Hi Michael:
On May 18, 2011, at 8:37 PM, Michael F Uschold wrote:

> Alan, I'm glad you made that suggestion.  I was also glad to see that Tim-BL 
> acknowledged that the URIs are just identifiers.  

Using URIs as identifiers does not imply they must not have human-readable 
components. Otherwise, there was no need for alphanumeric URIs and the DNS 
system at all.

> As you know, noone seems to be treating them that way, nor is there good tool 
> support to make it easy to do -- probably the main reason the practice 
> persists.
> 
> Part of Martin's argument is based on what he already did, given that, maybe 
> he is doing the right thing now.
> 
> Moving forward for the broader semantic web community,  the more interesting 
> question is, if Martin was starting from scratch right now, would there still 
> be any good arguments for having URIs with meaningful names?
The only thing I would do differently when restarting now would be a using 
shorter names for a few elements, and enforcing a bit of additional 
terminological consistency.

> Given the RDFa context:
>       • How much effort would it be, in terms of extra tool support, or 
> training users etc.
>       • Would it have even been possible to get GR off the ground in todays 
> market place w/o meaningful URIs?

It would be disaster to use URIs without meaningful names.


Martin



Reply via email to