I've looked at PerlMonks before, although it's been a while. As far as successfully automating Useless Python goes, I've got a few general thoughts. My vision: - Newbie (or non-newbie) decides to upload a source file and fills out a form to do so. Form elements would be the author's name, a description of the program (brief), and a field to browse their hard drive to select the file. An administrative email is sent to the small team of volunteers who approve/deny the submission. Denial would only occur in cases in which the submission is wildly irrelevant (such as it not being a Python program in any way). - For browsing files to rummage through, which is currently done by clicking on the numbered pages, I would like to add options. This could mean searching for keywords or modules used in the source files, which might make Useless Python a truly powerful repository. Alternatively, we could have a database of general categories that source files fit into (such as "GUI", "socket", etc., etc.), which is vastly more limited, but probably easy enough to implement. There appear to be a number of web hosting services available where we would have access to any number of resources, and costs appear to be non-prohibitive with some Python-friendly hosts. Any thoughts? Rob
-----Original Message----- From: Danny Yoo [mailto:dyoo@hkn.eecs.berkeley.edu] Sent: Monday, October 21, 2002 5:03 PM To: Rob Cc: edu-sig@python.org Subject: RE: [Edu-sig] re: Slighly OT: O'Reilly article [other resources for newbies?]
I feel that navigation is the second weakest element of the site, and have been giving a lot of thought to how to address the matter without messing up the experience too much. The site's single greatest point of weakness is that it lacks automation and is maintained manually by one person. Automating the site could well be the means to resolving both the navigation and management issues.
I wish I knew more Zope to help out here; I'm still trying to get my head around it. But from what I understand, perhaps Zopifying Useless Python may help with the content-management side of things!
The tricky part is that the original design concept was indeed for the user to rummage through the randomly-ordered source files. This didn't seem as problematic when there were only a few dozen source files, but with the next major update the number should increase to well over 200. At this size, I can certainly understand the frustrations expressed by a few. (Fewer than half a dozen to date, including today's comments.)
I know it's heretical to look at what the Perl folks are doing, but I can't help it. *grin*
The Perl folks have a site called http://www.perlmonks.org that combines aspects of both the Tutor mailing list and Useless Python. It's a large repository for Perl knowledge.
It might be nice to see what things perlmonks is doing right, and see if we can apply those things to Useless Python.