Re: [Monotone-devel] Re: current multiple heads (was Re: write access to my public server)

[Jon Bright]

graydon hoare wrote:

> there is an interesting problem which arises from this: disapproval 
> isn't quite as "final" as I had initially hoped. in particular, consider 
> this scenario:
>
>   A->B contains badness
>   A->C is OK
>
>   - I review A->B and disapprove of it

An interesting sub-aspect of this is the sequence A->B->C where A->B is 
bad, but B->C removes the badness with the result that C is good. 
Clearly, in this case, the badness shouldn't be inherited by C.  But 
there's the same case A->B->C where A->B is bad, and B->C just adds a 
debugging printf or so.  In this case, C is still bad.

Clearly it's not possible to handle these cases automatically.  It would 
seem to me that the only safe way to do this is for B->C and all 
subsequent descendant edges to initially inherit the disapproval which 
you originally applied to A->B.  If you then change your mind about what 
C looks like, you issue an approval for that edge - which is, unlike the 
disapproval, not inherited, but does curb further inheritance of your 
disapproval.

--
Jon Bright
Silicon Circus Ltd.
http://www.siliconcircus.com