Udi Manber, Google's
VP of Engineering, gave a brief 15 minute presentation at Supernova today entitled
Search is a Hard Problem. He explained that with an audience like Supernova, he
imagines we understand to some extent how difficult a problem it is, but it's probably a
harder problem then we even appreciate. He laid out three reasons why this is the
case:
I found the third point quite amazing. I would think with the number of queries that Google processes, they would have seen a much higher percentage of the queries before.
Next Udi explained that there are three levels involved in trying to deliver relevant information back to users:
Udi gave some examples of Google's ability to understand the different between two very similar queries. For example, Google understands that 'GM' stands for 'General Motors', while 'GM foods' is actually 'genetically modified.' If you search for 'B&B AB', Google knows that is 'bed and breakfast in Alberta', while 'Ramstein AB' is 'Ramstein Airbase'.
Google also will recommend queries that may deliver better results. For example, if you query 'Types of dogs' it will give results, but also suggests 'breeds of dogs' as a better search.
He then explained that they still can't find all the answers. As a fun example, he said the query "Why Search is Hard" is actually a very difficult query for Google to parse.
Next Udi reviewed some new Google search functionality, which while not live yet - will be soon. Apparently, Google is going to start trying additional queries based on certain user queries. For example, the query "How much does it cost for an exhaust system" will pull up results from "cost of an exhaust system." Beyond just removing certain general words, they are also interpreting the question as part of the model; for example the following two queries:
Finally, Udi talked about how results need to be different when the query is conducted in different locations. For example, the query 'government' needs to return results about your countries' government. I haven't tried this in other countries, but here in San Francisco the first result is for the US Government.
He also reviewed a tool at Google.com.eg, which actually takes a query in another language, translates it to english, runs the query, and then returns the results in that language. You can actually view the page in that language. There are a whole suite of language tools Google seems to be leveraging at: http://www.google.com/language_tools
I'm surprised there aren't more copyright issues here, but I'm not a lawyer. For an example, here is the Read/WriteWeb homepage translated into Arabic. Apparently, when Udi was demoing this for Larry Page, he asked why the images weren't translating. Obviously, there is still work to be done, but it is quite amazing.

After listening to Udi's talk, I must agree that while I thought search was complex, I probably underestimated some of the areas of real difficulty. It is amazing to step back and think about how conceptually complex this is. It sheds new light on many of our challenging experiences around the AltSearchEngine's Day without Google.
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/2296
Comments
Subscribe to comments for this post OR Subscribe to comments for all ReadWriteWeb posts
The 20-25% new queries stat, do you know if that is a percentage of total queries (that seems high, as you said) or as a percentage of the number of unique queries (i.e., 'football' searched 40,000 times counts as 1 query, the same as 'skippy peanut butter recipes with pictures' searched once is counted as a single query). The latter would be less shocking. :)
I'm based in Australia - Google's first result for me typing 'government' is www.australia.gov.au, the Australian government portal
Josh,
Good question. I'm not sure but I would imagine it is unique queries which makes much more sense. This also helps me reconcile some of the things Jason C has said about Mahalo regarding the 'head' of search.
- Sean
Sean, This looks like a repeat of Udi's presentation during Google's universal search PR event a month or so back.
The Arabic translation looks neat. Were you able to find someone who understand Arabic to confirm its accuracy?
Also, I can't imagine Google being able to translate images in a very long time (GIF and JPG images are pixels, and have no text values). Think about how long has it been brought up that FLASH SWF files was going to be indexed; and search results today of SWF files are close to nil.
The gm cars vs. gm food is nothing more than a synonym look-up. See the gm vs. genetically modified search examples in this post http://stack-up.com/stack/2007/05/22/searchology-no-magic-behind-googles-magic/
Re the automatic translation, you can also take an Arabic web page, e.g. from BBC's Arabic News Service and translate it into English. Makes it easier to assess than in the other direction ;-)
GM food
in GM food was served in paper plates..
produced by GM food containers leaked..
at GM food is the favorite pass-time..
statistics/popularity does not always work! This is not understanding! From 1000s of employees, don't they have one linguist to correct these statements?
Good Article :)
It is an interesting attempt to deliver better search results, however when Google tries to "guess" what you are searching for it will not deliver what you are actually searching for.