Re: [Gnu-arch-users] Status of global and tree aliases

[Aaron Bentley]

Tom Lord wrote:
> Still, I don't think that that's a semantic difference.  Rather, I
> think that "per-version" and "per-tree" are equivalently expressive 
> but optimize the syntax and implementation for different patterns of
> use.

To the extent that per-tree variables can be related to the namespace, 
they're more like per-revision variables than per-version.  That is, if 
you update tree variables and commit, the only users who will see the 
values are those who retrieve that revision or a later one.  (Assuming 
the later one hasn't changed the values again.)  But if you update 
version variables, users of any revision in the version are affected.

Still, I don't see them as revision variables, because they aren't 
necessarily archived.

I think it's worthwile to distinguish between the two types because they 
have different
- scope
- retrival requirements
- alteration requirements
- inheritance characteristics

I do agree that since they can be used for similar purposes, it would be 
foolish to treat them as unrelated.

To get back to the matter at hand, if we call them all "version 
variables", we can't talk sensibly about what James has or hasn't 
implemented.

Aaron

--
Aaron Bentley
Director of Technology
Panometrics, Inc.