- From: Martin J. Duerst <mduerst@ifi.unizh.ch>
- Date: Wed, 26 Feb 1997 18:41:09 +0100 (MET)
- To: Daniel LaLiberte <liberte@ncsa.uiuc.edu>
- Cc: "Gregory J. Woodhouse" <gjw@wnetc.com>, uri@bunyip.com
On Mon, 24 Feb 1997, Daniel LaLiberte wrote: > Gregory J. Woodhouse writes: > > I'm not so sure they are. True, existing applications use URLs like > > filenames, but then again, that's all we have. If there were no DNS, we > > would use IP addresses [...] > > Another way to describe this division is symbolic versus numeric > identifiers. URLs are already symbolic, and as such, they belong in > the column with filenames and domain names. Comming back to URL internationalization, I think the distinction between symbolic and numeric identifiers is a very useful one. One can see symbolic identifiers as things that mean something to the user, and numeric identifiers as things that don't mean anything to the user. [of course, there are numbers, such as 007, that mean something to many, but in general, they don't] Now ASCII identifiers can be highly symbolic for people used to the basic Latin alphabet. To others, who may have heard about them in school, but never much used them, these things will be highly "numeric". One core idea of URL internationalization is to allow to make URLs symbolic for those user groups that are mostly dealing with them (for the rare outsider, we have the "numeric" %HH fallback). Regards, Martin.
Received on Wednesday, 26 February 1997 12:41:48 UTC