FreeCAD Contribution Process

Have some feature requests, feedback, cool stuff to share, or want to know where FreeCAD is going? This is the place.
Forum rules
Be nice to others! Read the FreeCAD code of conduct!
chrisb
Veteran
Posts: 53945
Joined: Tue Mar 17, 2015 9:14 am

Re: FreeCAD Contribution Process

Post by chrisb »

adrianinsaval wrote: Thu Nov 17, 2022 1:50 am - Critical or security bug fixes MAY be merged directly by maintainers
That would again be open to discussion: what is critical. I mean we don't control nuklear plants with FreeCAD and if something really exceptional will happen, then I am sure an appropriate solution will be found. Looking back what happened in the past, this is so rare that I don't think it should be incorporated in the general process.
A Sketcher Lecture with in-depth information is available in English, auf Deutsch, en français, en español.
User avatar
adrianinsaval
Veteran
Posts: 5541
Joined: Thu Apr 05, 2018 5:15 pm

Re: FreeCAD Contribution Process

Post by adrianinsaval »

https://github.com/FreeCAD/FreeCAD/issues/7910

I would like to hear other people's opinion about what to do with issues like this: severely lacking information, very poor description, impossible to reproduce as is. IMO this should be closed and the OP notified that they must provide more information so that the issue is reopened.

Also, in the issue @Kunda1 recommended asking in the forum, is this something that we should be recommending for possible bug reports? In the case of help requests I agree users should be redirected to the forum, but in the case of actual bug/problem reports IMO it is counterproductive to our goal (to move issue and technical discussions to github) to keep asking people to go to the forum.

And about help request, eventually we will get help requests incorrectly posted as issues, IMO it would be a good idea to put some message on the issue creation page encouraging users to ask questions and make help request on the forums and to use the issue tracker exclusively for problem reports / feature requests.

About the current issue creation page:
Captura de pantalla 2022-11-30 153739.png
Captura de pantalla 2022-11-30 153739.png (29.16 KiB) Viewed 485 times
The link to the forum currently ask that problems be discussed first on the forum, I suggest instead to use this to clarify the difference between a help request and a problem report and that help request should be made on the forum. This link should probably be first on the list too.
There is a template for issue report and problem report, this doesn't make much sense, probably one of the two should be removed (also there's a typo, "File an problem report").
Do we want to keep the feature request template? It seems the idea is now to categorize everything as a problem.
chrisb
Veteran
Posts: 53945
Joined: Tue Mar 17, 2015 9:14 am

Re: FreeCAD Contribution Process

Post by chrisb »

adrianinsaval wrote: Wed Nov 30, 2022 6:44 pm I would like to hear other people's opinion about what to do with issues like this: severely lacking information, very poor description, impossible to reproduce as is. IMO this should be closed and the OP notified that they must provide more information so that the issue is reopened.
I agree, that this is a completely useless error report, but I would not be too strict with closing. It is technically ok, if it is reopened, but is also rather repelling. So perhaps at least one attempt could be made to get sufficient information from the opener.
but in the case of actual bug/problem reports IMO it is counterproductive to our goal (to move issue and technical discussions to github) to keep asking people to go to the forum.
I agree. It would split the information, and that is what the developers didn't want, because the want it all in one place.
I suggest instead to use this to clarify the difference between a help request and a problem report and that help request should be made on the forum. This link should probably be first on the list too.
I agree only partially, because it should be clear that the bug tracker is there for bugs. However, people searching the internet may land on the tracker and then a redirecting message could be helpful.
There is a template for issue report and problem report, this doesn't make much sense, probably one of the two should be removed
I have asked about this here: https://forum.freecadweb.org/viewtopic. ... 72#p643668. Still waiting for @sliptonic to weigh in.
A Sketcher Lecture with in-depth information is available in English, auf Deutsch, en français, en español.
GeneFC
Veteran
Posts: 5373
Joined: Sat Mar 19, 2016 3:36 pm
Location: Punta Gorda, FL

Re: FreeCAD Contribution Process

Post by GeneFC »

This was one of the primary cautions raised before the push to move all bug reports to github. A report such as this one would have been quickly dealt with on the forum. Either fleshed out into something useful or discarded.

Not saying the "old way" was better, but this will be a continuing problem.

Personally I would simply close the report with a statement that it is too ill-defined. The question is who should have the authority to close. Certainly we would not want random github users to arbitrarily close reports.

Gene
User avatar
adrianinsaval
Veteran
Posts: 5541
Joined: Thu Apr 05, 2018 5:15 pm

Re: FreeCAD Contribution Process

Post by adrianinsaval »

GeneFC wrote: Thu Dec 01, 2022 12:36 am This was one of the primary cautions raised before the push to move all bug reports to github. A report such as this one would have been quickly dealt with on the forum. Either fleshed out into something useful or discarded.

Not saying the "old way" was better, but this will be a continuing problem.
yes, but I think we were all aware of this, IMO this is not a big deal as long as we don't waste much time on them, this can be dealt with as quickly or quicker than in the forum on github, just close the issue and tell them that they must provide more details for the report to be valid, if they then provide those details the issue can be reopened.
The question is who should have the authority to close. Certainly we would not want random github users to arbitrarily close reports.
Of course not, to be able to close the issue you have to be given permissions for this (or be the creator of the issue), we probably need more people with this permission than now though.
chrisb wrote: Wed Nov 30, 2022 8:26 pm So perhaps at least one attempt could be made to get sufficient information from the opener.
I don't propose to close and shutdown all talk about the issue, if the OP then provides meaningful information the issue can be reopened, actually the OP can reopen it themselves.
I don't think it's a big deal to close an issue, keep in mind that the alternative of asking for info and waiting for response before closing has a higher maintenance burden because you have to ask then wait (how long btw?) and then somebody has to remember to go and actually close the issue, there's a very real chance that the issue will remain open for a long period of time until someone else spends some time on the useless task of reading the issue and finally making the decision of closing (instead of working on real issues).
On the other hand if you directly close and ask that they give proper info for a reopen then the maintenance burden ends there, because if the user then eventually responds and gives information they will reopen the issue themselves or create a new one, or if the user doesn't realize they can reopen it is trivial to make the decision because you already have the info, no additional effort needed and the useless issues that would go stale are easily filtered out by this.
Post Reply