Frustrated package submitters?

[Hans de Goede]

Patrice Dumas wrote:
>>  - '''MUST''': For each binary and noarch RPM generated, reviewers 
>>    must verify that the list of provides ("rpm -q --provides ...") 
>>    are acceptable for this package.
> 
> That would have avoided the issue, I believe. And it is good practice, so 
> let's enforce it. It raises the cost of doing a review, but I believe it
> is worth it.
> 

I'm not so sure, the longer the MUST list, the less thorough each step 
will be done. We need to strike a balance here. This would make a fine 
should though.

Regards,

Hans