Archive for January 26th, 2005

Old friend, Brewster Kahle, did a recent talk at the Library of Congress which is now online. Brewster is exuberant and ready to digitize the entire Library of Congress using perhaps his 300 page per minute digitizing scanner which is already in use in Toronto. He’s also talking about the Internet Archive’s Bookmoblie which is a print on demand van. Cost per book from the van is about $1. Harvard says it costs $2 per book to loan it. So why not print from the van?
Brewster discusses the problem of books that are out of print but still covered by copyright. These constitute what he calls “orphan” books since you cannot access them in any way.
Brewster is exuberant about the tape-trading groups like eTree.org having offered his “unlimited disk space online forever” to them and to about 700 bands for concert tapes.
A bit about the Archive’s video/music collections and the bandwidth they require. Especially Rick Prellinger’s archive. Rick saw a 40% increase in sales and eventually sold his collection to the Library of Congress after giving away movies on the Archive. Brewster is excited about Lego movies too. Lectures etc.
This is Brewster at his very best. A lot to tell and he tells it well. Each visionary musing is tied to a real ongoing and possibly do-able project at a fairly large and international scale.
The series, which has a link to the talk, is called “Digital Future” will also feature Ed Ayers, Larry Lessig, Brian Cantwell-Smith and others.

Here’s a direct RTSP link to Brewster’s talk.

Comments No Comments »

Although the Yahoo! Video Search has gotten less press than the more limited Google Video Search, it’s different and better in many ways.
While Google has a very limited number of videos — all from recent Bay Area TV — indexed by transcripts, Yahoo! uses www page contexts and file names (ala Google Image Search) for their database. This means you get a huge collection to search. It also means that you can actually watch the video once you’ve discovered it. With Google, you only get images snapped at about 30 second intervals.
Frankly the two searches attack very different problems. Yahoo! is about resource discovery and not much about the content within the video at any level of granularity. Google is much more about mining the content within a video stream — but mostly by using closed captioning as a transcript (or so it seems).
You can submit your videos to either system. Yahoo! tells you to use RSS enclosures:

Enter the full URL, including the http:// prefix (for example, http://www.yahoo.com/mymedia.xml) of the RSS feed (either with Media RSS or RSS 2.0 enclosures) you would like to submit to Yahoo! Video Search. Dead feeds will be automatically detected and removed on an ongoing basis. For additional information about Media RSS feeds, see the Media RSS specification. Your RSS feed will be crawled soon. After your feed is crawled, please expect a delay before your content will appear in Video Search results

Very cool and very clued.

Google uses only a web-based form for inquiring about adding your collection.

A great benefit of being “Paul Jones” is that you can seek and find many of your name-twins’ videos. All these and more found using Yahoo!’s Video Search.

Google video search for Paul Jones returns nothing.

Comments 1 Comment »