Re: Old, stale bug reports (bug triage)
On Sat 29 Apr 2017 at 14:29:19 +0200, Felix Dietrich wrote: > Jonathan Dowland writes: > > > On Wed, Apr 26, 2017 at 02:53:28AM +0200, Felix Dietrich wrote: > > > When closing, annote the bug with found/fixed version information. > > If you are using the bts command-line helper, you could do > > (for the ficftional bug number ABCDEF, which was found in a version > > 2016.15-4, and fixed in a version 2017.4-1): > > > > bts found ABCDEF 2016.15-4 , fixed it 2017.4-1 , done it 2017.4-1 > > I hadn't come across that helper program, yet. Thank you. > > > At the end of the day, if the maintainer disagrees, they can reopen the > > bug. > > I followed Ben Finney's advice and sent an email to the report > requesting information regarding the current state of it. I will wait > awhile for a response before I am going to close the bug report. While you are there you could consider closing the five wishlist reports. All are from 2009/2010. 7/8 years has passed and no one has stepped forward to respond significantly to these enhancement requests. They are highly unlikely to be fulfilled; the reports serve no purpose. Has upstream altered the software to accomodate them? How likely is it that anyone will? The reports will linger in the BTS for the forseeable future. Their disappearence will not reflect on the quality of Debian or the good intentions of the submitters. Get rid of the reports. Suppose 100 readers of this list did the same for 5 other packages each. That would be 2500 closed bugs. Then the same readers get carried away and look at the other bugs in their chosen packages. Where could this end? 25,000 closed bugs? It doesn't happen until users make it happen. Maintainers can only do so much. -- Brian.
Re: Old, stale bug reports (bug triage)
On Sat 29 Apr 2017 at 14:14:01 +0200, Felix Dietrich wrote: > songbird writes: > > > does installing package how-can-i-help help? > > Thanks for the suggestion. I will have to spent some time going through > that list and see if I can find anything that I am able to do. As Jonathan Dowland said > . there are far, far too many bug reports and not enough people > working on them. This is one area that people can really make a > difference in contributing to Debian. Never a truer word was said. The bug count increases and the bug record for well-used packages continues to increase. For a start, this makes it difficult for a user with a problem to sort the wheat from the chaff. Culling the number of reports is a task which does not necessarily require much time or deep understanding of the package or the way the software works. Just some commonsense and the support of the maintainer. The result would be a cleaner and more useful database of bug reports. For example, are reports about issues on Squeeze relevant today? The chances are many will have been fixed by now. Perhaps a lttle judgement would be required, but, in all probability, most could be closed. Wheezy too. Consult the submitter to see whether the bug is still reproducible in unstable. No reply? Close the report after a reasonable time has passed. Five-year old wishlist bugs? Read; come to a judgement; close (giving a sound, decent reason); most are past their sell-by date. -- Brian.
Re: Old, stale bug reports (bug triage)
Jonathan Dowland writes: > On Wed, Apr 26, 2017 at 02:53:28AM +0200, Felix Dietrich wrote: > When closing, annote the bug with found/fixed version information. > If you are using the bts command-line helper, you could do > (for the ficftional bug number ABCDEF, which was found in a version > 2016.15-4, and fixed in a version 2017.4-1): > > bts found ABCDEF 2016.15-4 , fixed it 2017.4-1 , done it 2017.4-1 I hadn't come across that helper program, yet. Thank you. > At the end of the day, if the maintainer disagrees, they can reopen the > bug. I followed Ben Finney's advice and sent an email to the report requesting information regarding the current state of it. I will wait awhile for a response before I am going to close the bug report. > Thank you for wanting to make Debian better! Thank you for doing so already. -- Felix Dietrich
Re: Old, stale bug reports (bug triage)
songbird writes: > does installing package how-can-i-help help? Thanks for the suggestion. I will have to spent some time going through that list and see if I can find anything that I am able to do. -- Felix Dietrich
Re: Old, stale bug reports (bug triage)
Felix Dietrich wrote: ... > I had considered the mentors list, but, while using the bug tracking > system might start to border on the purview of that list, a quick > skimming over its archive left me with the impression that the messages > are generally package sponsering requests or otherwise about packaging – > my question felt more like that of an interested user who has started to > explore the possibilities of contributing to Debian. does installing package how-can-i-help help? (i've not looked in a while to see what it now contains, but...) songbird
Re: Old, stale bug reports (bug triage)
Cindy-Sue Causey writes: > Is this also something where Felix would participate actively over at > Debian-Mentors? > > https://lists.debian.org/debian-mentors/ I had considered the mentors list, but, while using the bug tracking system might start to border on the purview of that list, a quick skimming over its archive left me with the impression that the messages are generally package sponsering requests or otherwise about packaging – my question felt more like that of an interested user who has started to explore the possibilities of contributing to Debian. -- Felix Dietrich
Re: Old, stale bug reports (bug triage)
On Wed, Apr 26, 2017 at 02:53:28AM +0200, Felix Dietrich wrote: > Looking through bugs in the Mnemosyne package I noticed a couple of > old and stale reports that could have already been closed. This is an endemic problem in Debian: there are far, far too many bug reports and not enough people working on them. This is one area that people can really make a difference in contributing to Debian. Having said that: be aware that, for some packages, triaging bugs is possibly not the best use of time. For example, a package which is not likely to remain in Debian (or, in someone's estimation, should be removed), effort might be better spent on the removal, or packaging an alternative, or figuring out a migration strategy for users from the old package to an alternative. > What is the etiquette dealing with seemingly forgotten bug reports? > Should I send an email to the maintainer that asks him to have another > look at the report and decide for an appropriate action; may I close the > report myself or set e.g. the "wontfix" tag; something else? If you are unambiguously sure that the packaged version of the software no longer has that bug, then you can close it yourself. Note that if it is fixed upstream, you still need to be sure that it is fixed in Debian too: sometimes Debian packages deviate a lot from upstream wtih patches which means it might not be clear. When closing, annote the bug with found/fixed version information. If you are using the bts command-line helper, you could do (for the ficftional bug number ABCDEF, which was found in a version 2016.15-4, and fixed in a version 2017.4-1): bts found ABCDEF 2016.15-4 , fixed it 2017.4-1 , done it 2017.4-1 At the end of the day, if the maintainer disagrees, they can reopen the bug. Thank you for wanting to make Debian better! -- ⢀⣴⠾⠻⢶⣦⠀ ⣾⠁⢠⠒⠀⣿⡁ Jonathan Dowland ⢿⡄⠘⠷⠚⠋⠀ https://jmtd.net ⠈⠳⣄ Please do not CC me, I am subscribed to the list.
Re: Old, stale bug reports (bug triage)
On 4/26/17, Ben Finney wrote: > Felix Dietrich writes: > >> What is the etiquette dealing with seemingly forgotten bug reports? > > You should start with a message to the bug report email address. That > will reach both the current owner of the report, and the package > maintainer address. > >> […] to have another look at the report and decide for an appropriate >> action; may I close the report myself or set e.g. the "wontfix" tag; >> something else? > > Closing the report should be done only if that is clearly the > conclusion. As a newcomer you should not do that except where there is a > clear statement that the report has nothing further to be done. > > Same with setting a “wontfix” tag; that is an expression of the > maintainer's position, and so it needs to be clearly expressed by the > maintainers. > > You can begin with just a “What's the latest on dealing with this > report?” message. More often that you might expect, that prompts someone > to give an update on new information that never made its way to the bug > report. Is this also something where Felix would participate actively over at Debian-Mentors? https://lists.debian.org/debian-mentors/ It does sound like a "maintainer" type question since that's on average who I usually see messing around with the bug reports over at Debian-Mentors... but I, of course, could always be wrong on that point. *grin* Just thinking out loud... :) Cindy :) -- Cindy-Sue Causey Talking Rock, Pickens County, Georgia, USA * runs with duct tape *
Re: Old, stale bug reports (bug triage)
Felix Dietrich writes: > What is the etiquette dealing with seemingly forgotten bug reports? You should start with a message to the bug report email address. That will reach both the current owner of the report, and the package maintainer address. > […] to have another look at the report and decide for an appropriate > action; may I close the report myself or set e.g. the "wontfix" tag; > something else? Closing the report should be done only if that is clearly the conclusion. As a newcomer you should not do that except where there is a clear statement that the report has nothing further to be done. Same with setting a “wontfix” tag; that is an expression of the maintainer's position, and so it needs to be clearly expressed by the maintainers. You can begin with just a “What's the latest on dealing with this report?” message. More often that you might expect, that prompts someone to give an update on new information that never made its way to the bug report. -- \ “If we could change ourselves, the tendencies in the world | `\ would also change.” —Mohandas K. Gandhi, _Collected Works_, 1913 | _o__) | Ben Finney
Old, stale bug reports (bug triage)
Looking through bugs in the Mnemosyne package I noticed a couple of old and stale reports that could have already been closed. E.g. the discussion in report #783929 ends with the upstream developer saying that Mnemosyne is working as intended and in the last message he simply writes "Done." – yet, the report has no been closed. What is the etiquette dealing with seemingly forgotten bug reports? Should I send an email to the maintainer that asks him to have another look at the report and decide for an appropriate action; may I close the report myself or set e.g. the "wontfix" tag; something else? -- Felix Dietrich
