Arguing in that manner isn’t cricket.
I happen to agree with hadw and kocio. The point that language and its translation is a “sticky wicket” has been made. The problem in OSM, by its very nature, brings that sticky wicket into play, as has been demonstrated.
Now, I need to let go of this problem.
My apologies, but I didn’t seem to be getting through to you in other ways.
Not necessarily (although I think it would probably be for the best). As I said before, if you can come up with a concrete example where it would actually be useful and be better than the alternatives, then it would be time to re-examine it. The fact that you didn’t do so indicated to me that you hadn’t really thought things through. Then again, in the past my subconscious has come up with ideas that I couldn’t immediately consciously justify, so maybe there’s something waiting to surface.
As it happens, because I’m a contrarian by nature, after I argued against your position I then argued against my own position. 
It initially struck me that using IPA would be unworkable because very few people know it in full, so it would benefit very few users. However, on an active device (such as a phone) rather than a printout, it could be handed off to a text-to-speech app to provide a spoken rendering. But few mappers will know IPA, so take-up would be minimal. In any case it’s easier to have the local name on a map and point at it to a local than set up a system to vocalize IPA (that few mappers will enter and may require an internet connection to get speech from a server).
hadw made a point about Latin alphabet transliteration and whether that should be local transliteration or English one. If you use local transliteration then the orthography may only make sense to the locals (who could already read the local name in their own orthography) and may be garbled if interpreted as English orthography. If you use English transliteration then you favour English over other languages using the Latin alphabet. Plus there are phonemes in other languages (such as the Welsh “ll”) that are not unambiguously representable in standard English orthography.
Again, apologies for being a bit brutal.