0.9.2 Branch and Beyond
Sunday June 24th, 2001
Chris Blizzard has posted an update on current tree management plans for the 0.9.2 branch and the 0.9.3 trunk. The plan calls for mozilla.org to drop the requirement of firstname.lastname@example.org approval for check ins to the trunk, while continuing it till 0.9.2 is finished on the branch. Chris felt that using 0.9.2 as a stability milestone was a success, partly thanks to the drivers requirement, but mostly due to better self policing by those who were checking in.
I just wanted to say a big thanks to the drivers specially, because I guess I just can't imagine the amount of mails they recieved during those two weeks. They did a fabulous job, in my experience I never had to wait more than a day to get a response to an approval request. Here's hoping, like Blizzard said, that we can keep the tree in such a good state till 1.0!
#2 Netscape staying on the Trunk worked well too
by TonyG <email@example.com.Yuk>
Sunday June 24th, 2001 5:43 AM
Not only did we get a much cleaner NS buid, but I expect it also helped the tree as all hands were on deck.
Hopefully Netscape will drop back onto the trunk after 0.9.2.1 as well...
Are we to assume that, as the tree likely to lose stability when 0.9.3 opens that Netscape 6.1 will be based on 0.9.2? Or will the 'upcoming release' be PR2, with Netscape returning to the trunk for the final version.
Personally, I think Netscape should release 6.1 without any more preview releases.
#4 MozillaQuest Publishing Rubbish Again
Sunday June 24th, 2001 1:33 PM
Our good friend Michaelangelo over at MozillaQuest <http://www.mozillaquest.com/> has just published an article entitled "Mozilla 0.9.2 Branches on Schedule, but with Many Bugs" <http://www.mozillaquest.c…_branched-01_Story01.html> that says that 0.9.2 is likely to be the buggiest milestone for a long time. Funny, as Chris Blizzard says that up to the branch of 0.9.2 the tree managed to "maintain the same quality of the time right before the release of the 0.9.1 milestone." <http://www.mozillazine.or…articles/article1955.html> Now, who are we to believe, a classical Renaissance artist or a member of the <firstname.lastname@example.org> team?
The MozillaQuest article points to the number of bugs targeted at 0.9.2 ("280+") compared to those targeted at 0.9.1 when it branched ("some 20 to 30") and says that this means that 0.9.2 will be very buggy. This is wrong because Mozilla continues to get better, whether 20 bugs or 280 are fixed between milestones (yeah, I know there's regressions, but I'm speaking generally). Many of the bugs that were targeted at 0.9.2 were probably targeted before it was announced that 0.9.2 would be a 'stability milestone' and haven't been retargeted to 0.9.3 yet.
The only reason I ever go to MozillaQuest is to see what rubbish it's publishing. I'm beginning to wonder if this is a deliberate ploy on MozillaQuest's part to get more visitors.
#5 Re: MozillaQuest Publishing Rubbish Again
Sunday June 24th, 2001 2:01 PM
Do you know why Mozillaquest doesn't have a talkback area? I know why, when I read such bs.
Who cares what Mangelo thinks? Anyone who has used NS6.0, and has used a Mozilla build from the past two weeks KNOWS how much more stable recent Mozillas are. Certainly all the Mozillas I've tried (Windows9x, OS9, RH7; haven't tried Fizilla yet) are more stable than current betas of IE6 or even (indeed, especially) IE5.1 for OSX.
#6 Re: MozillaQuest Publishing Rubbish Again
Sunday June 24th, 2001 2:02 PM
"I'm beginning to wonder if this is a deliberate ploy on MozillaQuest's part to get more visitors."
I stopped wondering months ago. I'm still waiting for an "article" that provides any useful and accurate information about the mozilla project or product.
In the early days of the website I was willing to overlook the factual inaccuracies, the grammatical errors, the completly unappealing writing style and the <insert something negative here> design, etc. because a couple of the hunks of text he put up (chatzilla and skinning stuff) seemed well intentioned and I thought that it might encourage new folks to check out the project and maybe even get involved.
I think that what happened to change that was that his first negative article about the revisions to the roadmap got some slashdot coverage driving traffic to his webpages. The combination of that and some unexplained obsession with schedules and release dates (as if Mozilla was some commercial product that has marketing pressures pushing for incomplete releases) has driven his coverage of Mozilla into a frantic slamfest giving front page coverage to typos and paintshop artwork.
But then again, I'm a Mozilla advocate and my views are obviously colored by the fact that I like this project and I like working long hours every day to help make Mozilla better.
These opinions are my own and do not necessarily reflect those of mozilla.org (or any other mozilla contributor for that matter).
#8 Re: MozillaQuest Publishing Rubbish Again
Sunday June 24th, 2001 4:50 PM
I love how he's skewing numbers to represent something without actually looking at what the numbers mean. I love his writing style (or lack of it, maybe): attempting to sound professional while his page design speaks for itself (I know we shouldn't judge a book by its cover, but I believe I'll make an exception this one time). And even reporting on things that any logical/normal person can see for themselves. I wonder why all the Linux distros decided to hold off on N6.0x for the time being...
It's not a matter of holding off. Most Linux vendors are Free Software supporters, and included NS4x only because they are pragmatic and needed a decent graphical browser in 199x.
Today's distributions often include Mozilla (as an optional extra) and they will remove NS4x as soon as their local policies dictate that it's appropriate to do so. In Debian's case this has already happened.
For the commercial distributors the requirements are more complicated, stranding NS4x users would be commercial suicide so they can't remove NS4x until they have a seamless replacement.
Netscape 6.x is probably closer to achieving 4x parity than Mozilla but not by enough to outweigh the additional maintenance and political price of shipping yet another round of non-free software with the core OS.
So Red Hat and other Linux distributors will wait for Mozilla to reach an "acceptable" level of parity and then replace Netscape once and for all.
If Netscape planned to SELL version 6.x on Linux I'd tell them not to bother, but since they're giving it away anyhow it can't hurt any more than Opera or other non-free browsers on Linux.
#9 Re: MozillaQuest Publishing Rubbish Again
Sunday June 24th, 2001 7:04 PM
>>Now, who are we to believe, a classical Renaissance artist or a member of the <email@example.com> team?<<
I don't see that there is any reason to put this in ad hominem terms. One side of the discussion is giving particular numbers and figures, whil ethe other is putting its response in terms of insults. I can tell you which side I would tend to believe in such a situation.
Blizzard's claim of stability is not backed by any statement about such basic factors as the rate of new bug reports or ther number of regressions -- it is simply a flat assertion, unsupported by any analysis of the bug database. The hated MozillaQuest, on the other hand, simply points out that the number of deferred bugs has increased.
#11 Re: Re: MozillaQuest Publishing Rubbish Again
Monday June 25th, 2001 12:14 AM
If you want numbers then go to bugzilla (or someone who knows bugzilla). The bug fixing rate has been steadily increasing over the last 4 months. the number of bugs fixed in the last 3 weeks is well over 900. compare that to the 1600+ bugs fixed in the 6 weeks of the previous milestone and (taking into account the additional slowdown associated with a 3rd level of review on each bugfix, something new to this shortened milestone) even the most remedial in math can see that bug fixing is going up not down. Take a few minutes to look over the talkback, topcrash and mtbf (mean time between failure) statistics and notice that hundreds of crashers and scores of topcrasher bugs have been fixed in the last two milestones and see that mean time between failures has more than doubled in the last two milestones. mangelo's querying abilities and understanding of both the process of bug triage and fixing doesn't even approach remedial. His statistics are both meaningless and misleading. And if you think that he (or anyone else) has a better grasp of Bugzilla, the bug counts, the milestone triage process, talkback and mtbf stats, etc. than I do then feel free to post some numbers to refute my accounts of the progress in these areas that we've made in the last couple months.
Anyone who has been paying close attention to the process even as long as mangelo has been posting his mozilla commentary could probably give a more meaningful and accurate account of the bugfixing progress that we've made. mangelo only recently started posting stats (when I talked to him about three months ago he said he couldn't query bugzilla becuase it was too hard and it would "involve him with the project" and "compromise his journalistic integrity) and the first three attempts he's made at posting bug stats have all been so far off-base as to be laughable. (it's really not that hard to get useful numbers from Bugzilla. it should be especailly easy for a computer savvy member of the press like mangelo)
I'd like to thank you for pointing me out to some of those stats links. It's been pretty easy to modify it the way I want. The one thing I don't dare to do is do a new query of bugzilla. I just modify the links you gave. :) Thanks asa.
#12 Re: Re: MozillaQuest Publishing Rubbish Again
Monday June 25th, 2001 1:30 AM
"One side of the discussion is giving particular numbers and figures, whil ethe other is putting its response in terms of insults. I can tell you which side I would tend to believe in such a situation."
Strauss, do you feel qualified to analyse mangelo's numbers and make a determination about the accuracy and applicability of those numbers to his arguement? Or are you comfortable assuming that if someone adds a number to an arguement that the arguement is superior to one which does not have numbers? Would you find one arguement more credible if it simply had more numbers than the other? What if it had bigger numbers? What if it used fractions and ratios? Does credibility, believability or simple mathmatical ability come into the equation for you at all?
I'm going to assume that you really didn't mean that you were more likely to believe one arguement over the other because it cited numbers which are not only unverified but easily refuted by any person who was willing to spend part of a day learning to query Bugzilla. I'm going to assume that you were making the more general statement that give two competing accounts, you're more willing to believe the one that contains credible statistics than the one that contains personal attacks. I'll also assume that you have (up until now) no reason to doubt mangelo's credibility, or ability when it comes to querying Bugzilla and making sense out of the results. I'd like to take this oportunity, then, to point out that mangelo lacks the ability to query bugzilla and generate any meaning from the results.
As recently as June 12th mangelo was posting commentary with such entertaining statements as "However, on 9 June 2001, which is when it appears that Netscape 6.1 PR1 was branched from the Mozilla development tree, there were some 2800 bugs in Mozilla. That's too many bugs for a commercial, end-user, product." Not only was he factually wrong on the date that Netscape "branched from Mozilla development" he was wrong about their branching at all. Anyone with a simple understanding of bonsai queries could have queried for the branch containing regexp _9_1 and told you that there was no netscape branch and that the netscape release was taken from the exact same spot on the mozilla branch (6/7 not 6/9) as Mozilla 0.9.1. But even more inaccurate, to the point of being ludicrous, was his claim (I assume he made based on a clear misunderstanding of chofmann's bug counts posted to the newsgroups) that there were "2800 bugs in Mozilla" (I wish. I mean come on. How can anyone be that far off). He simply has no understanding of how to retrieve useful numbers or what they mean.
I'm not slamming mangelo because he is anti-Mozilla. I don't think he is (he's jsut click-through hungry). I'm slamming him because he is woefully incapable of writing anything meaningful when the information is all readily available, up to date in realtime with easy web-accessible query tools. It's like someone handed him a deck of playing cards, telling him to report on the different cards in the deck and when he did he left out three of the suits and added 2 of his own. It's just laughable.
Strauss, I am an expert when it comes to using Bugzilla. I'm comfortable stating that there are very few people who understand as much or more than I do how to mine useful information from that database. I can promise you that mangelo is not one of those people. If mangelo was reporting on bugs in the linux kernel and linus torvalds told you that mangelo's numbers were wrong who would you believe?
#24 Re: Re: Re: MozillaQuest Publishing Rubbish Again
Tuesday June 26th, 2001 2:33 AM
Just curious, it seems like mangelo was at least correct in his assessment of targetted 0.9.2 bugs (looks like there are about 170 bugs as of Monday night). What's the reason for branching with a comparatively large number of bugs (if it even is large compared to previous releases)? Does it even have any consequence other than that it might take a few extra days to close that many bugs? I assume that all the bugs targetted for 0.9.2 will get closed before release or get retargetted.
Anyway, mangelo generally gets thing wrong, but I thought I'd see he had actually noticed something noteworthy in his observation that the branching is happening relatively early.
#28 0.9.2 Milestone plans changed
Tuesday June 26th, 2001 8:54 AM
The milestone plans for 0.9.2 changed as soon as 0.9.1 branched. See <http://www.mozillazine.or…articles/article1938.html> for more information.
Basically the bug list was left alone but there was less time to work through it, plus there was more QA on the bug fixes that did go in. This is not news, the thing that he didn't understand is that the 0.9.2 bug list did not receive a complete triage after the policy change so there were a lot of bugs targetted for 0.9.2 that had little chance of getting fixed. Also, most of these bugs had been in there for a long time - it's not like they are _new_ bugs, they are just bugs that the team had hoped to fix for 0.9.2 but didn't have time to do so. This also appears to be something he doesn't understand.
#29 Re: 0.9.2 Milestone plans changed
Tuesday June 26th, 2001 10:09 AM
>>they are just bugs that the team had hoped to fix for 0.9.2 but didn't have time to do so. This also appears to be something he doesn't understand.<<
I'm not sure what it is you think he doesn't understand. He observed that the build has fallen short on its bug fixing expectations. You have now confirmed that.
Through all the personal attacks here, and claims of ignorance, inexperience and stupidity, there seems to be little attention given to the facts of the case.
#30 Is this intended as a troll?
Tuesday June 26th, 2001 10:17 AM
The development cycle was shortened by something like 2 weeks, with additional QA, but the bug list was NOT shortened when that was done. Is it any surprise that they didn't manage to get as many of the bugs fixed as they had hoped to before the cycle was shortened?
Is this intended as a deliberate troll? It has every appearance of being such.
#33 Re: Is this intended as a troll?
Tuesday June 26th, 2001 1:07 PM
I don't see your point.
According to you, the build cycle was shortened by two weeks, meaning that many of the bugs slated for fixing in that round were not fixed. This appears to confirm the statement on MozillaQuest that the build would not meet its quality goals.
So what is it exactly that you're disagreeing with? It seems to me that you are agreeing with him on the facts and that the only disagreement is one of personality.
#34 The point is the goals changed!!
Tuesday June 26th, 2001 1:52 PM
The point is that the goals changed. If you want to discuss this intelligently it would be in your best interest to read some of the references at <http://www.mozilla.org> <http://www.mozillazine.org> and also <http://www.mozillaquest.com> - it doesn't sound like you have, so you're shooting from the hip so to speak.
The 0.9.2 milestone was changed from being primarily a "development" milestone to being a "stability" milestone. If you don't understand what this kind of change means in software development parlance, here's a summary:
In a development milestone, the developers are trying to add features and fix problems whose solutions might be more risky. For example, a "bug" might be that a particular part of the program is slow or inefficient but works correctly in the sense that you get the right answer. So you have a higher chance of regressions, that is, introducing a lot of NEW bugs since you're likely to be adding a lot of new code.
In a "stability" milestone, the concentration is getting the product to work reliably - even if there may be some inefficiencies or cosmetic problems or a few missing features. Bug fixes that involve a lot of risk of introducing new bugs will be discouraged unless the bug that's being fixed is fairly serious or the feature being added is considered essential.
Note that all bugs are NOT created equal, and shouldn't be treated as being of equal importance. Talking about raw bug counts does exactly that.
If you really want to discuss software development issues intelligently, some nearly essential background reading for you would be "The Mythical Man-Month" by Fred Brooks, also useful would be "Code Complete" by Steve McConnell and "Rapid Development" by the same author. There are many others. If you haven't read at least the first one, and preferably at least some others of the genre, it is like you are trying to fight a war when you are unarmed.
#35 Re: The point is the goals changed!!
Tuesday June 26th, 2001 3:14 PM
I'm not sure why you continue to resort to ad hominem. Just so you'll know, my first training in software engineering was directly from Dr. Brooks, about twenty years ago. I am a software manager, I have often acted as a project manager, and I have managed project managers, as well as working in close conjunction with QA leads and managers for many years. I'm tempted to quiz on you on some basic software process issues, but I'm trying to avoid desending to the personal level that has dominated this discussion so far.
I don't see the relevance of your observation. Stability milestones are for bug fixing. You can't allow in high-risk fixes during a stability milestone, but you also don't cut it so short that you can't reach your bug fixing goals. In fact, you're not even saying that those bugs were locked out because they were high risk (which would make sense) -- you're saying they were just dropped due to a changed deadline. Saying that it was fine to ship with scaled back bug-fixing goals because it was a stability milestone is very nearly an oxymoron.
#39 Re: Re: The point is the goals changed!!
Tuesday June 26th, 2001 4:21 PM
Sigh. I have been in software development for over 20 years. I have been the programming lead on a major software product that has been praised as being "the most stable of any application of its type, on any platform." No, I didn't take any classes from Dr. Brooks (my degree is from another school not far from there), but he was a well-known figure in the area and we often went to his colloquia. I'm hardly a newbie either.
If you have the background you say you do, then you would understand that when you lump every possible change under the rubric "bug" (as bugzilla tends to do), that you can't just claim that all of them are of equal importance, especially with regards to "stability" rather than "enhancements." And when I point out that you don't seem to be making any such distinctions, I'm not trying to make an ad hominem argumeent.
The obvious implication of having 0.9.2 being a "stability" milestone is that that was the point that a major vendor (Netscape) was planning to use as a branch point for their release. It would make a lot of sense for them to continue to make "stability" fixes beyond that, in their branch.
I do not feel that I am qualified to make a judgement on whether mozilla is down to a reasonable bug count for the stated purposes of this milestone - it would take far too much time for me to make such an assessment, and as I said I am not actively involved with it; just an interested bystander.
But I do know that raw bug counts that include everything from crashers to spelling errors are a highly unreliable statistic.
#46 Re: Re: The point is the goals changed!!
Wednesday June 27th, 2001 1:37 AM
I tell you that I want you to paint 6 rooms in my house and you have 6 days to complete the work. You agree to paint 6 rooms in my house. I then tell you that I only want 3 rooms painted and I need them painted in 1 day. You agree to work harder and get half the original work completed in 1/6 the original time. You don't update your bookkeeping because you're busy painting rooms. You get the three rooms painted and my neighbor comes over and says that you're a failure and I'm a failure for not getting the 6 rooms painted. I tell my neighbor that he is wrong and that I changed my plans about the other 3 rooms which are going to get wallpaper and not until next week. He continues to call you and me failures for not getting the other 3 rooms painted. My interior decorator tells the neighbor that it wasnt a shortcoming, it was a change in planning and that it actually meant that the rooms were going to all be renovated faster but the neighbor continues to call you and me failures. You and ask the neighbor to please bugger off.
See, we actually fixed more bugs per day this milestone than we did the last milestone. We were fixing bugs so fast that we didn't push bugs out to the next milestone (like we do every milestone) in daily triage of the buglists. We were so busy fixing bugs that we didn't do the bugzilla maintainence (bookeeping) we usually do until we got ready to branch. You have a fundamental misconception or misunderstanding of how we (Mozilla) manage our buglists. You're free to criticize our process but please learn what that process is before you start that criticism. If you really understood our process you would be impressed what we accomplished in the last 3 weeks.
#36 Re: The point is the goals changed!!
Tuesday June 26th, 2001 3:26 PM
Bruce, let's leave mister the software manager alone. I don't think we will ever reach agreement so we might as well stop right now. Apart from one or two people in here, we can all read Blizzard's post about the 0.9.2 cycle correctly.
#37 Re: The point is the goals changed!!
Tuesday June 26th, 2001 3:58 PM
One more note: It appears to me what is being implemented is not a stability milestone, but a release milestone. At a release milestone, code is frozen. All bug fixes are locked out except for those required to ship the product, because even apparently low-risk fixes might turn out to cause problems.
The reason for 0.9.2 being redefined as a release milestone is because Netscape plans to push out yet another premature browser version based on this still-not-quite-ready-for-prime-time code base. (No criticism implied -- it may be ready for prime time in a little while, it's just not there now.)
Given that we all know Netscape has seriously damaged the Mozilla effort through these premature releases, to the point where for a while all Netscape 6 articles were banned on this site, I'm surprised to see people defending this practice here. Mozilla needs to be going ahead full tilt to fix as many of the remaining bugs as possible. This artificial code freeze to support another bad Netscape 6.x release is not in Mozilla's best interests.
#42 Re: Re: The point is the goals changed!!
Tuesday June 26th, 2001 5:21 PM
Okay this is all very simple, academic, and a somewhat stupid argument. So let's try to explain why Mangelo is wrong simply.
1) Before 0.9.1 there were bugs in Mozilla targetted for 0.9.1, 0.9.2, 0.9.3, 1.0 and future. 2) Before 0.9.2 there are bugs in Mozilla targetted for 0.9.2, 0.9.3, 1.0 and future. 3) After 0.9.2 there will be bugs in Mozilla targetted for 0.9.3, (0.9.4, probably), 1.0 and future.
Look! Between 1 and 2 and between 2 and 3 a whole set of bugs have been fixed. Why? Because when considering how many bugs a milestone has we don't look at the bugs targetted for that milestone, we look at the bugs targetted for that milestone *and beyond*.
Sure, that gives 0.9.2 *MORE* bugs that Mangelo suggests there are, but it also shows that there are fewer bugs in the milestone than the one before.
#45 Not exactly a release yet...
Tuesday June 26th, 2001 11:17 PM
Not to harp on this too much, but if you've been following the recent developments closely, you might have noticed that it appears that Netscape will release 6.1 based on a code base that will be called 0.9.2.1, which will be a branch that will start with 0.9.2 and include stability fixes beyond that. This kind of branch is fairly typical if you want to prepare a commercial release while continuing active feature-oriented development on a product in a different branch.
I am not privy to how long a cycle they are planning for 0.9.2.1, nor what is targetted for it. That code point is clearly the point that ought to be called the "release milestone," not 0.9.2. Personally, looking at the outstanding bugs in bugzilla I find the number of stability-related bugs to be worrysome, but as I've noted elsewhere I don't have enough in-depth familiarity with the project to feel qualified to register much of an opinion on how important they really are or how difficult they will be to fix. Some of them have been targetted for 1.0 or even beyond, so for at least those bugs there seems to be a feeling that they can be put off for now.
As for whether this is a good idea, I'm not sure. It's certainly reasonable to try to get something better than Netscape 6.0 out there even if it's not what you want in the long run, especially if Mozilla 1.0 might still be several months away (as appears likely). On the other hand it does take away resources that could be applied towards Mozilla 1.0 and probably pushes that release back somewhat. Of course many of the bugs that will be fixed for Netscape 6.1 could presumably have their fixes retrofitted into the trunk, and it's not like they aren't issues that would have to be fixed anyway in order to release Mozilla 1.0.
Frankly, I don't think my articles have been defending Mozilla so much as they have been hammering MozillaQuest - but if you're going to present yourself as a journalist, it behooves you to try to have at least a little bit of understanding of what it is that you're writing about, and you invite others to judge whether you have done so. If it's politically incorrect and elitist to point that out then I plead guilty on both counts. :^)
#47 Re: Re: The point is the goals changed!!
Wednesday June 27th, 2001 1:44 AM
This was not a release milestone any more than every milestone is a release milestone (do you consider mozilla milestones like 0.9.1 to be 'releases'?). This was a "lets continue recovering from a major destabilizing few milesstones we've had recently (you remember, right? the ones where we landed major rewrites of just about everything. 0.8, 0.9.) miletone. <firstname.lastname@example.org> spent 3 weeks moderating checkins and working to keep risk at a minimum so that we could return to a point of stability sufficient to get a good idea of where were were on the road to 1.0.
#31 Re: Re: 0.9.2 Milestone plans changed
Tuesday June 26th, 2001 12:30 PM
just to reiterate one more time, just in case :-)
0.9.2 was shortened by two weeks. The people who triaged bugs for 0.9.2 at the begining of the cycle didn't expect two less weeks to fix them. This resulted in 280 bugs at the time of branching, but as Blake and Hixie noted in probably the most passionate argument ever in #mozillazine, the number of "bugs blocking 0.9.2" was only of 14. (they have "critical for 0.9.2" in the status whiteboard). Would the milestone be less buggy if we pushed all the 0.9.2 bugs to 0.9.3? Of course not, but M. Angelo would have said "this will probably be a very good milestone because there are only two bugs targetted at 0.9.2 at the time of branching".
This is nonsense and should be condemned.
#18 Misunderstanding of software development process
Monday June 25th, 2001 1:12 PM
The problem with the kind of "analysis" used by MozillaQuest is that it misunderstands the software development process. Many of these bugs that were targetted for the 0.9.2 milestone have in fact been in Mozilla for a long time, as can be easily verified with bugzilla. So the previous milestones had them as well, but just left them unfixed. This hardly means that the new milestone is "more buggy" than the previous one, simply that the team had hoped to fix more of the outstanding bugs for the milestone than they may actually accomplish.
This is quite typical in software development - you target the most critical and/or most visible things first. It's not at all unusual to ship something with "known problems" - it's typical. At some point you have to decide that the existing bugs are of a nature that they do not seriously impair "typical" use of the product and that holding the product back in order to fix them will either be likely to introduce more (and possibly worse) bugs into the product, or seriously delay the product introduction. Obviously you need to get more bugs out with a "final" release than you do with a "beta" but the point is the same. If you think there is a single software development house on the face of this earth that operates any other way then you are living in a dream world.
Unfortunately this kind of statistic has the (false) ring of truth, like Mark Twain's dictum of there being "lies, damn lies, and statistics." If you don't understand how software teams have to triage bugs then you are likely to totally miss the point, as the MozillaQuest author does and as he invites his readers to do.
And this doesn't even address the point that in Bugzilla (as in many -- possibly even most -- bug tracking systems), "enhancements" are lumped in there with "bugs," as are misspelled words in the dialogs and all sorts of other things. In this kind of system, you even have to be careful about making incautious conclusions based on the number of unresolved bugs (not just the number targetted for a particular milestone) -- over time, you can always think of enhancements faster than you can add them, and often you get bugs added into the system that are later found to be problems restricted to earlier releases or to misconfigured software or malfunctioning hardware. This is especially true if your number of users is going up. One example of a more reliable statistic that Asa mentioned is Mean Time Between Failure (MTBF).
Unfortunately most of this kind of distinction will sail over the heads of the general public, who are not familiar with the workings of software development efforts.
#19 Re: Misunderstanding of software development proce
Monday June 25th, 2001 4:30 PM
"Unfortunately this kind of statistic has the (false) ring of truth, like Mark Twain's dictum of there being 'lies, damn lies, and statistics.'"
Minor point, but it was actually Benjamin Disraeli who said that quote, though it was attributed to him by Mark Twain. :-)
#49 Re: Re: MozillaQuest Publishing Rubbish Again
Friday June 29th, 2001 6:40 AM
I don't care what statistics back 'Blizzard's claim of stability'. We don't need statistics to back it up (though asa provided them) - we have a *browser*.
I just convinced a colleague at work to check out Mozilla again today after ditching NS 6.0 for IE because it was just too slow and buggy.
I didn't do that because I'm some totally one-sided, unobjective Mozilla advocate who just wants to push the browser. (If you've seen my posts here, you'll realise that I am absolutely willing to criticise or point out problems in Mozilla, as I believe that pretending things are better than they are is completely unhelpful.)
I did it because Mozilla absolutely and demonstrably *is* better than it has been recently (and FAR better than Netscape's 6.0 release). It's very reliable (I browse a lot and it hasn't crashed in several weeks, before that it was crashing every few days), it's fast, it generally works nicely. There are a few bugs in this particular nightly build I'm using, but none of them seem like the kind of bugs that will lurk for weeks and months.
Basically the project is going nicely. There are things that suck still, but relatively few of them, and for quite some time (even when it was buggy) Mozilla has been my preferred browser not because I just want to use something that supports standards even though it's actually worse than IE to use, but because it's actually nice to use as well.
#13 Re: MozillaQuest Publishing Rubbish Again
by Brendon <email@example.com>
Monday June 25th, 2001 1:39 AM
Stupid question perhaps. Have you sent a warning of some kind to large news sites that link to MozillaQuest's articles, i.e. newsforge.org?
I doubt MozillaQuest itself gets much traffic.
#21 Re: MozillaQuest Publishing Rubbish Again
Monday June 25th, 2001 8:38 PM
Have you noticed the homepage's title? "MozillaQuest Magazine Front Page. GREAT, INFORMATIVE Articles & Info About Computer Software, Hardware, Linux, Windows, Netscape, Mozilla, browsers, dual boot download, tutorial, install help, open source, & lots more." Great and informative, huh?
(Disclaimer: The author of this comment is expressing his opinion (and disapproval of the MozillaQuest site) and no one else's, although others may share the same opinion.)
#10 Thanks for keeping us informed!!
Sunday June 24th, 2001 8:05 PM
I really appreciate articles like this. It's nice to know what's happening with the development of my favorite browser.
I say F**k MozillaQuest and lets get on to our lives again.... I find it depressing to see 21 POSTS for the article, as aposed to 7 the other day, and them to all be under the same topic... Doesn't anyone have a life to go on with.. bugs to kill, milestones to test... I for one do, and am having a blast doing so!
Anyhow, MozillaQuest SUCKS, and we all know that!
back to the topic: release number 0.9.2 sounds like it going to rock.. and 0.9.2.1 even better. Netscape has always found neet ways to polish the mozilla code.
Only thing I'm worried about, is the eagerness to NOT allow patched that could "damage" or lower the quality of the code... I understand the doing so, but many regresions are needed to get things right. But hell, I'm not complaing, the've done an awesome Job.. and ASA.. you rock man.. keep up the good work
#23 Re: Is anyone esle tired of this?
by gerbilpower <firstname.lastname@example.org>
Tuesday June 26th, 2001 1:20 AM
I'm tired of it too but many of us, myself included, are worried that people who don't know better (COUGHslashdotCOUGH) and those who could be potential contributors, bug reporters, etc, be getting the wrong idea about Mozilla and be scared off by MozillaQuest's inaccuracies.
But back to Mozilla, to move us to more important news: there is a discussion going on what the Mozilla 1.0 criteria should be (not specific bug numbers, just criteria that would help determine what bugs should be fixed for 1.0). Interesting stuff has already been said and more feedback would help. The NewZilla Alex <http://www.gerbilbox.com/newzilla/>
#26 How believable is MozillaQuest?
Tuesday June 26th, 2001 7:57 AM
I'm not actively involved in the Mozilla effort (other than running a number of the nightly builds). I keep track of it because of its possible impact (hopefully for the better) on various business projects I'm involved with. I also keep track of what's being done on it via this site and mozilla.org, including bugzilla.
However even if I knew nothing specific about Mozilla, I would find it difficult to take MozillaQuest seriously. It is blatantly obvious that he has never been close to any major software development project of any description (toy projects he might have done in school don't count). He tends to dwell on any kind of typographical error that he finds anywhere to the point of tedium. He seems to expect that every page he finds anywhere on the net in general or mozilla.org in particular would be updated at least daily -- highly unrealistic for most sites that aren't primarily in the business of providing news. And to top it off he appears to have a grandiose view of his own importance but the writing style of a junior high school student.
Frankly I don't find his site credible on its face, even without any other information. I'm sure I'm not alone, at least among individuals who actually know something about software development. Compare his news site to ones that are run by people who actually do have a clue, like NewsForge (<http://www.newsforge.com>) or The Register (<http://www.theregister.co.uk>) and the difference is obvious.
Unfortunately, judging from what I've seen, not everyone seems to be very sensitive to these things. A number of people seem to fall into the old print fallacy that if it's in print it must be true. P. T. Barnum was right, unfortunately.
#27 Corrected links....
Tuesday June 26th, 2001 8:24 AM
#48 Re: Is anyone esle tired of this?
Wednesday June 27th, 2001 1:57 AM
Jedbro, the trunk (as opposed to the 0.9.2 branch) is no longer under the additional driver approval guidelines. It is our hope that folks will use care and work to maintain the stability we've gained in the last couple milestones. There are lots of bugs which can only be fixed with major changes. We hope that these changes will be well tested (on experimental branches with binaries provided to the QA community for testing prior to landing) and well reviewed. We don't expect the kinds of major wholesale rewrites like we saw in 0.8 and 0.9 cycles so we should be able to get some good bugfixes in and react quicly to any regressions or destabilization.
I havn\\\\\\\\\\\\\\\'t have time to keep up on mozilla lately but I do sometime download nightly here and there to try it out. I found .9.1 to be very stable and useful. I useing it right now and it render all the pages I visit! still some bugs thought. I can\\\\\\\\\\\\\\\'t wait until the Release@
Tuesday June 26th, 2001 12:52 PM
Is there any setting that will cause the browser window to be in the same position each time it is loaded? Whenever I shut down the browser and reload it, I find it in a different place.
At least MozillaQuest offers some criticism of the Mozilla project. This site does nothing but praise a project that is years late and one of the buggiest pieces of software on the internet. Mozilla did more to kill it's own browser than MS ever could. If you want a browser that does everything Mozilla promised to do and isn't IE, download Opera 5.
The whole problem is that Mozilla IS NOT the browser. It's the development platform - like IE but crossplatform and more promissing. Although majority of people still think about them as browsers.
On the other hand - Opera is probably the best browser. But have you ever tried to develop any XML+DOM application for it? Impossible.
#43 Re: MozillaQuest
by gerbilpower <email@example.com>
Tuesday June 26th, 2001 6:35 PM
The complaints about MozillaQuest aren't because they are negative, but because there are based on inaccurate information and false assumptions.
Frankly this site is slanted towards praise but there has been a good amount of (constructive) criticism. For better discussion I've found that the newsgroups to be much better for that than here.
#44 Re: Re: MozillaQuest
Tuesday June 26th, 2001 8:51 PM
There are a number of things that concern me about where Mozilla is; for one thing, it is very late to market (yes I know the technical reasons given, that still doesn't change the fact that the market won't stand still for you).
For another, there are still quite a number of bugs in bugzilla that contain the dreaded word "crash" in their description. It is difficult for an outsider not intimately familiar with the project to determine what the exposure is for many of these bugs. For example, how many only occur with the embedded product, or only under extremely rare circumstances or only on specific platforms? How many can be reproduced in the current build? If they can't be, it might mean that the bug has been fixed by another correction or only under certain difficult to reproduce conditions.
It is obvious that Mozilla has made huge progress in the last 6 months. It's also obvious that there's a lot that still needs to be done.
#41 That's why this was Netscape 6.1, not Netscape 6.5
by Byter <firstname.lastname@example.org>
Tuesday June 26th, 2001 5:07 PM
Anything is better than leaving an extremely buggy and unfinished browser out there as Netscape 6.01. Netscape 6.5 should be based on Mozilla 1.0, but a usable replacement needs to be made for the jokes that were Netscape 6 and Netscape 6.01, and Netscape 6.1 should serve that purpose.