Yesterday we asked a question on the blog about transcribing NARA’s handwritten records.
Thanks to those of you who commented for the ideas and examples! We love the idea of crowdsourcing and have been thinking about how we could make it work for NARA. Projects like the one that Craig mentioned at the Australian National Library, as well as others like the Norfolk Transcription Archive in the U.K. and the US GenWeb group here in the States really demonstrate the relatively low-tech power of collaborative effort.
On the techier side, we’re intrigued by the reCAPTCHA project, which utilizes the common bot-blocking software CAPTCHA to allow users to help decipher difficult-to-OCR words in scanned texts. Plus, as Mark points out, there’s the added benefit that the work people are contributing is presented in the form of a quick, entertaining game! We’ve also been interested in exploring some of the innovative transcription projects that Jeanne over at Spellbound Blog discussed following THATCamp 2008’s “Crowdsourced Transcription and Collaborative Annotation” session.
Of course, as Craig points out, there are questions of motivation and return on investment that need to be contemplated before undertaking mass transcription projects, and while we’re still just exploring the idea, we’re certainly sensitive to those issues.
Thanks again for the great questions and ideas. Your feedback always gives us something to think about!