Oups, missed the - issue :/
IMHO, rules should never give us less freedom.. Can't it be the exception that confirms the rule ? Do we have to live without a _() documentation just for historical reasons ? Maybe we can break rules for one time.. What will happen if we add _.xml ?
BTW, do livedocs solve the problem ? (didn't have the time to look at it)
There are a *lots* of tools dependant on the _ to - conversion (including the online function search, some of the CHM features, many quickref tools built into IDEs and browsers, the PHP docref() feature, the revcheck file output, etc). This is not a rule we have a chance to think about changing...
Goba
