I am writing to you to bring to your attention some new tool that we (Daniel Burgarth and myself) have developed that has the aim of making science papers just a little more accessible. Its called Videoabstracts and consists of 'homemade' videos in which an author of the paper explains the key point of the paper in front of a whiteboard. The videos should not be longer than 5 minutes to force people to get to the point efficiently. We feel that these 5 minutes clarify the content and relevance of a paper much better than any abstract can do.
We have produced several examples that you may see on http://www.quantiki.org/video_abstracts. We did not strive for perfection as we feel that anybody should just be able to do these with a webcam and then upload them on QUANTIKI. The videos will then be stored on YouTube and at the same time a link will be created on the arXiv.
Cool. We need to do one about our latest paper arXiv/0905.0901.
As I said on Joe's blog: "Wow. This is brilliant!"
Yeah, I just watched a couple of them. Pretty nice.
I love this! Maybe someone could write an iPhone app for searching and rating these...
That's a really good idea. I'll be keeping an eye on these.
They've been doing this at a number theory journal for a while now, which I think is great. I'll do the video abstract for our paper while I still have this ridiculous mustache.
It's a cool idea for sure, but my skeptical side is wondering whether people will bother with it much. What is the motivation for doing this via quantiki if you don't already have a login for that site? After all, you still have to go and add the link to the talk to the arXiv record for the paper manually. You might as well just upload your video to youtube directly and add the youtube link to the arXiv record. Moreover, there are sites like SciVee which already do a similar thing on a larger scale and with a better user interface (BTW having the videos upload to SciVee as well as youtube would be a good idea).
These sites need to catch up with Web 2.0 and publish APIs to make it easier to move around between the different services.
I don't see any advantage of a video over a well-written abstract and introduction, but I am an old fuddy-duddy.
> After all, you still have to go and add the link to the talk
> to the arXiv record for the paper manually.
It appears to the contrary that arXiv is set up to ingest links to videos on quantiki directly, see the "blog links" in the right margin of the abstracts on arXiv
> You might as well just upload your video to youtube
> directly and add the youtube link to the arXiv record.
It may be more difficult for arXiv to search youtube for arXiv, the naive search brings up plenty of extraneous material, so in that case you're correct and the user would have to add the link by hand.
> Moreover, there are sites like SciVee which already do a
> similar thing on a larger scale and with a better user
> interface (BTW having the videos upload to SciVee as well
> as youtube would be a good idea).
A naive search of SciVee for arXiv brings up only a handful of videos, all associated to a single abstract. (That too is already linked from the arXiv site.)