CPANHQ - a community-driven, meta-data-enhanced interface to CPAN.
The current CPAN interfaces - http://search.cpan.org/ , kobesearch, and the rest of the CPAN services that various individuals have set up - are not integrated, are not based on enough meta-data, are not enough community-driven and are generally lacking. As a result of all this, some people started the "Rethinking CPAN" effort:
CPANHQ aims to be a community-driven, integrated, meta-data-enhanced interface to CPAN that will make finding a module on CPAN faster, easier and more fun.
The following new features will be added to the CPANHQ codebase:
Perhaps by making use of the MyCPAN::Indexer manpage .
CPANHQ aims to be a community-driven, integrated, meta-data-enhanced interface to CPAN that will make finding a module on CPAN faster, easier and more fun. I originally announced CPANHQ in a blog post:
More information about CPANHQ can be found at the following locations:
http://github.com/shlomif/cpanhq/tree/master - shlomif's CPANHQ repository at GitHub.
http://wiki.github.com/bricas/cpanhq - a wiki page about CPANHQ.
http://www.shlomifish.org/open-source/projects/CPANHQ/screenshots/ - CPANHQ screenshots.
The project will take 2-3 months. I can begin work on it immediately.
I am an active user, developer, and advocate of Perl and other open-source technologies. I maintain many modules on CPAN (http://search.cpan.org/~shlomif/), and have contributed to other Perl projects. I am also proficient in C, C++, Assembly and other languages and have been actively involved in C and C++ projects.
I have successfully completed the Perl Foundation's "XML-RSS Cleanup" grant:
( short URL: http://is.gd/1nYZM )
I maintain an active homepage at http://www.shlomifish.org/ which contains more information about me.
Sounds like a really good idea, but your proposal doesn't answer the problem that is proposed "The current CPAN interfaces - http://search.cpan.org/ , kobesearch, and the rest of the CPAN services that various individuals have set up - are not integrated, are not based on enough meta-data, are not enough community-driven and are generally lacking."
How will CPANHQ address those challenges? After a cursory review, it looks like Yet Another CPAN Search Interface.
Is there not an opportunity to build on the efforts of search.cpan.org and kobesearch rather than introducing another site into the mix?
I personally have had difficulties in finding the `right' modules using http://search.cpan.org.
After reading the points listed at http://wiki.github.com/bricas/cpanhq, it is certain that CPANHQ will bridge the gap between programmer and CPAN. "User-contributed labels/tags" is a great plus! After all, AI cannot simulate human judgement.
http://cpanhq.org/ looks attractive as well. That would certainly increase interest in CPAN and Perl.
I support this project.
While I think improvements to CPANHQ are neat I think that if the Perl Foundation wants to support the user interface to CPAN it could best do so in the short term by prodding the search.cpan.org author to release the source to that site.
A lot of people apparently want to make relatively small changes to search.cpan.org, but the only way they can do so is by rewriting the whole thing from scratch with projects like CPANHQ or kobesearch.
Obviously CPANHQ is aiming for something slightly different so this isn't really pertinent to this proposal directly. But it is pertinent to TPF spending money in this area.