[ZODB-Dev] summary of version discussion

[Guido van Rossum]

>   GvR> Current ZODB Versions don't do what you want.  They are
>   GvR> somewhat close to a desirable feature, but that feature would
>   GvR> be implemented differently if it were implemented from scratch
>   GvR> today.
> 
> How would such a feature be implemented differently?

I don't know; I have no use for it myself.  But it sounds from the
discussion so far as if there is something that some people need, and
that versions appear to be a solution of the "close, but no cigar"
kind.

>   GvR> The design of such a new feature should not be restricted by
>   GvR> the existing implementation of ZODB Versions.  So the fact that
>   GvR> there *is* a need for such a feature shouldn't stop us from
>   GvR> ripping out ZODB Versions.
> 
> Unless there's no time to implement the right thing.

But the wrong thing isn't useful, so why keep it?

Anyway, I'll leave this up to your judgement; I expect I won't be able
to convince you and you won't be able to convince me, so "whoever"
maintains the code can decide. :-)

--Guido van Rossum (home page: http://www.python.org/~guido/)