Mozilla Branches for 0.9.8

Wednesday January 23rd, 2002 today branched for Mozilla Milestone 0.9.8, as well as opened the tree for 0.9.9 checkins. Look for branch builds to start sometime later this week, as early as tomorrow. (While a certain site believes that without builds, you can't have a branch, that is not true.) Pre-0.9.9 trunk builds will start tomorrow, or possibly tonight. See our earlier story for what's new in 0.9.8.

#1 When will 0.9.10 come out?

by hfx_ben

Wednesday January 23rd, 2002 9:40 PM

... sorry, I just couldn't resist. ;)

#2 Re: When will 0.9.10 come out?

by mrflip

Wednesday January 23rd, 2002 10:21 PM

Not for a while. The next version after 0.9.9 will, of course, be 0.9.A; only after 0.9.F do we go to 0.9.10.


#56 YHBT

by tny

Monday January 28th, 2002 8:19 AM


#3 A harsh test on 2002012203 build

by dipa

Thursday January 24th, 2002 1:28 AM

This build comes a little before branching, so it might characterize the upcoming milestone release too.

More to the point: From time to time, I perform the following test, in order to check browser's stability: I start with site that contains a lot of spam. I open links on new windows. In, this way, I open at least 20-30 browser windows, containing different pages. When the memory usage increases to the point of heavy disk thrashing, I close some windows. Tenths of different sites are browsed.

This test always made NS3.x and 4.x to crash after a few minutes. IE5.5 (security set to medium) performed very well, but after 1-2 hours of testing the system often becomes unstable (not able to save files, open new windows etc). Older Mozilla versions (e.g. 0.9.4) were definitely better than NS4 but not as stable as IE. Build 2002012203 did it (the least to say) astonishingly well. I was able to use it for many hours without a single crash. Memory usage was acceptable. And no system degradation problems, unlike IE. In fact, I never had a so extended browsing session before, on any Win32 os and browser combination. I 'm not sure if this stability was caused by the javascript settings in Mozilla preferences (I disabled pop-up windows, window moves and resizes, status bar overwrites etc). Actually, I don't care, because all these annoying javascript techniques are useless, so any ability to block them is a plus for me.

The above "test case" is not mentioned as a means of comparison between IE and Mozilla. Nor it is a consistent and reproducible test for anyone else except myself (I suppose smoketests are intended for this purpose). Just figured out that current Mozilla is so stable that I could hardly imagine before.

A few big annoyances remain, though: a) Window closing time is still long, when many windows are opened (at least 1 sec). b) Cpu usage is relatively high, enough for delaying new window creation, when the parent window is still loading.

Machine used: K6-III/400, 192MB ram, Win2000.

#51 Re: A harsh test on 2002012203 build

by rcoscali

Sunday January 27th, 2002 7:10 PM

Have you got the possibility to play your test on Linux ? It could be interrested to compare mozilla stability on several plateform.

#52 Re:

by dipa

Monday January 28th, 2002 12:33 AM

Unfortunately, there's no Linux on my machines, yet. I have to wait for the next hard disk upgrade :-(

In Win32, there was a "trip point" from where Mozilla became more stable than NS4.7. That was somewhere early in 2001, I think. From then, it continuously accepts improvements so the distance between Moz and NS4.x is getting longer.

From what I see on Bugzilla, Linux version stability problems occur at slightly faster rate than in Win32. It's a good thing that Linux testers are so many that bugs are quickly detected . Macintosh versions seem to be more problematic, that's my impression from everyday bug reports.

#4 Re: A harsh test on 2002012203 build

by tslukka

Thursday January 24th, 2002 4:09 AM

Hmm. Since we are talking about stability, I think it is the right time to point out this:

It is something I wrote over a year and half ago to test browser stability. I think there are still no browsers out there that could handle it without crashing. IE versions 4.x crashed also WinNT4 with this page. IE 5.x versions only crash the browser. Mozilla is not doing much better either.

I am not saying that it browser is bad, if it can't handle this kind of abuse but it is still interesting to see how they crash. Maybe one day we will have some mechanism in Mozilla to prevent this kind of sites crashing browsers.

#5 Works on Opera

by rkl

Thursday January 24th, 2002 5:09 AM

Your \'browser killer\' test (basically it\'s frames within frames within frames etc. etc.) actually seems to work quite well on Opera 5.05TP1 on Linux (I have DNS problems with Opera 6 on Linux, so that\'s I\'m still on 5.05). Opera runs very slowly as you\'d expect whilst loading, but it actually seems to complete (with the odd \"can\'t retrieve URL\" cos the Finnish site is probably strained during the test :-) ) without crashing.

Mozilla 0.9.7 froze straight away and I had to kill it by hand.

#6 Re: Works on Opera

by tslukka

Thursday January 24th, 2002 5:22 AM

Well, I think that "odd can't retrieve URL" is not a fault in my server. see below: 2:10pm up 6 days, 16:37, 3 users, load average: 0.08, 0.03, 0.00

However I admit this this really dirty trick to kill browser.

#7 Re: Re: A harsh test on 2002012203 build

by illsleydc

Thursday January 24th, 2002 6:26 AM

Actually with Win2k 2002012304, moz stays responsive and I can click Stop when I realise what's going on. Perfectly satisfactory.

What exactly whould you suggest that a browser should do when faced with this test?


#8 Re: A harsh test on 2002012203 build

by tslukka

Thursday January 24th, 2002 6:52 AM

Well. Try to simulate this behaviour with static html page. It does not work (or didn't work when I tried it), since there is some kind of check. It should not be possible to pass that check by supplying some dumb parameters. Also it would make sense to prevent mozilla from loading more than 50 frames from the same site. (the actual limit could be configurable in user.prefs)

However it is good to hear your browser stayed responsive, because mine just died before I could hit the stop button. It should allways be possible to stop whatever is going on. The point in all this is that when I surf in the net, I don't know what html will possibly come from the server. I don't want that some evil server-administrator can crash my browser without me having a say on that.

#15 Works for awhile...

by CatamountJck

Thursday January 24th, 2002 5:31 PM

I've found that Mozilla works fine for quite awhile for me... more than long enough to have pressed stop. I managed to get Mozilla from using 25,000K to 95,000K of memory before the browser crashed... until that time (or shortly before - I gave up testing it constantly after about 10 minutes) it responded quite well. Although it did crash in the end, I'm more than satisfied with stability in that respect! Go Moz!

#26 Re: A harsh test on 2002012403 build


Friday January 25th, 2002 7:36 AM

You see, I used a different build than yours (on Win98). The page itself caused no problem and I had plenty of time to stop it. I waited till the frames in the upper left corner were so small, that the single sub-frames weren't visible anymore. Browser was responsive all the time, so I stopped the process. So far no problem, but when I hit the back button of the browser, it crashed immediately.

#38 Works on IE6

by mtrump

Saturday January 26th, 2002 4:16 AM

Oops, that's rained on your parade....

#39 Re: Re: A harsh test on 2002012203 build

by macpeep

Saturday January 26th, 2002 4:54 AM

Like someone already pointed out, it works fine in IE 6.0. The browser and rest of the system stay 100% responsive and after seeing a couple of hundred frames or so (no, I didn't actually count them, but the window was pretty full of them), I just clicked stop to end the loading. Resizing the frames also works very well.. No noticable slowdown.

While the frames were loading, I opened up another IE window and surfed over to Worked great, except that it took about 2 seconds to open the new browser window due to the high CPU usage from loading the recursive frames.

I'm not really sure what exactly this test is supposed to prove. "Too much of" types of crashes or slowdowns will probably happen to any software at some point. Make an 80 GB long HTML file and see how it renders in Mozilla. Probably will cause some slowdown and/or crash. But so what? However, this recursive frame test doesn't actually seem to cause much else than some slowdown until you give the browser mercy and end the loading. But even if browsers crashed, it only proves instability in extreme situations.

What's interesting is stability under NORMAL conditions. At least that's what I assume most users would care about. You know, not crashing when you drag a bookmark to another folder or browsing a perfectly normal and common site.. I think those kinds of crashers should have a much higher priority than something like this.

#40 Re: Re: Re: A harsh test on 2002012203 build

by klee

Saturday January 26th, 2002 7:36 AM

Macpeep wrote: > Like someone already pointed out, it works fine in IE 6.0. The browser and rest of the system stay 100% responsive and after seeing a couple of hundred frames or so (no, I didn\\\'t actually count them, but the window was pretty full of them), I just clicked stop to end the loading.

Mozilla crashed for me when I _don\\\'t_ press Stop. What happens if you don\\\'t press Stop in IE6?

#49 Re: Re: Re: Re: A harsh test on 2002012203 build

by macpeep

Sunday January 27th, 2002 5:42 AM

It keeps loading for a pretty long time.. I'm on a 2Mbps DSL and I get tens of frames after only a few seconds, so the pace is very fast. My DSL modem lights are still flashing as if IE is loading data after 3 minutes and the IE memory usage is up to 300MB. Other IE windows (same process) are compltely responsive, as is the system, and completely stable. After 5 minutes or so, the window loading the frames becomes so slow that it seems it's impossible to switch to it (possibly redrawing simply takes so long that it seems "jammed"). The other IE windows and the OS is 100% responsive and stable. No significant change after 10 minutes. The DSL modem lights no longer flash so it's done loading / not longer responding. No crashes and the rest of the IE windows continue to work. Memory holding steady at 300 MB.

#53 Agree

by dipa

Monday January 28th, 2002 12:47 AM

A recursive test like this doesn\'t mean much. At this point, Mozilla is very stable (especially when using milestones or proven nightlies) but it\'s ui is still slow with regard to competition. The above test stresses cpu usage, so ui becomes pretty unresponsive, but this is not a normal case.

\"not crashing when you drag a bookmark to another folder or browsing a perfectly normal and common site\". That is what I expect from Mozilla wrt to stability. And it\'s doing pretty nice, at least for the site browsing.

#55 Re: Re: A harsh test on 2002012203 build

by hfx_ben

Monday January 28th, 2002 7:59 AM

I just used your page on 2002011703 and it held up pretty well! By the time if was obviously not responding it had filled the 35Meg of free RAM and written 45 Meg to disk cache ... 4.79 snuffed it much quicker.

BTW, when I finished running ChkDsk *ayup, locked up hard* I found this in the .chk file: "incomplete dynamic bit lengths tree oversubscribed dynamic bit lengths tree incomplete literal/" ... doesn't mean much to me, but it might to someone who spends more time reading the entrails of chickens.


#9 Problems

by Tanyel

Thursday January 24th, 2002 10:56 AM

The bandwidth problems have not been solved. The site still says, "Unable to select database right now. Try again later," and, "Unable to connect to SQL server," every day. It also tends to say there are no forums available. The form I am using to post this message will not fit on the screen even though the browser window does.

#11 Re: Problems

by strauss

Thursday January 24th, 2002 11:53 AM

Yeah, the site is now broken so badly between the HTML and SQL problems that it's not worth my time to participate. Just to respond to your message I had to switch to threaded mode to get the Reply link to point to the correct message, refresh a few times to get past messages about being unable to connect to the database, type in a box that is wider than the window so I have to keep scrolling back and forth, and figure out how to work around sidebars sitting on top of both your message and my reply text box. Right now the Submit button is 75% hidden under a sidebar and the last time this happened I had to dink around for almost five minutes to press Submit. This is a ridiculous level of problems.

Yep, can't press the button, even the part of it that's showing. Let's see, what did I do last time? Sigh. Oh yes, tab over to it, that works. But what do I get? Unable to connect to database. Let's try again.

Lee Strauss

#17 Re: Re: Problems

by kerz

Thursday January 24th, 2002 10:20 PM

Feel free to not return.

#21 :-) Nice shot !

by dipa

Friday January 25th, 2002 1:07 AM

Kerz, I definitely like your style :-)

#46 site problems and strauss

by JayeshSh

Saturday January 26th, 2002 3:40 PM


I would like to offer my help with testing your css/html design. Do you need any help with this?

I understand that it takes a ton of effort to run a site such as this one, especially since I think there are only several of you working on thisone site.

#28 Re: Re: Re: Problems

by macpeep

Friday January 25th, 2002 9:02 AM

I know it's a volunteer effort and all that, but I still think that's a pretty sucky attitude. Everything that Strauss was pointing out was true. The site used to work nicely, except for a few (very rare) database connection problems. Lately, it has taken a turn for the worse. I know this was done in the name of making the site lighter, but the method of making it lighter is a little weird, since it means the site is broken on a lot of browsers. You probably remember that I was complaining about it earlier (maybe a week or two ago) and it has considerably improved since then, at least on IE 6. But for example the talkback editor textarea (where I'm writing right now) is *WAY* too wide, which makes it impossible to see about half of the textarea. Things like this would be really easy to fix, but would make the whole site a lot more usable for those that aren't using Mozilla as the #1 browser.

Strauss is a very good contributor with lots of good advice, ideas and opinions. While his views are not so commonly shared here, he presents his opinion and arguments very clearly. He is a much needed "second opinion". Losing him because the site has been broken on his web browser / platform would be pretty sad. Going "feel free to not return", I must say, is beyond sad. It's actually very lame coming from the admin of the site!

As far as the HTML and database problems go, I don't see why it's such a big problem. If there's only a limited number of database connections available, just do a wrapper around whatever code is now returning database connections, and have it wait if there's no connection available (a certain connection count is exceeded). A simple mechanism like that would not even slow down the site, except for the few situations where there are no more free connections.

I also have a hard time understanding how it's so hard to make cross platform and light HTML, because the site isn't really that complex in it's layout. If it was a will to do it "right" (using only CSS and no tables etc. for layout) then I could understand it. But the primary motivation was to make it lighter. It wouldn't be hard at all to change the HTML slightly and achieve both working layout on all browsers as well as lightweight HTML.

#29 Re: Re: Re: Re: Problems

by mozineAdmin

Friday January 25th, 2002 12:18 PM

Oh, I have an idea. Fix IE!

Christ, why didn't anyone think of that before!

#30 Re: Re: Re: Re: Problems

by kerz

Friday January 25th, 2002 1:15 PM

People get on my nerves. So sue me.

I now see the textfield problem in IE, it looked fine to me before, as i never actually tried to use it.. Many people asked that it be wider, so I did that. I'll be working on getting that fixed in IE, right after I post this. As far as the database problems go, we think we have increased the connections available high enough to avoid anymore errors for the time being. You may still see them, but that just means the site is busy, and to try reloading. We'll work more on that, but for now it seems to be ok. We actually had two goals with changing the site's HTML. We want to both reduce bandwidth, AND create a standards compliant layout. I don't see any problems in IE right now other than the textfield issue and a long line under the dates on the main page, which is an IE bug.


#32 Re: Re: Re: Re: Re: Problems

by macpeep

Friday January 25th, 2002 3:08 PM

"I don't see any problems in IE right now other than the textfield issue and a long line under the dates on the main page, which is an IE bug."

I actually thought that the long line under the dates was intentional. The only thing that isn't working properly with IE 6 (now that I see you have fixed the textarea - thank you!) is that the borders and background colors of the "boxes" around posts is sometimes missing. It's not a serious bug - just cosmetic..

However in a more general sense, I think it's pretty pointless to say "it's a bug in IE, not this site". I mean.. Yes, so it's IE's fault. It doesn't mean that one shouldn't try a work around. After all, IE *IS* the most used browser in the world, so asking that a site at least WORKS on it isn't too much to ask, I think.. But.. as I said, the site now actually seems to work fine in IE 6.0, so I have no complaints regarding that! Thanks for the IE fixes!

#35 Re: Re: Re: Re: Re: Re: Problems

by strauss

Friday January 25th, 2002 7:00 PM

I still have all the problems on Mac IE 5.1 that I first reported on Monday in the Site Suggestions thread and repeated here, except that the edit text field for replies has now been made so narrow that it doesn't require horizontal scrolling or hit the sidebars, which still sit on top of message content.

I can't read your whole messages, but I did see some kind words that happened to be far enough to the left of the sidebars to be visible, so thanks.

#36 This site best viewed with Mozilla

by Kovu

Friday January 25th, 2002 7:52 PM

If you're using Mac IE 5.5 and visiting a Mozilla-centric site - indeed, a site that's very existence is to propogate Mozilla users - that's your problem. Come back and use Mozilla. This is not "IE-zine", I think as long as the site works in Mozilla, Kerz has done his duty.

And some of you may want to note that MZ is pretty much a one-man show, one man who works for Netscape, has a girlfriend at home, and thus has some kind of life. He doesn't spend every waking moment on the site, nor should he.



#44 No excuse

by thelem

Saturday January 26th, 2002 12:29 PM

When I'm at home, I access the site using Mozilla. When I'm at school I have no choice of browser, so I have to access the site in IE (5, 5.5 or 6 depending on the computer). It should work on these browsers.

#45 solution

by niner

Saturday January 26th, 2002 3:32 PM

so maybe you could offer some help? I'm sure kerz will appreciate if someone takes the task of making the site compatibel with other browsers :)

#83 Chicken and egg problem

by tepples

Thursday January 31st, 2002 9:18 AM

> Come back and use Mozilla. This is not "IE-zine"

But it's also not "Mozilla-users-exclusive-zine." Some users come here to *get* Mozilla. If a site does not work in the other major browsers, how are users behind other major browsers supposed to find a good Mozilla recent nightly build (i.e. one without prominent smoketest blockers)? It becomes a chicken and egg problem. And what browser are they supposed to use if Mozilla starts crashing on them? Are they supposed to spend 15 minutes reinstalling an older build of Mozilla just so they can post about why the build is broken? And what about users on machines they don't administer, such as users at schools and libraries?

#42 The problem with Mac IE 5.1

by grayrest

Saturday January 26th, 2002 9:56 AM

The problem with Mac IE is that it\'s more standards compliant than Win IE but less than mozilla. The IE stylesheet was specifically made to work with Win IE since that\'s what the majority of non-mozilla browsers that visit this site are. The other broblem is that nobody who works on the site can test with Mac IE 5 because none have access to a mac. Unless you or somebody else running your configuration know CSS well enough to submit changes to the IE stylesheet and have it work on your browser and on Win IE, this issue will most likely not be fixed.

#31 Re: Re: Re: Re: Problems

by tradervik

Friday January 25th, 2002 2:35 PM

What is mozillaZine built on? CGI? PHP? If you want to do things like connection pooling, I think should switch to something like server side Java. If you think J2EE sucks, go with Turbine (see

#43 PHP n/t

by thelem

Saturday January 26th, 2002 12:25 PM


#47 Re: Re: Re: Re: Problems

by SubtleRebel

Saturday January 26th, 2002 4:08 PM

macpeep, the biggest problem with strauss is his attitude. Your post here defending him may directly oppose my view, but you at least have presented your opinion in a fairly reasonable manner.

strauss's post complaining that "the site is now broken so badly between the HTML and SQL problems that it's not worth my time to participate" is full of insult and sarcasm, but hardly anything constructive. He may express his opinion clearly, but that does not make it usefull.

#50 Re: Re: Re: Re: Re: Problems

by macpeep

Sunday January 27th, 2002 5:51 AM

"strauss's post complaining that "the site is now broken so badly between the HTML and SQL problems that it's not worth my time to participate" is full of insult and sarcasm, but hardly anything constructive"

It doesn't have to be constructive. He was pointing out that the site is so broken that he doesn't feel like wasting time fighting with it in it's current shape. Right after the side modification had been used, it was so badly broken for me too that I could not use it. I felt the same way. It seems very, very silly to force people to use a marginal browser to view this site. Yes, I know it's a site about Mozilla, but just because most readers are interested in Mozilla doesn't mean that they are / can be surfing with Mozilla as their primary browser. And yes, now, with the latest fixes, the site is again quite usable, at least on IE 6.0, even though there are some minor cosmetic glitches. But if it's not on Mac IE 5.5 that strauss is using, then his comments are very accurate and not sarcastic at all. More like there was a deep sigh in his message. I can understand it quite well..

The Mozilla attitude has always been to get rid of "Best viewed with ...." buttons on sites that have been made to only look good in some browsers. Well, MozillaZine did the exact same thing. The only justification was that it was standards compliant. However when the only browser that could view the site has less than 1% market share, it's definitely not a good thing to shut out other browsers.

I would *LOVE* to use Mozilla as my primary browser and primary email client. I just plain and simply do not find the quality to be good enough for me to switch. I still use NS 4.79 for email and IE 6.0 for the web because those are the two apps that work best for me. That doesn't mean that I'm not interested in Mozilla. So why would you want to force me to use Mozilla and have a degraded overall web experience? Just so I could view this one site? Doesn't sound very tempting..

#58 Re: Re: Re: Re: Re: Re: OT: Mozilla Logo

by JayeshSh

Monday January 28th, 2002 1:09 PM


I think you are confusing several issues:

1)This site's message-database is sometimes unresponsive ("bandwith problems").

2.)This site sometimes displayes incorrectly - both in Mozilla (with the page squeezed to one third of its proper size; this problem also makes typing the talback )textarea) formfields difficult) and in other browsers (Internet Explorer and Opera.)

3.)Whether it is okay to post messages which are inflammatory, sarcastic, or which attribute blame to a specific person (or persons) for bugs / lack of features in either Mozilla or Mozillazine.

In the first case, I think the "Bandwith problems" have been largely resolved, thus leading the "unable to select database" to occure much infrequently than previously.

In the second case - the Mozillazine site does/did not display correctly in Internet Explorer - don't forget that there are page display errors in Mozilla too. Also, a couple things to keep in mind about Mozillazine specifically and webdesign in general: firstly Mozillazine is run by only a couple of people (Jason (kerz) and Asa, if I understand correctly) and secondly, creating an interactive website which works for most browsers and platforms is really, really tedious and time-consuming. If you put the two of these points together, you'll see what I mean: Mozillazine is a volunteer site with limited human resources, and has to support high traffic, and a whole community users, each of whom has his or her idea about how the site should look and how Mozilla should behave. Looking at what Mozillazine adds to the Mozilla community and development process, I'd say that it hardly a waste of my time to participate in discussions here. If the site is unresponsive or if the pages show up squeezed to the left, I just press Reload, and most of the time that solves the problem.

As far as Strauss' comments go: his post was not written with the aim of objectively expressing his difficulties with the site. It was a subjective post, and while explaining that the site did not work, also implied that the site was not worth using (i.e. was *unusable*). Think about that for a second. I think the site is very usuable, since pressing reload solves most of the problems. Additionally, implying that he would never come back to Mozillazine because of a temporary problem was short-sighted at best, and incendiary and embodied non-constructive critcism, when looked at in the worst possible light.

None of us pay any subscription fees to take part in Mozillazine. Why should we expect that it works as reliably as a commercial service? And no one is compelled to take part in this site or in the Mozilla community. Participation from all seems to be welcomed here, but this is not forced upon anyone. And as far as Mozilla's market share is concerned - I think you are bringing in an irrelevant issue here: this is not about how many people use Mozilla. It is about what we expect from Mozillazine and what it means to be a constructive member of the Mozilla community.

It is important to be able to examine ourselves critically (and to be aware of our mistaked), yet I fail to see how anyone could find great fault with the people who run Mozillazine, since it a site run voluntarily, without commercial support, and with limited resources.

In short, please be patient, and please try to understand what I mean by *constructive criticism.*


- Jayesh

#59 minor correction

by asa

Monday January 28th, 2002 1:31 PM

>Mozillazine is run by only a couple of people (Jason (kerz) and Asa, if I understand correctly)

Scratch Asa from that list. I'm just a contributor. I contribute the BuildBar with a webform not unlike contributing an article to the site. The two guys behind the site are Jason Kersey and Chris Nelson.

#60 thanks for the correction, Asa

by JayeshSh

Monday January 28th, 2002 1:57 PM


thanks for the correction!

Regards, - Jayesh PS: Sorry for the wrong message title in my last post - the form manager had saved it, and I overlooked it before I hit "Submit".

#63 Re: Re: Re: Re: Re: Re: Re: OT: Mozilla Logo

by macpeep

Tuesday January 29th, 2002 12:33 AM

"As far as Strauss' comments go: his post was not written with the aim of objectively expressing his difficulties with the site. It was a subjective post, and while explaining that the site did not work, also implied that the site was not worth using (i.e. was *unusable*). "

He said it was unusable FOR HIM. And not worth using FOR HIM. Did you read what kind of problems he has / had with it? I frequent this site too and right after the redesign, the site was damn near unusable for me too. Strauss said that with this amount of problems, it's not worth his time (you know, work, wife, kids, dog, life.. all that).

"yet I fail to see how anyone could find great fault with the people who run Mozillazine, since it a site run voluntarily, without commercial support, and with limited resources."

Putting a site that was BADLY broken for 80% or more of all web browsers used on the Internet hurted a lot of readers and made it impossible for them to use their site. It had nothing to do with resources. It was bad planning, *TESTING* and implementation of the site update. It has nothing to do with Kerz having a girlfriend, not being paid to run MozillaZine etc. It's still bad planning, bad testing and bad implementation. Now it's fixed, and that's great. But at the time it happened, it was really awful and the amount of negative feedback from it shows it (check the other threads).

"implying that he would never come back to Mozillazine because of a temporary problem was short-sighted at best, and incendiary and embodied non-constructive critcism, when looked at in the worst possible light."

Where does Strauss imply that he will *NEVER* come back?

#81 Re: Re: Re: Re: Re: Re: Re: Re: OT: Mozilla Logo

by SubtleRebel

Thursday January 31st, 2002 6:57 AM

Mozillazine revised their website and the revision had some issues for some browsers. Why does that give strauss or anyone else a right to insult them?

If strauss found the problems to be unbearable then he had the option of not visiting the site; no one asked him to waste his time; no one asked him to waste our time with his bitching.

Although strauss implies that he has better things to do than visit a website that he can not view properly with IE 5, he does not seem to consider the fact that the people behind Mozillazine might have better things to do than work on making Mozillazine view well in IE 5.

If strauss were paying for Mozillazine then he might have some justification for his ranting. If strauss had offerred polite, constructive criticism then no one would mind his post. If strauss has a mental disorder that causes his rudeness then we could probably excuse it.

There were plenty of people who reported problems with the new Mozillazine layout without being obnoxious, why should strauss not be expected to do the same? Is he not capable of staying within the bounds of appropriate behavior? Regardless, why should anyone be expected to quietly tolerate his inappropriate posts?

#19 no problems here

by niner

Friday January 25th, 2002 12:42 AM

but I have to admit that I always use threaded mode, but still. The design fits perfectly, everything is usable and these SQL errors were there for at least a year from time to time but hitting reload helps..

#10 Another Browser Crasher

by jsgremlin

Thursday January 24th, 2002 11:18 AM

This is a JavaScript crash of your favorite web browser, whatever it may be (so long as JavaScript is enabled).

This is more of a hang than a crash really. Most browsers won't send up blue screens, they just never work right afterwards until you hard kill them. It has worked on all browser in the past, but I just tested it on NS 4.7x and Opera 6.0b. NS 4.7 appears to escape, but you'll notice some things not responding and then when you exit the browser just keeps going, sucking up more and more memory/virtual memory until you kill it or it eats your system alive. Opera hangs. IE has hung in the past. Mozilla hangs. All of them, meanwhile, just keep grabbing more memory. It's really a pretty cool kill. Somehow it avoids IE and Mozilla's warnings about scripts taking too long and Netscape 4.x and Mozilla's script termination on "too much recursion".

#14 Oh... My... God!

by jonasj

Thursday January 24th, 2002 3:22 PM

Very simple script which makes Mozilla hang. Must not be possible. Is a bug filed?

#23 Re:

by dipa

Friday January 25th, 2002 1:34 AM

The bug for filtering recursive events that lead to crash is 77271

Some known hangs from recursive events: bugs 112241 and 89300.

#22 Re:

by dipa

Friday January 25th, 2002 1:16 AM

What the hell of a code this page uses! I\' wasn\'t able to download it for investigation. Tried \"Save Link As\" and download manager. Both methods produce an unreadable page (not valid html).

#24 Here's what the page does


Friday January 25th, 2002 5:12 AM

function han(n,start,dest,aux){ if (n<=1){ stepsdone++; stepsout.value=stepsdone; output+= start+dest+" "; if (stepsdone==stepsto){ return output; } } else{ han(n-1,start,aux,dest); han(1,start,dest,aux); han(n-1,aux,dest,start); } }

And later on document.write(han(50,'A','C','B')); in the body, using script-tag. The rest of the page is harmless HTML and does nothing of importance.

#27 thanks (n/t)

by dipa

Friday January 25th, 2002 7:46 AM


#33 towers of hanoi

by jsgremlin

Friday January 25th, 2002 6:32 PM

It attempts to solve the towers of hanoi problem for 50 discs, a problem which requires more than 1.1258 E 15 steps (and as many recursive calls). If there are any computers that can solve it, they are in super-computing labs and certainly not using JavaScript to work on the problem. The same code with a number less than 15 or so can be used to actually solve the problem for x discs.

#25 Your link is harmless on Build 2002012403


Friday January 25th, 2002 7:18 AM

Just tried your link on Build 2002012403 on Win98 and the browser stops reacting for maybe 2 minutes, after that it reacts like normal and only an empty page is displayed. It does not crash the browser and it also does not waste enough resources to endanger the system (once the page opens, the memory usage raises by 5 MB, a decent system should survive that, even if it only has 64 MB RAM and running Windows with less than 64 MB is PITA). It's only annoying that you can't hit the stop button during those 2 to 3 minutes.

#34 patience is a virtue

by jsgremlin

Friday January 25th, 2002 6:34 PM

You have far more patience than I. It's possible that other browsers recover after that long too: I have never let them sit and try. As I mentioned, it's not just Mozilla, but NS 4, IE 5+, and Opera 6 as well. Opera 5, however, allows the user to retain full control, including stopping the page at will.

#61 Your script is all wrong

by astrosmash

Monday January 28th, 2002 3:30 PM

First of all, the han() function will never return any useful information. It only returns the "output" result when stepsdone==stepsto. stepsto is always 0 and stepsdone is never 0, so no data is ever returned.

Second, and most importantly, I compiled the script to C++ code to compare the performance. As C++ code, the han() function did not return after more than 10 minutes -- At the 10 minute mark, it had appended to the output string over 170 million times (I never waited for it to complete)

How would expect a web browser to handle a function like this?

-- darren

#71 but now it is right

by jsgremlin

Tuesday January 29th, 2002 10:29 PM

You are right that the script is messed up - that's what happens from cutting and pasting (from an old project) and editing without paying much attention. a corrected version is up at the same url. if you add ?x (where x is some integer) to the url, it will compute the towers of hanoi solution for x discs instead of 50. I have successfully run this as high as 20.

I know it won't work in C++. It won't work in anything. That's the point of the script. Nothing can compute more than 10^15 steps, and if it could, it couldn't store a more than 3000000 GB string. As to what the browser should do with this, it should realize the script is taking forever and at least give the user the option of nuking it. If you give MSIE a short novel and tell it to run a few thousand regexps on it, it will do this. If you try a more normal (but equally excessive) recursive function in Netscape, it will kill the script with the error: "Too much recursion". Whatever Mozilla does, it should not allow a javascript to lock up the entire browser indefinitely.

However, even if Mozilla continues to hang on this script, it's not in bad company. Opera 6, Netscape 4.x, and MSIE5+ all choke on it too.

Also, this little exploit means that if JS is enabled in email, I can kill your browser session without you having to do anything but try to read an email I send you.

#75 why mozilla does for me


Wednesday January 30th, 2002 4:22 PM

moz build 2002012608 gives the following dialog after about 1min: Script warning: A script on this page is causing mozilla to run slowly. If it continues to run, your computer may become unresponsive. Do you want to abort the script? OK / Cancel

#78 great! does it work?

by jsgremlin

Wednesday January 30th, 2002 5:23 PM

That's very good, much better than hanging totally, though it would be nice to maintain control of the browser (as in Opera 5) before that dialog comes up. This suggests that one of the bugs referred t0 earlier (by people other than me) is being worked on. I had managed to generate those dialogues with a script bubble-sorting some 9000 items, but they didn't work. Neither button gave me back control, and they kept popping up periodically after. So, are these OK/Cancel buttons more effective than the ones I experienced?

#99 WFM =)


Sunday February 3rd, 2002 1:03 AM

hm, well yes. that's what i said, right? after telling mozilla to stop the script it did stop the script. but you'll loose control while it is still working on the script, unfortunatly. but this would mean to thread the JS-interpreter and maybe even run it at a lower priority. i don't think there will be big changes for a while, since XUL and all the mozilla interconnection relys on JS. and still we want to ship 1.0 in april, don't we? maybe then an aproach could me made to take another look at the behavior. still it's unlikely for the near future. but we will see. we're can still be aiming for perfection for v2.0, right? =)

#12 mozillazine threading is wacked

by flacco

Thursday January 24th, 2002 2:27 PM

Database often unavailable, threading is all screwed up (replies go to the wrong thread).

What's up folks?

#13 Re: mozillazine threading is wacked

by kerz

Thursday January 24th, 2002 2:42 PM

Working on it right this very second.

#16 Stylesheets and Mozillazine?

by CatamountJck

Thursday January 24th, 2002 9:14 PM

Anybody notice that stylesheets on Mozillazine seem to load after the page has loaded - or sometimes not at all? Is this a mozilla bug or a mozillazine one? Not a terribly big deal, a reload usually fixes it but I'm curious if I should file a bug.

#18 Re: Stylesheets and Mozillazine?

by Lancer

Thursday January 24th, 2002 10:42 PM

Sidebar Link doesnt work too

#20 I'm seeing this, too

by dipa

Friday January 25th, 2002 1:01 AM

Other than bugs 111451 and 120370, I couldn't find anything. But those bugs are probably not related. Please file a new bug.

#41 Re: Stylesheets and Mozillazine?

by klee

Saturday January 26th, 2002 7:52 AM

The corresponding bugs:

Bug 121142 - intermittent layout bug

Bug 120946 - Navigation bar on Mozillazine mispositioned on first load

#48 Re: Stylesheets and Mozillazine?

by geertn

Saturday January 26th, 2002 5:01 PM

Maybe related, maybe not, but when I visit Mozillazine I always have to do a reload or I get the cached page. This doesn't seem to happen on other sites...

#37 mozilla now using XP's theme engine....

by wtmcgee

Friday January 25th, 2002 8:07 PM

is great. it\\\\\\\'s as close to native UI widgets as they\\\\\\\'re going to get, and the performance increase is incredible to me. also, it\\\\\\\'s a lot better looking on the eyes. mozilla/netscape 6x is definately going be a great browser when \\\\\\\'finished\\\\\\\'!!!

#54 Main Concern is IE only sites

by darex

Monday January 28th, 2002 6:11 AM

Most of the talkbacks on this site address what seem to be relatively trivial issues in comparison with my main concern - sites that - for whatever technical reason - you cannot perform major functions on with mozilla. How frequently does anyone else experience this?

#57 Re: Main Concern is IE only sites

by tny

Monday January 28th, 2002 8:29 AM

There is a bugzilla mechanism to cope with this issue: file an evangelization bug on the site at so the Mozilla evangelization crew can break out their Sunday-preaching-garb and sermonize the sinning web developers on the importance of cross-platform accessibility. We're all aware of it, but it is a web site developer issue, not a Mozilla issue.

This seems to happen most often in two cases: 1.) very small sites where the web designer simply doesn't have enough experience to know better, and 2.) very large sites where development and administration are separate functions and the developers don't see the web browser stats. The market penetration of Mozilla-based browsers sounds negligible until you actually look at your stats log and see how many hits you're dropping because your site is totally IE-specific.


by goodwatast

Monday January 28th, 2002 6:48 PM

So, for example, a site like where if you check out this link,1306,520|22199|1|,00.html your first problem is that the page falls apart, and from there you can't configure your car. I'm guessing that this happens with all of their vehicle configuration pages, not just the one for the new Z, but I haven't tried it.

Is this something that should be filed as an evangelism bug? And if so, what happens then? Do they contact Nissan?


#64 Re:

by bzbarsky

Tuesday January 29th, 2002 9:24 AM

> Is this something that should be filed as an evangelism bug?


> what happens then?

An evangelism person contacts Nissan, notifies them of the problem, and offers help in fixing it.

#65 Re: Main Concern is IE only sites

by bim

Tuesday January 29th, 2002 9:36 AM

I realy hate the sites that show a page telling me to use an other browser and when I set the user agent toolbar to impersonate IE, everything works. I normaly just search for a contact-address and mail them about the problem. It doesn't allways help however...

#66 Whats with this site?

by Lancer

Tuesday January 29th, 2002 12:42 PM

it looks horrible in mozilla

#68 Re: Whats with this site?

by WillyWonka

Tuesday January 29th, 2002 2:13 PM

I know looks horrible, but what can mozilla do about it? :)

I'm not seeing any differences between ie and mozilla here.

#69 Re: Re: Whats with this site?

by Lancer

Tuesday January 29th, 2002 4:23 PM

When i loaded in IE, appears fine. with no problems

#73 as usual....

by wtmcgee

Wednesday January 30th, 2002 12:07 AM

...99% of the problems experienced on websites when using mozilla, while annoying, isn't its fault. its bad coding by the web devolpers.

#84 Where's the documentation?

by tepples

Thursday January 31st, 2002 9:36 AM

>99% of the problems experienced on websites when using mozilla, >while annoying, isn't its fault. its bad coding by the web >devolpers.

And at times, they can be excused for not having good documentation available covering ways to work around bugs in the browsers. For example, what's the proper way to send XHTML to both IE and Mozilla? Sending as text/html makes Mozilla parse it as SGML rather than XML ( and ). Without a stylesheet, IE will draw XHTML sent as application/xhtml+xml as an outliner view (which looks like a collapsible "view source" instead of rendering it as HTML. How are web developers supposed to know that such a stylesheet exists, and that it's not just an inherent inability of IE to handle XHTML?

#87 Re: Where's the documentation?

by asa

Thursday January 31st, 2002 12:15 PM

<> <> <> <> <>

there's lots more out there. Use google.


#92 This MADNESS!!!!!

by Lancer

Thursday January 31st, 2002 7:30 PM


#67 where is .9.8?

by wtmcgee

Tuesday January 29th, 2002 2:05 PM

i thought it was due yesterday?

#70 Re: where is .9.8?

by fuzzygorilla

Tuesday January 29th, 2002 5:51 PM

In general, while the feature freeze and code branch are fairly dates that maintain a five week schedule, the release of the milestone is typically much more flexible. If you check the Mozilla Development Roadmap you will see that the "ideal release" date for 0.9.8 was 25-Jan-2002 (not yesterday). But the drivers are still trying to ensure as many of the "make Mozilla 0.9.8 not suck" bugs get fixed as possible. Expect 0.9.8 when it is ready (don't rush the Moz!)

#72 okay

by wtmcgee

Wednesday January 30th, 2002 12:05 AM

i'll just keep using the nightlies until then. just wondering where it was :)

#85 Asa, where are you?

by tepples

Thursday January 31st, 2002 9:40 AM

>i'll just keep using the nightlies until then. just wondering where it was :)

Heck, I wonder where the *nightlies* have gone. There hasn't been a new Windows nightly since 20020124. Asa, where are you?

#88 Re: Asa, where are you?

by asa

Thursday January 31st, 2002 12:17 PM

Sorry, busy working on the Milestone. Hopefully I'll be back to the buildbar soon.

Latest trunk builds <> Latest 0.9.8 branch builds <>


#74 Mmm...

by Tanaaln

Wednesday January 30th, 2002 9:46 AM

Well, all I can say is that I've been using mozilla as my primary browser and mail client since, oh, a year ago or so for the browser and since august for mail. The only problem I've had was a corrupted profile when I upgraded to XP from ME, but I doubt that has much to do with Mozilla. Recently, I've noticed that mail has become much faster... and actually works faster on IMAP than outlook (I haven't used outlook in a while, but I have what came with IE6, so it can't have changed that much.)

The only reason I use IE now is for kerberos authentication for my university's online records system, and to test my site layouts to make sure they don't explode or anything. ;)

Maybe my computer is just special, but Mozilla rarely crashes. One thing I don't really like is these textareas, though. They're better than they were a while ago, but still... weird stuff happens randomly, and there is a delay between when I type and when the text appears. Oh well. Go 0.9.8! :)


by Lancer

Wednesday January 30th, 2002 5:18 PM

I want to see 0.9.8 now!


by wtmcgee

Wednesday January 30th, 2002 5:21 PM

should be here any time now...

#79 Quit demanding

by PaulB

Wednesday January 30th, 2002 8:46 PM

Wehave no right to demand anything from Mozilla. They are providing one of the greatest browsers for free! Stop demanding---relax. 0.98 will be here when it is ready, not before like some other browser. ;)

#82 I get it!!! Or ?

by zreo2

Thursday January 31st, 2002 8:09 AM

Yes, I know what you mean. Damn that Opera browser. That's what you mean right ? ;)

#89 Feel free o fill in the blank

by PaulB

Thursday January 31st, 2002 4:09 PM

My original thoughts were that Mozilla certianly will not be released before it is ready. BTW I haven't used Opera; what is it like? I'm on the Mac and from what I have heard is that Opera is not ready for prime time on the Mac?

It would be nice to see a comparison (not necessarily a full blown review, but this would be nice) of all the modern browsers available on Windows, Mac and Linux. Why? It would be nice to get a rough benchmark as to how Mozilla compares to other browsers. There would at least be three main categories for comparison: 1. Feature comparison (How easy is it to manage bookmarks, manage mail, perform a web search, etc.) 2. Viewing Web pages (does the site render and function as intended? a. Does the browser render pages in a useable form? The browser need not be 100% W3C standards complient, but if the page renders well and the page is very useable without very obvious rendering mistakes (more than simply a couple of pixels out, but text omited or pictures or javascript not working) this is an acceptable tolerance range. b. Does the browser work with third party plugins such as Quicktime, Shockwave, Real, MediaPlayer, Java on all of the platforms for which the browser is written? 3. How does the browser performance feel when compared a. to other programs on the same platform? Do windows render slower, do fonts look better/ worse than other programs, etc. In other words doe the browser appear to be a well wrfitten program for the platform it is written for? b. to other browsers on the same platform (Mac browsers compared to Mac browsers, Linux to Linux.)

I am not seeking hard numbers, but these are nice. The end user's experience is really what counts. when comparing a browser, it would be nice to know if the browser appears inferior to other browsers available on that platform.

Why? I believe it would be nice to know what other browsers' use is like compared to Mozilla? Where does Mozilla shine? Where is Mozilla's tarnish, requiring more polish? I this is known then we will know wat to focus on for future Mozilla versions. Some tarnish might have to wait until post 1.0, but some could be focused on right now. Much of this triage has been done already, just not by focusing on the performace of other browsers to Mozilla in order to highlight areas of tarnish.

Any thoughts?

#94 You have a good point

by zreo2

Friday February 1st, 2002 12:55 AM

I agree but it's very hard to do that kind of comparison. Especially because alot of people only wants the browser, not the mailprogram or the composer. For those who are intrested in what browsers who supports CSS1, CSS2, ECMA-262 etc best there are alot of sites about that. For example: (old comapison though).

And I do also think it's hard to compare the rendering. I believe that's were both Mozilla:s strength and weekness lies. Because of the use of XUL and a layout engine (or whatever it is called). But that also makes the browser easy to add features etc. Just look at all the projects at!

If you look at or they are Mozilla based browser but can you really compare them to an ordinary Mozilla release ???

Well, I hope you get my points!

#95 Re: Native Widgets

by Tanyel

Friday February 1st, 2002 9:27 AM

I think Internet Explorer renders webpages the best overall. I think whatever browser is considered to be the most popular will be the one that renders pages the best. I think Opera renders pages the worst.

I think Opera renders webpages the fastest. I had to load webpages from a hard drive in order to prevent bandwidth problems from affecting my tests. I think Mozilla renders webpages the slowest but that may be due to differences in the way it handles unavailable Internet connections. The webpages are on the hard drive but the advertisements were unavailable because they have to be loaded from a server. Internet Explorer and Opera load the pages without interruption. Mozilla hangs until it times out and then finishes rendering the page. I think Opera would still be faster if this was not a problem.

I think Mozilla has the best features. It would be better if all of them worked. Ever since the tabs were added, I stopped using Opera more than Mozilla. Now Opera seems only good for curing boredom like a Rubik's Cube. If I acquire a Rubik's Cube then my use of Opera will probably decrease even more. Internet Explorer handles bookmarks the best. Mozilla handles them the worst. I use for searches so the search features of the browsers are irrelevant to me. My Java plugin works with Internet Explorer, Opera, and Netscape, but not Mozilla. When I type, it says, "Click here to get the plugin" even though the browser is set to enable Java and all of the files from the Netscape plugin folder are copied into the Mozilla plugin folder.

I use the Mozilla browser most but Internet Explorer is the only browser that works with every site I visit.

#96 Re: Re: Native Widgets

by johnlar

Friday February 1st, 2002 10:52 AM

This is due to the installer not recognizing mozilla, mozilla could go out and find the plugins from the netscape directory, (which it does for certain plugins) but it doesn't. Anyways if you want java to work in mozilla, here are the instructions. After you install the jre you should have a \program files\javasoft\bin directory. go there. Find all .dll files with the letters NS infront of them (NS meaning netscape plugins) and copy them to the mozillla/plugin directory. It will now work.

#102 Re: Native Widgets

by Tanyel

Sunday February 3rd, 2002 11:07 PM

I always copy every file from the Netscape plugin folder. The Java plugin is still not working. Sometimes the quicktime plugin does not work. I once found a build that handled all of the plugins...

#98 Re: Re: Native Widgets

by asa

Saturday February 2nd, 2002 2:26 AM

Tanyel, you can set google to be the default search engine then you can search directly from the URLbar. Or you can do what I do and create a bookmark with the URL give it the custom keyword \"g\" then typing \"g foobar\" and hitting enter will do a google search on \"foobar\".



by thegoldenear

Thursday January 31st, 2002 2:11 AM

#86 almost there :)

by wtmcgee

Thursday January 31st, 2002 10:38 AM

just a few more to go..

#90 [OT] Spellchecker

by zevious

Thursday January 31st, 2002 7:04 PM

Does anyone have a working link to download the spellchecker? The one on seems to be hosed.

#97 Re: [OT] Spellchecker

by jsgremlin

Friday February 1st, 2002 12:10 PM

email me (click my name) and I will email it to you.

#91 [OT] Spellchecker

by zevious

Thursday January 31st, 2002 7:10 PM

Does anyone have a working link to download the spellchecker? The one on seems to be hosed.

#93 [OT] Spellchecker

by zevious

Thursday January 31st, 2002 11:06 PM

Does anyone have a working link to download the spellchecker? The one on seems to be hosed.

#100 DHTML too slow

by pepejeria

Sunday February 3rd, 2002 7:44 AM

I been using Mozilla since m13 i think, and what bothers me the most is not sites that uses IE-only code, its the dhtml-speed. As a web developer I always code so its works in Mozila too. But its really depressing seing how slow it is. For example, try these site:

Most of the links there totally frezzes Mozilla.. sad..

#101 Re: DHTML too slow

by PaulB

Sunday February 3rd, 2002 6:42 PM


\"Most of the links there totally freezes Mozilla...\" HUH! None of the links freeze Mozilla running on Mac OS X. Since I have no other browser other than NS 4.7x on my computer, how do these examples perform using other browsers...are these examples which tesat the limits of DHTML or would most modern browsers run these examples slowly? How well does Opera or IE handle them? Can we have some more feedback as to how speedily other browsers handle these examples? Thanks.

#105 Re: Re: DHTML too slow

by Tanyel

Monday February 4th, 2002 12:20 PM

The ones I tried worked well with Internet Explorer. They caused Mozilla to crash. I did not try them with Opera or Netscape 4.8 because I assumed they would not work since nothing else does.

#104 Yes, it is

by dipa

Monday February 4th, 2002 12:57 AM

See for the dhtml performance tracking bug.

#103 Mozillazine

by Tanyel

Sunday February 3rd, 2002 11:09 PM

Are all of the links and the search form supposed to be under the messages, or is there something very wrong with build 2002012903?

#106 please ignore

by gemal

Saturday February 16th, 2002 2:45 PM

testing bug...

#107 test

by moztest

Saturday February 16th, 2002 2:53 PM


#108 test2

by moztest

Saturday February 16th, 2002 2:53 PM


#109 test3x3x23x123

by moztest

Saturday February 16th, 2002 2:54 PM