Comments on: GCDS summary

By: amondo

amondo — Thu, 10 Sep 2009 08:04:50 +0000

Midgard2 is the future

http://bergie.iki.fi/blog/why_you_should_use_a_content_repository_for_your_application/

By: Jan

Jan — Sun, 19 Jul 2009 18:54:46 +0000

@pvanhoof CouchDB can store these references. it just can’t get you more then one hop of a target doc to a referenced doc in one query by default. *by default* is the key here. Adding a graph-aware indexer to CouchDB is trivial. Please do get in touch if you want to help out with that or like more information.

By: antimonio

antimonio — Fri, 17 Jul 2009 23:19:13 +0000

Totally agree with what you say about RMS: and that each fucking person in the community/planet is blogging about it, is completely a waste of time (of him and of the readers).

By: Tim

Tim — Thu, 16 Jul 2009 18:34:53 +0000

No no no, please don’t cut down on the talks. Even though I do follow GNOME development to some extent (desktop-devel list and other lists), there’s still plenty of stuff I’m missing or only following lazily. The talks form the cornerstone of guadec to me, they’re a good opportunity to catch up with what’s going on elsewhere in GNOME (and beyond). Having said that, I found the scheduling a bit unfortunate as well. I think BoFs and talks should be interleaved with the talks rather than both being clustered in the way they were in Gran Canaria. I don’t think there’s a problem with having more ‘core days’ or having talks scheduled outside of the core days.

By: pvanhoof

pvanhoof — Thu, 16 Jul 2009 17:30:44 +0000

CouchDB can’t store references as object of a triple. So regretfully it’s not very useful for storing Tracker’s triples in. Having references is essential to build a graph. Having rich graphs is essential for Nepomuk as ontology. Nepomuk is what all desktop metadata projects are agreeing on. So it’s kinda important…

CouchDB could, however, synchronize itself with Tracker’s data, of course.