Re: Identity modelling

Marshall Spight wrote:
> x wrote:
> >
> > Let's try to use RM for this little problem:
> > - the URL is a value
> > - the content of the resource is a value
> > - there is a relation At(URL, Content) where URL is the primary key with
> > At(u,c) meaning that at this instant the URL u point to the content c.
> >
> > So the URL is a key after all. :-)
>
> I have to disagree. Note that you can't map your above example
> onto keys in a relational table, since there is no analog
> to the separate (key, content) relation.
>
> The distinguishing characteristic of content-based addressing
> is that data is addressed by some part of the data. Note
> that nothing *in* the html file identifies the html file.
> Most html files don't contain their url. So the URL is a pointer.

I disagree. If one is modeling this data (and I think the referent of "this data" might be the problem) and taking a very traditional key-value approach with your data model, then the URL is surely a key. I think your use of "pointer" here is very extreme -- miles away from memory locations, for example. There is nothing hidden or behind the scenes about a URL. One issue I see is that you opted to talk about modeling the html file, rather than all of the data under consideration. So, perhaps if you add in the URL to the data you are modeling, you will agree that it is, indeed, a key?

cheers! --dawn Received on Wed Aug 31 2005 - 23:45:46 CEST