Hi everybody, I'm pleased to announce that Petr Ročkai's project on optimising darcs has been accepted for Google Summer of Code.
More details can be found here: http://socghop.appspot.com/org/home/google/gsoc2009/haskell Petr's project fits into an ongoing effort by the darcs community to improve our performance so that one day we will able to recommend darcs for medium sized and larger projects. Until early 2008, we have been focused on darcs 2: which provides the darcs 2 theory (to solve most of the exponential merge problems), and also the "hashed" format repositories (including darcs 2). The project for this summer will allow us to complete the work on hashed repositories. In their current state, hashed format repositories make darcs a lot more robust and also allow us to fetch repositories faster (with darcs get --lazy and with a global patch cache). But in their current form, hashed repositories can can also slow down day-to-day operations like darcs whatsnew and darcs record. Petr has identified the cause of these delays has provided a work plan which we think will get hashed storage right. If his project is successful, future versions of darcs should make operations like darcs record and whatsnew much more comfortable in practice. Congratulations and good luck to Petr! And many thanks to the darcs and Haskell communities for their support. -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> PGP Key ID: 08AC04F9
pgpnAFiBOvhlW.pgp
Description: PGP signature
_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
