“Google’s co-founders Larry Page and Sergey Brin, addressing a technology conference in Monterey, California, on Saturday, said they encourage their employees to pursue their own ideas and projects. The company tracks the results with an internal list known as the “Google Top 100,” they said.
“For 20 percent of your time, if you’re working at Google, you can do what you think is the best thing to do,” Page, 31, said.”
– Reuters, Google Founders Keep ’Top 100’ List of New Ideas, Sun Feb 29, 2004
The webmaster of German “www.googl.de” claims this Google-typo has led to his page 728,553 times.
First, there’s VoteLinks. You may know that as far as Google’s PageRank is concerned, a link towards a page increases a page’s relevance. Now this can be any type of link; maybe someone telling “Go check this out, great site” – maybe someone telling others “don’t buy here, I was cheated” or “this site will have nasty pop-ups and crashed my browser”. Now a VoteLink for URLs you like is written like this;
<a rel="vote-for" href="http://www.greatsite.com">Great Site</a>
And a negative VoteLink uses the rel="vote-against” attribute (the third and last option is “vote-abstain”).*
**Yet another approach to judge link value is to check for words like “great”, “good”, and so on appearing in the link text. From my experience experimenting with the Google Web API to judge if a movie is good or not it turned out certain words (within a page, not within a link) are more suited than others for this job. E.g. many people might say “this movie is not good”, but not many people would say “this movie is not lame” or “this movie doesn’t stink” – in other words, occurrence of the word “lame” is a better indicator than the word “bad” to judge if a movie, well, stinks.
“The act of using Technorati to search weblogs should be known as ’blogrepping’ from blog and grep.”
An API is a programming interface, such as the Google Web API which lets me interface Googlebot’s library – I go to the counter and ask politely and he returns the right info (the not-so polite way of doing this is called screen-scraping and directly grabs the HTML, searching it for familiar but flaky patterns).
And Technorati of course is the greatest backlinks checker* out there (with a heavy albeit not exclusive focus on the blogosphere).
*If only Google, or another search engine, would be as nice to show all links they already store (Google, AllTheWeb et al won’t, and it’s like putting a big road block on Developer’s Street).
If you register as Technorati member, you get your free API key (for non-commercial use, up to 500 times a day). You can grab the Technorati XML file by using an URL like the following:
http://api.technorati.com/cosmos? url=blog.outer-court.com&type=link& start=0&key=[YOURKEY]&format=xml
“[YOURKEY]” must be changed to your key. The format “xml” parameter doesn’t need to be used as there is only a single output type format at the moment (RSS might follow in the future). The XML resulting from this call looks like this sample for <blog.outer-court.com>.
Now all you need to do to finish is convert the XML to some usable output like (X)HTML to display, or do some other type of calculation with it. (There are already a lot of Python and Perl interpreters for the “TAPI” out there, though I couldn’t find a ready-made PHP script posted on the wiki.)
FindForward has a new search option: All-around. Using this will cluster-grow your search into a variety of topics, which then yield results from more related topics. E.g. see FindForward’s All-around search for Love.
Meta backlinks: You can now also check links pointing to a URL by choosing “Backlinks” from the drop-down menu (or enter the URL here).
>> More posts