Hi Tom,
right now there is no interaction between "my"whiteist and the kspam configuration db.
the whitelist is implemented in the users mailfile and kspam configured to deliver all mails ( even with 100% probability ). the agent checks if the mal sender ( or domain ) is whitelisted and if the mail will be placed in the inbox, if not and the prob. is higher than the watermark that the usr is able to set, the mail will be placed in junkmail.
we have choosen that approach because its pretty hard to block mail, if the user do not have the possibility to whitelist things and in some companies we are even forced to deliver all mails.
works out pretty good so far.
Best regards
TW
PS.: i know its hard, but after some time a user based whitelist its more handy than the traffic you gain if you do not deliver "spam".
a webinterface and all the features from tools like proofpoint can not be expected from kspam, bu people here compare theses products which s some kind of compliment for this project as well ;-)