Links Toolbar Landed

Wednesday October 3rd, 2001

Gervase Markham writes: "The Links Toolbar from bug 87428 has finally landed, bringing us ever-closer to full support for HTML 2.0. You'll see it in this morning's builds. The auto-show is still in development over in bug 102832. Good places to try it out are Bugzilla buglists, the W3C, and many machine-generated manuals or documents, such as the GNU Make manual."

#114 Re: Re: Re: Re: Cascading Bookmarks Possible?

by strauss

Saturday October 6th, 2001 11:39 AM

You are replying to this message

>> Third parties have released stable browsers based on Mozilla code. Public forums and news outlets have proclaimed that It Was So. <<

That's just not true. All the Netscape 6.x releases have been awful in terms of bugs and performance, and their bugginess has been noted in the press in every review I've seen. At best, lately we have been getting some remarks that's it's improved from previous versions (which it has), but not that it is actually good.

The instability of Netscape 6.x is well known both in the web developer community and among those end users who are aware of its existence at all. That's why less than 1% of web surfers have adopted it, even though it's been freely available for so long -- because of its well-deserved bad reputation for bugs and bloat.

Again, you may not like what I am saying, and I may not be doing anything to sugar-coat it for you, but it's not something I just made up. You really ought to know this already. If you don't, you should thank me for telling you.

With respect to Linux Mozilla packages, I'm not too concerned if they've gotten a positive response. Every single time I have been told by Linux users that a program was wonderful (GIMP, cvs, etc.) and I've gone through the multi-page setup instructions and tried it out, I've found it to be far below the standards of professional software. Linux users have very low critical standards and seem to be happy to get anything that just sort of works, as long as it's free. That's why Linux on the desktop has less than 1% of the non-technical end user installed base.

As for your penultimate paragraph, I appreciate your thoughts and the clarity with which you express them, but I think you're ignoring some basic facts about software projects. The reason that feature freeze is necessary to attain stability is not because of release schedules. It's because of the complexity issues in software systems. New features create new problems. As long as you are introducing features, you will be introducing problems. Because features interact with each other, the introduction of new features creates a multiplicative rather than an additive set of new problems -- that is, as the system size increases, the potential for problems with new features also increases, because they interact with more other features. To attain stability, you must reduce problems, not increase them. To reduce problems, you must hold the system size nearly steady. That's as true of incrementally released open source projects as it is of commercial software.

If you don't want to hear from me any more, I suggest you discuss the issue with Bruce instead. His simple, rational post seems to have been all but ignored in all this. If you get a good conversation going there, I'll gladly bow out.