Re: [Asrg] Summary of junk button discussion

[Jose-Marcio Martins da Cruz <[e-mail]>]

Ian Eiloart wrote:

>> I don't like the name "Add a junk button", as this mechanism can
>> be used to signal a junk message when someone want to complain,
>> but it may also be used to do feedback about messages : both
>> false positives and false negatives.
> 
> Yes, we've already discussed that problem, and several ways to mitigate 
> the problem, quite extensively. I've added this text to the wiki: "Also 
> out of scope are other ways in which administrators respond to such 
> reports, however it's worth noting that any response mechanism should be 
> robust to false reports, whether they're malicious or accidental."

This mechanism being described and discussed may be used more largely. Let me give an example.

You and me, we both work in an university. So, some thing useful in environnements like the ours is 
a way to have users sending feedback to us (us = mail admins) : both false negatives AND false 
positives. These feedbacks may be useful both to tune our filters, to feed statistical learning filters.

The underlying mechanism may be the same as the "junk button", although I agree with you that there 
are weaknesses related to false reports, and noise. But this is a trust problem which may probably 
be handled in a higher level, or adding some more precise information.

I find it too restrictive to call this a "junk button" as this mechanism can be used more widely 
than just this way.


-- 
  ---------------------------------------------------------------