On Mon, May 25, 2015 at 8:34 AM, Mike Connor <[email protected]> wrote:
> > > On 25 May 2015 at 11:18, Eric Rescorla <[email protected]> wrote: > >> On Mon, May 25, 2015 at 8:10 AM, Mike Connor <[email protected]> wrote: >> >>> On criteria, let's not get too far into the weeds here. Hard cases >>> generally make bad law. >>> >> >> But we already know that there are hard cases that we need to deal with. >> Indeed one such case is the origin of this thread. >> > > The origin of this thread was from a legacy entry that mis-identified > Kosovo as a part of Albania. I don't think that passes any reasonable > sniff test. > And yet people clearly want to not change it. Long experience in other fora shows how controversial this is. > I'll take good judgement (and reasonable discussion over disagreements) >>> over trying to introduce a finely-detailed rule set. >>> >>> So, as a way to move forward, I'd propose that we do the following: >>> >>> 1) Use the Mozillians API in all cases to apply a uniform standard >>> across all Mozilla sites. >>> >>> 2) Standardize on "Country or Region" or "Location" or similar for >>> fields to avoid sovereignty implications >>> >>> 3) Use ISO-3166 as a base, and put the Mozillians owner (or their >>> delegate) in charge of deciding on any variations (Taiwan, Kosovo, etc) >>> from the base spec. >>> >> >> I don't agree with this for the reasons Adam Roach so eloquently observed. >> > > Adam's argument seems generally rooted in a desire to avoid making > decisions. I do not believe blame avoidance should be a primary goal here. > Avoidance of long divisive discussions on topics where we have no particular expertise is not at all the same thing as blame avoidance. > As Tim and others have pointed out, the list is flawed because of UN > politics, > All lists will be flawed. > so all we really get is someone to blame. > No, we get to rule these discussions out of scope. It's different. 4) Identify .governance as the point of escalation if someone wants to have >>> a wider discussion on an owner decision to include/exclude a proposed entry. >>> >> >> A big public discussion about whether X is a valid region. What could >> possibly go wrong? >> > > We'll have it no matter what we choose as a policy. > We don't actually need to have that if we defer to ISO. People can of course argue, but since those arguments won't have an actionable result, they can be safely ignored. This feels like the least bad option, and provides an >>> >> > escalation path towards a wider discussion on the merits of an individual >>> situation, rather than strawman arguments about where a theoretical line >>> should go. >>> >>> Thoughts? >>> >> >> I think this is a bad idea. Much better would be to just use 3166 as the >> entire prepopulated list and let people write in whatever they want, but >> not add it to a list we ratify. >> > > And that's not going to work for technical and abuse reasons. > I don't think that's at all obvious. The technical reason seems to be that people won't be able to find each other, but that's just not true: people can converge on the same value in the same way that write-in ballots in elections work. As for abuse, in a writein field it's much easier because we just need to filter for obviously objectionable content in the same way we do in (say) people's names, not make political decisions. > And leaves the original, clearly wrong example on the table, except we'd > get to point the finger elsewhere. > Not having this be our problem seems to have a lot of merit > Do we care more about inclusiveness or blame avoidance? That's where the > argument seems to hinge at this point. > I don't think that's an accurate summary at all. -Ekr _______________________________________________ governance mailing list [email protected] https://lists.mozilla.org/listinfo/governance
