> > Many of these could be addressed by more sophisticated use of
> > caching control parameters and by having server side include and
> > more general CGI processing synthesize a Last-Modified-Date based on
> > the real content,
>
> And even that doesn't really do the trick.
>
> I maintain my web page in CVS, so I use $Date$ keywords as a cheap and
> easy way to build 'Last Modified'.  But I modify pages all the time to
> fix up spelling and grammar mistakes, or clarify labyrinthine
> sentences.  If I didn't make use of templating and cross-includes, I'd
> be modifying every page every time I made a change to the site
> look&feel or some of the common code.
>
> All of these would be changing the 'Last Modified', whether by
> filesystem mtime or the revision $Date$.  But none of them say
> anything about whether I checked that the content was up to date.
> There's no way an automated system can know whether I checked or
> updated something like that, so it can hardly be anything HTML (or
> HTTP, or some server-side programming language for that matter) can
> address.

Why couldn't a system be put into place in HTML that would allow
things like this to be tracked? I have documents and XML files that
specify when things were modified and that usually means non-errata.
So I don't change the modified date when I make a spelling change or
fix up a sentence (I often have labyrinthine sentences myself), but it
would allow one to specify when something was added, removed or
substantively changed. I would posit that any additional or removal of
a section is substantive.

--

Orion Adrian