Adding binary to the Haskell Platform

[Christian Maeder]

Krasimir Angelov wrote:
> I know that I can use compression to reduce the size of the output but
> this will make the deserialization only slower, not faster.

Couldn't the binary package try to create a ByteString with maximal
sharing directly? I.e. every ByteString gets an index, which is marked
and used, whenever the index is shorter (as ByteString) than the
ByteString it stands for.

(A similar approach is used for shared ATerms, ie. see
http://www.haskell.org/pipermail/glasgow-haskell-users/2005-December/009485.html)

Wouldn't that be faster than the separate compression and decompression
phases and faster than reading and writing such large files?

Cheers Christian