Re: gimp 3.0

[Thomas Klausner <wiz%NetBSD.org@localhost>]

On Tue, Apr 08, 2025 at 11:38:51PM +0100, Jonathan Perkin wrote:
> * On 2025-04-08 at 22:54 BST, Thomas Klausner wrote:
> 
> > Please let me know where you have a different view.
> 
> I will always prefer the previous behaviour.

The one where the bulk build didn't start?

> It is very important to me
> that bugs are never hidden or masked in any way, and are exposed as loudly
> as possible so that they are fixed as quickly as possible.

From the outside (i.e. if you are not the person running the bulk
build) the old behaviour (not starting) was only detectable by not
getting an email from a bulk build where I don't even know if it was
started or when. So, basically invisible unless the person running the
build manually sent an email about this.

Now I get a bulk build result where the packages with problems are
listed explicitly as standard result of a bulk build.

I know which one I find louder :-)

> I would rather
> not have a build published than potentially ship broken packages to my
> users.

But in all three cases there are no binary packages for the packages
with problems (and their dependencies). So there is no difference
here.

> However it's clear that I am alone in this opinion, so I stopped objecting
> to pbulk changes that I disagree with.  If you are actively monitoring and
> fixing scan failures from now on then we're all good.

Ok.
 Thomas