> > To do this would require several things the current Dynamic library
lacks:
> [...]
>
> I think it's much harder. How would you store and restore actual data?
In general, I agree: a lot of work and thinking would be required for
a full solution. But a lot of thinking has been done already, and as we
currently have no persistence for Haskell, I am tempted to see any
progress on the implementation front as a good thing. Even incomplete
implementations would allow for some experimentation, and may
provide partial and temporary solutions, so if Tim has some concrete
plans that are good enough for his application, I would be interested
to read about them.
As for the `real' thing, I think the token for working on Haskell
persistence is currently in St. Andrews (which probably doesn't
mean that they would say `no' to any help offered?-). See:
A.J.T. Davie, K. Hammond and J. Quintela
"Efficient Persistent Haskell"
In Preparation, February 2000
http://www-fp.dcs.st-and.ac.uk/publications/2000/phaskell.ps.gz
Some of the claims there are a bit strong, actually ("first-ever
implementation of orthogonal persistence for a compiled purely
functional language.."), but may be correct if you take the very
long list of qualifiers very literally. Shows that there have been too
many papers about this subject already, so those who do the work
to get these nice ideas running for other languages (such as Haskell)
have difficulties to publish their progress..
Just for completeness,
Claus
