Slashdot Mirror
Larry Page: Google Was an Accident
DarklordJonnyDigital writes "Ars Technica is reporting that Google founder Larry Page has admitted that the Google project wasn't originally intended to be a search engine at all. "It wasn't that we intended to build a search engine. We built a ranking system to deal with annotations." ' Of course, happy accidents have often been the cause for advancement, technologically or otherwise.
many great inventions/discoveries are accidentally invented/discovered.
Newton's Law, gravity constant, etc
Archimedes' buoyancy Law
Accident or not, I'm glad it happened. Search engines at that time left much to be desired. Google was simply magic. If I wanted something, it would magically appear on the first link.
The guys who created the Expand Accellerator were actually trying to develop a new encryption method when they stumbled across a method to increase virtual bandwidth.
Jerry Yang's original set of links was a Sumo wrestling enthusiast's page...that for a time was valued at $120 billion dollars (!).
Here in Google groups..
Now can someone find the first mention of searching Google looking for the first mention of Google in Google?
Not really, previous search engines did well what they were intended to do. They searched the web focusing in each site as isolated in the web.
But used the wrong point of view, they didn't see the web so interlinked that searching based in how much linked a site is could be a measure of how much desirable could be find that site.
Sometimes the better solutions are just viewing a hard problem from another point of view.
After reading all the information on google's technogoly, I wonder how many lines of code pagerank really is.
Do you figure it's 50k+lines, or something very simple, and only a few hundread lines
For some reason, I don't think the pagerank algorithm is more than 1000 lines of code... I know lines of code isn't really a defining characteristic of anything, but it's still interested...
From the article
Larry Page: "It wasn't that we intended to build a search engine. We built a ranking system to deal with annotations. We wanted to annotate the web--build a system so that after you'd viewed a page you could click and see what smart comments other people had about it. But how do you decide who gets to annotate Yahoo? We needed to figure out how to choose which annotations people should look at, which meant that we needed to figure out which other sites contained comments we should classify as authoritative. Hence PageRank.
"Only later did we realize that PageRank was much more useful for search than for annotation..."
Now think about blogging with page ranking applied. Might be much more useful than normal blogging. As search engines with PageRank are compared to normal search engines.
Bye egghat.
-- "As a human being I claim the right to be widely inconsistent", John Peel
Mmmm I should check Google Labs before saying something that looks so obvios, they already doing it in Google WebQuotes
From my experience with meta search engines, they end up taking up more of your time. Each query gives you the first 10 responses from each engine. That's fine if you are looking for something pretty easy to find (but then you probably don't need a meta engine anyway).
On the other hand, if you are looking for a needle in a haystack, as I seem to be usually, using a meta engine just means you have to wade through that many more pages before getting to the stuff you want. After a few months of using Dogpile, I ended up deciding that my time was better spent going deep into Google. YMMV, natch'.
Congratulations! Now we are the Evil Empire
I wonder how many lines of code pagerank really is.
Try one equation, iterated a few times:
However, the PageRank value is only one aspect of Google's ranking; for brand-new pages that haven't had time to gather links yet, Google seems to use straight textual ranking.
Will I retire or break 10K?
Google's not only an accident, but also a misspelling: It should be googol.
Although I'm kinda glad it got misspelled though, because google is much cooler that googol.
Interesting googol fact from whatis.com:
Later, another mathematician devised the term googolplex for 10 to the power of googol - that is, 1 followed by 10 to the power of 100 zeros. Frank Pilhofer has determined that, given Moore's Law (which is that computer processor power doubles about every 1 to 2 years), it would make no sense to try to print out a googleplex for another 524 years - since all earlier attempts to print a googleplex out would be overtaken by the faster processor.
Who said Freedom was Fair?
...the Information Retrieval (IR) geeks reckon there's 2 major factors. You are correct that one of those is relevance, which is known as precision. And the other is recall. Think of recall as getting all the relevant results.
One of the tricks that can be used to cull irrelevant results is to cut down the total number of results. The IR dudes quickly started playing the numbers. Showing the best 20 results is better than showing the top 100 with 60 of those being irrelevant.
I like to think of these as accuracy and completeness.
I used to occasionally browse through TREC. Seems like they have locked up the past results nowadays...
So does Anonymous Coward have good karma?
A friend of my brother-in-law was suprised to hear that there were other search engines in existance.
He thought that Google was just a standard, like HTML, FTP, Gopher, or NNTP.
That was quite the little accident they had.
Google have a top-notch system but the whole indexing thing is still laughable. They are not really taking advantage of structured markup in evaluating keywords - they extract the same information as if it were a plain text file sans markup. Yeah, sometimes top-level headers and link text is used, but that's it really.
Its good, however, to see that Google aren't resting on their laurels, as Google Labs amply demonstrate. I like Google sets, which makes good use of list markup, like when the shuttle crashed last week I was trying to remember the names of all the space shuttles, so entering Colombia, Challenger and Enterprise into Google Sets gave me the names of the other three shuttles, Discovery, Endeavour and Atlantis -- a useful tool indeed.
Considering Google's purchase of Blogger announced this past weekend, I'm looking forward to more semantically based search abilities - since blogs are by their nature very structured (especially those with RSS or XML feeds).
The fact that these guys accidently created a search engine that blows all the other ones away kinda says something about the laughable state of search engine technology before google, don't it?
I think it says more about the business of search engine technology at the time. All I ever used before Google was AltaVista, and I started using it back when it was a demo for DEC's 64-bit Alpha chip (side-note of irony is that the much better Google search engine uses all 32-bit Intel architecture). AV started out as 5 or 6 Alphas networked running Ultrix, and it simply indexed the web. I still use it for exact phrase matches, simply because it does a better job at that.
But when Google came out, AV had been split out into its own company, tried to become a "portal" (screw that, I just want search results), and was shamelessly selling top-billing in its search results to anybody with money. This was the norm for search engines at the time.
So Google stepped in and simply offered honest search results with no ads. I remember reading the Scientific American article before the site started, but I anxiously awaited it after that. But the thing that brought people to it in legions was the simple, honest results and lack of ads.
After building a reputation, they still needed revenue, so they brought in ads but they didn't give up their honesty for it. The ads are clearly marked as such, and nobody minds. It's probably too late for the other search engines to try to make up the lost business.
Anyway, they make an honest living, it's an interesting way to differentiate yourself in a market (and says something about that market).
MDC
Do you have ESP?
Mosaic had annotations. They would appear in a small frame beneath the content.
I just found something really really neat while browsing among the first mentions of google in the Google news archive circa summer '98.
There was a guy interested in getting a licence on Google technology... And this guy is from deja news... which eventually was bought by google itself... And I learned this in the dejanews archive on GoogleGroups... Ironic, uh ?