Re: git config user.email

On Thu, Jul 22, 2010 at 5:41 AM, Magnus Hagander <magnus(at)hagander(dot)net> wrote:
> On Thu, Jul 22, 2010 at 11:33, Peter Eisentraut <peter_e(at)gmx(dot)net> wrote:
>> On tor, 2010-07-22 at 09:18 +0100, Dave Page wrote:
>>> On Thu, Jul 22, 2010 at 9:11 AM, Magnus Hagander <magnus(at)hagander(dot)net> wrote:
>>> > On Thu, Jul 22, 2010 at 10:04, Dave Page <dpage(at)pgadmin(dot)org> wrote:
>>> >> On Wed, Jul 21, 2010 at 5:54 PM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>>> >>> We need to decide what email addresses committers will use on the new
>>> >>> git repository when they commit.
>>> >>
>>> >> Are you are aware that we already have a list of "approved" addresses
>>> >> for the committers?
>>> >
>>> > Are you referring to the mapping list for the git mirror, or something else?
>>>
>>> Yes, the mapping list.
>>
>> The mapping list was originally composed by me on a whim based on what I
>> thought people's email addresses tended to be.  It wouldn't hurt to
>> ponder Robert's points at this time.
>
> Agreed. And per the discussion at the developer meeting, even if we
> don't limit what can be used, we should at least give committters a
> chance to pick a different address from the one they are on that list
> with today.
>
>
> *Personally*, I'd prefer to keep using my main email address for
> commits. This is what I use for all other projects (postgresql or
> others) that I commit or contribute to. It's an address on a domain I
> own, and fully control. It's a pretty clear indication of my
> "identity" in the opensource world, whereas close to nobody would know
> who mha(at)postgresql(dot)org is. Plus, email to it tends to be delivered
> much quicker and more reliably than the @postgresql.org one - though
> that has improvied significantly lately.
>
> But I can also see Roberts point. If a committer doesn't have a
> *stable* address, we won't be able to track this committer through
> time. Say if he changes job and gets a new address, we can start using
> that one for new commits, but not for old ones. And since we grant
> commit status to the *person* and not the representative of a company,
> using a company email address doesn't quite match up there.
>
> When it comes to using generic @gmail.com or whatever addresses,
> that's somewhere in between. For a lot of people, those can definitely
> be considered stable, because a change in employment, a move to a
> different country, things like that, won't affect the email address
> (which it would be if it was an ISP-specific one for example - that
> might not transfer to a new country or even a new city).

As for me, I'd much prefer to be rhaas(at)postgresql(dot)org than
robertmhaas(at)gmail(dot)com(dot) While it's true that I'm unlikely to lose
control of robertmhaas(at)gmail(dot)com, I might decide I'm no longer happy
with their service, or whatever. Assuming I stay on the sysadmin
team's good side, rhaas(at)postgresql(dot)org can always be repointed.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise Postgres Company