Proposed package blocking due to FTBFS

[Ralf Corsepius]

On 12/08/2010 09:50 AM, Adam Williamson wrote:
> On Wed, 2010-12-08 at 01:05 +0000, Bastien Nocera wrote:

> I agree it's a bit questionable whether we should block packages for
> FTBFS,

IMO, there can't be any doubt about FTBFS's to be "must fixes" and them 
to release blockers for packages being affected.

Rationale:
- It's important packages are buildable at any time, to be able to 
quickly react on bugs.

- Packages, which are hit by FTBFS issues often suffer from other but 
"mere technical issues", e.g. maintainers having gone AWOL, the package 
being of low quality, maintainers not being sufficiently skilled etc.

- Packages, which are hit by FTBFS issue often reveil hidden packaging 
issues (e.g. broken deps having silently being introduced), which should 
be addressed as soon as possible to prevent other packages from being 
infected with "work-arounds" (e.g. redundant package deps or 
configuration hackery).

- FTBFS issues occasionally reveil global issues, which so far have 
managed to get away unaddressed/unnoticed (e.g. compiler bugs, toolchain 
issues etc.)

Ralf