That's correct, because the reflow branch will be merged onto the trunk, rather then the 1.8.1 branch. You can expect Firefox 3 to pass Acid2. My point is merely that the code for Firefox passing Acid2 is written and available if you want it. It's not vaporware, in other words.
That sounds like normal memory use. The memory leak becomes apparent when you can't get memory usage to go below 200 MB. Besides, are you actually having any problems with memory usage, or just complaining about a number? You really shouldn't be watching the numbers at all...
I don't think the point of the CVS cop-out is that you shouldn't discuss features available in non-final builds. The point is that you shouldn't dismiss problems in final builds because they're fixed in non-final builds.
My biggest beef with Firefox is that it still crashes frequently and has massive memory leaks that require me to quit and restart the browser on a daily basis. It doesn't take much to get Firefox to grow to 1GB in memory footprint and start causing my system to thrash. A fundamental flaw is that it does not release memory back to the OS, so when you close tabs and windows, the process doesn't shrink. While this isn't directly Firefox's fault, there are lots of ways around this that they refuse to implement. On the other hand, the true memory leaks ARE their fault.
Any clues to how we could reproduce this extreme memory leak so we can write up a bug report and get it fixed would be greatly appreciated!
Garbage collectors can collect only objects that cannot be reached, and therefore cannot be used. If a pointer or reference in the code hangs onto an object so that it might be used later, a garbage collector cannot collect it. Many Firefox extensions have problems like this — they keep a reference to many pages so they cannot be collected, even with the most sophisticated garbage collector. A web page can also leak memory by continually allocating objects and holding onto references to them.
That's exactly my point. End users should not monitor memory usage, and therefore should not see the effects of memory leaks. They're only noticeable if you watch closely over days.
The Firefox 2 Beta 1 builds are all candidates until the final is announced. The latest is Firefox 2 Beta 1 RC 3, but there might be more changes before Firefox 2 Beta 1 is officially released.
Yes, there are hundreds of leaks in Firefox. Gmail triggers several of them. The good news is that all but one of them is fixed on the trunk, so Firefox 3 should leak much less on Gmail. The other good news is that you generally need to run Firefox for several days before the leaks become noticeable, if you monitor the memory use number closely.
Firefox's success has to do with marketing, plain and simple. Firefox had a major marketing push, as a completely free browser. Opera could obviously never reach as many users as Firefox since it was adware/payware.
Opera's failure to gain usage share dates back to 2002 when Mozilla 1.0 was released. That's when a large percentage of existing Opera users seemed to switch from Opera to Mozilla. Opera was able to gain more users up until that time, but since the release of Mozilla 1.0 Opera hasn't been able to sustain any increase in usage share.
People who thought that Opera's status as adware/payware was holding them back claimed that Opera usage would soar quickly after removing the ad banner/making Opera free in 2005. Many people who made that claim have now realized it didn't happen. It looks like users really do prefer Mozilla browsers.
There's a good test available at present, and the experiment is being performed all over the world. People can have both Firefox and Opera free, and they choose Firefox. They choose Firefox even though Firefox is the still the most unstable program in common use.
(The 1.5.0.4 version of Firefox is quite stable when the FlashBlock extension is installed, but still, after days of many windows being loaded, Firefox is so unstable that the Microsoft Windows operating system must be re-started to return to original performance. My experience has been that Linux remains stable, but that all Firefox windows must be closed to regain performance of Firefox.)
Although I don't doubt that Firefox is unstable on your computer for some reason, I don't think many Firefox users are seeing that kind of instability on Windows. In the past several years of hanging out in the MozillaZine forums, I can't remember even one person saying they had a problem so bad that they needed to reboot Windows. Serious problems are generally the result of bad extensions, plugins, hardware problems, driver problems, or other problems not with Firefox. If you want to continue to claim Firefox has a common and very serious bug in it, I'd like to see the Bugzilla bug report on it.
Of course! They've written the Leak Monitor extension to help extension authors to find leaks in their extensions. As far as I know, there's no way to limit the memory an extension uses without causing additional problems, or to kill an extension that's using too much memory. If you have a detailed suggestion for how extensions or plugins could run in their own processes, perhaps you should explain it.
If a browser manages to bring a 2 Ghz system with 768 MB to its knees in a mere half hour of browsing there is something wrong. Unfortunately this often-heard complaint does not seem to get the attention it deserves.
I agree. Write up a bug report, complete with instructions for how to reproduce the problem, so it can get the attention it deserves.
Plugins, especially Flash, have also been known to cause problems such as high memory use and 100% CPU use after waking up from hiberation. Be sure to get the latest Macromedia, Java, and Acrobat plugins.
But the web is supposed to be a cross-platform environment from the ground up. For the most part, I treat it that way, and just let pages look suboptimal in IE. If someone complains how a page looks in IE, I just "dumb down" the page so IE can understand how to format it better.
thecounter.com reports about 5% of users still use Windows 98. That percentage will be much lower by the time Firefox 2.x stops getting security updates.
Security bugs that have not already been publicly disclosed cannot be viewed until users' systems are updated with the patches. From a standpoint of trying to keep users' systems secure, it makes no sense to give malware authors easy access to information on how to write exploits. Mozilla is being responsible by keeping the bug reports secret for now. They will be opened later, after users' systems are patched.
Slashdot Mirror
That's correct, because the reflow branch will be merged onto the trunk, rather then the 1.8.1 branch. You can expect Firefox 3 to pass Acid2. My point is merely that the code for Firefox passing Acid2 is written and available if you want it. It's not vaporware, in other words.
That sounds like normal memory use. The memory leak becomes apparent when you can't get memory usage to go below 200 MB. Besides, are you actually having any problems with memory usage, or just complaining about a number? You really shouldn't be watching the numbers at all...
I don't think the point of the CVS cop-out is that you shouldn't discuss features available in non-final builds. The point is that you shouldn't dismiss problems in final builds because they're fixed in non-final builds.
Garbage collectors can collect only objects that cannot be reached, and therefore cannot be used. If a pointer or reference in the code hangs onto an object so that it might be used later, a garbage collector cannot collect it. Many Firefox extensions have problems like this — they keep a reference to many pages so they cannot be collected, even with the most sophisticated garbage collector. A web page can also leak memory by continually allocating objects and holding onto references to them.
That's exactly my point. End users should not monitor memory usage, and therefore should not see the effects of memory leaks. They're only noticeable if you watch closely over days.
The Firefox 2 Beta 1 builds are all candidates until the final is announced. The latest is Firefox 2 Beta 1 RC 3, but there might be more changes before Firefox 2 Beta 1 is officially released.
There are already builds of Firefox available that pass Acid2.
Yes, there are hundreds of leaks in Firefox. Gmail triggers several of them. The good news is that all but one of them is fixed on the trunk, so Firefox 3 should leak much less on Gmail. The other good news is that you generally need to run Firefox for several days before the leaks become noticeable, if you monitor the memory use number closely.
Opera's failure to gain usage share dates back to 2002 when Mozilla 1.0 was released. That's when a large percentage of existing Opera users seemed to switch from Opera to Mozilla. Opera was able to gain more users up until that time, but since the release of Mozilla 1.0 Opera hasn't been able to sustain any increase in usage share.
People who thought that Opera's status as adware/payware was holding them back claimed that Opera usage would soar quickly after removing the ad banner/making Opera free in 2005. Many people who made that claim have now realized it didn't happen. It looks like users really do prefer Mozilla browsers.
Of course! They've written the Leak Monitor extension to help extension authors to find leaks in their extensions. As far as I know, there's no way to limit the memory an extension uses without causing additional problems, or to kill an extension that's using too much memory. If you have a detailed suggestion for how extensions or plugins could run in their own processes, perhaps you should explain it.
There are lots of extensions with memory leaks and other serious problems. Be sure you're not using an extension on that list if you're having problems.
Plugins, especially Flash, have also been known to cause problems such as high memory use and 100% CPU use after waking up from hiberation. Be sure to get the latest Macromedia, Java, and Acrobat plugins.
But the web is supposed to be a cross-platform environment from the ground up. For the most part, I treat it that way, and just let pages look suboptimal in IE. If someone complains how a page looks in IE, I just "dumb down" the page so IE can understand how to format it better.
PHP:
$x["a"]["b"]["c"] = "d";
Perl:
$x[a][b][c] = "d";
You can put also write it the PHP way (with all the quotes) in Perl, if you really want to.
thecounter.com reports about 5% of users still use Windows 98. That percentage will be much lower by the time Firefox 2.x stops getting security updates.
Yes, i know, those who don't watch Showtime won't get it.
I found this article helpful when I was shopping for an SSL certificate.
Security bugs that have not already been publicly disclosed cannot be viewed until users' systems are updated with the patches. From a standpoint of trying to keep users' systems secure, it makes no sense to give malware authors easy access to information on how to write exploits. Mozilla is being responsible by keeping the bug reports secret for now. They will be opened later, after users' systems are patched.