Re: extra pseudo-class for anchors

> You meant URI.

Now that you mention it. Yes

> It has been suggested before.  The basic problem, I believe, is that,
> in general, the URI used to access the page may not agree with the
> Location header returned, or with the URI derived from a self link
> the page. 

I feel that if a browser can check the entries in its history to compare
them to an URI just to tell if its been there before, it should be able
to tell if an URI is equal (enough) to the current URI

> This is likely to be very common with many commercial sites, which
> to redirect half a dozen times before finally ending up on the page,
> it is also an issue with mirrored sites and with content negotiation.

If the URI in the anchor is relative then it shouldn't give too much
problems with mirrored sites as long as they use the same structure. If
the URI is absolute then it indeed poses a problem. 

> > Content-type: text/html; charset=us-ascii
> Please don't do this on public lists.

I completely forgot. Sorry.

Received on Monday, 12 January 2004 11:35:47 UTC