[Development] Trouble getting reviews in months and years

[Ilya Fedin]

On Mon, 14 Oct 2024 09:15:45 +1000
Lorn Potter <lorn.potter at gmail.com> wrote:

> Other open source projects are the 
> same. You might have to ping people on the mailing list, 2, 3 times,
> or one of the several developer chat places like discord and irc. 
> Developers are often busy with their own things.

Yeah but such projects are usually unmaintained (I mean they could
still have developers but having no one acting as a maintainer).

There were such answers on this mailing list though:

On Tue Apr 9 03:29:09 CEST 2024
Chris Adams <chris.adams at qinetic.com.au> wrote:

> In my opinion, 6 months or so delay on some reviews is fairly
> acceptable (there can be higher-priority items which take attention,
> it's understandable).

On Wed Apr 10 06:04:25 CEST 2024
Thiago Macieira <thiago.macieira at intel.com> wrote:

> Don't wait months and years: you're 
> entitled to getting a review within one month. 

It sadly doesn't seem to work...

On Mon, 14 Oct 2024 09:15:45 +1000
Lorn Potter <lorn.potter at gmail.com> wrote:

> It happens to me heaps, but I am partially to blame, as I am the 
> developer trying to get changes into a huge code base. Not every 
> reviewer will know a lot about the bug or code/module being fixed, as 
> well. Especially something like X11 code that goes way back.
> 
> Ya just gotta keep trying. If someone disagrees with the premise you 
> either have to just drop it, explain things better with more detail,
> or go back and work on the bug fix more.

Qt sadly has reputation of having the worst (after Electron perhaps)
Linux support. It had big troubles with Wayland for a long time and
still has noticable troubles such as struggling with providing API
fitting the Wayland way of popup positioning. It has poor
xdg-desktop-portal support. Weird bugs in X11 support other toolkits
don't have.

The near-impossibility to convince maintainers, requirements for
bureaucratic paper trails and silence in reviews don't really help on
that. Qt maintainers seem to have no capacity/interest to make those
improvements and don't let others to make them.

Even when maintainers don't disagree, the silence itself is enough for
a patch to get lost if the author needs it less than Qt needs it
(i.e. has no reason to do pings for years or trying to go to mailing
list).