Discussion on cp2k-release-supported-features-wishlist

Axel akoh... at gmail.com
Wed May 21 20:47:30 UTC 2008



On May 21, 2:05 pm, Matt W <MattWa... at gmail.com> wrote:
> To appeal to as large a user base as possible I'd suggest that non-
> orthorhombic cells and cell optimisation (no matter opinions on best
> practice) should be on the list.

matt,

this is not about a large user base. this is about being able
to "deliver". worse than having feature 'X' not included is
having it not working and users getting disappointed and then
claiming that cp2k is not a good software.

i know it is hard to leave out the stuff that is the reason
for many people to try out cp2k in the first place, but
you have to consider that the user base that a release would
attract is less knowledgable than the people hanging out here.

for the others there is always the cvs version...
if you want to attract people, perhaps a list of
"coming attractions" combined with the hint to use
the current cvs version might be a viable compromise.

that doesn't mean i would categorically opposed adding to the
list, but i would put the bar pretty high:
if there is an adequate documentation, sufficient "real world"
tests and a couple examples/tutorials that explain how to use
a feature correctly and how to spot common mistakes in layman's
terms it would be worth considering.

however those features that are slated for rewrites
would be the least likely (it is better to wait). since
there has not been a release at all in all these years
there is not point to rush. like teo said. it is better
to start small and then support in each release a few
more features.

cheers,
   axel.


>
> Matt


More information about the CP2K-user mailing list