darcs

Issue 1343 promote the use of a default from address for more reliable sending

Title promote the use of a default from address for more reliable sending
Priority feature Status resolved
Milestone Resolved in
Superseder Nosy List darcs-devel, dmitry.kurochkin, jaredj, kowey, thorkilnaur, twb
Assigned To twb
Topics ProbablyEasy

Created on 2009-02-04.01:03:02 by simon, last changed 2009-08-27.14:20:32 by admin.

Files
File name Uploaded Type Edit Remove
unnamed simon, 2009-02-04.01:02:53 text/html
Messages
msg7268 (view) Author: simon Date: 2009-02-04.01:02:54
I think it is pretty common for a simple "darcs send" to fail for new  
contributors to a darcs-using project, for this reason: the project  
uses a subscribers-only mail list, and the developer is sending from a  
machine which sets a from address different from their list address.

This has happened recently on the happstack project, where it seems at  
least the first two or three contributors ran into this problem and  
didn't find the solution (specifying the correct address with --from).  
Instead they started planning a patch-forwarding subscriber (which  
might work but would probably lose attribution, making discussion  
threads less useful, and is an unnecessary additional setup hurdle).

I think this is/will be a pretty common scenario, and it would be nice  
if darcs-using projects and their new contributors didn't have to  
remember to add --from all the time. What to do ?

It might be worth promoting - somehow, somewhere - the use of a  
default from address setting. Ie, encourage users to add "send from  
my@address" to their ~/.darcs/defaults (or platform equivalent) on  
machines they are sending from. And/or, to suggest this to darcs-using  
projects as something that they should mention to their contributors,  
if they use a closed list.
Attachments
msg7270 (view) Author: simon Date: 2009-02-04.01:10:37
One simple thing that might help is to have darcs send display the  
from address that it used. This would help users notice the issue. It  
could also print a notice: "If you are sending to a subscribers-only  
list, set from address with use --from or defaults as needed".
msg7635 (view) Author: kowey Date: 2009-04-09.12:12:24
Is this worth doing, Trent?
msg7662 (view) Author: twb Date: 2009-04-10.06:34:00
For this ticket, I propose two actions:

Firstly, the user manual describes the darcs send workflow.
This section of the manual should also warn about this issue.

Secondly, I think the easiest way to make the sender aware of
which From address they are using is to simply include all
outbound headers in the text file that --edit-description opens.

This would be similar to mutt's "set edit_headers=yes".
msg7675 (view) Author: kowey Date: 2009-04-10.21:56:16
On Fri, Apr 10, 2009 at 06:34:02 -0000, Trent Buck wrote:
> Firstly, the user manual describes the darcs send workflow.
> This section of the manual should also warn about this issue.

Ok.  Is this something you plan on working on?

> Secondly, I think the easiest way to make the sender aware of
> which From address they are using is to simply include all
> outbound headers in the text file that --edit-description opens.
> 
> This would be similar to mutt's "set edit_headers=yes".

Sounds like it would be ProbablyEasy
msg7678 (view) Author: twb Date: 2009-04-11.02:27:45
On Fri, Apr 10, 2009 at 09:56:18PM -0000, Eric Kow wrote:
> On Fri, Apr 10, 2009 at 06:34:02 -0000, Trent Buck wrote:
>> Firstly, the user manual describes the darcs send workflow.
>> This section of the manual should also warn about this issue.
>
> Ok.  Is this something you plan on working on?

Inasmuch as I plan to overhaul all our documentation, yes.  The real
question is: when?  I dunno.
msg7862 (view) Author: twb Date: 2009-05-31.15:52:19
The following patch updated the status of issue1343 to be resolved:

* Resolve issue1343: mention darcs-users is subscriber-only, use --from. 
Ignore-this: e38c4cd6a447767a4b0bd95301a3f68
Introduce a HACKING stub that incorporates the useful data that the
issue1312 fix removed from building_darcs.tex.  Also tells people to
use Cabal for functional and unit testing, and --in-reply-to.
History
Date User Action Args
2009-02-04 01:03:04simoncreate
2009-02-04 01:10:39simonsetstatus: unread -> unknown
nosy: kowey, simon, thorkilnaur, dmitry.kurochkin
messages: + msg7270
2009-04-09 12:12:26koweysetpriority: feature
nosy: + twb
messages: + msg7635
assignedto: twb
2009-04-10 06:34:02twbsetnosy: kowey, simon, twb, thorkilnaur, dmitry.kurochkin
messages: + msg7662
2009-04-10 21:55:51koweysettopic: + ProbablyEasy
nosy: + jaredj
2009-04-10 21:56:18koweysetnosy: kowey, simon, twb, thorkilnaur, jaredj, dmitry.kurochkin
messages: + msg7675
2009-04-11 02:27:48twbsetnosy: kowey, simon, twb, thorkilnaur, jaredj, dmitry.kurochkin
messages: + msg7678
2009-04-11 02:32:00twbsetnosy: kowey, simon, twb, thorkilnaur, jaredj, dmitry.kurochkin
2009-05-31 15:52:23twbsetstatus: unknown -> resolved
nosy: kowey, simon, twb, thorkilnaur, jaredj, dmitry.kurochkin
messages: + msg7862
2009-08-25 17:40:27adminsetnosy: + darcs-devel, - simon
2009-08-27 14:20:32adminsetnosy: kowey, darcs-devel, twb, thorkilnaur, jaredj, dmitry.kurochkin