[aur-general] Idea for AUR improvement

[Hugo Osvaldo Barrera]

On 2012-06-02 11:55, Xyne wrote:
> Hugo Osvaldo Barrera wrote:
> 
>> I think a list of "packages I've contributed to" (similar to "my
>> packages", but also includes packages you've orphaned) in AUR would
>> solve this, and be helpful for other stuff.
>>
>> If a user leaves Arch for some reason, and comes back, IF he's
>> intereseted in re-adopted his orphaned packages, he'll just see that
>> list, and adopt them.
>>
>> Currently, it's pretty hard to know what packages you've contributed to
>> in the past, and it is something nice to have.
> 
> That poses two problems already raised in this thread:
> 
> 1) privacy issues: not everyone will want to be permanently associated with
>    packages

That's why I said "package *I*'ve contributed to"; each user can only
see him own contributions.

> 
> 2) backend complexity: each package would have to store a list of contributors
>    in the database

It's not really that complex.  You'd need a new table
("former-maintainer"?) for mapping users<->packages.

> 
> 1 would not actually be a list of contributors, only a list of current and
> former maintainers, as those who contribute via comments will not be tracked in
> this way. It thus defeats the goal of giving credit, but it would still work to
> track previous maintainers.
>

Yes, the list would actually be "packages I've maintained".

> 
> I lean towards the privacy argument on this and would prefer that we don't
> track every maintainer, but I don't see it as a big deal.
> 
> I also think that tracking the last maintainer would be much more useful than
> the submitter. Currently someone could easily adopt orphaned packakges, insert
> malicious code and then orphan them again. A last-maintainer field would enable
> use to determine who did that and deal with it.
>

Yes, that's exactly the point.  I've maintained packages in the past,
and I'm curious as to what happened to them.  Since I actually adopted
them (not submited), I've no way of easily listing them.


> 
> Now, switching submitter for last maintainer might be easy enough to do on
> the backend.

Yes, it makes much more sense; the last maintainer is way more relevant
than the submitter.  Complete rewrites are not uncommon, and the
submiter is irrelevant in those cases.

-- 
Hugo Osvaldo Barrera