[MLton] web update

[Stephen Weeks]

> And I didn't know where on mlton.org I should copy the files from.

The right thing is to keep the web directory of the CVS checked out
somewhere on mlton.org, and to copy from there.  That's what I'd do in
a script anyway.  Since the files are small, what I actually do right
now is copy them from my home machine, where I have the web CVS
checked out.

> "Search Archive" would be fine.

Done.

> That seems to be a little harder, since HyperArch.py is generating archive
> pages for arbitrary lists.  Note that in order to narrow google enough,
> when you search the  mlton  list, then I pass  -inurl:user  to filter out
> the  mlton-user  list, while to search the  mlton-user  list, I pass
> +inulr:user  to filter the other way.

Yeah, I have no better suggestion than to hardwire if tests for
specific list names into the code.  I don't object to that.

> Well, my affiliation remains Cornell; I'm just physically located at
> Harvard.

Ahh.  OK.