Hi all,
Here's my idea for the google summer of code - it's a resubmission
from last year. We had a couple of interested students, but their
proposals weren't quite competitive enough to get funding.
Regards,
Hadley
Summary: Create an information portal for the fast growing list of R
packages, integrating package documentation and journal publications
with feedback from the useR community.
Required skills: Familiarity with Ruby and R, and web development in
general (html, css, js, ...). knowledge of RSS/Atom could be useful.
Description: The CRAN archives contain over 1300 source packages of
very high-quality, and BioConductor has another 200+. For many data
analytic tasks there is more than one package offering a solution,
often using different names for similar functionality, because
different research communities may use different vocabulary for
similar methods (e.g., statistics vs. machine learning jargon). Hence
it is often hard for users to find the packages they really need. CRAN
task views offer some help, but are edited by designated maintainers
and hence cannot cover all fields and aspects.
crantastic.org implements a first step in complementing this with
package reviews from the useR community, but it still lacks many
features we would like to have: integration with task views,
vignettes, and journal publications; match authors and users based on
email address; list top users and authors on respective pages;
implement a rating system; event streams for packages and users;
integration with CRANberries RSS feed, etc.
If you're interested in this project take a look at the current code,
available at https://github.com/hadley/crantastic/tree, and include a
couple of concrete suggestions as to how you would improve the current
code, and how you plan to implement the additional features.
--
http://had.co.nz/
