How Alice and Bob got telephone/SMS spam on their Moko.

[Tilman Baumann]

ian douglas wrote:
> Tilman Baumann wrote:
>> In my eyes, trust should be calculated automaticly, not by manually 
>> defining trust (like pgp). Using the system could improve it.
> 
> Perhaps number of successful phone calls, and length of phone call to 
> add a number of trust 'points' ... the more successful calls over some 
> length of time, the more trusted someone becomes.
> 
> Still, having individual profiles as someone else mentioned, would be 
> ideal too. For example, I'd want my wife's phone call to ring no matter 
> time of day or busy status (meetings, etc). A call from my 
> mother-in-law, however, should never play any ring tone during business 
> hours, and unknown caller ID values should never play any ring tone and 
> route them immediately to voice mail.
This would be a locally managed call managing system. Definitely a thing 
we need to. But not quite the same thing as we discussed.
But i am absolutely with you, wee need this. And it would probably be a 
good idea to design a system like this first. And then extend it in the 
future for distributed network blacklists.

> 
>> I have better things to do with my life than managing trust relations 
>> for my spam filter...
> 
> Well, even tools like SpamAssassin need training to teach it what is 
> considered spam or not.

Yea. But this is limited to a simple question 'was this call spam?'.
What i mean was that you should have 'the list[tm]' that is 
automatically the right thing for you. And not havong to decide actively 
whoms reports you trust.