[teampractices] [Wikitech-l] RfC on Product Management Tools and Development Toolchain

[Guillaume Paumier]

Hi,

On Tue, Apr 15, 2014 at 6:19 PM, Dan Andreescu <dandreescu at wikimedia.org> wrote:
>
> It's my understanding that the current Phabricator instance is temporary.
> Indeed, it includes mostly jokes and throw-away testing tasks, comments,
> boards, etc.

Actually, we're using the test instance to project-manage the
Phabricator RFC itself, by tracking missing features compared to our
existing tools, and planning for the possible migration if the RFC is
successful. So it's not just jokes and throw-away content :)

That said, that project is probably the only one that we can get away
with hosting there, because if we end up keeping the status quo, we
won't have a need for that project any more and we can safely delete
the test instance.

>  Could we stand up an instance to which we would potentially
> migrate to?  We would have to reserve task ids 1-100000 to allow us to port
> from Bugzilla, but do we have any other blockers?

I understand where you're coming from, but I would caution against
this course of action. "Testing" shiny new tools with real projects is
one of the reasons why we now have so many tools. If a Phabricator
instance is created for "real" use, and the RFC is unsuccessful, then
we'll have yet another tool on our hands (obligatory xkcd reference:
https://xkcd.com/927/ ).

At the moment, it's difficult to see if there are other blockers than
just reserving IDs. I agree that having a production instance now (and
migrating later) would be easier for the RFC, but in order to have
resources allocated to plan and work on the migration, there needs to
be an explicit show of interest in the RFC. I realize it's a bit of a
snake-eating-its-own-tail situation; the throw-away test instance if
the best option we've come up with for now.

-- 
Guillaume Paumier