Mailing list archives

[akuchling <andrew at a...>]

There are mailing list archives all over the Web. Some archiver
programs make the sender of an e-mail message available as a mailto:
link, though this is becoming less common because of spammers
harvesting addresses.

There's a FOAF application here. Archivers could easily compute the
SHA1 hash of an e-mail address and make it available in some way. A
smart user agent could then use some FOAF lookup service to get more
information about the user (such as an actual name, home page, etc.)
A neat property would be that you'll never find an outdated e-mail
address in an archive; assuming users keep their FOAF up-to-date and
list all their e-mail addresses, you could always get current contact
info even for archived messages that are a decade old.

But how to bootstrap this?

One approach might be to just embed a little RDF snippet in the
HTML-rendered form of each mail message that said:
<#> dc:author 
[ foaf:mbox_sha1sum "1234... ];

A more general approach might be to figure out how to put the info
into an appropriate LINK tag.

But users wouldn't notice this capability until their browser looked
for this info, and supported a FOAF lookup. That would require
modifying Web browsers, a long and difficult task. Anyone have ideas
on a way make the new capabilities available to unmodified browsers?

One I just thought of: you could browse through a proxy that expanded
such references and modified the HTML, though proxies might be too
difficult to set up for naive users. Maybe a little Mozilla add-on
could add a 'Tell me more about this author' button when there's an
appropriate link tag.

--amk (www.amk.ca)
LEAR: I fear I am not in my perfect mind.
-- _King Lear_, IV, vii