MozillaZine

Mozilla 0.9.6 Released

Tuesday November 20th, 2001

mozilla.org today made available for download binaries of the Mozilla 0.9.6 Milestone. The builds are available on the releases page, or you can get them directly from the ftp site.

New to this milestone are fixes for about 1,600 bugs including support for site icons in both the url bar and tabs (expect IE's favicons to show up in 0.9.7), displaying both Windows Bitmap (.bmp) and Windows Icon (.ico) files inline on all platforms, a new print preview implementation, Page Setup improvements on the Macintosh, Mail message 'labels' (Correction: This feature is not fully complete, only parts of the backed have landed.), and a new select and search context menu item, among others. If you are interested in more information on any of these new features, be sure to check out the release notes.

Also, mozilla.org last week released the source code on which Netscape 6.2 was based.


#161 Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: What happe

by bcwright <bcwright@ix.netcom.com>

Friday November 23rd, 2001 1:15 PM

You are replying to this message

Asa,

I can't speak for Strauss, but what I'd call the "stability situation" (for want of a better term) isn't really how often the product crashes or the like. I think we can all agree that Mozilla crashes a lot less often than it once did, and is a whole lot better than it once was. Rather what is most worrisome is how often lots of new problems get introduced (or old problems get re-introduced), especially when adding new features. Code stability, in other words, rather than how often the product crashes.

The issue about what bugs are targetted for what milestone is more of an issue of understanding where you are than it is of whether "enough" bugs are being fixed for each milestone. I would think that the current situation would make your lives more difficult even apart from whether it makes it easier for Mangelo to track his bug counts :-). You even allude to this in your own reply.

The other issue is the number of bugs targetted for all releases between now and 1.0. Obviously that will have to be a generally decreasing function of time or you'll never get there except by dumb luck. Until recently this had not been the case. Now I don't know enough about the details of why this is so - how many of these amount to desired enhancements that got added rather than actual errors that were found - but it's clear that for quite a while the situation was somewhat out of control in the sense that these numbers kept going up. After all, the entries targetted for 1.0 (which is supposed to be an actual _release_) matter a lot more than those targetted to just a single _milestone_. My impression is that a lot of them amounted to enhancement ideas that were contributed, which seems to indicate too much of a reluctance to say No. However if most of them were errors instead, then clearly it indicates a situation where either code changes were not being tested and reviewed enough, or where the bug reports were not being reviewed fast enough to figure out which were duplicates and user errors and the like rather than program bugs. Neither situation is good, at least if you ever want to get to an actual 1.0 release.

One way or another you have to get a handle on that figure if you're ever going to get to 1.0. Fortunately it's finally been moving in the right direction over about the last month, so there is now some hope that you'll get there. But it's still going to require some discipline and self-restraint.

I do think Mozilla has gotten a lot better, I certainly won't dispute that. What I think it needs now is more attention to fixing the remaining "errors" (of whatever description, not just "crashers") and not so much on new "features" - really nothing except extremely low-risk or essential missing features. The tabbed browsing and link features are examples of things that were, in my opinion, very marginal and risky additions at this stage. Not that they aren't nice, but this kind of churning makes it harder to converge; and neither one is essential (they are in neither 4.x nor IE). Things like this can always be added for 1.1.

Looking forward hoping for a great 1.0.

--Bruce