Re: Issues, Issues, ???

On Tue, Dec 19, 2000 at 05:55:16PM -0500, Geoffrey M. Clemm wrote:
>    From: Greg Stein <gstein@lyra.org>
> 
>    On Tue, Dec 19, 2000 at 03:50:36PM -0500, Geoffrey M. Clemm wrote:
>    > 
>    > (1) Should version history URL's be in core (i.e. be required)?
> 
>    Prefer no.
> 
>    I don't have the mindset for a "core" server, so I'm not sure why this would
>    be needed.
>    [ as Boris pokes fun :-) ... yes, I need it for a REPORT, but that is it. ]
> 
> Well, actually, anyone that implements versioned collections will
> end up providing version history URL's because each member of a
> collection version is a version history.

Only if you allow "slashing-through" a collection version resource. :-)

This does imply that a core server would not implement collection versions.

> So if /repo/ver/73 is a collection version with three members named
> a, b, and c, then /repo/ver/73/a, /repo/ver/73/b, and /repo/ver/73/c
> are all version history URL's.

That's the theory :-). Honestly, I probably won't be allowing that in the
first revision. I've got enough to do as it is...

> With this in mind, do you want to reconsider your "prefer no"?
> (I.E. why would you prefer we not require something you are planning
> on doing :-).

We're talking about putting them into core, not whether I'll be implementing
them. :-)

Version history seems a relatively complicated beast, so making it core
seems like it would need a justification. Therefore, my "prefer no" and a
request for why a core server would *need* them.

>...
> (And also note that working collection members will be initialized
> to be version history resources, just to bang that drum one more time :-).

Sure, but that isn't the question at hand, now is it :-)

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

Received on Wednesday, 20 December 2000 02:21:41 UTC