[darcs-users] Conflict marking

Thomas Zander zander at kde.org
Thu Oct 13 12:31:38 UTC 2005


On Tuesday 11 October 2005 06:45, Stephen J. Turnbull wrote:
>     David> Indeed, both versions presented may be in states that no
>     David> user has ever seen before, but I still think we're not too
>     David> hosed.  [local] just means that "local changes would have
>     David> led to this version."
>
> If no user has ever seen them, it's likely that labelling it "local"
> will be of no help to the user in figuring out which is which without
> doing the careful semantic analysis.

I respectfully disagree, based on talking with a lot of people from a 
usability point of view (due to my background) I know that the concept of 
'local' is pretty clear in peoples mind.
"Me" or "Mine" as a psychological idea is also very clearly defined for 
all people.

For something like cvs/svn the [local] means a conflict of an un-committed 
patch and practice shows most people understand this just fine.
For Darcs this is extended to local-but-recorded-patches, but most 
importantly its the *opposite* of the remote repository we are pulling 
from.  Marking it [not from the remote version] would be the information 
most people need in order to make a quick decision based on roles instead 
of content of which patch is the right one.  Luckily the [local] tag 
conveys the same emotions and is just as good for making those decisions.

If you feel you need to know exactly which patches lead to the conflict 
I'm pretty sure the interface we already have is insufficient for you 
anyway and you should probably have a different tool for your needs to 
extract that info from Darcs.

I just know that the majority of users will appreciate the propsal as is.

-- 
Thomas Zander
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.osuosl.org/pipermail/darcs-users/attachments/20051013/833da36b/attachment.pgp 


More information about the darcs-users mailing list