[Development] Remove OSX 10.6 Build?

Alexis Menard menard at kde.org
Fri Jan 24 11:36:06 CET 2014


On Thu, Jan 23, 2014 at 6:35 PM, Jan Farø <jan.faroe at gmail.com> wrote:

>
> On 24/01/2014, at 03.46, Alexis Menard <menard at kde.org> wrote:
>
>
>
>
> On Thu, Jan 23, 2014 at 5:16 PM, Jan Farø <jan.faroe at gmail.com> wrote:
>
>>
>> I don’t think anybody has mentioned the lack of ability to upgrade
>> hardware - mostly because of financial issues, I suppose. 10.6 is as far as
>> I know the last Mac OS to support 32 bit systems. Previous versions of my
>> own software supported PPC and down to Mac OS 10.4, which gave me a
>> considerable user base from that segment. Percentages aside, there’s still
>> a LOT of people sitting with old hardware, simply because they cannot
>> afford to upgrade.
>>
>>
> But is that a significant part of the Mac OS X users or users of Mac OS X
> Qt applications? I seriously doubt so. Let's be realistic, less and less
> software are supporting PPC nowadays, the best you can get is a 32/64 bits
> binary for Mac OS. Last machines from Apple with 32 bits only processor :
> 2006.
>
> One other point is that Qt5 is about QML and is pushing towards its usage
> on the desktop with better components for it with a modern GL scene graph.
> Running on outdated graphic cards with outdated graphic drivers is also not
> something people want to bother testing and fixing.
>
>
> I completely agree in regards to PPC support.
>
> In regards to users of Mac OS Qt applications: I’m am extremely confident
> that more Mac OS applications would be/have been written in Qt, if the
> priority for native looking widget support was higher. Mac OS users are
> notorious for their attention to detail and noticing a non-native L&F.
> Forcing application developers to resort to Objective C/Cocoa/style sheet
> hacks/whatever in order to make the UI look and behave more native sort of
> defies the notion of a cross platform framework.
>

Right but your point is about people with no condition to upgrade. You're
asking then the Qt project to support a fraction of users who have 32 bits
only machine so 8+ years old machines (if these are still alive).

Others can and should upgrade to Lion (it'd 29 USD if it's not free
nowadays). If they don't want to upgrade there is nothing we can do about
it but we're talking now of supporting a very little amount of users. Maybe
few of them will make the 29 USD switch if Qt doesn't support the old OS
they're running on.

Sorry I just can't make sense out of it.

Code freeze makes no sense also as it will stop improvements in Qt the way
it does today. I don't see how this improves the current situation : the
actual maintainers would like to simplify the entire Mac OS port. Freezing
these code paths is postponing future refactors or improvements Qt really
needs (ARC, ....).



>
>
> Again let's balance the cost of the maintenance of the code of 10.6 vs
> supporting few users stuck in the past? If they must stick in the past for
> various reasons (financial or others) then they can just use Qt4, it works
> just fine for Mac OS 10.6 or even Qt5 released versions. Why such users
> would care of modern Qt5 applications?
>
>
> Qt4 looks suboptimal on Mac OS. It still has problems with some of the
> list widgets. Among other things. Qt5 has several showstopper issues on Mac
> OS, some of which seems to finally being taken seriously (5.2.1?). You
> can’t ship a quality application on Mac OS with Qt5.0 - Qt.5.2.0.
>
>
>
This is another topic.


-- 
Alexis Menard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/development/attachments/20140124/c3b0c0bd/attachment.html>


More information about the Development mailing list