[petsc-dev] Fwd: [petsc-maint #124771] bug report

[Barry Smith]

On Jul 19, 2012, at 6:07 PM, Satish Balay wrote:

> On Thu, 19 Jul 2012, Barry Smith wrote:
> 
>> 
>> On Jul 19, 2012, at 5:46 PM, Jed Brown wrote:
>> 
>>> On Thu, Jul 19, 2012 at 5:38 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
>>>> This is what I do for myself [with pine]
>>> 
>>>   Doesn't solve the problem because the user still has in their header the developers email address.
>>> 
>>> This is something that a developer using certain hosts can do to make the issue moot, not something that involves the user.
>> 
>>    I do not understand this. If the user knows your email address they can still send you email. You are telling me you can send me an email that does not contain your email address somewhere in the header? 
>> 
>>   yes  reply-to is a good start but it is not a finish. People by-pass that.
> 
> I think the primary issue is folks doing 'reply' instead of
> 'reply-all'. [so a mechaism that adds a reply-to:petsc-maint] would
> suffice such use case.
> 
> Any other bypass mechanisms are not worth dealing with.

   So say you.  

    Users intentionally make sure that they are sending a personal message to one of the developers thinking they will get more attention!  Call me paranoid but this is the truth, it does not happen through inadvertent wrong kind of reply buttons.

   Barry


> 
> 
>>>> And then there will be no more private
>>>> messages [between us] on petsc-maint :)
>>> 
>>>    Why? petsc-maint server will always bounce everything sent to it back to the developers.
>>> 
>>> How often does some not-to-be-named developer remove the user from the Cc list and make a snide remark?
>>> 
>>> If you answered Never, you earn two points.
>> 
>>    This is easy. The not-be-named developer simply modifies the subject line in any way. The match doesn't work so the message does not go back to the user.
> 
> e-mails have other hidden headers that are used by RT to
> identify/process such e-mails. I don't think just changing subject
> line will suffice.
> 
> Satish