Ideas & Involvement

This page describes project ideas for some of the open source projects we work on. If you’re a student or outside developer interested in working on these projects, just start now! Of course, a little bit of support might be nice. Feel free to email us directly at code [at] creativesynthesis.net. If you are interested in some kind of support, we *might* be able to provide it, but please include the application details (as below).

Project Ideas for Graph Gear

• Add support for other graph types. We currently use a pretty simplistic force directed layout. There are lots of more complex options that could be explored if they are appropriate enough (See: http://www.aisee.com/cgi-bin/examples?topic=2). The goal isn’t to support “every kind of graph”, rather every kind of graph that makes sense. Before implementing other graph types it would be nice to have a good sense of how realistic the layout is computationally (we’re pretty limited).
• Add support for live data formats. Graph Gear currently depends on a pre-constructed xml file to load graphs from. It also doesn’t ‘really’ support tiered loading. Both of these limitations leave out hugely useful opportunities. It would be great it graph gear could automatically follow and build graphs of solid web properties (facebook, flickr, etc.) and open standards (social graph api, foaf, etc.). Ideally any explorations could be saved as a graph xml file for later viewing.
• Ports. Graph gear could stand to be ported to a few different things. An actionscript 3 (flex or not) port certainly makes sense. A javascript / canvas port would be exciting if it could achieve comparable performance, but would certainly be more difficult. (Java is not really a reasonable port - there are a lot of heavy java graph solutions)
• Add other graph structure elements. Possibilities include other node types, adding direction to the graph (graph gear supports direction only in the slightest way), and better support for visualizing multiple (separate) graphs on the same layout.

Project Ideas for Weblab

• Widgetize weblab. Weblab is built out of open source components and web services. Some of them have support for ‘widgetized’ views and sharing, but many of them don’t. Ideally there are lots of views for sharing this information in ways that benefit the lab.
• Add semantics/microformat support. While weblab is a hosting environment (and not really designed as a highly distributed collection of services) we’d like it to be able to branch out from the server in smart ways. A lot of the formats with the most utility to us are still in early planning stages. Developers should be prepared to make gutwrenching choices and code for future changes.

Application for Support

If you’re really great and would love to help out, but need some support you can consider filling out the following application info and mailing it to us.

Name: Email: and Project Title: These should be pretty self explanatory.
Synopsis: A short description of your project.
Benefits to the Collaborative Mission / General Public / Etc.:

Deliverables: Quantifiable results. Also keep in mind that we like reusable education things as well. E.g. “At the end of my project, GraphGear will support directed graphs.” and optionally “I’ll create a screencast to teach other students about force-directed graph algorithms using GraphGear as an example.”

Project Details: A more detailed description of the project you want to do.
Project Schedule:How long will the project take? When can you begin work? Do you know of any planned absences or other major conflicts (summer classes, vacations, etc.)
Bio:Who are you? What makes you the best person to work on this project? If it makes sense here please include any appropriate scholastic information (school/major/etc.).
Amount Requested: Pick the smallest amount you could possibly need, then divide it by some number.

Mentors: Matthew Hockenberry for Graph Gear. Matthew Hockenberry / Leonardo Bonanni or Connor Dickie for Weblab.

Other possible project ideas: Anyone interested in working with the mousetracking code should get in touch with Ernesto Arroyo. Pez is a project we don’t personally control, but is based off of the personal zeitgeist (right now actually and always at least conceptually), if it is worth exploring for a student, we could consider funding that support if it jives with the developers.