Hi Patrik, I am not doing a good job in explaining my idea. I am working on actually setting up the entire system (for testing and data gathering purposes) and I will let you know when it is completed so that you can see exactly what I mean.
Thanks Ben ----- Original Message ----- From: "Patrik F�ltstr�m" <[EMAIL PROTECTED]> To: "ben" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Saturday, October 27, 2001 2:54 AM Subject: Re: [idn] An ignorant question about TC<-> SC > --On 01-10-26 15.13 -0400 ben <[EMAIL PROTECTED]> wrote: > > > However, what I am certain > > is that you have illustrated the fact that > > applications/clients/users/servers/etc can be made to take advantage > > of this explicit labeling of what "script" an IDN is in and over time > > (with people writing appropriate applications) can be developed into a > > very powerful and useful system. > > No, you misunderstood. It is exactly the contrary. > > Comparison of characters from two different charsets is extremely > difficult. It can not be done effectively for all combination of charsets. > > Only way of dealing with this problem is to map each of the two charsets to > one, and then use comparison rules for that charset. > > This is the model which works in DNS, and that is why this wg have chose > this path. This has already been discussed, and discussing this again > doesn't help this wg make progress. > > IF you want to start to tag a string (in whatever charset) with meta > information such as language, geographical location of creator of the data, > date and time when the string was created, and use this information when > you do a matching decision, then you talk about something different than > DNS, something which looks like a keyword system. John Klensin has > described such systems in drafts which have circulated on this list. > > paf > > >
