> It would help in assessing "sensible" if you would say which ones you > prefer and why.
I think the well suited hierarchy of options may be: 1) New fake_id for each post: In my opinion this approach won't be able to establish the basic trust needed for the conversation to flow. eg.I queried something to mr. A on mailing list and he replies it with id B. And it won't allow replying mechanism to work easily as we will have to store multiple id's for each user at a time. This will considerably increase the database requirement 2) New fake_id for each Thread: This approach aims at maintaining the conversation flow when a person is talking on a particular topic. Drawback is the user wishes he can't change his user name during a thread (although that can be spoofed) 3)New fake_id for each list: It is quite a compromise over the small amount of security we are aiming to provide by this feature. As it makes the user more traceable then in above approach case. For eg. I don't want my company to hunt a single fake id and find me out, if the secrets were exposed from different addresses they would be more anonymous and there would be no proof if a single person did that or many. 4)Same fake_id for all lists: It sounds like maintaining a dictionary each user and her corresponding fake name but again it makes the user more traceable as in the above case. > Plus a couple of > different dimensions: change ids if posting from different IPs, and > change IPs' at intervals of time. I'm not sure which of those (and > their combinations) make sense. I plan to clean out the headers and leave no trace about the IP the mail came from. Also For eg a...@example.com posts a mail on list I pose him as xyz@list_name.com where xyz will be randomly generated id. So I don't think there is a need to change IP address each time I am more inclined towards no. (2) mentioned above, because it seemed to me most sensible on the anonymous list feature I plan to implement. > Finally, I suspect that different list owners will have different > preferences, as will different subscribers. The same individuals > might even have different preferences depending on list. But then we can't satisfy every need ;) we can only implement the most general case. Please correct and guide me further. Regards Rashi _______________________________________________ Mailman-Developers mailing list Mailman-Developers@python.org https://mail.python.org/mailman/listinfo/mailman-developers Mailman FAQ: http://wiki.list.org/x/AgA3 Searchable Archives: http://www.mail-archive.com/mailman-developers%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-developers/archive%40jab.org Security Policy: http://wiki.list.org/x/QIA9