Forums » Suggestions
How about a proper Bugzilla ?
In my opinion the current bug submission interface is lacking at best. A proper bugzilla (or similar approach) would be beneficial to the community in a number of ways...
1) would allow the bug reporter to monitor the progress of bug resolution & status changes
2) would allow people experiencing the same bug to quickly find said bug and view bug status or potential bug resolutions if the bug is not to be fixed by VO staff (example: a distro-specific bug for the linux client)
3) would (hopefully) reduce duplicate bug reports
4) would allow people with similar system setups to potentially confirm a bug that isn't reproducable by VO/Guild staff and offer additional information that my lead to bug resolution
5) would offer a publicly accessible record of open and closed bugs, and offer an explanation of why a bug may recieve a 'won't fix' status or cannot be resolved in the near future
That's just off the top of my head, other things to consider:
1) bugzilla would NOT be used for feature request
2) bugzilla would NOT be used as a tech support forum (example: "I can't get GLX direct rendering to work" is not a bug)
3) bugzilla should be properly searched before filing a new bug to ease load on devs by reducing duplicate bugs
4) bugzilla could be extended by categories to include (but not limited to) graphical bugs, physics bugs, website bugs, mission bugs, etc... Omit billing system problems since personal information obviously shouldn't be shared on a publicly accessible forum.
I really can't think of a reason not to implement a proper bug reporting interface other than the initial setup time required
1) would allow the bug reporter to monitor the progress of bug resolution & status changes
2) would allow people experiencing the same bug to quickly find said bug and view bug status or potential bug resolutions if the bug is not to be fixed by VO staff (example: a distro-specific bug for the linux client)
3) would (hopefully) reduce duplicate bug reports
4) would allow people with similar system setups to potentially confirm a bug that isn't reproducable by VO/Guild staff and offer additional information that my lead to bug resolution
5) would offer a publicly accessible record of open and closed bugs, and offer an explanation of why a bug may recieve a 'won't fix' status or cannot be resolved in the near future
That's just off the top of my head, other things to consider:
1) bugzilla would NOT be used for feature request
2) bugzilla would NOT be used as a tech support forum (example: "I can't get GLX direct rendering to work" is not a bug)
3) bugzilla should be properly searched before filing a new bug to ease load on devs by reducing duplicate bugs
4) bugzilla could be extended by categories to include (but not limited to) graphical bugs, physics bugs, website bugs, mission bugs, etc... Omit billing system problems since personal information obviously shouldn't be shared on a publicly accessible forum.
I really can't think of a reason not to implement a proper bug reporting interface other than the initial setup time required
There is something to be said for having increased feedback to the userbase on bug status.
However, some of the "benefits" are not really beneficial to us: duplicate bug reports, for instance, demonstrate how widespread a particular problem is. This is why we ask people to report any/all bugs, especially crash related issues that actually spawn the bug reporter, even if they've submitted the bug before or know others have. Hardware and driver-specific bugs are much easier to track down when you can identify commonalities from a large number of reports.
The other side is that our current bug system is custom-written by us, and makes use of our automated trace reporting and so on, allowing us to sort by particular aspects of a bug. We could certainly shoehorn our custom functionality into bugzilla, or we could just expand our own thing to include more of the features you mention.. exposing our existing bug system would be pretty trivial. Either way, it probably won't be on the priority list for quite some time (until later this year, perhaps), although it's not a bad idea. We just have too many irons in the fire right now.
We did look at Bugzilla and various workalikes before we built our own thing. We could have used it, but it was kind of a.. very large and all-encompassing package that did vastly more than we wanted, and was a little over-complicated for our needs at the time. So we threw together a lightweight custom thing that we could easily change and customize to match our shifting requirements. It's less than ideal in a lot of respects.. it's really pretty bare-bones, but it works for now.
Thanks for the suggestion though, the ideas are well received.
However, some of the "benefits" are not really beneficial to us: duplicate bug reports, for instance, demonstrate how widespread a particular problem is. This is why we ask people to report any/all bugs, especially crash related issues that actually spawn the bug reporter, even if they've submitted the bug before or know others have. Hardware and driver-specific bugs are much easier to track down when you can identify commonalities from a large number of reports.
The other side is that our current bug system is custom-written by us, and makes use of our automated trace reporting and so on, allowing us to sort by particular aspects of a bug. We could certainly shoehorn our custom functionality into bugzilla, or we could just expand our own thing to include more of the features you mention.. exposing our existing bug system would be pretty trivial. Either way, it probably won't be on the priority list for quite some time (until later this year, perhaps), although it's not a bad idea. We just have too many irons in the fire right now.
We did look at Bugzilla and various workalikes before we built our own thing. We could have used it, but it was kind of a.. very large and all-encompassing package that did vastly more than we wanted, and was a little over-complicated for our needs at the time. So we threw together a lightweight custom thing that we could easily change and customize to match our shifting requirements. It's less than ideal in a lot of respects.. it's really pretty bare-bones, but it works for now.
Thanks for the suggestion though, the ideas are well received.