Niklas Bergh wrote:

The initial "virgin" estimator should give worse estimates than all but
the most dismal experienced estimators. This means longer estimated
retrieval times.
For all keys or just most?

Initially - all keys.


Good question.. I am basing that theory on the fact that tessiers
(month-long+ ?) forced datastore specialization ought to have affected the
networks opinion on his node if some aspect of the current NGR algorithm
wasn't actively countering his actions. This was before estimator
pass-around though, maybe it would behave differently now...

The problem I described in my original email amply explains why tessier's node would despecialize in this situation.


Ian.

_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to