hi everyone!
i think we really should have a hackathon to speed up lenya development
and give the developer and user community a boost. wdyt?
a meeting place in the ruhrgebiet could easily be arranged, and even
cheap or free private accommodation, but from what i know of the other
lenya devs, somewhere in switzerland or southern germany would probably
be more practical...
personally i'd be willing to take a few days off and fly my humble ass
to almost anywhere in europe, provided there's going to be enough
advance notice to get a cheap ticket. what about you other folks?
i'm volunteering to coordinate the whole thing, but depending on the
location, somebody who actually lives in the area needs to do the grunt
work of securing a place.
as to timing, i suggest something in the second quarter of 2007,
probably in april or may. we could also meet during the apachecon
europe, which takes place from may 1 to 4 in amsterdam, or better yet,
before or after. if you think amsterdam is an ok place, i'll get in
touch with arje of cocoongt fame - maybe we can find a place in the same
building that has hosted the get together for the last 2 years.
regarding the agenda, let me first state that i'd like to see 1.4.0 out
well before the hackathon, so that we can have some breathing space for
more visional ideas.
topics that come to mind are:
* improve and enlarge the lenya community
* resyncing with the world - users use 1.2, devs work on 1.4, and the
two fractions barely seem to know the other exists
* syncing with cocoon 2.2, talk about spring
* writing down lenya coding best practises, discuss and adopt
"convention over configuration"
* improve the storage backend, provide a migration path between lenya
versions
* define what lenya is good at, and seriously figure out reasons why the
world needs another cms :)
* make lenya more data-friendly, loosen the document-centric structure.
* hack and code and test and break stuff and have fun.
as a side note, i'd also like to revive the discussion of "code
ownership". i agree that "ownership" is bad, but "code shepherding"
could be a good thing. code shepherds would be people who feel
particularly responsible for particular sub-systems and have (by merit)
a certain authority on new additions and architectural decisions. such
"code shepherds" could then give talks at the hackathon about their pet
code and maybe even tutor newbies to help them become developers.
as it is now, there are traces of really nice "best practises" all over
the code, but they tend to be swamped by code contributions that do not
honor them (which is natural, since those practises are not documented).
a code shepherd could demand consistency and suggest improvements to
patches which are not in keeping with good practice.
an additional benefit of this scheme is that we can identify sections of
code that are unmaintained and not thoroughly understood. ideally, each
subsystem should have its own shepherd. (i know, this is not very
realistic atm given the size of the dev community, but it's a worthy
goal to strive for.)
i'd also suggest to ask around on the user list for advanced users who
are interested in learning about particular parts of the lenya code and
become devs. we would then have to organize special tutorials for those
users to give them a kick start, but it should certainly be worth the
effort.
that's it for now, i'm looking forward to your comments.
best,
jörn
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
