[Development] Error database cleaning continues

[Giuseppe D'Angelo]

Il 14/09/26 11:57, Jouni Äijälä via Development ha scritto:
> Error database cleaning process is going forward and Qt 4 bugs are now 
> closed. The next step is to continue the process gradually with Qt 5 
> bugs starting from 5.0.* Qt version. If you know that a bug reported by 
> you can be reproduced also with supported Qt version please add the 
> correct “Affects versions” label. Otherwise it will be closed. Note that 
> the bugs can be reopened later on if needed but if the bug is reopened, 
> evidence needs to be provided that the bug is valid with supported 
> version. “Not_in_maintenance_scope” label will be added for further search.

I must say I'm not pleased at all with this process. From the 
notifications I'm getting, a ton of widgets-related bugs are being 
closed "simply" because they're old. With QtWidgets being "done" both in 
Qt 5 and Qt 6, it's extremely likely that many these bugs are still 
completely valid. We're just shoving the dust under the carpet, open bug 
reports counter goes down, everybody happy?

The reasons for the lack of activity on the bugtracker are IMNSHO 
completely independent from the validity of the issue. They're more to 
be found in the general lack of upstream response to these bugs (or 
willingness to fix them, given again, widgets are "done"); and/or in the 
general huge disengagement of the community from Qt, which is now seen 
first and foremost as a commercial product, so why should one spend the 
effort to improve it (... notice how this very email thread was sent on 
the _wrong_ mailing list -- not the one(s) for the users of Qt; nor it 
was cross-posted to Qt's blog, newsletters and other announcements); 
and/or in the fact that the quality of web searches has plummeted in the 
last few years, so people hitting a bug don't even see that there is a 
bug report opened about their issue; and so on.

Also, it's good netiquette *not* to comment under a bug report just to 
say "this is still valid". We don't put undue pressure on maintainers. 
They are the ones supposed to check if a bug is still valid, not the 
other way around.

If anything, this exercise could've been a perfect use case for using an 
AI agent, e.g.:

* have it check if a bug has a proper description of the issue;
* have it check if it has a minimal reproducer attached, or have it 
*create* one if missing (or minimize the existing one if not attached);
* have it build and run and see if the issue is still there;
* have it create patches adding the relevant autotests into Qt proper 
(either passing or XFAIL);
* leave comments / add labels as needed, depending on the results of the 
above.


My 2 c,
-- 
Giuseppe D'Angelo | giuseppe.dangelo at kdab.com | Senior Software Engineer
KDAB (France) S.A.S., a KDAB Group company
Tel. France +33 (0)4 90 84 08 53, http://www.kdab.com
KDAB - Trusted Software Excellence
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4889 bytes
Desc: Firma crittografica S/MIME
URL: <http://lists.qt-project.org/pipermail/development/attachments/20260925/1c0d1afd/attachment.bin>