From: Ralph Babel (rbabel@babylon.pfm-mainz.de)
Date: Fri Oct 12 2001 - 05:25:55 CDT
Brad Templeton wrote:
> If somebody forges an article in my name,
> I should be able to cancel it quickly and easily.
Sign your articles. Then there's
no need to cancel forgeries later.
> If somebody posts an article from a site,
> the site admin should be able to cancel it.
Please define "site", "site admin" and "to post
from a site" unambiguously, and please explain the
implications for parallel injection of articles.
Besides: why did the "site admin" give the
poster permission to post in the first place?
> If somebody forges a spam attack appearing to come
> from my site, I should be able to cancel it.
If the so-called "spam attack" doesn't carry a valid
signature, everyone can recognize that it didn't
come from "your site". Not a cancel issue.
> If I was not running a brand-new newsreader/poster when
> I posted an article, I should still be able to cancel it
The first three items you listed didn't provide
any justification for cancels anyway, so why would
your newsreader need such a feature at all?
> As above, without having to install a
> new newsreader/poster to do the cancel
See above.
> It's nice if most cancels can mostly be acted on without
> needing to get the original first, both because it may not
> have arrived and to avoid the disk I/O of fetching it if
> it has.
See above.
> The amount of material added to the net
> to support cancels should be minimized.
Great, "zero" _is_ pretty minimal.
> It should be possible to authorize 3rd parties
> to issue cancels in various subsets of the net.
Why?
> I have solutions to _all_ the above problems.
If all you have is a hammer ...