Slashdot Mirror


After 9 Years, Bugzilla Moves Up to 3.0

BuggyUser writes "Bugzilla, the popular application to track and manage software development bug reports, has moved up to version 3.0. The 2.x series has been in service for the last nine years. From the article: 'According to the Bugzilla 3.0 release announcement, some of the new features in this version include custom fields, support for the Apache mod_perl module, per-product permissions, an XML-RPC interface, and the ability to create and edit bugs via email. A demo site has been set up where users can test the new version before downloading.'" Linux.com and Slashdot.org are both owned by OSTG.

99 comments

  1. features... by frakir · · Score: 5, Funny

    "...the ability to create and edit bugs via email."
    Love it.

    1. Re:features... by kongit · · Score: 3, Funny

      Windows has had this ability for some time now, catch up oss.

    2. Re:features... by Anonymous Coward · · Score: 0

      I can create bugs via email!

      Sending patches to people does the trick nicely.

  2. What if bugzilla has a bug... by Anonymous Coward · · Score: 0

    that when reported via bugzilla, prevents the report from being seen?

    1. Re:What if bugzilla has a bug... by __aaclcg7560 · · Score: 1

      Then you send a nasty email to the maintainer about why your bug and the other bug wasn't fixed.

  3. What, why? by PietjeJantje · · Score: 1

    What does it do that make it deserve a mention as opposed to the other zillion bug tracking systems? The site says they don't support their own language choice any longer.

    1. Re:What, why? by Knuckles · · Score: 5, Informative

      What does it do that make it deserve a mention

      It's the bugtracker that is used by most major FOSS projects.

      --
      "When I first heard Daydream Nation it quite frankly scared the living shit out of me." -- Matthew Stearns
    2. Re:What, why? by orra · · Score: 2, Interesting

      used by most major FOSS projects.

      This is kind of worrying. See, unlike Firefox and Thunderbird, Bugzilla is only licensed under the MPL. Debian considers the MPL non-free, at least because it requires that source code be available at least six months after you stop distributing a binary, if you distribute the binary over a network. This is considered too onerous a restriction, as unavoidable circumstances (e.g. a Slashdotting) might prevent the availability of the source code. [Note that the GPL does not require this: if you distribute source code side-by-side with binaries, you can stop distributing both at the same time].

      I've not got anything against Bugzilla as an application besides this freeness issue. I've used it for GNOME bugtracking, and it seems quite, er, good.

    3. Re:What, why? by cortana · · Score: 2, Informative

      Debian does not consider the MPL to be non-free: Debian includes MPL-licensed software like Bugzilla and Firebird in their distribution.

    4. Re:What, why? by spiffyman · · Score: 2, Interesting

      A little clicking around in the GP's link renders the following: "If it's MPL *only*, you'll have to reject if it's targetted for main. If, like firefox and friends, it's multi-licensed, then it's fine." Also, this: "It is, in fact, not distributable as an executable by Debian."

      I'm not involved in Debian, but that's a pretty resounding set of rejections. If you read the whole thread (here, find "MPL") you can see one or two dissenting opinions, but the "reject" option does seem to be the consensus view. And since BugZilla is MPL-only, it looks like distributing it as a binary will not be an option for Debian maintainers.

      One thing I'm not sure about is whether Debian can go ahead and distribute the source. This would be a PITA for developers, but it's better than nothing.

      --
      So you can laugh all you want to...
    5. Re:What, why? by cortana · · Score: 2, Informative

      debian-legal is just a discussion mailing lists. The messages posted there do not represent the views of the Debian project.

      Actions speak louder than words, and the continued presense of Bugzilla, Firebird and other MPL-licensed packages in main indicate that the project does not consider the MPL to be non-free.

    6. Re:What, why? by larry+bagina · · Score: 1

      Debian is a bunch of jackasses. Remember firefox/ice weasel?

      --
      Do you even lift?

      These aren't the 'roids you're looking for.

    7. Re:What, why? by Anonymous Coward · · Score: 0
    8. Re:What, why? by Ant+P. · · Score: 0, Flamebait

      They're jackasses for respecting the trademark licensing agreement for the Firefox name and logos? If not that, then what?
      Are they supposed to break the law to get in your goody-goody list? Fuck you.

    9. Re:What, why? by ben+there... · · Score: 1

      Perhaps more significant is that it is GPL incompatible. Also, see the Mozilla Relicensing FAQ for details about their decision to triple license (MPL/GPL/LGPL), rather than modifying the MPL, to make Mozilla source compatible with the GPL.

      AFAICT, the MPL is not a bad license, other than its GPL incompatibility.

    10. Re:What, why? by Avatraxiom · · Score: 1

      The site itself doesn't say that at all.

      -Max

      --
      Everything Solved, High-Quality Bugzilla, Perl, and Linux Services
  4. I found a major bug in bugzilla by iamacat · · Score: 4, Funny

    But it crashes when I try to submit it. Oh well.

    1. Re:I found a major bug in bugzilla by Soltys · · Score: 1

      Bugzilla should use other bugs report software than own

  5. Their System? by Anonymous Coward · · Score: 0

    I bet they are still using v2 for the bug tracking for v3

  6. Compared to test director.. by k1980pc · · Score: 4, Interesting

    [rant:begin]
    I find bugzilla lacking in polish..Been using the test director for quite a long time at work and it seems very slick. I have used bugzilla only a few times [have raised some of the early ubuntu-vmware issues ]. Interestingly, feature by feature, it holds ground against the more [very] expensive counterpart. Bugzilla works well with firefox and safari - which I guess Test director may not due to few activex dependencies.

    The reason why I felt this was I suggested bugzilla to a colleague in a different organisation and they were far from satisfied. A more intuitive gui and some pleasing css works would have saved the day for bugzilla.

    Come to think of it, I could say that against many of the projects(FOSS in particular)... A bit more effort on UX could make a world of difference. Been testing office 2007 last few weeks and I'm very impressed. Just one of the apps in recent times whose UI made me feel why didn't I think of it. Just a pity that the guys who made Office 2007 were not more involved with Vista.
    Now off to some much needed sleep....
    [/rant:end]
    PS - Most of the comments above are subjective and anecdotal - Your experience and opinions might differ and I can live with it.

    1. Re:Compared to test director.. by nightgeometry · · Score: 1

      The advantage of Bugzilla is that you can easily change large chunks of it, it is far easier to customise than TD...

      I would say that if you want to use the other parts of TD (Test Plan, Test Lab et cetera), then it is just about the best thing since sliced bread (even given that it *can* be fucking annoying to use, and reporting is a bit shit). BUT - if you just want bug tracking then I do think Bugzilla is the way to go, even if only due to the price :) As said, you can customise it a lot, and perhaps their new templating system will help out with usability issues?

      --
      The best is the enemy of the good
    2. Re:Compared to test director.. by jsebrech · · Score: 1

      Come to think of it, I could say that against many of the projects(FOSS in particular)... A bit more effort on UI could make a world of difference. Been testing office 2007 last few weeks and I'm very impressed. Just one of the apps in recent times whose UI made me feel why didn't I think of it.

      It's not a matter of effort, it's a matter of vision. The nature of how OSS works (many eyeballs) is counter to how good GUI's are designed (central visionary designer). To make matters worse, in most OSS projects the core developers are engineers, not designers, and so even if they have a centralized concept of the GUI, it's built targeted towards engineers, not towards the actual userbase.

    3. Re:Compared to test director.. by moranar · · Score: 5, Insightful

      Ahem. The normal userbase of a bug tracking program is not composed of coders and engineers?

      --
      "I think it would be a good idea!"
      Gandhi, about Internet Security
    4. Re:Compared to test director.. by ismakhan · · Score: 2, Informative

      Compared to test director, bugzilla is cross-platform. HP Test Director [now part of Quality Center] needs a ActiveX control on client to work. This may not be problem for you but is for many people.

      Test Director also has a very bad reputation for security problems. Almost all security for it is done on the untrusted ActiveX client and not on server! Do not use test director if users are not 100% trusted.

    5. Re:Compared to test director.. by cerberusss · · Score: 2, Interesting

      Yeah well if you start talking that way.... I'm using the version control utility subversion at work which doesn't have much of an interface by default (except a commandline one).

      Now I've heard about the graphical client SvnTortoise or something, but my point is: I'd rather stab my eyes out that going back to that slow piece of shit Perforce.

      So I ask you: please, UI is important, but it's only one of many features.

      --
      8 of 13 people found this answer helpful. Did you?
    6. Re:Compared to test director.. by carou · · Score: 2, Insightful

      Ahem. The normal userbase of a bug tracking program is not composed of coders and engineers?

      Who is it discovers bugs and submits reports in the first place? Magic leprechauns?

    7. Re:Compared to test director.. by moranar · · Score: 1

      Normally, users "evolved" enough to even know what is a bug tracker and how to follow the instructions.

      Sorry, but something that is obvious to you (and that you asked rethorically) isn't to me.

      --
      "I think it would be a good idea!"
      Gandhi, about Internet Security
    8. Re:Compared to test director.. by carou · · Score: 1

      The point is only that end-users see it.

      If you install Bugzilla on your product's web site, and users think the Bugzilla user interface sucks, then they'll most likely associate that suckiness not with Mozilla but rather with you. That's why it is in your own interest to use a bug-reporting system which has an interface targeted at your end users rather than at yourself.

      And it's nothing to do with users being "evolved" or having the mere capability to read instructions - their ability to complete the task is no excuse for making it unnecessarily painful. Your users are doing you a favour in filing reports so that you can get your bugs fixed. Don't put stumbling blocks in their way, or make it any more difficult than it absolutely has to be. If you confuse them you won't get bug reports, and the resulting un-patched bugs will lose you customers.

    9. Re:Compared to test director.. by spitzak · · Score: 1

      We put "submit a bug" in as a menu item and it talks to Bugzilla (by sending email). The user never sees the bugzilla web pages.

      However I also think the bugzilla pages are pretty ugly and sparse. Can't they get rid of about 3/4 of those fields so you don't have to scroll to see the text of the bug?

    10. Re:Compared to test director.. by moranar · · Score: 1

      You assume that normally, users have access to the bug tracking interface, which is only generally true of FOSS software, not of commercial software in general.

      Don't get me wrong, there are many things to be improved in Bugzilla's interface. Still, the normal public of bugzilla is either motivated enough (as I am) to work with it, or is not, in which case they tend to use IRC channels, email or other tools to submit their reports (as I sometimes do: I don't always want to register on a bug tracker database just to report one single bug.

      Making the interface easier to use also has drawbacks: for example, developers complain about duplicate bugs as it is. Imagine if an easier to use program allowed many more people to submit their bugs with just a cursory glance at previous submissions.

      --
      "I think it would be a good idea!"
      Gandhi, about Internet Security
    11. Re:Compared to test director.. by Anonymous Coward · · Score: 0

      Not true. All development teams I have been part of also comprised some sort of management/team leader roles. Those guys did not necessarily have a CS background, but of course where _very_ interested in following up the bugs/defects situations ...

    12. Re:Compared to test director.. by Anonymous Coward · · Score: 0

      404 File Not Found The requested URL (subversion.tigris.org) was not found. If you feel like it, mail the url, and where ya came from to pater@slashdot.org.

    13. Re:Compared to test director.. by John+Whitley · · Score: 1

      Ahem. The normal userbase of a bug tracking program is not composed of coders and engineers? Sure, but expecting coders and engineers to be good interface designers is like expecting lumberjacks to all make violins like Stradivarius. After all, the violin's just made of wood, right?
    14. Re:Compared to test director.. by turbidostato · · Score: 1

      "I suggested bugzilla to a colleague in a different organisation and they were far from satisfied. A more intuitive gui and some pleasing css works would have saved the day for bugzilla."

      Was the guy to provide developer time or money to Bugzilla? Probably not.

      As you already said Buzguilla can stand feature-to-feature against solutions more expensive and (probably) with their own sets of problems (like being privative, for instance). So in the end, your colleague dismissed bugzilla not on a technical background but on aesthetics thus missing the chance of using a powerful, cheap and easily tweakeable tool for the task. On the other hand, bugzilla neither won nor lost no matter if your colleague used it or not.

      No sir. A more intuitive gui and some pelasing css would have saved *your colleague's day*, not bugzilla's.

    15. Re:Compared to test director.. by Angostura · · Score: 1

      The tools are not only useful for external communities of users. In a previous job, several years ago now, I had to look after the development of a fairly complex site. This comprised a public facing side, but also a number of back-end tools. As we developed the site, and post launch we needed a bug tracker so that members of staff could report problems as we found them. The Bugzilla interface was just too inflexible and overly complex for our users to cope with. We ended up using Mantis instead. And were very happy with it.

    16. Re:Compared to test director.. by sconeu · · Score: 1

      Normally, users "evolved" enough to even know what is a bug tracker and how to follow the instructions.

      Now, however they're unintelligently designed so as not to be able to do so.

      --
      General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
    17. Re:Compared to test director.. by Anonymous Coward · · Score: 0

      The problem with bugzilla is that after 9 years it still has the same horrible interface.
      You shouldn't have to go to massive customisation efforts to get a useable interface out of something.
      There are no excuses, just like the OP said, all it would take is a little touching up. Surely after 9 years you can expect someone to have looked at that last 10%?

    18. Re:Compared to test director.. by samkass · · Score: 1

      We use Perforce at work and I'm loving it. (They want us to move to ClearCase to integrate with the rest of our (huge) company, which scares me.) Perforce also integrates really well with our Bugzilla system, in that I can attach the resolution of a bug to a Perforce changelist directly, so when I look up the source code changes later I can immediately see which bugs that checkin fixed. Handy.

      At my last company we used StarTeam, and although it sucked badly as a version control system, it had a nice tie-in with the bug and build systems. We could mark a bug "fixed in next build", then when the build system fired off, it would tag the build and update the bug database to read "fixed in build XXXX". It made tying together regression testing, functional testing, and bugfixing really easy. But lack of cross-branch integrates or atomic commits made it painful for VCS.

      --
      E pluribus unum
    19. Re:Compared to test director.. by Avatraxiom · · Score: 1

      You know, I'm one of the primary developers of Bugzilla, and I agree with you! It absolutely could use some UX love.

      One problem is that UX engineers are kind of hard to come by. Most of them get paid a lot of money and work all day, and I know few of them who work on open source projects in their spare time.

      We have some good programmers on the team, but we're not all UI designers. I have some UI design experience myself, but I'm usually caught up in doing architecture and backend. We do consider improving Bugzilla's UI to be a high priority, though.

      -Max

      --
      Everything Solved, High-Quality Bugzilla, Perl, and Linux Services
    20. Re:Compared to test director.. by Anonymous Coward · · Score: 0

      If you're using Windows, then I strongly suggest you evaluate TortoiseSvn as soon as possible. It is one of the slickest Explorer integrations ever; it is almost as if the operating system natively supports version controlled files.

      I too agree that Subversion is so much stronger than Perforce, especially if you have the internet between you, and your server. Pretty much anything people say Perforce is great for, Subversion is at least as good.

      Posted anonymously as this has gone 'off-topic'
      Prehensile

    21. Re:Compared to test director.. by Anonymous Coward · · Score: 0

      Yeah well if you start talking that way.... I'm using the version control utility subversion at work which doesn't have much of an interface by default (except a commandline one).

      Now I've heard about the graphical client SvnTortoise or something, but my point is: I'd rather stab my eyes out that going back to that slow piece of shit Perforce. I'm lost - are you saying that svn sucks, or that the GUI sucks, or what?

      And how is svn relevant to bugzilla, anyway? They're two completely different pieces of software ...
    22. Re:Compared to test director.. by shellbeach · · Score: 1

      Don't get me wrong, there are many things to be improved in Bugzilla's interface. Still, the normal public of bugzilla is either motivated enough (as I am) to work with it, or is not, in which case they tend to use IRC channels, email or other tools to submit their reports (as I sometimes do: I don't always want to register on a bug tracker database just to report one single bug. But that's the problem with large OSS projects - there's often no contact information for submitting bugs other than the link to the bug-tracking software.

      And the problem with Bugzilla is that (a) you can't submit anonymous bugs, and (b) the search engine to check if anyone's submitted the same bug before you is particularly sucky. Which means that a lot of bugs don't get submitted because users can't be bothered, and the ones that do get submitted often turn out to be dupes.

      Sourceforge.net has a much better tracking system, IME. It makes it as easy as possible for a user to submit a bug, rather than trying to discourage them from doing so.

    23. Re:Compared to test director.. by moranar · · Score: 1

      "that's the problem with large OSS projects - there's often no contact information for submitting bugs other than the link to the bug-tracking software."

      Sorry, that hasn't been my experience at all. Most big projects I know of have at least a mailing list and an IRC channel, in addition to the bug tracker. Maybe we're thinking of different areas of OSS?

      --
      "I think it would be a good idea!"
      Gandhi, about Internet Security
    24. Re:Compared to test director.. by turbidostato · · Score: 1

      "The problem with bugzilla is that after 9 years it still has the same horrible interface."

      Yeah, but that's the problem to whom? It seems obvious it's not a significative problem to those developing bugzilla at least or else, they probably would have expent some time/effort on that issue as they have done on whatever they've been devoloping all this time. If if it's not *their* problem, maybe it is *your* problem, isn't it? But then, what have *you* done about it? Maybe, hummm... nothing?

      "There are no excuses"

      I'll take your word for it.

      "all it would take is a little touching up. Surely after 9 years you can expect someone to have looked at that last 10%?"

      Now I know there are no excuses, what are *your* excuses not to do even a "little touching up" on a long "9 years time span" so bugzilla shows not "such a problem" to *you*? Please put your money/effort where your mounth is.

      Anyway, it really doesn't matter: bugzilla has shown a very good track record about returning money to disatisfied customers. I suggest you ask them for your license money returned plus a 50% on top of it for mislead advertisement and another 10% for each day you lost on it due to their mislead advertisement.

  7. I'll even it out for you by BuR4N · · Score: 1

    http://www.fogbugz.com/ -- best bug tracker ever

    --
    http://www.intellipool.se/ - Intellipool Network Monitor
    1. Re:I'll even it out for you by Anonymous Coward · · Score: 0

      Not free, however.. (In either sense)

  8. Bugs and their Creation by N3wsByt3 · · Score: 0, Redundant

    "and the ability to *create* and edit bugs via email."

    Is that *really* such a good idea, I wonder? ;-)

    --
    --- "To pee or not to pee, that is the question." ---
    1. Re:Bugs and their Creation by smurfsurf · · Score: 1

      Works for Debian all right for a long time now. See http://www.debian.org/Bugs/Reporting

  9. Don't rewrite from scratch by RedMagic · · Score: 3, Insightful

    Bugzilla's 9-year-road to 3.0 is a good example of why code should very rarely been rewritten from scratch and even if, then never the whole codebase. The more ambitious the goal one tries to achieve by that the harder the task - especially if one needs to keep updating the old codebase. There is no code which cannot be iteratively improved to achieve whatever the fresh code is suppose to.

    1. Re:Don't rewrite from scratch by jsebrech · · Score: 2, Insightful

      Bugzilla's 9-year-road to 3.0 is a good example of why code should very rarely been rewritten from scratch and even if, then never the whole codebase. The more ambitious the goal one tries to achieve by that the harder the task - especially if one needs to keep updating the old codebase. There is no code which cannot be iteratively improved to achieve whatever the fresh code is suppose to.

      So, if you had a bug tracker written in assembly you'd keep iteratively improving the assembly instead of rewriting it in a higher-level language? Different programming environments don't have equal productivity levels, so if you want to add features to an application it sometimes can help you to rewrite it to get where you need to be going faster.

      I agree though that in bugzilla's case a rewrite probably doesn't make sense. The benefit of other languages over perl is probably marginal at best, and to think that a man alone could rewrite bugzilla, without defects, in 18 months, like the developer in the article proposes, is a bit naive to say the least.

    2. Re:Don't rewrite from scratch by ushering05401 · · Score: 2, Insightful

      If nothing else they broke the 'release early and often' cycle. That's worth something.

    3. Re:Don't rewrite from scratch by heinzkunz · · Score: 1

      > So, if you had a bug tracker written in assembly you'd keep iteratively improving the assembly instead of rewriting it in a higher-level language?

      If resources are tight and you have commitments to your customers, you definitely should not try to rewrite from scratch.
      Rewriting from scratch while maintaining a production branch takes a lot of effort - some times too much. You don't have to look far to find a perfect example: an earlier effort to rewrite Bugzilla from scratch failed - the rewrite never caught up with the progress of the original branch.

      By the way: I guess that most programmers repeat this history at some point in their career.

      You better migrate in small steps. In case of the bug tracker written in assembly, you could introduce a bridge to a higher level language, then move parts of the application over.

    4. Re:Don't rewrite from scratch by Anonymous Coward · · Score: 0

      Bugzilla's 9-year-road to 3.0 is a good example of why code should very rarely been rewritten from scratch and even if, then never the whole codebase.

      But it was rewritten: it was originally Tcl. Are you saying it would have made it this far faster if they had stuck with the Tcl codebase?

    5. Re:Don't rewrite from scratch by Bearhouse · · Score: 1

      Yup. MGPP (Multi Generation Project Plan) is useful here.

      See http://tinyurl.com/2l2vdn [ isixsigma.com, not goatse ;-) ]

      It's not 'rewrite fast, release often, fix the bugs, restart'.

      More, you define where you want to be in 10 years time, then focus on delivering something that takes you nearer to that goal every - say - 3 to 9 months.

      Think NASA and going to the moon...

    6. Re:Don't rewrite from scratch by darkwhite · · Score: 1

      If nothing else they broke the 'release early and often' cycle. That's worth something. How is that a good thing?
      --

      [an error occurred while processing this directive]
    7. Re:Don't rewrite from scratch by ushering05401 · · Score: 1

      All too often releases are made for reasons other than having a release ready product. While I notice this mostly in the proprietary world, the last release of FF jumps to mind as an OSS example.

      I love FF, but felt the last release was rushed to prevent IE from getting all the headlines. I seem to recall some high profile bugs and gripes about the release feeling more like an incremental... none of which helped FF's profile in the non-OSS world.

      As for proprietary examples... um... where would I start?

      'Release early, release often,' is quoted regularly by developers on a major MS development board to which I belong. The practice is considered standard by those developers, but I find it repugnant. Then again, I aspire to design products that need only be upgraded as their environment changes, not to add/fix things that should have been implemented and extensively excercised in the *.0 package. But then, I also have the benefit of being a niche developer servicing industries that possess a slower adoption rate, which gives me more time to mature my products without excessive external pressures.

      Anyhow, if it takes nine years to produce the software that will continue to build on Bugzilla's previous success than I am glad they took nine years to do it.

      Regards.

    8. Re:Don't rewrite from scratch by shellbeach · · Score: 1

      All too often releases are made for reasons other than having a release ready product. While I notice this mostly in the proprietary world, the last release of FF jumps to mind as an OSS example. But "release early, release often" refers to bleeding-edge development builds, not to stable products! The whole point of the concept is to get as many users as possible providing constructive feedback and testing, so that when a stable release does happen it's as complete as possible.

      (Of course, in these happy days of SVN, it's pretty much been made irrelevant, since everyone has easy access to the development version of OSS projects.)
    9. Re:Don't rewrite from scratch by ushering05401 · · Score: 1

      That is why I had a hard time coming up with OSS examples. The philosophy works well for incremental OSS releases because community participation is essential to the development process. *.0 releases are a different story, and the last FF .0 release was not up to par.

      Proprietary vendors using the same release philosophy is akin to selling a car that needs a shitload of work, but selling it at a new car price anyhow, knowing you will make money off bringing that car up to standards because the customer has no option but to pay you to fix it.

      I believe we are in agreement in general.

      Regards.

    10. Re:Don't rewrite from scratch by shellbeach · · Score: 1

      I believe we are in agreement in general. Yes, I think so :) I'd certainly agree that FF 2.0 was a rushed job that shouldn't have been released on the masses at the stage it was at.

      Mind you, the masses were so used to IE bugs that they don't seem to have cared overly much, so I guess it all evens out in the wash.

    11. Re:Don't rewrite from scratch by Gerv · · Score: 1

      That's rubbish, for two reasons. 3.0 wasn't a from-scratch rewrite from 2; the team did the normal incremental and iterative development. And so the "9 years" says nothing about whether software should be rewritten from scratch or not. In fact, the main reason that Bugzilla hasn't been labelled "3.0" before is because one developer a few years ago went off and started an abortive complete rewrite which he called "Bugzilla 3", and we didn't want to confuse things by reusing the number.

      This is actually explained in the release announcement.

  10. Terry Weissman Rocks! by Anonymous Coward · · Score: 0

    It is amazing that the quick and dirty bug tracking tool that Terry wrote for the browser group would last so long, and that the design has not really change through all the years.

  11. Where are the perlheads? by jsebrech · · Score: 4, Interesting

    The article mentions the fact that bugzilla's release manager wants to see it rewritten in some other language because in his opinion perl is no longer a good language to be writing large applications in. I expected to go into the comments and see nothing but outraged reactions from perl lovers, because that's what I would have seen 5 years ago.

    Where has all the perl love gone?

    1. Re:Where are the perlheads? by Anonymous Coward · · Score: 1, Insightful

      I'm pretty sure they've grown up. Perl _is_ a great language for what it is. It _isn't_ for everything.

      There's no reason to think that a rewrite of any large perl program wouldn't retain some perl bits or that most programs of this nature wouldn't benefit from some perl here and there. The point is that perl need not and should not be the core of a large project like this.

      It's nothing against perl.

      It's like building a stereo system: there's a few key components you want to go balls to the wall on (eg, speakers) but the bulk of the system doesn't have to be killer if you do that part right. In this case you balls out on perl for key parts and a more general implementation for the bulk.

    2. Re:Where are the perlheads? by Marcion · · Score: 1

      Seems that some of them want to port it to Ruby on Rails or similar. I do not do Perl or Ruby so I do not really care either way (I'm a Pythonista), other users of the software probably do not either, it is just a tool to get other (more interesting) things done. Bugzilla is of historical interest because it was the first major open source tracker, which then helped to create the notion of using an open online tracker; however these days there are lots of alternatives, as well as hosted canned services etc.

      Having said that, using a web framework is interesting, it would make it more flexible and you can then integrate it more with tools from that framework, however, it is also another whole chunk of things that people have to learn.

      Unlike some proprietary companies where the brand is more important than the code, Free/Open Source is not big on porting stuff for the sake of it, and instead are more likely to create a new bug tracker in Ruby or whatever and slowly migrate to it.

    3. Re:Where are the perlheads? by renoX · · Score: 3, Insightful

      Well for me, the more I used Perl, the less I liked it so I'm not surprised that Perl's popularity has faded..

      When I was taught Perl, I thought 'cool a better, more powerful, portable shell' and then I had to maintain Perl's code, some written by beginners and some by 'experts'.. And I discovered what a mess Perl is..

      Sure it's portable but the language don't give you the correct defaults so beginners code is usually awful AND experienced Perl coders let them sucked by TWTDI which makes their code hard to maintain by anybody else..

      It takes *a lot* of self-discipline to write maintainable/readable Perl, so not surprisingly lots of Perl code is junk.

      Hating Perl, I looked for another language and found Ruby which unfortunately I don't use that much as my boss won't let me do it (not widespread enough for him), *sigh* such a beautiful language and having to use shell or Perl instead..

    4. Re:Where are the perlheads? by Plugh · · Score: 1

      I love Perl and have coded in Perl almost exclusively for nearly a decade. I think Perl has failed to deliver on Perl6/Parrot. Perl5.x has some great features, but they're pretty hacked-up. Perl has pointers (references), structs (hashrefs), is extensible (XS) and OO (close enough for government work). If you want something portable, and you find that Java is inflexible, resource-hungry, and has all the comfort of a well-worn straightjacket, use Perl and you'll do great.

    5. Re:Where are the perlheads? by Anonymous Coward · · Score: 0

      Probably because many of the Perl lovers have probably seen the Bugzilla code and are already embarrassed Bugzilla is associated with Perl. Its so badly written its not a big surprise some maintainers blame the implementation language, not the poor design choices (which in fairness, they've inherited from others). As the post you cite notes, Bugzilla started in TCL and was ported to Perl by someone apparently learning the language.

      The first time I worked for a company trying to use Bugzilla we, like many others in the same position, wanted to add "just one extra feature" and assigned an experienced Perl programmer to the job (I was his development manager). I remember listening with incredulousness as he explained that Bugzilla kept the current database result set in a global variable accessed everywhere. This poor design made it very difficult indeed to add the one extra query our feature needed. What should have been a day of work turned into weeks as he kept turning over rocks in the code and finding some new horror. Eventually we rewrote the whole thing (in Perl).

      Since then I've worked at two other companies who have tried and then dropped Bugzilla for similar reasons. Once we switched to a commercial GUI app in C++, another with a homegrown web-based Java app. Both worked better than Bugzilla, but I'm not so foolish to believe that implementation language made them better. They just weren't design disasters like Bugzilla.

      Avatraxiom makes some good points about Perl's failings in the design-by-contract area, and its possibly true that another language might have led to better future maintainability. But anyone who has seen the Bugzilla code has to wonder if it wouldn't have been just as bad in Python/Java/whatever. Bad code is just bad code.

    6. Re:Where are the perlheads? by Error27 · · Score: 1

      The guy obviously knows Perl and he's right about what's wrong with Perl.

      The python stuff is pretty bogus. You don't need a special editor to recognize indent blocks. (Hint: Look for things 4 spaces further away from the margin than the line before). That said, python is slower than perl for many things. I've written internal webpages in python but I've not seen many comercial websites done in Python. There is probably a reason for that.

      Ruby. Don't know. Joel on Software says it's slow. Otherwise, it's very popular.

      Java. He complains about mod_perl but there isn't even an apache mod_java available to complain about so far as I know. I really doubt the "fast" bit.
      They did a rewrite of sourceforge in java and it was slow and sucky.

      PHP. Don't know. Very popular. Probably a decent choice.

      The rest of the languages were not serious options.

      Rewrites are a pain. The article basically says they tried the rewrite trick for 7 years to try create the 3.0 branch and then they abandoned it. You'd think they'd learn. :P You'll always lose some existing crontributors if you switch to a new language. Still 80k is not that much so it could be done. It would be worth it if you gained more new contributers or if the remaining contributers could write faster in the new language.

      BTW. I'm the bugzilla maintainer at work and I've found the code to be pretty easy to read. It was pretty easy to upgrade my sandbox system to 3.0. It even preserved the custom fields we had added.

      PS. Please don't rewrite it in java.

    7. Re:Where are the perlheads? by Breakfast+Pants · · Score: 1

      "That said, python is slower than perl for many things. I've written internal webpages in python but I've not seen many comercial websites done in Python. There is probably a reason for that."

      Check out reddit.com

      --

      --

      WHO ATE MY BREAKFAST PANTS?
    8. Re:Where are the perlheads? by Avatraxiom · · Score: 1

      Well, it was more of a post exploring whether or not there would be advantages and if it could all be done iteratively without a re-write. I'm big on iterative development--to a large degree, I spearheaded a lot of the recent iterative improvements to Bugzilla.

      -Max

      --
      Everything Solved, High-Quality Bugzilla, Perl, and Linux Services
    9. Re:Where are the perlheads? by Just+Some+Guy · · Score: 1

      Hating Perl, I looked for another language and found Ruby which unfortunately I don't use that much as my boss won't let me do it (not widespread enough for him), *sigh* such a beautiful language and having to use shell or Perl instead..

      Have you tried Python? It seems to live in the same problem space as Ruby, but has the boss-wowing factor of being a paid project of Google.

      --
      Dewey, what part of this looks like authorities should be involved?
    10. Re:Where are the perlheads? by renoX · · Score: 1

      I've spent quite some time evaluating Python and Ruby as I was trying to find the "best", my conclusion was that these language were twins, they had exactly the same qualities (with a little preference for Ruby), but Python doesn't suit my boss either..

      The thing: with the number of tools written in shell&Perl, you cannot avoid having those two installed, and advocating a *third* interpreter is a hard sell (plus he didn't like the space indentation of Python).

    11. Re:Where are the perlheads? by Nevyn · · Score: 1

      The python stuff is pretty bogus. You don't need a special editor to recognize indent blocks. (Hint: Look for things 4 spaces further away from the margin than the line before).

      Speaking as someone who writes pretty much all of his non-C/sh in python nowadays, I call BS ... Python has a lot of things going for it, but making invisible characters part of the syntax isn't one of them.

      Maybe you don't think you need real begin/end markers to read code, but I find it very painful to not have them ... and even more painful when I have to reindent blocks of moved code (as my editor can't do that properly using the same method I've been reindenting code with in every other languages for 15 years or so).

      --
      ustr: Managed string API with ave. 44% overhead over strdup(), for 0-20B
  12. Trac is da bomb by cerberusss · · Score: 4, Interesting

    Nowadays, the sysadmins have installed Trac for us. Works very good, with integrated Wiki and all that jazz. I don't know how it stands up featurewise against Bugzilla, but Trac has a very flat learning curve. For instance, searching is one box. One search box. Compare that to the humongous Bugzilla search screen.

    --
    8 of 13 people found this answer helpful. Did you?
    1. Re:Trac is da bomb by Anonymous Coward · · Score: 3, Informative

      Bugzilla has two search screens, one simplified and one advanced. Though I can tell you I am one of the few people who actually searches for bug report for a major (and I mean major) Free software project. Most users just file the bugs and I (and sometime when I am sleeping, others) close bug reports as being duplicated. The major problem I have found with bug tracking systems so having too many of them.
      At work, I use 4 different bug databases, 3 of them are bugzilla and one a home grown one. Moving bugs from one database to another is annoying, though I only have to move them from one bugzilla to another (actually to upstream). The developer support team moves the one from the home grown database to bugzilla (though sometimes forgetting one important piece of information that I have to go searching for).

    2. Re:Trac is da bomb by Anonymous Coward · · Score: 1, Informative

      Bugzilla's advanced search screen is indeed humongous, but the default simplified screen is definitely not hard to grasp. The plus side of trac seems to be very nice SVN integration.

    3. Re:Trac is da bomb by AlXtreme · · Score: 3, Interesting

      I completely agree with you on this one. After trying Trac out a few years ago, I haven't looked back. Besides for personal and university projects, I've been using separate Trac-installs for clients. Everything from support requests to development projects end up as tickets, with wiki pages for additional details and background information.

      The main reason why Trac is so successful is indeed the flat learning curve / simple interface. Two sentences along the lines of "If you add a new ticket here, I'll get right on it" are enough. Clients happy because they can easily bug me and have an overview of past tickets, I'm happy because of the decrease of postit's on my monitor.

      Oh, Bugzilla you ask? I've had to deal with it in the past for a few OSS projects. I'm still scared of it, and wouldn't let my clients near it even if they begged me. It might be more useful when you have hundreds of technical users working on a single project with needs like reporting and time tracking, but for anything less a more flexible alternative is preferable, IMHO.

      --
      This sig is intentionally left blank
    4. Re:Trac is da bomb by rgravina · · Score: 2

      Not only that, but Trac integrates with Subversion *very* well (and I think CVS and other version control systems too). You can see diffs between versions, reference commits in the comments of tickets and vice versa, browse the source (of any version). Really really nice.

    5. Re:Trac is da bomb by foniksonik · · Score: 2, Interesting

      I'm still scared of it, and wouldn't let my clients near it even if they begged me. It might be more useful when you have hundreds of technical users working on a single project with needs like reporting and time tracking, but for anything less a more flexible alternative is preferable, IMHO.


      This is so interesting.... something must have changed at say 2.2 or 2.3 cause back at 2.16 of Bugzilla I as a designer with about 6 months or perl exposure (but plenty of html + css + javascript) went in to the Bugzilla UI pages (search, reports, etc.) and completely customized them for a company with about 35 devs at the time. I was 'the web guy' at the time. They all coded in C++ primarily.

      I created a bunch of canned queries in a select menu, plus a quick search that had better defaults. A better default search page with info and a layout that actually meant something for the devs rather than being generic but consistent with other Bugzilla installs. Added color coding and collapsible lists to the results pages, and more..... it really wasn't that complicated to make changes... the html in perl is still html so creating a better UI was a matter of looking at source for Bugzilla output in the browser and then doing a global find on the Bugzilla web dir with an html snippet or css class or something as my search.

      --
      A fool throws a stone into a well and a thousand sages can not remove it.
    6. Re:Trac is da bomb by slyborg · · Score: 1

      Has some nice features. I like the Roadmap. But the Searching, feh. Only by strings? And I need to write SQL to create a report?

    7. Re:Trac is da bomb by loshwomp · · Score: 2

      We tried out Trac and found it annoying creating separate Trac instances for each project. With bugzilla, we have some degree of knowledge sharing with everything in one place, and issues can move between products if they're miscategorized, or if they apply to more than one.

      Bugzilla is too hard-coded as a software issue tracker, but Trac is even more so. Both begin to break down when used for tracking hardware problems, customer inquiries, etc. We're still searching...

    8. Re:Trac is da bomb by jhol13 · · Score: 1

      Trac is somewhat limited. Having used Quality Center I appreciate integration of requirement specifications and test cases (to bug database).

      After all, a defect should be findable by a test case (else you should create new one), and calculating defect count for a feature/requirement is a good thing to have.

      Integration to SVN is a very good feature (which QC lacks).

    9. Re:Trac is da bomb by Kent+Recal · · Score: 1

      One word: Mantis. (be sure to look at the demo)

  13. Re: lacking by m0llusk · · Score: 1
    I find bugzilla lacking in polish.

    Specifically in this case it seems that of the 17 available localizations only Japanese in an initial release form is currently available for the 3.0 version. The most recent available Polish versions are for 2.20 and 2.18 which is as fresh as any other localized package apart from Japanese. This is another example of localization trailing development instead of being integrated with it. (Reference: bugzilla download localizations)

  14. Re: lacking by Anonymous Coward · · Score: 0

    I am confused here. Are you joking or you really thought it is about localization? The GP meant the look and feel. But if it is meant to be funny, good one mate......and apologies for slightly spoiling it :)

  15. it's wonderful if they let it actually work by r00t · · Score: 1

    Every damn Bugzilla insists that I make an account. I don't want any more accounts!!!

    I don't have any sort of account with Debian, yet I can file bugs and comment on bugs. It's easy, so they get lots of good bug reports from me. Likewise for the kernel, because they take bugs on an unrestricted mailing list.

    With Bugzilla and restricted mailing lists, usually I just don't bother. Screw them. If they wanted bug reports, they wouldn't have made filing bug reports a pain in the ass.

    So... will it be easy like Debian? I'm betting not. An account is surely required.

    1. Re:it's wonderful if they let it actually work by Firehed · · Score: 1

      Hm... maybe they should consider support for OpenID, or some other form of unified login concept.

      I know that on more than one occasion, this kind of thing has scared me off from filing a bug report. Enough accounts aside, I just don't feel it's worth my time or effort to sign up for something else just to tell them about a glitch.

      --
      How are sites slashdotted when nobody reads TFAs?
  16. Re:im doin it now by Anonymous Coward · · Score: 0

    It's only an AE away from GOATSE. Coincidence?

  17. Maybe Duke Nukem Forever is next by chasisaac · · Score: 1

    Could it be that Duke Nukem Forever will be released soon?

    --
    -- A computer without Windoze is like a choclate cake without mustard
  18. Perl is more sublime by bill_mcgonigle · · Score: 1

    Where has all the perl love gone?

    Unlike Python or Ruby on Rails, one is not required to swear a blood oath to evangelize Perl at every opportunity to be a Perl hacker. So, the Perl lovers are probably off writing code instead of doing songs and dances here on Slashdot. Its evangelized in a more sublime way - working code. Yeah, that's probably not as effective marketing in today's "ooh, shiny" society.

    I'm not sure what the release manager is thinking. Perl 5 hasn't changed drastically in quite a while so his 'no longer' thing doesn't make alot of sense unless he's contrasting with something else he has in mind (and thus he should just say so).

    The 'killer-app' for Perl is mod_perl, which this version of Bugzilla finally uses. Without checking the dates, I'm pretty sure mod_perl was available when Bugzilla 2.0 was released, but the Bugzilla code-base was such a steaming pile of hackery (perhaps the right design given requirements) that Bugzilla has been unusable on mod_perl for all of its 9 years. I think I've been watching the "clean up Bugzilla so it runs on mod_perl" bug since I opened a Mozilla Bugzilla account in the late 90's, and it's only received any attention in the past couple years. Kudos to whomever finally 'got it'. If somebody can let me know who that was I'd like to send flowers, chocolate or beer.

    So, the Bugzilla code is now cleaned up, designed for proper OOP for the first time, today. It's not even reasonable to talk about a project being easy to maintain by a large number of people until it has a rudimentary OOP design, so for him to blame the language when it was being used in a 15-year-old manner, is just absurd. He should really evaluate how Bugzilla 3.1 shapes up. I might go try to write a plug-in for it, assuming there's some API documentation. I think many others will now also be unafraid to tackle enhancements. I tried a couple before and ran away in fear (that is fear of the amount of time it would take me to understand 'hello world' in the old regime).

    At this point the other option is Java. Python doesn't have a strong web framework yet and is slow (somebody needs to write a LISP compiler for it...), the Ruby interpreter is too slow and mod_ruby is immature, PHP is just a heck of a poor language in the current version (different functions for 'normal' and 'weird' strings, etc.) and even Zend isn't as fast as mod_perl (plus the whole security headache...). Interestingly, if Perl 6 ever shows up, it should be able to run any of the above languages well.

    So Java - it's great for deploying large applications, but part of Bugzilla's success has been that it's a small application that's easy to deploy. Maybe they can figure out a way to actually create a .war that you can just drop into tomcat and go, but java application packaging has traditionally been a really sore spot for admins.

    Without being unappreciative of any of the work he's done, the release manager should really be more careful or explicit about what he's saying on this topic. After all, if I think the whole thing is likely to be scrapped, I'm not likely to go start writing plug-ins. A conspiracy theorist might think that was the whole idea.

    --
    My God, it's Full of Source!
    OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
  19. self-discipline and languages by Krishnoid · · Score: 2, Insightful
    It takes *a lot* of self-discipline to write maintainable/readable Perl, so not surprisingly lots of Perl code is junk.

    It takes a lot of self-discipline -- a lot -- to write readable, well-structured English -- even more so than Perl, because it doesn't even have to pass a syntax check -- so not surprisingly, lots of English is junk. Perl purports to be accessible to beginners and experts, and to make easy things easy and difficult things possible, and as a result, it's more feasible to turn a poor design or no design at all into working code (which then serves as a working proof-of-concept for the next redesign).

    Then again, maybe this isn't really that unexpected?

    P.S. Perl::Critic attempts to alleviate these sorts of problems.

  20. Flyspray is another nice option by Anonymous Coward · · Score: 0

    Flyspray is a lightweight and easy to use ticketing system worth checking out. We chose it after reviewing the main choices (bugzilla - too complex, mantis - not powerful enough, etc) available at the time and have been very happy. It has continued to improve nicely since, although development seems to be dragging a bit lately.

    http://www.flyspray.org/

  21. Perl community to Bugzilla community... by merlyn · · Score: 2, Insightful

    "We'd actually prefer if you STOP using Perl. You seem to be giving it a bad name. KTHX, Bye."