2016-03-05 16:09 GMT+01:00 Maarten Dammers maarten@mdammers.nl:
Hi Luca,
Op 5-3-2016 om 14:30 schreef Luca Martinelli:
Probably the threshold we set up for the conversion is too high, and this might be one of the causes why the whole process has slowed down to a dying pace.
You call https://www.wikidata.org/wiki/Special:Contributions/Maintenance_script a dying pace?
Instead of complaining here people should participate in https://www.wikidata.org/wiki/User:Addshore/Identifiers/0 . Still plenty of easy properties that are clearly distinct, unique and have an external url. It doesn't make sense to discus the more complicated cases if we haven't gotten the easy cases out of the way yet.
Point taken, I apologise for using too dramatic tones.
Nonetheless, I stick to the point that probably a ">99% unique identifier" threshold is too high. Just to make another example (disclaimer: I asked for this property since it is yet another catalogue that my institution runs), P1949 has not been converted to identifier because it has "only 98.82% unique out of 507 uses", that translates in only *six* cases out of 505 items which have two P1949 identifiers.
More, I did not intervene because of my blatant conflict of interest AND because I do not know with who discuss this and where, not even the general "what is an identifier" discussion. Probably there is a place where this discussion is going on, and I apologise again for not knowing (though I have some pretty good excuses), and I'm serious when I say that I'd be thankful to you if you please can point me in the general direction of where this is happening. :) (https://www.wikidata.org/wiki/User:Addshore/Identifiers maybe? Though that discussion seems to be pretty blocked)
L.