Re: [GNU-linux-libre] Amendment proposal for incoming distros page

[bill-auger]

On Thu, 2018-05-03 at 20:26 +0000, André Silva wrote:
> Seems Hyperbola endorsement process (to be evaluated by community) is
> stalled for now.
> 
> It's so early to be officially considered as stalled, however the first
> thing that came to my mind was in which if it persists for a long time
> (months, years, etc), there isn't an amendment for those cases. What is
> the next step that should we do for it? Shouldn't we create an amendment
> that is needed find a review manager to handle it (eg. the same
> application manager)?
> 

the 4th bullet point on the "Incoming_distros" page was meant to address the
case of a stalled review[1] - it was added as a exit plan for if a distro has
known FSDG issues that are not addressed in a timely manor by the maintainers -
in which case they should be moved to the "does not comply" section and the
review manager can be relieved of over-sight

it does not specifically speak to the potential case where there are no known
problems found with the distro; but the review has simply stalled or the review
manager is unresponsive - in which case it would be incorrect to move it to the
"does not comply" section; but there is no other category to move it to

there is no clear remedy either other than finding another review manager, much
the same as if it were the first day; so probably there would be no new wording
to add to the procedure - i think it already says something to the effect: "...
unless a new review manger can be found"

surely, it is not a perfect system - all reviewers are volunteers and it is not
at all clear how many or few there are at any time - the real solution would be
to assemble a healthy team of volunteers

this thread has made me realize what a big task a thorough review really is - if
the requirement was a 100% comprehensive audit if the entire distro, all
software documentation, and rebuilding from complete corresponding sources; i
would think it would take a team of reviewers working full-time a good part of 
ayear to complete - the actual requirement is not so stringent, but there is no 
full-time team either - to do a reasonably thorough job with a very small team 
of part-time volunteers, i would expect to take a very long time with several 
jerks and lulls along the way

i will add my personal opinion here, that i would not see it as unreasonable if
the review process were mandated to last exactly two years, no more and no less
- that could allow ample time for the review process and for the distro to
address any FSDG issues; and it would also ensure that the distro and it's
management were capable of lasting two years before being endorsed - i could see
some extra value in that of the confidence that the distro has been in
communication with and in good standing with the FSDG reviewers for something
that resembles "the long term" before being endorsed - there is surely less
confidence when the review process is brief, or if the guidelines are not
applied perpetually - that was the topic of a rather lengthy discussion recently
- and i will add that confidence is diminished in distros that do not
participate in these discussions before, during, and after they have been
endorsed

probably the only way to prevent a total stagnation would be to keep some
conversation going during the lulls - i dont think it is reasonable to put any
time limits on the actual progress points without a solid dedicated team of
reviewers - i would only suggest extreme measures (such as replacing the review
manager or cancelling the review) in cases where there has been absolutely no
communication for several months


[1]: https://libreplanet.org/wiki/Incoming_distros

signature.asc
Description: This is a digitally signed message part