- From: Carlos Iglesias <carlos.iglesias@fundacionctic.org>
- Date: Wed, 21 Jun 2006 15:13:32 +0200
- To: "Shadi Abou-Zahra" <shadi@w3.org>
- Cc: <public-wai-ert@w3.org>
Hi, > > IMO if we have a clear use case for non-URI identifiers > then we should > > choose "identifier", if not we should choose "URI" > > For earl:WebContent it is clear that we need a URI to > describe the resource. However, I can imagine some use cases > for dc:identifier too. For example a hash or checksum of the > resource to compare it at a later fetch of the same URI. > > As to earl:Software, it seems to be the opposite approach: an > identifier such as the build or version number may be the > best identifier, and URI may sometimes be available too (as a > location to download or find more information about the software). The only problem I see is that identifiers are supposed to be unambiguous, and a build or version number (or a checksum) is not because it could be useful to identify a piece of Software if you know other information (e.g. name, developer, etc.), but not by itself. For example: - WatchFire Bobby v5.2 can identify specific software - Microsoft IExplorer v5.2 can also But v5.2 can not (Bobby or IExplorer?) Additionally the fact of having several identifiers for something could be a bad idea because people are used to unique identifiers (one per thing) Regards, CI.
Received on Wednesday, 21 June 2006 13:14:12 UTC