[Qgis-developer] QGIS 2 64bits, is it stable ?

[Matthias Kuhn]

I think we have to be fair with the current state of the project in 
this discussion.

The stability of the current release is heavily influenced by 
increasing the major version number. What I mean with this is, that 
because for the next release only the minor number will increase, I 
expect it to be to be more stable and not less. Regardless of testing.

I'm by no means opposed to testing. I would like to encourage everybody 
to write test-cases. But, recently it appears to me, that testing is 
proposed as cure to all evil, which I don't believe. It can surely help 
to get feedback about broken functionality fast, but in the end it is 
also additional effort to implement these tests and keep them up to 
date. Let me quote Martin Grässlin "I rather use a working system 
without unit tests than a system with unit tests that doesn’t work" [1].

So right now, I think it is important to fix bugs. If somebody pays a 
developer to fix a bug, please feel free to include a regression-test 
in this contract. It will be appreciated. But even more important: 
Please DO consider hiring a developer to fix bugs at all.

Matthias

[1] http://blog.martin-graesslin.com/blog/2013/05/mir-in-kubuntu/

On Fre 04 Okt 2013 12:01:53 CEST, Jonathan Moules wrote:
>     > maintainance costs, for which it is always difficult to find funding.
>     > We need more unit test, more code quality dashboards and much
>     stricter rules
>     > relative to what code has to be accepted into master.
>
>     I think this is a matter of balance: a large part of QGIS success
>     is due to the huge
>     number of new functions and developers that keep on coming.
>     Setting up to srtict
>     rules will dry up our main source.
>
>
> I have to respectfully disagree with the premise behind this. I'm sure
> ease-of-committing has facilitated much of the development, but at a
> certain point in a popular project's life it should become "mature" -
> unit tests, code reviews, etc. This will make it harder to commit that
> cool new tool that someone hacked up over the weekend, but you know
> that when it is committed and approved, it is a lot less likely to
> break something else.
>
> Yes QGIS is almost undoubtedly the most popular FOSS GIS on the
> planet, but it will struggle to maintain that reputation if lots of
> users start encountering regressions and bugs and crashes.
>
>
>     > A PostgreSQL-like code inclusion workflow, with commitfest and
>     review, could be
>     > something interested, to be discussed.
>
>     IMHO a desktop program is a totally different beast from a server
>     one. Unit testing
>     for atomic functions is relatively easy, it can be a nightmare
>     when you have very
>     complex user interactions.
>
>
> Maybe the question should be - how do other successful Open Source
> desktop applications do it? Could QGIS not find some other projects
> that release regular relatively bug-free builds and ask them what
> their process is? QGIS isn't the first project to encounter these
> problems; it can learn from others. I don't know the answers which is
> why I'm posing them as questions - but maybe someone else does.
>
> Cheers,
> Jonathan
>
> This transmission is intended for the named addressee(s) only and may
> contain sensitive or protectively marked material up to RESTRICTED and
> should be handled accordingly. Unless you are the named addressee (or
> authorised to receive it for the addressee) you may not copy or use
> it, or disclose it to anyone else. If you have received this
> transmission in error please notify the sender immediately. All email
> traffic sent to or from us, including without limitation all GCSX
> traffic, may be subject to recording and/or monitoring in accordance
> with relevant legislation.
>
>
> _______________________________________________
> Qgis-developer mailing list
> Qgis-developer at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer