Good insight from RWW on Yahoo Search plans:
Next year, Yahoo will introduce new technology to augment their Yahoo Search results: abstracts of key information alongside URLs. Instead of just offering a list of links, Yahoo’s search results will include machine-extracted information that is relevant to the URL returned. Sound familiar? The technology is very much like SearchMonkey, except for one thing: this time the technology is being built in-house and not by independent third-party developers.
Being able to see what is behind the links will be a great feature to augment Yahoo’s search engine. However, you don’t have to wait a year to have a tool like that because searching behind the URLs is exactly what this new product called ChunkIt does – and it works on any search engine, webpage or online document. You can use it with Yahoo and see what is going on underneath the URLs right now, there is no need to wait a year.
I’ve been using ChunkIt for the past few months and love it. ChunkIt goes beneath the URLs a search engine provides and pulls out the information you are looking for. It will change the way you search, and the best thing is that it is a free download. You can check it out at http://www.getchunkit.com.
Is this some kind of text-mining engine they are developing inhouse?
Isn’t the speed of modern text-mining engines far behind the speed of crawlers?
I’ve been using ChunkIt for the past few months and love it. ChunkIt goes beneath the URLs a search engine provides and pulls out the information you are looking for.
John:It could be that ChunkIt! already does what Yahoo may be contemplating, searching behind the links.Take the Wikipedia page on the Chesapeake Bay.There are about 200 links.So,if one is writing a paper on pollution in the Chesapeake Bay,type “pollution” in the ChunkIt! search bar and then hit”Chunk Links”.Only seven paragraphs get extracted containing the word “pollution”;ChunkIt! saved you the time it would have taken to click into 193 links and read.Additionally,if you are collaborating on the writing of this paper, all the relevant research Chunks,paragraphs containing the word “pollution”, can be stored in Save and Share which is ChunkIt!’s folder system where the user can have an unlimited number of individually named folders, perpetually storing web research on the ChunkIt! server.Save and Share is only about a week old and I’m up to 25 folders,all containing different topics.Unlike bookmarks, which just store the URL,Save and Share archives the web page you saved today.We all know that a bookmarked URL might have dumped, in the future, what was important to you on that web page you thought you saved today.Also,by giving your id and password to your colleagues,they can contribute to your “Chesapeake Bay Research Paper” folder.Joe
John:It’s Joe Pratt again because since the time of comment 4, above, ChunkIt! has added a collaboration feature (in private beta) to its Save and Share folder system.From there, one can email friends/associates inviting them to open an account at http://www.sharechunks.com which will provide them access to designated folders to facilitate collaboration. It seems that the folks at Tigerlogic(TIGR) who created ChunkIt! might have read the November 22,2008 New York Times article, “The Online Search Party:A Way to Share the Load” in which Gary Marchionini,a professor of information technology at UNC,Chapel Hill, and Microsoft consultant, defined the collaboration problem Microsoft is attempting to address with new tools in development, “Right now I have my screen and I search. My colleague does the same thing, and we have clumsy ways for sharing the results. Why shouldn’t we collaborate on search from the get-go? We can partition the job out, have a place to put joint results and then annotate them in summary sections. So we can say not only “This is what I found” but ” these are the most salient things about it”. That will make us more productive.”