Even though it is Friday again, I will try to respond without getting too emotional.
Graeme Geldenhuys wrote 5 lines
Vincent Snijders het geskryf:
> I only want feature requests that stand a reasonable chance to be
> implemented before version 3.0 (due after 2025).
And who decides the fate of such feature requests?
He who implements, decides. This is open source, you know ;-). And with all those distributed SCM systems, core developers stand no change impeding feature requests, if there is a developer who implements them. :-)
Show quoted text 5 lines
If you did not
notice, I first posted the idea here in the mailing list to be discussed
and get a feeling of what other developers think. I made it clear that I
do not know how others feel about this feature, but that I thought it
would be helpful.
I have no objection to a feature request, but a feature request without somebody who wants to implement is.
Show quoted text 7 lines
I did not want to "pollute" the Mantis tracker with
features only I want.
Mattias and Michael commented, without explicitly saying: NO THIS IS
RUBBISH! So based on their responses, I understood it that others also
feel this could be beneficial, but none of use have the time to
implement it right this minutes.
Hence, the Feature Request in Mantis -
so the idea is documented and not forgotten.
Still in Mantis it will be forgotten is my experience. Adding them to mantis has no additional value. So http://wiki.lazarus.freepascal.org/Feature_Ideas is as good.
BTW: This is the first time I hear that Lazarus developers do NOT look
at the Feature Request section in Mantis. It's news to me.
I can only speak for myself and how I see it used. Other Lazarus developers have to tell what they do with feature requests in the wiki. Take for example "Using navigation keys in a form at design time" :
http://bugs.freepascal.org/view.php?id=1738. Yesterday Paul committed some changes to implement this feature. I asked if he had seen this issue in mantis, he said no, but a colleague of him wanted it to be implemented very much. He didn't even know about it. If I would not remember it, I bad it would be there for a year or more, until somebody noticed that this was implemented a long time ago.
Show quoted text 5 lines
> You can. But IMHO this one is a bad example of such a feature
> request.
In your opinion maybe. By like I said, who decides what is good or bad?
Note the difference: This discussion is about what is a good or bad feature request. It is about what is a good or bad place to document a feature request. With assignee -> bug tracker, without assignee: http://wiki.lazarus.freepascal.org/Feature_Ideas
Example: a good feature request for the bug tracker is: http://bugs.freepascal.org/view.php?id=14618
Dmitry is maintainer of this component and the issue can be assigned to him.
As a Lazarus developer who uses the bug tracker a lot and wants to keep it manageable, I may not decide about it, but I fancy the idea I can make suggestions about its usage.
At least I posted my idea in the mailing list - to be discussed. You
might not have noticed, but I do this often before I report features
requests or bugs in mantis.
That is good, keep doing so. I read all those emails. I am sorry I could not respond between 11:26, the first answer to your feature request and 12:36. At that time I hoped "I'll simply add it as a feature request" meant adding it in the wiki. I was simply too late.
Show quoted text 7 lines
> I fear I offend people, if I close this issue with "No change
> required", "Won't fix" (it is not true, patches are excepted).
Why not simply allow the creator of the feature request or bug report to
delete their own reports.
I would be reluctant to that, maybe only for issues not yet resolved. It may cause valuable information to be lost. In svn log messages we often refer to issues in the bug tracker.
Vincent
