Kids today.

One of my Lucid Emacs co-authors has been banned from reporting any more bugs against Mozilla -- for being too darned grumpy.

It's like they just don't speak the language of my people any more.

Tags: , , , ,

120 Responses:

  1. What the...? That bugreport seemed totally reasonable to me, and, considering the subject matter of randomly popping up shit, rather restrained...

    • ix says:

      It's actually quite agressive. It scares off casual contributors. It's not very polite. All good reasons to disable his account.

      I know the default in many open source projects is communicating in this way and there's nothing inherently wrong with that, but frankly I prefer to stay away from people who do so, as I find it somewhat grates on me.

      Context is important as well. A blog post or general article like the bug report above I'd find funny and if I was one of the developers I'd take it as is. The same post on my bug tracker, much less so. This is something people pour their heart and soul into, if you're gonna tell them it's shit to their face, at least be polite about it.

      • Lyle says:

        This is race to the bottom mentality. I don't know when it became acceptable to shaft longstanding members of a development community in favor of "casual" (aka: sporadic) contributors, but it sets an alarming precedent.

        • ix says:

          It' s actually pretty democratic. Oh wait, that's what you said. My mistake. :-)

          • nobody says:

            Does anybody else find the "oh wait" style of response to be totally infuriating? It doesn't matter whether the person is right or not, it immediately marks them as a complete douchebag who's more interested in scoring Internet Argument Points than they are in actually, you know, having a worthwhile discussion.

        • Blake Winton says:

          Who's the longstanding member of the development community you're referring to? I checked the hg logs (which admittedly only go back to July 22nd, 2008), but I didn't see any patches from Richard…

          Or did you mean that Mozilla shouldn't shaft the people actually trying to fix the bugs by letting Richard continue to sporadically and casually attack them?

      • DFB says:

        Tell me more about why this would scare casual contributors. I think the opposite is more likely.

        I agree it could have been more polite, but the emphasis doesn't seem to be chosen poorly here. The default for the modal dialog is strongly biased in favor of preventing the recovery of accidentally deleted messages, which is potentially a very serious mistake.

        • ix says:

          If I were a mozilla dev I'd feel personally attacked every time I'd have to read a bug report like that. As a thought experiment, imagine someone accosting you at a conference and starting a rant like that. I don't think that would make you all fuzzy and warm inside. I don't see how that would actually attract contributors, but maybe I'm missing some bit of human psychology here.

          It's incidentally also not a very efficient way of communicating. All the ranting detracts from the message it should be giving. Anyway, it's not the only bug report like that.

          • CTD says:

            But in this case, that would actually be a good effect. Developers who added annoying pop-ups about internal operations, as well as developers who compact folders whether you want that to happen or not, are exactly the sort of people you want to discourage from contributing.

          • Maybe you should get out of the business, and/or get less precious about your stuff. Any dev worth his salt already knows everything he/she creates is buggy unusable shit, and should be nodding right along to reports like this, and thanking them for pointing out such an awful mistake. Their continuing mission: make it less buggy and shitty, the mission that never ends. You don't get good software from people whose feelings are so easily hurt.

            • ix says:

              I know my software is buggy unusable shit (and am not particularly touchy about it), I just don't think that means I need to get shouted at all the time. It's macho bullshit that doesn't help you write better software, IMHO.

              You do not agree, but all that means is I will continue to seek out projects where that kind of behaviour is discouraged, while others will gravitate to projects where it is still the norm (most GNU stuff, I'm told). The two can exist side by side, it's just hard to move between them because you are used to the rules of one community (be as abrasive as you can be) and then move into another (mozilla devs probably put their rants on their blogs).

          • Iain Adams says:

            You're completely missing the point. The reporter feels the software is being far ruder than his bug report. People who write code that behaves like that need to see how it feels to be on the other end of their UI otherwise they'll never learn. If they can't learn, they shouldn't be writing code.

      • Chris says:

        I would murder five users who send in marginally useful (yet ultra-polite) bug reports to get one user sending in quality bug reports like those.

        If supporting "casual contributors" has gotten the mozilla projects to where they are today, they might consider changing tactics. Cf OpenBSD.

  2. Peter Hollo says:

    I think he's actually been banned for Speaking The Truth.
    Bah humbug.

  3. Ryan Beasley says:

    Huh. Account disabled by Mozilla Messaging CTO who was with Netscape in the 90s. Don't know the guy, so I'm not slamming him, but I would have expected him to tolerate much, *much* worse.

  4. Lloyd says:

    At least he's not trying to report bugs in The GIMP.

    (What is it with the holier-than-thou attitude of all gnome developers, btw?)

    • Nick Lamb says:

      "I couldn't find out why the code management toolset was changed, but I imagine fashion has an awful lot to do with it."

      This is very amusing coming from someone who claims to be a software engineer. Hmm, what possible reason could a large distributed software project have for upgrading to one of the game-changing distributed version control systems? Must be fashion.

      You're right though, despite his youth Lloyd is a prototypical grumpy old man of the sort JWZ is talking about. Nothing has ever been right and if it was he'd be unhappy about it. I imagine he was already shaking his fist and talking about "damn kids" back in the 1990s when he could still pass for an undergraduate.

      • Lloyd says:

        The GIMP development was upgraded from cvs to svn -- which is basically cvs completely and deliberately redone, with many of the same cvs features or mistakes, depending on your point of view -- and then to git. Only one of those upgrades could be called moving to a 'game changing distributed version control system'. Two upgrades in short order is just staying fashionable. (I believe they're just about to move to mercurial, too.)

        They were requesting I fix their bug, even though their introducing-how-to-develop-for-gimp documentation was still based on cvs while they were gits.

        I don't know what GNOME development is supposed to be, other than very hard to get into based on available documentation, but is it software engineering? Based on their handling of bug reports, I suspect not.

  5. Edouard says:

    Remember kids - you've always got to be polite in the Open Source communi...hey, what?!

    Seemed like a perfectly reasonable report to me too. Seriously, that's a appalling piece of UI design - I mean, what next? A dialog asking if you'd like to rebalance your binary search tree?

    • Eric Fitton says:

      I was thinking of looking at Thunderbird again, but if random dialog boxes are the order of the day I'll pass. Glad to know that they won't fix poor behavior and shoot the messenger. .

      • LionsPhil says:

        It's worth installing Thunderbird 5 just to see the rest of the UI trainwreck.

        Which is impressive given they actually made it worse than Thunderbird 3. Now all the icons have lost their colour cues and the main menu and toolbars on Win7 are backed with a semitransparent "glass" effect. Which is to say that you can't read the bloody things.

        • Matt says:

          This says it all.

          • LionsPhil says:

            No, this is something worse: UIs designed by artists.

            MacOS X users might at least think all that 'glass' area is at least a bigger drag target for moving the window. Nope. Only the usual narrow strip is functional, only now there's no graphical distinction as to its bounds.

            • Eric Fitton says:

              Between that screen shot and the bug report "conversation" about why they refuse to capitalize letters, I'm taking my 10 foot pole and backing slowly away without touching Thunderbird.

              I'm amazed that Mlynarik stuck around so long filing bug reports. I can't stand software that treats me like crap.

          • Lloyd says:

            Wasn't this inspired by jamie's adventures with really-bad-attitude?

        • Lloyd says:

          Speaking of UI trainwrecks, Eudora has finally given up the ghost and become... a skin for Thunderbird called Penelope. This is very much a case of putting lipstick on a pig. I've tried Thunderbird a couple of times, and retreated rapidly from it.

          Apple's own Mail is probably the closest to an heir to the original Eudora, with decent, fast, searching and okay filtering.

  6. Edouard says:

    Ha! Now I'm reading some of his old reports - they're classic!

    But the experience is tinged with a certain sadness that someone who is actually willing to contribute that much in the way of feedback has gotten shafted...

  7. I came expecting fire and brimstone. I read the bugreport and... what the hell? How does the guy survive switching on his TV or leaving his house?

  8. Noah Friedman says:

    "Fuck you" should always be a dialog option.

  9. Noah Friedman says:


    (progn
    (while (not (y-or-n-p "Would you like to garbage-collect now?"))
    (message "Ok")
    (sleep-for 1))
    (garbage-collect))

  10. If this is a typical bug report from Richard, he would serve himself and the communities he participates in by injecting a little more self-deprecation and humor into his expressive bug report. This tone in feedback gets really tiring, because although it does an effective job of expressing useful criticism, it does nothing to balance that with a sense of empathy for the challenge that developers have in providing discoverable features without totally annoying users.

    That said, the Mozilla organization and its representatives need to take the high road and suffer snarky feedback, if they care to make great software and identify as many true flaws as possible. Obviously the report in question has valuable feedback, you just have to look past the tone and (probably more importantly) any historic context of dealing with the particular person, in order to incorporate it properly.

    • grue says:

      Meh. That's a pretty tame bug report to get upset enough to whack someone's account over.

      If the developers don't want useful feedback unless it meets some sort of style filter, that's fine, and I'm sure they and thier project will get the quality of feedback they deserve.

    • MEpton says:

      Yeah, cause that's the purpose of bug reporting: providing developers with a source of empathy...

    • Feature discoverability is what menus are for, though. Asynchronously popping up dialog boxes are extremely annoying because, among other things, they unexpectedly change the keyboard focus. Feature discoverability by unexpected dialog boxes is a UI choice that should have died with Clippy.

  11. Fazal Majid says:

    "Shoot the messenger" is a prescription for mediocrity in software and irrelevance in the not-so-long term. I don't find the bug report offensive at all, and the Mozilla admin's reaction, on the other hand, is simultaneously sanctimonious, preachy and high-handed. No wonder I've flipped the bozo bit on Mozilla.

  12. Karl Shea says:

    Somehow I'm not surprised. This is the same group that *insisted* that there were no memory leaks in Firefox for *years* and then released a major version that touted how many memory leaks were fixed.

  13. Chris D says:

    So, to be fair to dmose, this may be a good example of the larger pattern he cites:

    You know, every time I see a comment from Mark "WONTFIX" Banner I just want to reach out through the intertubes and cut off his damned fingers to prevent him ever writing any code -- or worse, writing "specifications" that define bugs out of existence -- ever again.

    I'm the user. DON'T TELL ME THAT MY BUGS DON'T HAPPEN.

    People like you are why free software is completely doomed, and why I (the second person after Stallman to ever write any GNU code) gave up a decade ago.

    Seriously, screw you.

    Whether the bugs are valid or not, any volunteer project, especially a highly visible one, is better off without someone whose volume goes straight from "There's a bug" to "JESUS CHRIST YOU'RE AN ASSHOLE."

    • Dan Sylveste says:

      As a Thunderbird user (who upgraded from 2 to 3 only to downgrade after two weeks of grief) this second one is pretty familiar.

      They've got one keyboard shortcut bound to two features, deliberately. I'm getting the feeling that there's two camps within the project and they're unable to prioritise their features.

      Also 'I have the quick filter bar hidden to save vertical space, how can I open it quickly' - 'press the quick filter button, which isn't a regular toolbar button, it's pinned to the tab bar' - 'oddly enough I have the tab bar hidden also to save vertical space, can I add the quick filter button to the toolbar' - 'no'

      Just because others are throwing years of usability conventions out the door with non-topmost menu bars and non-customisable toolbars ('ribbons'), doesn't mean you have to follow, guys...

    • Edouard says:

      Oh. Stallman.

      Imagine if he posted bug reports. That guy's "personality" is such that he'd get banned in no time. I wonder if everyone's arguments be the same then?

    • Eric Fitton says:

      If anyone ever breaks command-F on a piece of software I regularly use, I'd want to cut off their damned fingers too.

  14. anonymouse says:

    I know of this Richard Mlynarik, and yeah, he's kind of an asshole, one who tends to occasionally make good points, but in the worst possible way. But you know what? He's a user, and it is the nature of users, especially ones who are filing bugs, to be pissed off. He's not trying to be a part of the developer community, he's just reporting an issue he has with the software. Why do we all have to be one big happy family just because we use the same software?

  15. Thomas Lord says:

    Here is the proper voice to use in open source (the SNL Fops):

    http://www.hulu.com/watch/19178/saturday-night-live-the-norm-macdonald-fanclub

  16. Ysabel says:

    It bugs me that folks are calling his general dickishness "bullying". It's like when someone gets a bad price on something and says, "I was raped!" No, you got a bad price on something. Stop devaluing the word. "Bully" means something, and it's not this. *sigh*

    Also, in other contexts the things folks banning Richard are saying in defense of their actions are a derailing technique known as "the tone argument".

    • Ysabel says:

      As someone pointed out, I should clarify why I said what I said: Bullying requires a power imbalance. If anything, in this situation it appears the power goes the other way. Being a raving asshole is not the same thing as being a bully, and that distinction matters if you want to combat bullying.

  17. Juanjo says:

    Clearly the OP did not recognize a patch for annoyance-compatibility with Outlook's "AutoArchive" dialog.

    Hilarious :')

  18. Richard Mlynarik says:

    Thunderbird bug 478468 tells you all you'll ever need to know. WONTFIX! Doomed!

    • CTD says:

      Wow. That is the poster child for developer-driven development. Or possibly flatus-driven development.

      • Waider says:

        It's not a million miles away from a bug I watched, out of train-wreck ghoulishness, wherein a user's expected behaviour was deemed aesthetically displeasing in some way. The discussion dragged on for several *years* before eventually being WONTFIXed.

        The point that these people seem to be missing is that their efforts to mollify oversensitive developers doesn't necessarily draw any more actual contributors, but it very much does put off both regular users from reporting bugs, and capable (if occasionally abrasive) developers from becoming contributors.

        • Adam A. says:

          I'm one of those. I'm a pretty capable engineer, with a pretty thick skin (developers these days -- so sensitive!), and these epic fights over trivial crap ending in WONTFIX are exactly why I stay the hell away from contributing to big open source projects.

          It's such. a. waste. of. time.

          Why would anyone want to contribute their hard-earned skills when some wet behind the ears code monkey is going to decide that he knows better than decades of prior art and arbitrarily change it and then say piss off?

          And let's not even get started on the bugs that are closed because of the ground-up re-write that rendered them irrelevant because the code no longer exists, except that, of course, the rewrite is only 75% functional, and 25% of that is buggy as hell...

          Yeah, it's shit like that that keeps me pretty much only writing code when I get paid well for it. My time - it really is worth something to me.

    • MickS says:

      Quite an enlightening read. I pretty much gave up on open source at the start of the century when it became a hobby/political thing rather than a make it work thing. Doomed indeed.

    • Karl Shea says:

      Wow, that pretty much is all you need to know.

      I had heard that there are now more people working on Thunderbird than there were, which was exciting because it has been heading downhill in a big way. Guess I'll have to switch back to Apple's Mail though, which I'm not a huge fan of. But I don't see Thunderbird getting better anytime soon.

    • Edouard says:

      Breathtaking. Comment 42 is a instant classic too.

      • anonymous coward says:

        Note also comment 45 where the same doofus who shut off mly's account blatantly ignores every single point made in 42 and, apparently characteristically, threatens to shut off comment-42er's account because of similar butthurt. That's the guy in charge? I am amazed if there is anyone even marginally competent who would work for him.

    • LionsPhil says:

      If only it were an outlier.

      Bugzilla is where defects to to be denied.

    • Sean B says:

      I've figured Firefox, at least, has been Doomed basically ever since they broke the "right-context-menu 'back'" thing ages ago (I know jwz posted to that bugzilla thread). There was a clear lack of big-picture / coherent thinking about the UI in it.

      They also seem to underestimate the cost to users of UI changes. From Firefox 3 to Firefox 4 they changed the top context menu for links from "open in new window" to "open in new tab", so for weeks after that I was constantly accidentlaly opening new tabs when I wanted new windows. (This choice is bad in two different ways, the relevant one here being the fact that it's a change breaks everyone's muscle memory, the other one being closely related to the 'back' issue above.)

      What kind of amazes me is that given how terrible they are about this stuff that anyone even bothers filing bugs. UI issues? Expect it to be wontfixed because their view of UI is different (read: terrible) compared to yours. Technical issues? After ten years, I still can't view .c files in the browser: https://bugzilla.mozilla.org/show_bug.cgi?id=57342 (well, after seven years someone gave up and wrote an extension, but come on -- and it still won't happen automatically).

  19. Anonymous Coward says:

    Hey, some of those bug comments are pretty awesome. I wish I had the time to beat my head against a wall to bother reporting/suggesting/complaining when stuff that worked fine is broken by programmer dudes with no taste and no management. It looks like a good decision not to bother. You read them in order they start off all helpful and full of nice suggestions, and end up with blood leaking from his ears because of just how awful each new release is.

    My experience with Thunderbird is that things sort of worked in version 2.

    Then I too made the mistake of upgrading to version 3 where they broke filtering messages (Mlynarik complains about that I see, kind of sort of nicely, then with increasing frustration) and they also broke searching within messages. Simple filtering of a mailbox by subject/sender has got to be about the fourth or fifth most common thing I do, so that was pretty awesome work by the Thunderbird Open Source UI dudes to wreck it. Plus the lower-case "subject" business that got his goat is enough to make you want to kill somebody, for reals.

    Now I read (not even going to think about downloading it!) that their new upgrade they've broken getting new messages and deleting messages, which are the second most common and most common things I do. Is there some sort of Software Fail Award they're trying to win?

    • gryazi says:

      Arbitrary mailbox filtering is the one thing Evolution can do without crashing!

      • LionsPhil says:

        For a little while, I swore by Evolution as a MUA that actually bloody worked and hardly caused me to swear at all.

        Unfortunately, this meant using Linux. (The Windows port is tantalisingly functional in the same way that a chocolate gateux with cat vomit on it is almost delicious---you could eat around the problems, or you could just settle for the fruit cake.)

      • Jon says:

        Funny you mention Evolution.

        I was on my laptop and wanted to have a look at that weird capitalization bug in Thunderbird mentioned above only to find out that Thunderbird is not installed. I forgot that I'm still using Evolution on that computer because I was too lazy to ditch it and actually it hasn't crashed in a year or so, which apparently is another reason why I forgot I wanted to get rid of it. Leads me to think - Evolution actually evolved.

        • Blake Winton says:

          I've heard from a number of people that Evolution has gotten _much_ better recently. I think they re-wrote their IMAP handling, and just generally spent a lot of time and effort making it work better. If people haven't tried it in a while, it's probably worth a second look.

  20. fantasygoat says:

    I've noticed the decreasing thickness of skin on the intertubes for awhile now - blame Generation Y and their interest in form over function.

    • Patrick says:

      Yeah, I kind of sort of remember the huge flamewars of years past that would just go back and forth, and which were usually so damn esoteric that only people who really understood it were the participants.. Good times.

      9 years and counting for Keychain support? I knew there was a reason I avoided FF on my Mac, now I'm glad for it. I mean, keychains aren't esoteric OSX-only constructs, sheesh.

      And if you dig around, you can find our gracious host grumbling about Mork: 241438

      • I've talked to a couple of Firefox people about the Keychain issue, because it baffled me, too.

        It turns out that Firefox's internal password and form data storage operates on slightly different assumptions than other browsers, and keys its data in a different way, largely as a result of being developed when password services on lots of major platforms were still in the "nice idea" stage. (Does Windows have a password store for apps yet? Is there a standard one, or a standard API, for password storage on Linux yet? Serious questions.) Mozilla and Firefox have also traditionally gone for "self-consistent across platforms" rather than "consistent with the platform on each platform", and password storage APIs just haven't been consistent enough to write uniform behaviour over.

        Getting "nearly right" interop with Keychain apparently wouldn't actually be all that hard - but "nearly right" has a nasty way of being "still wrong", so they'd rather get it "right" or not do it at all.

        All of that said, it's one of the reasons I'm not using Firefox these days, either. I can come away from a conversation with a greater respect for the reasoning behind the decision and still think the decision is a pain in my personal neck. :)

  21. Wayne says:

    A. The bug report is in fact (IMO) invalid and it is hard to believe the reporter didn't know it. Yes, Thunderbird behavior changed in release 5. But there are two easy/trivial ways to avoid the "annoying pop up": 1. uncheck the box, 2. change the compact option back to pre-v5 behavior to not automatically compact.

    B. The ban action doesn't seem completely out of line for the TONE of the bug report, given A above, and put in context of past warnings to the reporter for some past postings (which aren't, obviously, in the bug report being cited in this blog).

    • Edouard says:

      "change the compact option back to pre-v5 behavior to not automatically compact"

      I KNOW! That's what I was screaming down the telephone to my grandmother. You'd think she didn't even have a basic understanding of on-disk structures. If you can't be bothered at least paying attention in undergrad CS classes, then don't whine when you can't understand the dialogs!

      • You ignored 1. entirely for your sardonic straw man.

        • Edouard says:

          Yes, because my grandmother will understand what "compacting folders automatically" means. When all she wanted to do was delete an email. Bad GUI! Bad!

          • Do you want to see this message again?
            Yes >No

            • Edouard says:

              So you haven't actually read the bug report then? Or viewed the attachment that clearly shows a dialog that does not says "Do you want to see this message again? Yes >No".

              OK. Understood.

            • Matt says:

              I believe his point was that his grandmother would be confused as hell by any kind of message period discussing compacting when all she did was delete a message. The concepts aren't linked, the actions do not logically relate to anyone unfamiliar with general file store behavior, and the text is phrased poorly.

              The fact that you can stop it from happening a second time by unchecking a box doesn't change the fact it has already happened.

            • Chris says:

              Not "I don't want to see this message again, " but "Screw with my mboxes without asking me first."

              Definitely a bug.

    • Richard Mlynarik says:

      The bug report is in fact (IMO) invalid and it is hard to believe the reporter didn't know it

      I'm not going to get into a pissing match with this one-named Wayne on somebody else's blog: but just listen to what he's saying! Mozilla developers seriously believe that their software end users engage in:

      Step 1: pretend to have problems and then jump through hoops to report imaginary bugs, positively revelling in the rewarding activities of transcribing dialog boxes, making and clipping screen shots, attempting to create reproducible test cases, trying and failing to locate any release notes, trying and failing to locate any relevant UI options, etc;

      Step 2: ?

      Step 3: Profit!

      BTW One of the pleasures of reading this fine blog has been that I learned the new, often-applicable, and handy phrase, "I hope you die in a fire." And for that I am grateful to JWZ.

    • You may have missed the point. Asynchronous dialog boxes are bad anyway, because they divert keyboard focus. Being able to limit the number of occurrences of the dialog to just 1 still isn't enough, because we still have the first annoying instance.

      I couldn't count the number of times I've been typing away, and a dialog has popped up and been dismissed by some character in the text I'm typing (this happens mostly to non-touch-typists), leaving me wondering what the hell the dialog box is for and what option I just selected.

  22. Jonn says:

    Oddly enough, I reported a weird bug in FF4/5. No matter how many times I click it, the "Stay Signed in Box" on iGoogle website keeps reticking itself. The only way to counter it is to either a)check every time I sign in, which I invariably forget to do, or b)set Google's cookies to automatically wipe each time I close the browser, which screws up several other behavoirs. No one ever answered, on Mozilla or Google's forums, and I can't find anything about the bug on, ironically, Google.

  23. Thomas Lord says:

    Perhaps it has already been said but if so it surely bares repeating:

    IT IS VERY IMPORTANT TO SCARE CASUAL "CONTRIBUTORS".

    To do otherwise is to lie to them and, well, that would be to do them a disservice. Which would be a fine way to treat volunteers you pompous mozilla-f'ing twits.

    • BrendanEich says:

      Nice collective blame there, Lord -- combining "mozilla-" with the prim "f'ing" when the topic is Thunderbird.

      I'm Mozilla from day one, I don't like any of this crap, it is not happening where I'm focused (Firefox and Gecko, plus future stuff -- feel free to mail me exceptions). And now that I've heard what has gone on, I'm in the bugs. So lay off the lizard.

      I agree with you about bleeding-hearts-for-n00bs going the wrong way. I don't agree violent language is the right response in something not life or death, but that doesn't excuse the Thunderbird bugs.

      This is a sad day for grumpy old hackers.

      /be

      • tegeran says:

        Maybe -- MAYBE -- it's not happening right now where you're focused, but it took YEARS of increasing complaints for Firefox people to stop dismissing users out of hand and admit that there might be some memory problems. That alone says a lot (none of it good) about Mozilla, and combined with the epic and ongoing fail of Thunderbird 3+, makes his description pretty accurate.

        I suppose I could make it sound more polite. How does "arrogant elitists" sound?

        • BrendanEich says:

          No, we knew there were memory leaks. I commissioned a cycle collector based on Bacon-Rajan PLDI 2001 in 2006. We worked on fragmentation in Firefox 3, shipped in mid-2008, did better than the competition of the time (Chrome launched in September). We have always known about our memory problems (what, you think we're idiiots? Why are we even talking?).

          The fact is Firefox 7 uses a lot less memory, absent leak bugs, than other browsers up until those with good browser kernels detect the (inevitable and endless) WebKit leaks and kill the rendering process. And we're doing process isolation too (not for Firefox 7).

          The other big bet we made in 2004 (2003, actually) was that add-ons were a net positive. But our model allows add-ons to do a great many things to your one big rendering/UI process, including leak. We're fighting this too, by putting the top add-ons under continuous test for leak as well as startup time and other performance bugs. Also promoting a better model: Jetpacks (out of process addons).

          Of course, add-ons mean user X sees great memory profiles for Firefox while user Y sees terrible leaks due to "bad-ons". Anyone watching memory bug reports sees good community triage based on what add-ons and plugins the reporter uses. Yet according to you, we were just denying these bugs existed instead of triaging and fixing them.

          Whining about "Mozilla's attitude" is itself just bad attitude. Either file bugs and cc: me or another committed hacker on them, or go use some other software. There, that wasn't too arrogant of me. But I remain an elitist,

          /be

          • tegeran says:

            If you knew there were leaks, why was every blog post and other public comment blaming plugins, extensions, and users opening too many tabs?

            And it's not just "according to me". It doesn't take much googling to find lots of annoyed users sounding off about Mozilla ignoring memory problems and acting like it's all the user's fault.

            Ben's February 2006 post was probably the most telling and representative of the general attitude over the years -- he condescendingly wraps scare quotes around "memory leak", then proceeded to try and blame the whole thing on tab history caching, until it was pointed out to him that it didn't work like that. He didn't actually know what was causing user complaints, he just assumed it was the dumb little users not understanding a feature he himself did not understand.

            Then we got the babble about how things were so much better with 3.0... then 3.5.... then 3.6... And in each and every case it got worse, but the party line was "everything's fine! disable your extensions!". Somehow, you guys never quite understood that most people weren't using extensions in the first place.

            Do I think you're idiots? No. I think Mozilla collectively is just incapable of assuming that other people aren't. It shows through in blog posts like the one I linked to, in the reactions to bug reports like Richard's, in the UI design decisions, and to a lesser degree in your own responses here.

            You, personally, may be more reasonable than most Mozilla devs I've seen, but even still, your reaction to criticism is somewhat defensive and hand-wavy, bringing out pretty much the same party lines as everybody else. And then you bring out the most annoying line I've ever encountered (and sadly, it's one I've seen happen in a few other projects): "cc me or another committed hacker".

            How the heck are users supposed to know who the committed hackers are?

            If you only want reports from people who already know all the intricacies of the Mozilla culture, that's fine, but then stop marketing your software to the masses -- it's clearly not for them.

            • BrendanEich says:

              No, not "every" blog post was blaming add-ons or tabs. This is not a simple "all Mozillans (except maybe brendan) claimed memory leaks were someone else's fault" story. Anyone reading planet.mozilla.org for years knows that.

              Most users ignore the hyper-vigilant fan blogs that tell binary "sucks! rules!" lies about a project. Users can tell whether hackers are committed by judging the fruits of their labors. We're still in the game; check out Firefox 7's memory use.

              As for "Mozilla culture", beware you aren't making the kind of smear that won't pass in other contexts. Cultures do differ and reasonable people can analyze and argue about those differences (where PC allows). But criticizing live cultures is tricky, and if the goal is to improve any given culture, approaching the people sharing that culture in a groupthink/group-blame way does not help.

              Treat people as individuals where you can make progress with reasonable actors. That is what I've always done. If some clique or subgroup won't be reasonable, they may have to go. That's also on my mind.

              /be

  24. tegeran says:

    This (all of this, broadly, the general brokenness of Mozilla and reaction to feedback) is why Thunderbird has not touched any of my mail accounts since 2008/2009ish, and I switched to Chrome close to a year ago.

    Despite its IMAP brokenness, I've mostly been using Mail.app. "Develop cross-platform IMAP-capable GUI MUA that doesn't suck" is very high on the list of projects I wish I had time for.

  25. BrendanEich says:

    The developers did not add a new and annoying popup. The popup is an old bug, https://bugzilla.mozilla.org/show_bug.cgi?id=286888. The problem is an out-of-order move that invites bug reports such as Richards, to change the default for compaction so the old dialog bug inevitably bites. Fix the dialog bug first, then change the default (would have been better).

    I'm now in two of Richard's bugs (both on Thunderbird), defending the substance and not the style of his reports. Hackers have to look through style to substance, or else. Yeah, ban toxic people when necessary, but do not cover up the valid parts of bugs they report.

    Doing things out of order, sweeping bugs under the rug, dismissing user feedback, all are guaranteed to make some grumpy reporter go ballistic. That does not make the reported problems invalid or won't-fix bugs.

    In the bug jwz linked, I claim the valid part, and the only problem expressed in the bug summary, is the old bug 286888, still not fixed -- it should have been fixed before compact-all-the-time was turned on. So the new bug is a dup.

    In the other bug where Richard went off (https://bugzilla.mozilla.org/show_bug.cgi?id=579372), about how cmd-F on Mac no longer does find in message, and cmd-G seems broken, I cannot find valid bugs tracking what seem like real problems, not to be dismissed by marking this other bug WONTFIX.

    So sure, hackers shouldn't have thin skins. That does not entitle reporters to rant or to use violent language. Two wrongs are both wrong, no excuses.

    Nevertheless, other than banning bug reporters, we Mozilla hackers can only improve ourselves. Agenda item #1 on that front is never to sweep bugs or out-of-order changes under the rug. Dup the bug, morph it into the valid bits, file a new and clean bug. But do not throw baby and aggro-reporter bathwater out the window.

    Sad to say, I'm not as worked up about this as I would be about something like it affecting Firefox, because I gave up on Thunderbird years ago. Then, it was too crashy, it lacked full text / OS integrated search, and message threading was still done wrong.

    The way these recent bugs were handled (ignoring the reporter being rude) does not make me want to come back.

    Anyone tarring my work in Mozilla with the same brush, sucking up to Chrome over Firefox: have fun in your Google+ portal front end.

    Grumpily yours,

    /be

    • anonymous coward says:

      They broke command-F ??!? That cannot be defended. But then to defend it with yet another etiquette rant is just beyond reprehensible. It's pretty clear that the thunderbird team is sorely in need of adult supervision.

        • Waider says:

          So even given a developer on a sister product making a reasonable point about expected behaviour, the response is /still/ "no, YOU'RE wrong!" What sort of user focus is that, exactly? I can see how it'll encourage developers to join the project - it'll be those developers who always get negative code reviews in their day jobs and see thunderbird as an outlet for their unrecognised creative genius.

          What I find really annoying about this and similar bugs mentioned here (and I think it was mentioned in another bug report, or possibly up thread) is that someone made an apparently arbitrary change with huge impact and, again apparently, no consideration for what that impact might be, and now that it's done no one is willing to say, "oops, my bad, I'll roll it back".

          • BrendanEich says:

            I've been invited by one Thunderbird maintainer to issue an order to roll back this change. How would that go down? It would have at least three effects:

            1. Satisfaction for users who want cmd-F compatibility.

            2. Some fraction of those behind the change pissed off.

            3. Me in the loop and likely to be summoned again.

            I care about (1) and (3), and not about (2) _per se_. If a large-enough fraction of those behind the change can agree without being ordered to revert, then net win on all three points. I'm working on this route.

            /be

            • Ben Brockert says:

              Thank you for your posts here. I had just thought that Thunderbird was fucked up, which was ok as I don't use it. But if you're a Firefox guy, I should start looking for a new browser.

              • BrendanEich says:

                Why, because I didn't order the Thunderbird maintainers to reverse their bad decision?

                You want a dictator with a trigger finger, there are many to choose from. Knock yourself out.

                /be

            • Eric Fitton says:

              That is quite a dilemma. Honest, I don't think I'd bother one way or the other if I was in your shoes. It is great that the Thunderbird developers respect you and that you're making inroads on a number of bugs. However, as soon as you stop looking over their shoulder and are back working on Firefox, I have a hard time believing that they won't immediately revert to their previous behavior. And the previous behavior of breaking compatibility, ignoring their users, marking everything as "won't fix" and hiding behind etiquette is pretty ugly. If you want to babysit, that would be great and I'll give Thunderbird another go after you've sat on them for a while. But I'm guessing you don't really want to be the babysitter and I can't say that I blame you.

            • Ronan Waide says:

              Honestly, I don't have a horse in this race as I only run T-Bird to keep an eye on non-office mail while I'm at work, and the discussion on this page (which isn't news to me, really) makes me less inclined to continue using it. More power to you if you make a difference here, but a shame it takes this sort of approach.

        • Dan Sylveste says:

          Heh. I like how he's not mentioning that it's _still_ bound to find-in-message - if you press it twice!

          Has anyone else ever seen that anywhere? A keyboard shortcut that does something different the second time you press it?

    • Edouard says:

      Well that seems like an entirely reasonable approach to the problem. Thank-you Brendon.

  26. Madrox says:

    In any community (and yes, bugzilla is a community) you have to decide what behavior is considered acceptable. That's what etiquette is. Whether that definition is appropriate for that community is not isn't something you only do once you've received the ban-hammer after, apparently, repeatedly being asked to play nice.

    I wonder if this would be news if it weren't Richard. Or, if it were news, would people be saying, "meh, it's just some random internet douche. Who cares?"