Anthere wrote: As an outsider to this argument, I know little other than what I have read on the mailing list and in articles, so I apologise in advance for any inaccuracies, but I think it goes something like this: To a close approximation, Moldovan == Romanian, but written in Cyrillic, and is used in Moldova and Transnistria, which border Romania, and are closely historically and culturally related. There are considerably fewer mo: speakers (3.4 M in Rep. Moldova, 0.5M in Transnistria) than ro: speakers (24 M). See [[Moldovan language]] for details. However: Romanians have had an unhappy history with all things Russian (see [[History of Romania]]) and the historically-recent carving up of the [[Principality of Moldavia]], and many feel _very_ strongly, for political, historical, and cultural reasons that they don't want to have _anything_ to do with the use of Cyrillic to write Romanian, to the point that they took the Cyrillic character off the Wikipedia globe in the logo for the ro: Wikipedia. This is a _major_ nationalist issue for many Romanians. So, politics and script system appear to be very tightly coupled. There only seem to be three practical ways forward: 1 Maintain the status quo, and don't create an mo: Wikipedia 2 Create a separate mo: Wikipedia in Cyrillic, effectively duplicating the ro: Wikipedia in a different script. 3 Two front ends, one database. Add dual-script interface support to the ro: Wikipedia, but: unlike with the zh: Wikipedia, a dual-script interface within a single domain will _not_ appeal to those who are against having even a single Cyrillic letter on the logo, so there will have to be two domains, ro: and mo:, the former with a Latin-only interface and article display, and the latter with either Cyrillic-only or dual-script support, both using the same back-end database, which must, I believe, remain entirely Latin-based for technical, historical, and diplomatic reasons. My personal preference is option 3, since option 1 disenfranchises almost 4 million people, and option 2 effectively forks the ro: Wikipedia, and will in any case most likely simply be filled up with machine-transliterated articles from ro: However, to implement option 3, there will need to be not only a lot of consensus-building, but also the implementation of a reliable, round-trippable, Latin <-> Cyrillic transliteration system for Romanian, which does not yet exist. There are several options which are _nearly_ workable, but work would be needed to make this happen, and would require expertise from both the mo: and ro: communities. Whichever option you choose, you will make some people _very_ unhappy. -- Neil