"Amitai Schleier" <schmonz%schmonz.com@localhost> writes:
> On 12 Jan 2021, at 16:19, Greg Troxel wrote:
>
>> It seems obvious that there might be a Free maintenance fork, but not
>> much else is obvious. I'm curious if anyone knows anything concrete.
>
> FWIW, I asked a colleague who has spent a lot of time with Qt, and he
> said "As far as I recall, this has how it's gone for most of the major
> releases. It sounds like there are issues around the state of 6.x and
> why folks have not ported to it over the past 6+ months that aggravate
> the inflammation. That's unfortunate. It will hinder folks that want
> to move to 6.x, but need some module or feature that was not carried
> from 5 to 6."
>
> It's not clear to me either what open source packagers are expected to
> do here. I guess we keep watching the situation until the choices
> become more evident.
I think what's really upsetting people is that 5.15 is the most recent
LTS and it is still the version that ought to be used for production
use. Taking fixes for ancient LTS proprietary is not such a big deal
compared to the LTS that everyone in the Free Software world ought to be
running. At least that's how it strikes me.
(Of course, users may wish to avoid programs that depend on qt as they
are able to do so reasonably, and upstreams may choose to stop using qt
or refrain from starting, as they are able, but that's outside the scope
of what pkgsrc should do.)
From the pkgsrc viewpoint, I think 5/6 is enough of a change that we
should
Keep qt5 as 5.15 indefinitely
If there is a Free fork of 5.15 for fixes, (and there's exactly one
that there is consensus about), repoint our qt5 to it. But that's
wait and see and should be discussed before it happens.
Add qt6 at some point (currently in wip). This may just be waiting
on round tuits.
Think about what to do when 6.2 arrives and then leaves support, but
that's years away so figure it out then.
Does that sound like the right approach?
Attachment:
signature.asc
Description: PGP signature