On 2 December 2016 at 10:04, Tomasz Sterna <to...@xiaoka.com> wrote: > W dniu 19.11.2016, sob o godzinie 15∶19 +0300, użytkownik A napisał: >> Very easy. Popular clients just incorporated simple anti-spam >> measures to perform human-testing for any new senders.
> There also is an IM2000 proposal [1] which does it way better and more > clever. I don't see how this is any better than rate limits on the sending side. It's not like a spammer will need to store 10000000 copies of the same message to send it to 100000000 recipients, if they use their own server. Likely public servers would perform de-duplication also. And if they don't, it's not really different to the server just applying rate limits. Maybe there's an argument that server admins wouldn't bother to set rate limits, but they would if every message cost them storage space. On the other hand, storage space hardly costs anything nowadays. I just don't see the benefits of this approach. Am I missing something? Regards, Matthew
