This project is designed to be able to provide avid readers of RSS feeds with the ability to quickly seperate high-quality (interesting) articles from the rest of the content.
Currently, methods of filtering utilizing the Naive Bayes method of categorization are being tested for applicability.
Comment on 'RSS Feed Training'
A quick little web script that generates all the possible tea coupons made available by Traditional Medicinals.
This is a partial mirror of one of my favorite Internet pastimes: lowbrow.com. This mirror of over 15,000 entries was made over the course of a night, until I got in trouble by their ISP. Nonetheless, here's the results - plus with the ability to link to a specific post, something the official one lacks.
Tags: db, web, humor, perl, mysql, scrape
This application exists as a layer on top of Google Local, which can then be used to socially network with other users, generally in an attempt to find new friends with similar interests.
Most of the features are complete with the notable exception of being able to add friends. If there's enough interest in it, I'll be happy to finish it up.
2 Comments on 'Google Local Social Network'
I was disappointed by the lack of lightweight XML-based template presentation apps out there, so I whipped one up real quick and this is the result. This was written while on Co-Op at the Lab for Applied Computing. The premise for this application is that you have a XML file containing all of your slides for a presentation, then a template file (HTML::Template) which is applied against each slide and then rendered out in a CGI application. Very easy to use and get going.
Comment on 'Template Presentation'
This web application was designed to be used by CSH members who needed to keep track of interviews and read them again later. Currently this application is available to strictly CSH members.
Comment on 'CSH Interview Manager'
This site was created as a fun utility of sorts when an auto-updating listing of web sites on the rit.edu web server was located. (For anyone familiar with Apache, this particular feature is commonplace.) A nice aspect of hunting down pictures using this method is that pictures crop up that would normally be unavailable to web spiders - providing others with access to fun, and generally hidden, information. Essentially, the purpose of this program is to display fun pictures located in (typically) student accounts on the rit.edu domain. By browsing these images you get a wide variety of RIT flavor (anime, math, programming, more anime, etc.) - available in a slideshow format. Using the options available on the preferences page, one could easily make a custom screensaver for their computer, or even turn it into a simple drinking game of sorts (take a sip whenever you see an animated gif, a shot for a math formula, etc - the possibilities are endless).
Note: The online demo is currently offline.
Tags: db, web, perl, pgsql, rit, scrape
This was a fun little hack that I wrote as both comic relief and as a test of rule-based scoring systems. With this application I deveoped a database-driven rule-based engine with which to categorize a body of text (specifically, a web log). The application worked well and was quite popular over its lifespan, with over 5000 people using it.
The site use to be located at emotastic.com, but I allowed the domain to lapse this year as its usage had begun to trickle off and didn't see a need to keep it running. Instead, the site can now be found at its alternate URL.
Secret techniques of top JavaScript programmers.
The best techniques for professional JavaScript. Published by Apress.
Infrequent, short, updates and links.
