[dist-bugs] RFC buginfo microformat (handling disagreements)
Joey Hess
joey at kitenet.net
Thu Jul 10 17:03:16 EDT 2008
As I mentioned, I am not entirely happy with this, and of course it comes
down to something equivilant to solving the spam problem. This just provides
a few basic tools that can be built on to develop more useful/custom stuff.
Sometimes a third party might assert something about a bug that you don't
agree with. There needs to be a way to indicate that you disagree. This
creates two divergent states for the two views of the bug, which will be
tracked separately.
For example, A states that a bug is open, then B states A's bug is closed:
\[[buginfo X state=open sig="gpg-A-1"]]
\[[buginfo X state=closed ref="gpg-A-1" sig="gpg-B-1"]]
But A disagrees, and notes that it is ignoring B's buginfo:
\[[buginfo X ignore="gpg-B-1" ref="gpg-A-1" sig="gpg-A-2"]]
If A later changes its mind, it can unignore B's buginfo:
\[[buginfo X unignore="gpg-B-1" ref="gpg-A-2" sig="gpg-A-3"]]
Of course, B could choose to post a buginfo doing the same, or just repeat
its statement that the bug is closed. Either would re-close A's bug.
\[[buginfo X unignore="gpg-B-1" ref="gpg-A-2" sig="gpg-B-2"]]
\[[buginfo X state=closed ref="gpg-A-2" sig="gpg-B-3"]]
Or B could generate a new key, B', and use it to do the same. Which suggests
that there's no point in having a way to ignore *all* things signed by B.
If B forces the issue this way, A might decide to lock its bug state.
This would prevent any further changes to that branch of the state, unless
they were signed by A.
\[[buginfo X locked=true ref="gpg-A-2" sig="gpg-A-4"]]
If A wants to let a particular buginfo from someone else make a state change,
A would issue another buginfo that merges with it. (A could choose to do
that automatically for buginfos signed by anyone A trusts.)
--
see shy jo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://kitenet.net/pipermail/dist-bugs/attachments/20080710/1637dd12/attachment.pgp>
More information about the dist-bugs
mailing list