Web Browser: ---
Summary: Allow initiating offlist email through comments
Product: GCD Online
Version: 0.4 little nemo
Platform: PC
OS/Version: Mac OS
Status: NEW
Severity: normal
Priority: P3
Component: workflow
AssignedTo: gcd-...@googlegroups.com
ReportedBy: hh...@cornell.edu
Estimated Hours: 0.0
Since approvers can no longer see email addresses without asking an admin, we
should add a checkbox to the comments fields along the lines of "send my email
address and invite an offlist exchange."
This would include the commentor's email address in the cc field and/or the
message body, with a note saying "Please contact me at this address to have a
private discussion about this change", or something like that. We do not want
folks to feel tempted to put their email addresses in the public comments (or
much worse, for an approver to ask an indexer to do so, giving the impression
that it is safe).
We should probably put a note saying that you don't want to put your own email
address in the comments, while we're at it.
--
Configure bugmail: http://dev.comics.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
Jochen Garcke <gcd...@garcke.de> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |gcd...@garcke.de
--- Comment #1 from Jochen Garcke <gcd...@garcke.de> 2010-10-04 08:50:39 PST ---
I would think the better (or alternative) and more direct way would be to be
able to send an email from an users profile. That way the user gets the email
directly (with content and from the email address of the sender), but without
the sender knowing the email address of the recipient.
Donald Dale Milne <dond...@att.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |dond...@att.net
--- Comment #2 from Donald Dale Milne <dond...@att.net> 2010-10-04 13:31:57 PST ---
I have often wanted to send an e-mail to an indexer whose e-mail address I did
not know and my first stop has been the profile. I think it would be great to
send from the profile, without seeing his address.
Ralf Haring <ralf....@gmail.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |ralf....@gmail.com
--- Comment #3 from Ralf Haring <ralf....@gmail.com> 2010-10-04 16:49:54 PST ---
Why wouldn't these cases be handled by adding a "also send as email" checkbox
to the comments field?
--- Comment #4 from Henry Andrews <hh...@cornell.edu> 2010-10-04 17:02:50 PST ---
I think it would be great to offer both options to approvers. But certainly
from the indexer to the approver direction it needs to go through the comments
as they should not be expected to find the profile pages for this. Also, even
if they could it would mean that all approvers would be hooking up their email
to a publicly visible form, which I suspect folks would rather not do. The
form going to the indexers from their profile page would be available only to
approvers, who are at least nominally screened for some sort of sanity ;-)
--- Comment #5 from Jochen Garcke <gcd...@garcke.de> 2010-10-04 20:04:58 PST ---
(In reply to comment #4)
> I think it would be great to offer both options to approvers. But certainly
> from the indexer to the approver direction it needs to go through the comments
> as they should not be expected to find the profile pages for this.
One could just click on the link to the approvers profile which is visible on
the queue, or ?
> if they could it would mean that all approvers would be hooking up their email
> to a publicly visible form, which I suspect folks would rather not do. The
> form going to the indexers from their profile page would be available only to
> approvers, who are at least nominally screened for some sort of sanity ;-)
Hmm, don't really see the concern. The sender of the email (who needs to be
logged in) doesn't get to see the email unless the recipient replies. And one
always can send an email to an indexer/approver by adding a comment to any
changeset they are involved.
--- Comment #6 from Jochen Garcke <gcd...@garcke.de> 2010-10-04 20:06:40 PST ---
(In reply to comment #3)
> Why wouldn't these cases be handled by adding a "also send as email" checkbox
> to the comments field?
>
Because not everything needs to (or should) be stored publically and eternally
in the comments field.
--- Comment #7 from Ralf Haring <ralf....@gmail.com> 2010-10-04 20:12:09 PST ---
(In reply to comment #6)
> (In reply to comment #3)
> > Why wouldn't these cases be handled by adding a "also send as email" checkbox
> > to the comments field?
> >
>
> Because not everything needs to (or should) be stored publically and eternally
> in the comments field.
If it's about the changeset (the only reason to communicate with someone
through the GCD), I disagree.
--- Comment #8 from Henry Andrews <hh...@cornell.edu> 2010-10-04 20:14:29 PST ---
(In reply to comment #5)
> (In reply to comment #4)
> > I think it would be great to offer both options to approvers. But certainly
> > from the indexer to the approver direction it needs to go through the comments
> > as they should not be expected to find the profile pages for this.
>
> One could just click on the link to the approvers profile which is visible on
> the queue, or ?
The point is to make it easy on the indexer. One less click is a significant
thing. I don't understand why you object to this so much. This doesn't
necessarily involve adding a new comment that just says "contact me offlist".
It's about putting the button where it's needed.
> > if they could it would mean that all approvers would be hooking up their email
> > to a publicly visible form, which I suspect folks would rather not do. The
> > form going to the indexers from their profile page would be available only to
> > approvers, who are at least nominally screened for some sort of sanity ;-)
>
> Hmm, don't really see the concern. The sender of the email (who needs to be
> logged in) doesn't get to see the email unless the recipient replies. And one
> always can send an email to an indexer/approver by adding a comment to any
> changeset they are involved.
I'd prefer to avoid making it that easy, given some of the experiences many
folk have had with certain people over email. Something in the context of an
approval workflow is preferred.
--- Comment #9 from Jochen Garcke <gcd...@garcke.de> 2010-10-04 20:34:15 PST ---
(In reply to comment #8)
> The point is to make it easy on the indexer. One less click is a significant
> thing. I don't understand why you object to this so much.
I don't object, I just think both are useful. In particular if the initial
email should contain more than the 'contact me'
And if Ralf really thinks that 'the only reason to communicate with someone
through the GCD is if it's about the changeset' we do have a severe different
view on the project.
--- Comment #10 from Jochen Garcke <gcd...@garcke.de> 2010-10-04 21:13:57 PST ---
To make it more clear what I think we should have:
- something like proposed in the opening post
- for approvers and mentors the possibility to send an email to indexers
currently assigned to them (active changes, active mentee) via the profile page
- for indexers to their mentor and (maybe) as well to the approvers of the
currently active changes (again from the profile pages of these)
If indexers can't send an email to everyone neither should approvers be able to
do so.
--- Comment #11 from Henry Andrews <hh...@cornell.edu> 2010-10-04 21:57:44 PST ---
(In reply to comment #10)
> To make it more clear what I think we should have:
>
> - something like proposed in the opening post
>
> - for approvers and mentors the possibility to send an email to indexers
> currently assigned to them (active changes, active mentee) via the profile page
>
> - for indexers to their mentor and (maybe) as well to the approvers of the
> currently active changes (again from the profile pages of these)
>
> If indexers can't send an email to everyone neither
> should approvers be able to do so.
Approvers are GCD staff (or should be- I'll skip the whole argument about
approvers who don't do approvals but get upset at the idea of taking an
"emeritus" status, as it never seems to end well).
* Allowing staff to contact you over changes is fine. It's "necessary for the
'contract' between the parties", as noted in privacy law. Avoiding email
address disclosure is preferred.
* Allowing an indexer to contact their mentor or the approver of their change,
without disclosing the approver's email address, is fine. Likewise necessary.
We should at least check with the approvers as a whole before doing this,
though.
* Allowing any indexer to contact an arbitrary approver, even without
submitting a change, is not something that we allow currently. It is not a
privacy concern as it could simply be part of what you accept as a member of
the staff. However, the Board or the approvers need to state that they want
this. As tech folks, we can't just make this happen unilaterally.
* Allowing any indexer to contact any other indexer is absolutely a violation
of good privacy practices, even if we do not directly expose the email
addresses.
Henry Andrews <hh...@cornell.edu> changed:
What |Removed |Added
----------------------------------------------------------------------------
Version|0.4 little nemo |-.- future
--- Comment #12 from Henry Andrews <hh...@cornell.edu> 2010-10-31 20:41:16 PST ---
The Board still needs to sort this out, and that won't happen in this release's
time frame. The removal of addresses from the approver's view was never
checked in due to controversy.
Jochen Garcke <gcd...@garcke.de> changed:
What |Removed |Added
----------------------------------------------------------------------------
Depends on| |76
--- Comment #13 from Jochen Garcke <gcd...@garcke.de> 2012-02-10 21:35:27 PST ---
We can use
https://bitbucket.org/albertoconnor/django-email-form/overview
to send emails between users, in particular mentor/indexer or if someone needs
to contact an indexer due to a change.