I've seen FF crash
As you state FF crashes in XP constantly, the problem may be partly with XP
As from my experience as a windoz user, everything crashes daily, from the main OS to the applications, although XP is much improved over past windoz versions. It still melts down from time to time
The FF delayed mouse event problem, I have seen though. I was wondering what was causing that .. now I know
"Suppose you were an idiot...and suppose you were a member of Congress...but I repeat myself." Mark Twain
You don't have the problem, so it doesn't exist?
I'm not saying it doesn't exist, I'm saying that in own personal experience (and, apparently, a lot of other people on this board's experience), I haven't found anything remotely close to what is being claimed.
Did you read the crash reports? They are automatically generated. Are you saying they didn't happen?
People also have pictures of iPhones literally catching on fire or even exploading. That doesn't mean the millions of iPhones out there are bombs in disguise.
Did you consider that maybe your use of a browser is lighter than that of others?
I use 11 addons, and tend to have anywhere from 10-30 tabs open at a time in TWO different browser windows at a time (closer to 10 if general browsing, closer to 30 if I'm working on my website).
If anything, my use of Firefox is HEAVIER than most (or on par with "power users", many of whom on this board also report experiencing none of the issues you have mentioned).
Considering I'm running a relatively week Athlon X2 5400+ and 4 gb of DDR2800 Corsair XMS memory, I should be experiencing the problems you describe left and right...but lo and behold, I don't.
Living With a Nerd
Strange, 2.x and below were massive memory hogs and somewhat unstable in my experience, when 3.x came along and they started worrying about memory usage and plugging the memory leaks of old by firefox experience did improve, especially on older hardware. Chrome is the biggest memory hog of current browsers, this is very much by design.
Shiretoko? isn't that the codename for an old unstable/testing version of firefox? I would certainly expect that to be unstable and crashprone.
Not that i don't think firefox is starting to lag behind, the XUL GUI is just feeling impossibly slow at this point, but less FUD and more substance next time, please, ok?
Not only is it week, it's also weak!
Oh, and for the record: the two ubuntu 9.10 systems I run Firefox 3.5.x on? One is an Athlon 64 3000+ single core system (HTPC), and the other is a Dell Mini 9. If I was going to have stability issues with Firefox, I'm sure at least one of those two would have it.
Living With a Nerd
You'll have to admit a detailed accounting of the apparently more than $200,000,000 Google has given to the Mozilla Foundation would be interesting.
Maybe, these aren't public funds or donations. I don't really care how they use their funding, I only care about their product. Firefox does cause me some issues but the positives continue to outweigh the negatives.
However, I think it's probable that Chrome will overtake them within the next couple of years.
Eagles may soar, but weasels don't get sucked into jet engines.
I can't find official documentation on this subject. However, based on the updates that I get, there are 4 numbers in a given Firefox Version:
A.B.C.D
A= Major revision
B= Minor revision
C= Small feature revision
D= Bug / security fix
It now appears that features that was going to be in 3.7 will now be put into 3.6 feature by feature. So you may see an update like 3.6.0.2 which is just security/bug fixes from 3.6.0.1. When you see an update like 3.6.1.0, it means it has a new feature that would have been in 3.7 but was put into 3.6 instead.
"WHY ARE THESE MINES EVEN HERE???"
http://www.collegehumor.com/video:1770138
I don't know what problems the other poster was talking about, but for a several month period up until the last two updates, Firefox was crashing on me basically every night, unattended, with near 100% CPU usage and hugely bloating memory. Furthermore, if I tried to restart it, and it found that one of my plugins needed updating, it wouldn't attempt to restore the previous set of tabs that it was displaying when it crashed. Hugely annoying. I suspect it was some weird interaction with the Zimbra web app. Still, FireFox shouldn't crash like that, no matter what an app is doing.
I have trouble believing this. Firefox was causing you that many problems on a daily basis...yet you continued to use it for several months? ::sniff sniff:: I smell bullshit.
Living With a Nerd
You know, sometimes the architecture that you originally designed (and that was great and the right thing back then) does not fit your current needs anymore. You get slower and slower, everything becomes bloated and messy, and starts to look like an upside-down pyramid (Windows ME syndrome).
And that’s the time, where it’s good to think about not just making the next version. But about making the next generation. Like a complete rewrite, but not. More like forgetting everything and designing a good and more future-proof system from the ground up. Which usually results in not much loss of work, because you notice how much falls into that new design as if it were made for it, because you lose the coding around that you previously had do employ. (Which also is the indicator that a new generation was the better decision: When it is less work than what the other choice.)
Has anyone else the feeling, that we’re pretty close to that with Firefox right now?
It’s strange how many experienced developers think they can just pile up version after version of major new goals onto the same architecture.
I myself would at this point make two branches: One called Firefox. And one called Firefox Two / Firefox II / SomethingCompletelyDifferent. (As in “SomeMovie 2”, not as in “SomeSoftware 2.0.” One level higher.)
I hope the team makes the best decision.
Any sufficiently advanced intelligence is indistinguishable from stupidity.
Mod parent up! Super informative post...I knew they were stretched across many projects, I didn't realize it was THAT many -_-;;
Living With a Nerd
Uuuuhhhhh...dude? Got five boxes here, all running Firefox. Did the little about:crashes thing? There ain't any. Zip zero nada squat. Maybe they are running a funky extension, which you can't blame the company if a user decides to PEBKAC the thing. And it isn't like I'm running without extensions here-Distrust, Downloadhelper, downloadstatusbar, FEBE(must have), iMacros, ABP, Noscript,nightly tester tools, and Forecastfox.
That is NINE extensions, running on THREE different OSes (Win2K, WinXP Home/Pro, Windows 7 X64) with a range of....how many years since they released the 3.5 branch? Because I installed it clean on the Win2K. The other are a mix from a couple of years to 4 months for the new Windows 7 box. And ALL are stable, ALL are doing fine with regards to memory, and ALL are crash free.
Not saying you don't have a problem, I'm saying I'm willing to bet one of those more funky extensions are the culprit. The programming extensions like Firebug from what I understand are more than a little unstable. But I have been running those same above extensions almost since the day each individual one was released and been running Firefox since the 1.x days, and can say I could probably count the number of crashes I had on one hand, all during the 2.x branch, which was seriously unstable with regards to memory.
So while I don't know which extensions you are using, if it is any that isn't on my list above you might want to disable them for a couple of days and see what happens. Or don't and just put up with the occasional crash. I personally like my extensions enough that even if they did become unstable in the future I would probably just return to a previous version or put up with it. I like having my browser MY way too much to go back.
ACs don't waste your time replying, your posts are never seen by me.
Can we have a show of hands?
Does anyone here believe this AC?
You are welcome on my lawn.
Helpful Hint of the Day: There are other browsers. Use one.
Seriously, if it's not working for YOU, use something else. It works for everyone else here, so it must be something with you.
Also, it's open source. Please submit your patches directly to Mozilla or ask them for a refund in the amount of your purchase price. Either way this is not the place for it.
I believe they are either seriously misguided or a horrible liar.
ZING!
Living With a Nerd
So, is it okay that Mozilla Foundation disclaims responsibility for extensions being allowed to make Firefox unstable?
Um, yes.
I don't take responsibility for any code you write, and I don't expect you to take responsibility for any code I write.
Why in the name of sweet holy fuck would you expect the Mozilla Foundation to take responsibility for any code that some random developer completely unassociated with them writes?
The far and away priority one feature should be Multithreading. Each tab and each plugin should have its own process and its own memory space, so that a crash of one tab/plugin, or one tab/plugin using loads of CPU power, should have practically no effect on my other tabs/plugins on my 4-core CPU.
So I don't care about copying Chrome's GUI. But copying Chrome's sandboxing and multithreading architecture I very much care about!
There is a Mozilla project to implement this, but the project page hasn't been updated in months, as far as I can tell.
SO damn slow
If Google really cared they would fix Android Chrome to reflow text, instead of discriminating
I had to Google how... about:crashes
One. I have one. 20/08/2008. About 7 months after I bought this laptop. Two years, one crash. I didn't believe it at first.
Add-ons: ChromEdit Plus, Download Statusbar, eBay Sidebar for Firefox, Flashblock, Fuzzy Time, Google Gears, Greasemonkey (which gets a lot of abuse), ImgLikeOpera, ReloadEvery
Happened when saving a file. No it wasn't jpeg porn.
// probably the default file name is too long or contains illegal characters!
This post contains benzene, nitrosamines, formaldehyde and hydrogen cyanide.
You have to end-of-life old versions at some point. Yes, it's good to support old versions if there's a valid reason to be doing so (Apache 1.3, maybe. Firefox 2.0, no.), but there is a limit.
It's all rather moot with free software anyway. If you really think something should still be maintained, then just do it. (Or pay someone else to.)
Mine too:
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.7) Gecko/20100106 Ubuntu/9.10 (karmic) Firefox/3.5.7,
no crashes reported. I normally have 50+ tabs open (several hierarchical sets using tree-style tab, most of which are reference info I like to have available.)
Addons installed:
Adblock Plus, All in one Sidebar, Better Privacy, DownThemAll, FireGestures, FoxyProxy, Gmail Manager, Image Zoom, Leet Key, Morning Coffee, NoScript, Moonlight, Nuke Anything Enhanced, PDF Download, RSS ticker, Session Manager, Targeted Advertising Cookie Opt-Out, Tree Style Tab, Update Notifier, XUL Profiler, Youtube Comment Snob.
Flash plugin occasionally crashes, when that happens I kill the npviewer.bin process & reload the page. I've never seen it actually bring firefox down, at least not with the latest versions of each.
Not a sentence!
Because Safari 4 is out, and Chrome 4 is in beta, Having Firefox move to 4.0 sooner rather then later makes a great deal of sense based on that alone. It also seems that the features planed for what used to be 3.7 and the stuff shortly included in 3.6 when its released seems to be the second half of the 3.5 version jump justification.
Anyone know what features are being planned for Firefox 4?
From the rumors I've heard, it's likely I won't want to upgrade to Firefox 4.
I've heard things about replacing the menu and toolbar with a ribbon. I despise the ribbon concept, finding it eats up my screen real estate, and for me is much harder to find things in.
And there were a couple of other things I heard which turned me off, but I don't remember what they were right now.
I don't. Here's a list of my firefox crashes from one week:
7/27/2009 9:21 PM 9:31 PM 9:34 PM 9:34 PM 9:36 PM 9:44 PM 9:53 PM 9:54 PM 10:12 PM
7/28/2009 1:16 AM 4:05 AM 4:36 AM 12:29 PM 1:41 PM 1:55 PM 5:44 PM 6:55 PM
7/29/2009 11:17 AM 12:28 PM 1:39 PM 6:19 PM 8:24 PM 8:25 PM
7/30/2009 12:24 AM 12:58 PM 1:14 PM 5:22 PM 6:49 PM 7:01 PM 7:30 PM
7/31/2009 11:24 AM 5:35 PM 8:29 PM 8:32 PM 8:44 PM 8:55 PM 9:02 PM
8/1/2009 2:50 AM 11:36 AM 1:31 PM 9:48 PM 9:58 PM
This continued up until 10/9/2009, when the crashes just stopped happening. Since then, I average 1-2 crashes a month.
System specs/# of tabs open/addons?
Numbers mean nothing without context....although the close proximity of the timestamps (especially on 7/27) would indicate something really screwy was going on that involved more than just the core Firefox software.
Living With a Nerd
Unless I'm missing something?
Well, two things you are missing is evidence that the Mozilla foundation will (1) "withhold security updates until there is a feature ready to go", and (2) "withhold features until a security update is necessary".
How else are you going to roll out features alongside security updates? Just hope really really hard that every feature update is completed at the exact same moment as a security update?
Either feature updates will have to wait for security updates, or security updates will have to wait for feature updates, or they will not be coming out "alongside" each other.
You don't need "evidence" for that.
Nothing special about my system specs, really.
Windows 7 64-bit
Asus P5B Deluxe Motherboard
E6600 CPU
4GB Corsair
Back then was a 4x300GB Raid-0 array, switched to a 2x1000GB Raid-0 / Raid-1 Matrix array sometime in August, currently toss in a couple Intel SSD's as of a month ago
Nvidia 295 video card
External 1TB e-Sata drive
Number of tabs varies between 1 and 15 typically, sometimes as many as 20-25 though across 1-3 windows. .NET Framework, Move Media Player, Skype.
Addons I use area Adblock Plus, FiddlerHook, Firebug, HtmlValidator, LogMeIn,
Sorry, it's a E6700, not E6600. And I switched the corsair ram out for OCZ a while ago, don't remember when.
Good luck using the Hosts file in a computer where you don't have admin privileges.
Also, the Hosts file can't be auto-updated, like adblock.
And Hosts file block all the domain - good luck blocking that big annoying image or flash animation that's served from the same domain as the content.
And as far as I know Adblock removes the ad *before* Firefox tries to load it, so I doubt it even sends a DNS request. But with Firefox you can use both Adblock and Hosts, and get the advantages of the two. Chrome can't.
Dilbert RSS feed
I don't know for sure, but chances are an older version of one of the development addons you were using didn't want to play nice with the others...that'd be my best guess. Again, the frequency of the timestamps (especially the first one you showed) would lead me to believe something crazy was going down. Even if it were memory related, it would take longer than a minute or two for Firefox to bloat up to the point of crashing (and if you were trying to reopen 30+ tabs at once...well, then what'd you expect)
Living With a Nerd
LMAO. You owe me for the coffee sprayed over my screen.
Pray tell, dear linguist, is there a style of writing, of syntax, grammar, and of vocabulary that defines a programmer as being "MS"?
Thousands of ears are curious. I await your reply.
I doubt it will have merit, though...
If the browser had a decent plug-in architecture, a crash of a plug-in would crash ONLY THE PLUG-IN. The rest of the browser would keep running.
It follows that FireFox does not have a decent plug-in architecture.
Furthermore, a good application design would allow a single page/tab to crash without bringing down all the other pages/tabs. Again, FireFox's architecture is lacking in this regard.
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
It is a serious flaw in FireFox that a crashign plug-in brings down the entire browser and all tabs. Yes, applications and plug-ins are going to have bugs. Software architects should take this into account when designing things. FireFox's architects seem not to have done so.
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
They are taking that into account for the future now. There are two projects to address that.
https://wiki.mozilla.org/Content_Processes
Jetpack
https://wiki.mozilla.org/Labs/Jetpack
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
And they're currently working on exactly that. Do note that prior to Chrome, nobody had that particular idea so everyone else needs to catch up. You can argue that Mozilla is taking a long time to do so (and they certainly are) but you can't argue that their plugin architecture is unacceptably substandard when in fact the "standard" has changed too recently for everyone to be implementing it already.
Mozilla is slow but they're not idiots.
USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
Quick, let me tell you home many of those people actually give a shit about: 3, possibly 4
Firefox, Thunderbird, Fennec, and if you count devs, Bugzilla, which is a pretty stagnate project at this point. Fennec is debatable as well at this point.
Perhaps its not so impressive to see this massive list but more a sign that they don't have any direction and are infact pulling the exact same failure mode that Netscape did. Good job guys, maybe if you're lucky after this time around you can go for a threepeat on doing a bunch of shit no one wants while stroking your own egos and utterly failing to provide anything actually useful.
I'm just jealous, I want to get paid to sit around writing shit that suits me rather than producing something people actually want.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
And better memory management. It's annoying that it doesn't seem to do garbage collection when I close tabs and pages in Snow Leopard. I end up closing it and having it reload my pages and tabs at least once a day to shrink its memory footprint.
When you sympathize with stupidity, you start thinking like an idiot.
Awe look, someone who thinks they know threading talks about how to do it ...
By default, you have no reason to make those API calls unless you want to limit to a certain CPU or set of CPUs, out of the box, by default, all CPUs are fair game for any thread. Unless you have a reason to stay on a particular core or set of cores, than you really shouldn't be prevent the OS from doing its job. You can only make a multithreaded app use less processing power with those API calls, but in a few cases where you know you're going to stay in L1 cache, this is good, as a general rule, a web browser running JavaScript isn't going to fall into this category until our cache sizes get to some massive size no where near where they are now.
FireFox's issue is a common Windows threading issue (holds true with OSX as well I believe), the first thread that does any GUI work, handles ALL GUI work.
A web browsers job is almost entirely GUI work, so having a bunch of threads has limited usefulness when you have to shift EVERYTHING back into the single GUI thread. Welp, there went all your threading improvements, you know why? The GUI is the hardest part of the work load in this sort of app. Heavy DOM parsing and JavaScript hurts, sure, but displaying it is still harder, and lets face it, pretty much everything you do in a browser comes back to displaying. That is, after all, its purpose.
The solution? There are three:
Optimize your drawing so that its done outside the GUI thread, without actually using the native GUI components and then blt'd into the GUI as needed, but this is hardly ever faster.
Use multiple processes, now you have multiple GUI threads to handle the different displays. (Hello Chrome and IE8!)
Optimize your drawing so that its done efficiently ON the GUI thread by not drawing crap that you don't have to. This won't happen because a lot of this requires that you wait for the DOM to load and get to a solid state before displaying it, and people would much rather the page load slower but they get to see parts of it as it streams in. Even if they did jump and go this route, its freaking HARD to do that optimization on a code base the size of Gecko this late in the game. You might want to consider starting over (Don't say that too loud, the MozDevs are all about throwing out and starting over without a good reason)
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
It must have been something you assimilated. . . .
I'm not going to dignify your FUD trolling with much of a response, but to answer your question, yes, win 7 was out of beta at that time, perhaps your kid, the Windows 7 beta tester, knows.
Firefox was the only application crashing, of all the ones I use daily, which includes four browsers (IE, Chrome, Opera, Firefox), Dreamweaver, Visual Studio, Apache, and about 30 others. When it is only one application, and it crashess repeatedly everyday, it's pretty clear what the problem application is.
And seriously, as a web developer, I need those extensions, and I use the daily. I had all of them disabled for a while, and firefox still kept crashing. The .NET Framework add in isn't a known problem child, but zealots like you, hate it. Just because I didn't go into details on all the things I tried to resolve it, doesn't mean I didn't do any. See some of us are technically inclined, and some of us do this for a living.
No, there was no windows 7 patch, nor an nivida patch at the time the problem went away, but there was a firefox patch. Again, pretty clear what fixed it.
soo right... I use 3.6 NIGHTY. Every night it updates. 36 Extensions. My profile is 131MB big, 5 years old. Constantly ~20 tabs open. And i get crashes something like one time a month. All tabs recovered every time. And it uses half the memory of chrome.
If it's number 2, and I suspect it will be, they will do what Drupal does. Drupal releases NO bug fixes until a .X release. They ONLY release .X releases if they have a security issue to fix. There for a while there were months a months without security holes, which is nice, but then again there were simple bugs that just sat languishing. The response was always: Install the dev if you really need it, but don't ask us for help if it breaks something because you really shouldn't run anything that is in dev.
bear in mind that Drupal 7 was supposed to launch last year. They're just now getting around to thinking about a alpha release.
I only mention this because once 7 drops, 5.x is unsupported, and you'll be told to upgrade to 6 or 7. Which you'll do, and then a bug will be found, and perhaps you'll fix it and submit a patch, and then 9 months later, it'll get committed. I hope your boss understands this. Your new project can wait 9 months, right?
For all you Snow Leopard users...
In case you did not know, you can download optimized Mac versions of a number of browsers from here
Specifically, one of the browsers available is a 64-bit optimized version of FF 3.7 for Snow Leopard.
I finally installed it the other night, after eyeing it warily for the last month or so (as I worked through the latest 3.6 optimized builds). I finally installed it last night, and have to say that it's the biggest improvement to FF that I've came across.
It loads faster, uses less CPU & memory than previous builds, and it's mega fast. My impressions are that it's now as fast as Safari is on a Mac.
It's now my main browser. If you run Snow Leopard, you should check it out.
Most distros disable the crash reporter, such as Ubuntu.
These people look deep within my soul and assign me a number based upon the order I joined. -Homer Simpson
about:crashes only shows submitted crash reports... so if you didn't submit it won't be there
I KUT J00 M4NG!!!
I have it running on Ubuntu 9.10, OSX 10.6, and WinXP, and no, I have no problems whatsoever. Neither does anyone I know. Neither does anyone I've ever heard of, except for around here.
Actually, you're wrong. My DNS request wouldn't be logged, but my request to Slashdot's IP would, so it doesn't matter.
Example: Slashdot resolves to 216.34.181.45, so I would make a HTTP request to 216.34.181.45:80. The "spy" could know using two ways:
1) If the traffic is not encrypted, he could just read the "Host" HTTP Header, which is mandatory in HTTP1.1.
2) Even if the traffic is encrypted with HTTPS, he can just use Reverse DNS. Go to the first reverse dns site, and enter 216.34.181.45. Result:
As you see, the HOSTS file doesn't guarantee privacy in any way.
Dilbert RSS feed
Well, duh!
If I say a "computer where you don't have admin privileges" it doesn't mean a PC where you don't run as Root or Admin all the time, obvisouly. It means a computer where you *don't* have the admin password in any way.
So work places, public computers, cyber cafes, schools, colleges, etc, all places where you can run Firefox + Adblock and you can't modify the HOSTS file.
Oh, I'm sure, I didn't realize this was a competition. Next time I'll bring my jogging shoes.
Dilbert RSS feed
If you don't bother to read, stop replying.
Example: Slashdot is a site with great content. But it also has a HUGE BANNER, that is served from the same domain.
How can you block only the banner with the HOSTS, and still access the content? (hint: you can't).
So to overcome the resources Adblock wastes... you build a full clients server stack that's always running. And with a server that needs access to both the Internet *and* system files, so you're exposed to external attacks. /facepalm
Dilbert RSS feed
I tried the URL's listed, and so far, after about going through 2/3rds of them, nearly all work as expected. No animation. The fonts won't look identical as in the png version unless you have your system resolution set correctly (most people don't), and have the fonts that are used in the tests, loaded onto your system. I didn't have all the fonts, and I'm pretty sure I have more than most people.
Do your your fonts display identically in all your other applications? I.e. when you
display a page in multiple browsers, -- how about when you print to PDF and display in Acrobat? If your fonts are consistent aross all those (might as well throw in your favorite WYSISWYG editor as well). Oh -- and finally -- if you print it out and hold copy up against your screen, are the fonts identical on paper as on screen?
If all that is true, then you probably have your fonts configured right on your system reasonably well. Even then, you have to have the right fonts loaded onto your system -- specifically the ones used in the tests.
So give FF a break on its SVG tests regarding the fonts -- because getting those "perfect" will take the most configuration on your part. Some of the tests even say that you should ignore differences in font sizes and styles because they are likely to be misconfiguration on your machine for use as a test verification device.
Please make sure your DPI are set accurate and try to look through the sources to find what fonts you will need to have things look correctly. These are tests designed to operate under specific circumstances.
Some things are obviously broken -- like text displays one place, but not another,
and the animation stuff not displaying at all. Not good. But other stuff...my only
main gripe with SVG so far is execution speed on more complex images. But getting it working is probably the best first priority.
The SVG font tests I saw looked fine on my monitor. But I do have alot of fonts loaded on my machine (~2059 families).
Do you have a color profile installed in your Firefox?
If so -- uninstall it.
png's are not color-managed. So no matter what profile you have installed, if the svg profiles read as color managed, they will look different from the png's.
Second, do you have a generic profile installed in FF or have you profiled and calibrated your monitor and have a specific monitor profile for your monitor installed in FF (which is what you probably should have if you have any). sRGB and sRGBv4 don't appear suitable, *to my eyes*, for web images designed for display on good monitors -- especially not LCD's. LCD's have considerably more color range than what sRGBv4 was designed for (not to mention FF doesn't grok v4 color profiles). Applying sRGB to an image, vs. say the one for my monitor, dims out the colors and reduces the gamut size -- perfect for lower quality monitors and, perhaps printing to home printers, that have smaller color-range capacities (gamuts), but for looking at them on monitor, you'll be ripping off the color you could be experiencing with a profile suited to your monitor.
So before you complain about FF's SVG being screwed up, make sure your system is ready to display accurate ouput -- have your DPI set correctly for your monitor and get your monitor profiled (and calibrated if you can). The Win7 built-in util is adequate for generating a reasonable color profile. You can use that to install in FF (because FF doesn't use the system's profile, it will default to sRGBv2 if unconfigured (at least that's what it's documented to do...) which will usually be wrong if you've profiled your monitor.
Sorry if I snarked, must have been something in my throat. *cough*... SVG isn't super trivial -- learning it could take between now and when Mozilla gets a perfect implementation. :-) Check out the 'inkscape.org' website -- it's a free downloadable creator/editor for SVG -- and the focus is more on artwork than text. You can see lots of examples there, and it's amazing how many already work -- there's an entire website of clip art examples at www.openclipart.org. Also, if you google up SVG clipart, you'll find a bunch of stuff that already works in FF. So it's not a matter of 'when FF will be ready to display and use SVG', there are websites based on it already.
But are you discussing if *you* should use it? I was discussing the pro's and con's for all PC users, and requirement of admin privileges is a valid point.
Dilbert RSS feed
OK, keep talking to yourself instead of replying to my messages.
I say your http requests are still logged, and you can be traced via http headers or reverse dns. You reply with DNS again. Nice.
I should know better than to start discussions with shills.
Dilbert RSS feed
So your response to someone saying that he expects better quality of a program with an annual budget of 68 million dollars is to... fix it himself?
Firefox doesn't need access to system files, it can be Sandboxed for safety. Your "server" can't.
So you either lose your security, or the auto-update functionality. Hey, great improvement!
Then you don't block the damn banner.
Show me how you can block a banner and still access the rest of the content from the same domain, using the Hosts file.
Dilbert RSS feed
You're a shill because you start a technical discussion to surreptitiously promote your own application. It's not an argument, it's a fact.
Secondly, I have never tried to make Adblock a superior alternative to Hosts. I refered some advantages it has over the Hosts files, to add to your post.
Unlike you, I don't have any personal interest in trying to show what's the "best". I don't even use Adblock.
But I'll reply:
1) Sure, if you don't run any auto-updater, especially a daemon like yours. :)
2) Good Adblock lists are also available; their equivalent.
3) Right. But the auto-updater might be. Again, you can sandbox Firefox. You can't sandbox the updater (Yes, you *could* IF you wrote the driver, which you didn't)
4) Sure, I've used it in multiple occasions.
5) Yes, but if you want to auto-update, you have to allow writes to it.
6) Again, that's completely true. In the times of CoD1 you could use it to fake the key server
7) If "they" can log DNS, they can log HTTP requests. If you're worried about privacy you should use Tor, at least
8) Again, that's completely true.
9) Well, you can also use multiple DNS servers (I use my ISP and OpenDNS), which will give you access to all the sites, not the few you have hardcoded
10) Sure it does, you get nice GUIs, right-click context for images and it even overlays little buttons over Flash content to block it.
Again, Adblock and Hosts, both have their advantages, and they're best used together. But Firefox supports them both, and Chrome doesn't.
Dilbert RSS feed
Well, that was a useless 12000 character post. All you have done is failing to rebut my rebuttals, and saying "AHAH!!1!" when I agree with you that the HOSTS file approach has advantages, as I ever said it did.
I'm glad you "won", not sure what, but ok.
But it's not hard to find your email to "request" you app, is it?
Suffice is to say, if you weren't shilling, you would post the app name.
Oh, and by the way, just a heads up: Kylix has been discontinued for quite a while now. The last version doesn't even work with the latest glibc version.
Dilbert RSS feed
No my response to him is to find an alternative program that DOES meet his needs. Mozilla is doing just fine for me and many other people. I did not pay any of that money and I do not own stock in Mozilla, so I do not presume that I have the right to dictate to them how to spend that money. If it turns out that they are doing something unethical that I find unacceptable, my only recourse is to stop using the browser. Why does that poster feel that they have the right to tell Mozilla what to do? As I said, he/she can find another browser, submit patches to fix the problems, or deal with it same as the rest of us. Maybe the poster would find it helpful, or at least personally satisfying to file a bug report.