Here are the details from our first darcs hacking sprint ever!
So far, we are offering venues in the UK, USA and France. There will be jobs for both new and experienced hackers. Join one of the teams, or come hack with us on IRC #darcs (freenode).
- CONFIRMED - Brighton
- CONFIRMED - Portland
- CONFIRMED - Paris
Site: Confirmed! University of Brighton, Moulescoomb campus, Watts Building, Lewes Road, Brighton, BN2 4GJ
Access: Either the 25 or 49 in the Universities or Moulescoomb directions, or train to the Moulescoomb station. EricKow’s mobile number available upon request
Food: There are pubs within walking distance of the University for lunch. We would probably go into central Brighton for dinner at least Saturday night and maybe Sunday depending on how people feel.
Please bring: Laptop (network access will likely be provided)
Site: Confirmed! Galois Inc, 421 SW 6th Ave. Suite 300 Portland, OR 97204
Access: This is in downtown Portland. Please get in touch with JasonDagit
Please bring: Laptop (wifi on offer)
- DuncanCoutts (this or Brighton)
- TimChevalier (possible, but will probably participate in the influence-the-presidential-election sprint instead)
Site: Université de Paris 7, 175 Rue du Chevaleret [please bring ID]
Contact: NicolasPouillard (note: we need a full list of participants by Wed 2008-10-22)
- Laptop (switch will be provided)
- NicolasPouillard - Paris
ReinierLamers - Utrecht, NL
Site: #darcs on irc.freenode.net (maybe you could organise your own venues?)
During this first sprint, we shall be focusing our attention on the day to day performance issues that darcs users commonly face.
This is what we are reaching for:
- Fast network operations. We want to make it very pleasant for users to darcs get a repository and pull some patches to it over http and ssh. Git does this very well, and we plan to learn from them.
- Cutting memory consumption. We want to profile the heck out of operations like darcs record, darcs convert and darcs whatsnew. What’s eating up all the memory? And how can we can cut it down to size?
- Responsiveness. Sometimes basic darcs commands can take long enough for programmers to lose their train of thought. We want to track down these lost seconds and kill that dreaded context switch.
- Even better UI (lots of nice features to assign individually, for example, issue75 and issue126) Of course, if you are interested in other areas, then you can work on those instead.
- http://bugs.darcs.net/issue390 whatsnew lstats every file (Reinier?)
- http://bugs.darcs.net/issue80 record takes 2x memory (Jason and Don?)
- performance regression suite
- identify good 5 StandardDarcsBenchmarks that can be run on the buildbot, and add them to the repository (Eric)
- integrate Ian’s memory suite into darcs? (open!)
- make it possible to view only the benchmarking results in buildbot (open!)
- summarise benchmark results in buildbot (open!)
- new performance features
- on disk chunky hunk format (would be a new compatible repo format like old-fashioned vs hashed) (IanLynagh, JasonDagit, DonStewart)
- packs, as a network (latency) plus a responsivenes (less seeking of files) issue? (NicolasPouillard)
- cache for darcs annotate (GaneshSittampalam,BenediktSchmidt)
- poking at darcs
- profile darcs for memory usage on StandardDarcsBenchmarks (needs fleshing out) (open!)
- audit bytestring use in darcs (needs fleshing out) (open!)
- all easy jobs
- very easy jobs
- dist –tag
- send –gpg-command
- tag names
- unpull messages
- record -m x (taken by Reinier!)
- less easy jobs
- show status command?
- splitting hunks (!)
- tour of the darcs source (Nicolas and Jason)
- early - it would be nice if we could help our new hackers, Gaetan and Reinier get up to speed with the darcs code
- (talk about) iteratee/enumerator refactor? (JasonDagit and DonStewart)
- working on camp (IanLynagh)
- Reinier (thinking of issue 390)