On Nov 2, 2007, at 17:35 , Andrew Coppin wrote:
These are the things I'm thinking about. Is there some deep theoretical reason why things are the way they are? Or is it merely that nobody has yet had time to make something better? ByteString solves the problem of text strings (and raw binary data) very nicely, it's just a pitty we can't apply some of that know-how more widely...
I'm under the impression that several of the things that make ByteString faster (e.g. smarter fusion) are either implemented within (newer) GHC already, such that other list-like types can take advantage of them directly, or being worked on as part of a new Data.Stream module that anyone can use with their own types instead of simple lists. So in the long run it *won't* just be ByteString; that's just what's driving the development. -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery@kf8nh.com system administrator [openafs,heimdal,too many hats] allbery@ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH