mozilla.org Releases Mozilla 0.9.8

asa writes: "Today mozilla.org released the Mozilla 0.9.8 Milestone. New to this release are improved Address Book functionality, page setup(for printing), MNG/JNG support, native-style widgets on winXP and OS X, dynamic theme switching, improved BiDi support, speed, stability and footprint improvements, and much, much more. www.mozilla.org and www.mozillazine.org have the full scoop." The build I'm posting with (2002020305) is a little crashy, but most aspects are shaping up very nicely.

Comments

Mozilla needs to focus on correctness, not feature

[vondo]

It seems to me that recently mozilla has been making less and less progress towards a really useable release. It seemed to make good progress up until 0.9.4 or so, but is now languishing.

Now, I use mozilla as my regular browser, and have since M18 (before Netscape 6.0), but lets face it, it's still very much alpha-quality software. There are so many little annoyances and things that don't work, I find myself constantly making excuses to my co-workers. 0.9.7 is, IMO, pretty weak with constant crashes and freezes.

The problem, in my opinion, is lack of good leadership. What this project needs is a nearly complete feature freeze, only allowing things already in the UI to be added and any features (and there are a lot of them) still missing that exist in Netscape 4.7.

As an example, look at the recent dust-up with favicons. They were put in, caused regressions in the code that weren't fixed for weeks, and never really worked very well. Now, they are mostly turned off by default, but in the process wasted at least some effort that would have been placed elsewhere. All this for a feature, that as far as I can tell is mostly eye-candy with very little, if any useability benefits to the user.

Now in 0.9.8, we have the ability to get a mapquest map of people in your address book. Is this really the critical kind of feature needed for 1.0? Is this really something mozilla.org wants to start taking bug requests on at this point?

Another example. Tabbed browsing is cool, but there are bugs there too that make it look less than professional. Besides which, I'd give all that up to get a decent printout (shortly before 0.9.8 branched, several very old linux printing bugs were re-targeted for 1.1 or 1.2), a text edit widget that worked perfectly, or to be able to compose mail with an editor that works.

In positive news, it looks like a spell checker might actually be included in 0.9.9. Yet another example, the Mail/News people made things much faster for 0.9.7 but at the expense of introducing more bugs. Threading was broken even more, messages fail to show up. Mozilla has never been as good as 4.7 in the mail/news client department, so this is a major problem. In my brief look at the 0.9.8 pre-releases, it looks like it might be even buggier now than it was in 0.9.7. Another step down, and it might become unuseable.

So, back to management, the drivers should reject any patch that adds a new feature as they push towards mozilla 1.0. Or encourage people to split off an unstable, development branch for feature addition. Maybe they agree with me about a lack of good management since they've brought on Peter Bojanic of OEOne to do project management. Of course, if you look at the mozilla 1.0 manifesto, they've been saying the right words for a long time now:

As we've said often, we're not looking for new features; we want stability, performance, best-available standards compliance, tolerably few bugs, and good APIs.

Features cost us time directly (opportunity costs born by those implementing the features, who likely could instead help fix 1.0 bugs) and indirectly (collateral costs on code reviewers, expert consultants, and other helpers). If you think you must have a feature by 1.0, please be prepared to say why to drivers, and be prepared to hear "we can't support work on that feature until after 1.0 has branched" in reply.

But, they've pretty much ignored this. Let's hope this time its better and they really mean it.

Before I finish, I'll address the two arguments people are most likely to make against my complaint:

1. Mozilla is an open source project, so you can't expect organized development. People are scratching an itch.
2. Mozilla isn't intended for end users, but as a base for companies to release a product

1. The majority, maybe the vast majority, of work on mozilla is still funded by Netscape and to a lesser extent other companies (RedHat, IBM, Sun). This should influence what bugs get fixed. Of course, this can't stop patches with lots of regressions from getting in if mozilla.org has as much autonomy as they say.

2. True, perhaps, but if the base has problems, its impossible or a waste of effort for several companies to run around fixing the same bugs. And then there are the linux distributors who will distribute mozilla as an end-user product.

So, I'm no longer as hopeful about mozilla's prospects as I once was. I hope I'm wrong, but I'm going to be waiting and trying mozilla 0.9.8 for myself before I install it for people on our systems.

Re:Here's your spellchecker

[CaptnMArk]

>and it would seriously slow Mozilla down, since >spawning a process is usually pretty slow.

bullshit!

you mean, spawning a "mozilla" process is pretty slow.

rendering improved?(after 9.4)

[Glanz]

Remember the Moz 9.4, where rendering was at least normal??? Since then thousand of geeks all over the world have been going blind keeping up with Mozilla "improvements"..... If thay can't get a SIMPLE thing like rendering right, they are welcomed to KEEP their glitchy browser.