This is Claude Almansi's Typepad Profile.
Join Typepad and start following Claude Almansi's activity
Join Now!
Already a member? Sign In
Claude Almansi
Recent Activity
Thank you for this analysis and the comments on it, mweller,Jesse and Dominik. It's great that we'll be able to call real MOOCs, in the original connective learning sense, MOOCs, and not cMOOcs anymore. I suppose that the signs had been on the wall, in the case of Coursera, for quite a while - but I at least didn't understand where the pattern was leading. One of those signs was their zany mismanagement of their "internationalization" of course content via crowd-sourced participative subtitling on Universal Subtitles (now Amara), which they made such a hullabaloo about at first: see my "Amara autocaptions for Coursera videos" https://groups.google.com/d/msg/universal-subtitles-deaf-hoh/Ud45MJsk_KQ/rGEOSWxo0dMJ Feb 23, 2013 post. Then at the end of February, Coursera abruptly stopped adding videos and automatic captions to their Amara team, and told volunteers, but only the private wiki only accessible via Coursera login, they could copypaste the SRT files of their subtitles in pages of the same private wiki, as it doesn't allow the upload of SRT files. Meanwhile, subtitling by those who hadn't seen that message impressively picked up on the remaining videos of the Amara team, now that Coursera was not interfering with it anymore. In spite of that, sometime between April 8 and April 11, Coursera deleted the team, and on May 15, they announced in https://plus.google.com/+Coursera/posts/4vY33EELMFY that they were "very happy to announce that we are teaming up with organizations around the world to create the "Global Translation Partner Program". This means select courses will have all lectures fully translated into Arabic, Japanese, Kazakh, Portuguese, Russian, Turkish, and Ukrainian subtitles in the coming months! " and the linked blog post specifies that these translations will be done a) for free; b) on Transifex. What this boils down to is that - the number of translation languages has been drastically reduced, excluding e.g. French, Spanish, Chinese - which are official UN languages - and Korean, 4 languages where subtitling had been particularly active on Amara - translating subtitles will become much more difficult on Transifex (as was possible on Amara), because translators will have no way to fix Coursera's original automatic captions or to check them against the video. So their claiming that this is "a giant leap forward toward making high-quality education accessible to anyone, anywhere — regardless of what language they speak" is at best wishful thinking, but possibly just window dressing meant to draw more universities to offer courses on Coursera. Last thing, on the bright side: in spite of the deletion Coursera's Amara team, all the Amara subtitling pages for Coursera videos are still retrievable, via the activities of the two profiles that used to post the videos and autocaptions there: - http://www.amara.org/en/profiles/profile/stanford-bot/ (until the end of December 2012 - 204 pages of 20 activities each) - http://www.amara.org/en/profiles/profile/jngiam/ (until the end of February 2013 - 400 pages of 20 activities each). Couldn't this abundant material be of use for research projects about crowdsourced subtitling/translation? Isn't there an algorithm for data-mining it?
Claude Almansi is now following The Typepad Team
Jun 11, 2010