 |
Jon Carnes |
at Nov 12, 2003 at 7:09 pm
|
⇧ | 
| |
On Wed, 2003-11-12 at 11:21, Bruno Postle wrote:I've been made-aware that as well as indexing pipermail public html
archives, google indexes all the raw mbox files too:
http://www.google.com/search?q=pipermail+mbox+X-Mailman-VersionThese are also a resource for email-address harvesters, so I've
configured apache to redirect these requests to the private
archives:
RedirectMatch permanent ^/pipermail/(.+)\.mbox/(.+)\.mbox$ http://example.com/mailman/private/$1.mbox/$2.mbox
Can anyone see anything wrong with this approach?
..or should mailman be modified to allow only authenticated user
access to the mbox archives?
I like your way of doing it. That should work fine.
I run a script against my mbox file nightly (and the html pages) which
scans for specific email addresses and substitutes a specific
replacement for that email address - all this provided by users who
don't want their addresses showing up in the archives.
It was a nice add-on feature and easy to do. The names and
substitutions are stored in a simple database; the script is a simple
"for... done" that executes a sed statement for the substitutions.
Since I run it nightly, it's pretty fast. It would also be fairly easy
to add this to the Pipermail archiver. On some of my lists I run
Pipermail, on others I run Mhonarc, so I haven't bothered to integrate
it.
Jon Carnes
