Help me crowdsource an archival project for my public radio show.

December 6, 2011 5:05 PM

I am trying to create a search-able, browse-able archive for the public radio show I work for. I would like ask our audience to help me populate this archive, because there are many tech-savvy people in our audience -- and it seems like a project that could easily get done through crowdsourcing. I'm looking for software/websites/advice for managing a large crowdsourced project to make sure the archive is accurate.

My goal is to reduce emails to the show saying things like "I remember hearing an interview with an author who did this, but I don't remember who." The archive, as envisioned, would be browse-able by year, month and week.

I have a friend who is helping me create a database populated with material from an API query for the past few years. However, many of the variables for older shows would need to be filled in by hand because they don't exist in the API or aren't fleshed out the way recent entries are. I'd like to ask our audience to help with this part, if they'd like.

I also have a dream of connecting the material somehow to other variables for guests, so that people could search our show by categories like: topic, guest birth date, show host, month of broadcast, etc. I don't know how to do this.

I am looking to connect with people/software/websites that will make this easier. I know the NYPL did something crowdsourced with menus. If you've worked on a crowdsourced project -- or something similar, I'd love to hear your thoughts.

posted by melodykramer to computers & internet (3 answers total)

Ideally part of the database would link to information in the DBPedia for things like guest birthdates. (I.E. in my head, you could find every guest on the show that shares your birthday) But that's way, way, way down the line and way after we get the entire database up and running.
posted by melodykramer at 5:10 PM on December 6, 2011

Mefi did this with tagging the backlogs of this site and Metatalk; perhaps you can ask the mods for ideas.
posted by divabat at 6:12 PM on December 6, 2011

A Wiki is by far the easiest solution to implement, and provides a well tested solution for the problem. It would be fairly easy to seed the Wiki with data scraped from the API as a starting point. If you want, instead, to build your own database and front end, you need someone who knows databases, a scripting language, and something about web design. Not only do you need to present the information in the db layer, but also provide forms for user input to update the missing fields, a mechanism for such changes, and handling disputes/vandalism. I don't think you are going to find any solutions that will solve this out the box, although many web frameworks will make it easier (RoR, Django, etc). I would be looking into hiring a web developer.
posted by sophist at 11:50 PM on December 6, 2011