View Single Post
  #3 (permalink)  
Old 15th July 2006, 02:49 AM
Giant Giant is offline
Senior Member
Join Date: Dec 2005
Location: Canada
Posts: 1,806
iTrader: (19)
Rep Power: 735
Giant is an unknown quantity at this point
Re: Verisign Encoding Issue

Originally Posted by blastfromthepast
1. The devanagari encoding mystery has been solved: some registrars were not doing nameprep correctly. New registrations are now going through correctly. If you purchased the domain early in the game, (like Laura Snow did with Sari) you may have gotten an incorrect domain. New typins should resolve to the correct domain if the browser also implements nameprep the right way, which they should. Google searching is another matter. They need to combine results for what we can term identical Indic words with different character order of entry. They will in the future, just like they combine results for cafe and café if you search without ++"".

2. I'm not sure about your Bengali issue. It may be that in this case nameprep is not being correctly used like example 1. It may also be an error in nameprep, in that case, it is not Verisign's problem. Verisign deals with .com, .net., TLDs. Punycode is a technology standard that was approved for the whole internet. It is necessary to focus the example and use the best punycode converters to investigate. If you still find an error that can be duplicated, share it, and let one of the Indian IT specialists on this forum bring it to ICANN's attention, if this is indeed an encoding problem.
The punycode scheme still does not encode all characters yet, maybe it never will, because they continue adding characters to the Unicode table. This is not "problem", that's the way we design formulars. More and more characters will be covered, but send ICANN or the punycode team what you discovered to save them time. The best solution is NOT to reg such characters, the chance you will lose them is high.

Dot Com is King. will soon be king.
Reply With Quote