Blog Post

Media Ecology Project - Ideas for Crowdsourcing?

I'm just back from the Association of Moving Image Archivists annual conference where I spoke as part of a panel on the Media Ecology Project, an effort initiated by Mark Williams at Dartmouth to increase scholarly access to archival media. MEP is something I've mentioned in these posts before, but the video below gives a much better idea of what we're trying to accomplish with the project. We'd love to get some feedback from HASTACers!

Media Ecology Project: Sideloaded archives from John Bell on Vimeo.

The video goes into more depth but, as I mentioned, the core goal of MEP is finding ways to increase scholarly access to and use of media held in archives around the world. It does that by making the media held in those archives more discoverable through enhanced metadata and citation capabilities. We're also tying archival media into analysis and publication tools like Scalar and MediaThread more directly than has been possible with pre-semantic web software.

But the question remains: if we are increasing the amount of metadata scholars can use to find and index media, where is the new metadata coming from? As more and more digitization projects come online the quantity of video available has outstripped the amount of time archivists can devote to making it discoverable. A key concern of any attempt to generate third-party metadata about archival collections is a question of what qualifies as good information. Crowdsourcing is an option that we'd like to make more common, but there are issues of provenance any time you try to crowdsource basic data.

We've explored several potential solutions to the provenance question, ranging from algorithmic trust systems to OBI badges awarded to those who produce great metadata. But what would HASTAC do? Federated login systems? Badges? Mechanical Turks? Expert-only input (who is an expert)? I'm very interested to hear what kind of ideas HASTAC readers and scholars have for fact-checking crowdsourced metadata.




We've partnered with some of the folks at EUscreen to develop Open Journal Systems and it sounds like they might be an interesting source for you as you seek access to more media (particularly European media). From their About:

The EUscreen project aims to promote the use of television content to explore Europe's rich and diverse cultural history. It will create access to over 30,000 items of programme content and information, and by developing a number of interactive functionalities and dynamic links with Europeana it will prove valuable to the widest range of cultural, educational and recreational users.


Thanks for the link, Paul.  It looks like a wide ranging collection with lots of good content.  Do you know how they're handling access rights?  From this FAQ it seems like there isn't any kind of programmatic access or embedding allowed; how have you been using the EUscreen content?


Your rights question is a good one and I am not at all sure how that is being handled. It appears that big content owners (Deutsche Welle, Dansk Radio, BBC, and so on) are partnering with them and I am assuming that means that there is some limited licensing happening. It would be interesting to know how the US model of Fair Use is related or compares to what they're doing in the EU.

I haven't used EUscreen's content to this point. I help make an online journal called continent. with some friends and EUscreen is a friendly peer that we met through a discussion at the Open Journal System's development board. Since then we've done some light collaborating, including a session at the Beyond the PDF2 meeting earlier this year

As for your question about the crowd source provenance problem (not that this is necessarily the best informed opinion, but), it seems to me that if crowd-sourcing is the approach used then the problem of who gets to profit off of the copy rights will be off-set by the money a firm can make by harvesting the crowd.

Isn't that the old saying? If a firm is giving away their software it's because the software isn't the product, the user is? At some point the digital content provider accepts this as well or the Internet as we know it changes dramatically. Both seem like equally realistic options to me right now. The Internet becomes a series of walled-off gardens or we are all "copper tops" a la The Matrix (thanks, Facebook for making that so pallitable).