- From: Rick Jelliffe <ricko@topologi.com>
- Date: Thu, 1 May 2003 00:07:03 +1000
- To: <www-tag@w3.org>
> We have to be very conservative here, > and not take any chances of having the rug pulled out from under us. > For example, what would happen if someone made a convincing argument > that percent escaping in IRIs should be based on UTF-16 rather than > UTF-8? Not that I expect that to happen, mind you, but I don't want > to commit IRIs until the possibility has been ruled out by an adopted > spec. IRIs would only be adopted if they basically agree with the escaping method already used throughout the XML family specs. We don't have to be conservative: users can evaluate the trade-off, just like any encoding-related issue. Additionally, Larry Masinter's RFC (previously quoted) proposed UTF-8. UTF-16 is a red herring. So UTF-16 is not a reason not to support the Charmod policy of allowing IRIs everywhere. However, deciding to support IRIs everywhere (as long as IRIs reflect the Charmod and XML-family practise) does not mean that an IRI draft should be normatively referenced from W3C specs while it is still a draft, just as a procedural matter. But tt does not need to be: XML and the other specs can be, for the time being. Cheers Rick Jelliffe
Received on Wednesday, 30 April 2003 10:03:17 UTC