Re: removing permissions for long unused accounts?

[Bruno Haible]

Hi Paul,

> Also, by "active" do we mean "authored a patch", "committed a patch", or 
> "pushed a commit to Savannah"? I assume pushing is what counts. Dunno if 
> that's easily measured, though.

Pushing is what counts, yes. There is nothing to do for contributors who
send us patches for us to commit.

It is easy to measure, e.g.
$ name='Joel'; git log | grep -C 1 "Author: $name " | head -n 3

> how about if we remove people who haven't contributed in a 
> year to any GNU project?

Any GNU project - this is harder to measure. You might get some infos
through https://www.openhub.net/people .

But we don't need to have a fixed rule. Exceptions are also OK; it's
only the approximate reduction of risk that matters.

Who wants to do it, you or Jim? (I am not an admin on Savannah.)

I would find it important, out of courtesy, that after establishing
the list of people and before removing their write access, we notify
them in a mail, with emphasis that they are welcome to contribute again.

Bruno