One of many questions I have

Why does Google have different URLs for different geographies?  Why are the search results different for Google.co.in and Google.co.uk?  I always thought the net was going to be the great leveller but apparently not so.  At least this written communication subtracts the differential due to accents, but we can’t seem to get away from the curse of geography.  Please correct me if I am wrong.

Advertisements

8 Responses to “One of many questions I have”

  1. Hemant Puthli Says:

    I could be wrong, but I suspect this could be an architectural decision … to distribute the search indexes by geography, rather than have a centralized index which all searches (from anywhere in the world) might hit. Architects need to make these trade-offs when designing for highly scalable / high performance systems.

    There could also be other reasons behind this, which might have less to do with performance efficiency and more to do with politics. A decentralized architecture would facilitate country-specific content control (think China vs. Google).

    The good thing is that (in terms of your example) you can choose to search at google.co.in or google.co.uk 🙂

  2. sxray Says:

    But Hemant, why are the search results different?

  3. HyperActiveX Says:

    After reading your post I went and tried searching at google.co.uk and google.co.in for my name (vanity!) and found that the search results were more or less the same (in terms of the pages returned) .. except the order of the pages fetched by the search varied a bit. Haven’t tested this extensively, so I can’t say that this would be the case for all searches.

    My guess on why this happens at all is that, assuming the decentralized architecture, the indexes are sorted different at each “node”, and that could be because they are constructed, updated and maintained independent of one another. As opposed to a centralized program updating all the distributed indexes (or indices, if you prefer). The other content in the local node probably adds grist to the mill and increases the “scatter” for want of a better word. Decentralized databases will always grow differently, even if at some time early in their life-cycle they were mirror images of one another.

    Anyway … this is just speculation. Frankly, I have no idea 🙂

  4. sray Says:

    Man you’re a techie. You know much more about these things than I do. To me it is all very magical (any technology, sufficiently advanced, etc) and mysterious.

    I think I’ll revert to my original practice (that I had pitched upon when I was working with VANS in Bombay) of using several search engines for the same search. I used to use ALLTHEWEB or FAST search most of the time, before switching to Google under general social pressure.

  5. HyperActiveX Says:

    Which is what you could still do, with Bing and Yahoo to supplement your Google search. There was a tool called Copernic that I used to use long ago … not sure if its still around – it would fire multiple search engines (the usual suspects plus a few others) in parallel, fetch the results, collate and aggregate them and sort out duplicates etc.

    On a different note … traditional search is a bit old fashioned or else used by newbies (to a particular subject). These days, if you’re into a subject, you’d go find the sources of news / updates, pull the RSS feed from there and add it to your feed aggregator and have updates flashed on your dashboard.

    I’m really not a techie (any longer) but I play around with these things a bit.

  6. Sandeep Says:

    I think the reason is not technical. Its social. The context of the search term is affected by the social/regional context. In Google speak – the search categorization is different….

    ——————————————-
    Web Search Interest: t20
    United States, 2004 – present
    Categories: Computers & Electronics (25-50%), Sports (10-25%), Lifestyles (0-10%) , Entertainment (0-10%), Photo & Video (0-10%), Local (0-10%)

    Web Search Interest: t20
    India, 2004 – present
    Categories: Sports (50-75%), Lifestyles (0-10%), Entertainment (0-10%), Local (0-10%) , Computers & Electronics (0-10%), Society (0-10%)
    ——————————–

  7. sxray Says:

    @Sandeep

    What is the data you have provided? I’ll have to googleize the whole thing. What is “t20”?

  8. sxray Says:

    I think I get a glimmer of understanding. The list of stub items seems so full of overlapping terms that it was hard to interpret at first.

    Society means High Society? If so, how is it different from Entertainment? Just kidding.

    @Hemant: You’re right; even I have started relying on the RSS feed. I think it is perfectly feasible to define areas of information which will not be of even the remotest interest to one (cricket for me is like that) and filter that stuff out. Trouble is there is a risk that you might unintentionally filter stuff out that you didn’t imagine could be relevant. As for me, at work I am creating a sort of glossary for about 10,000 terms so web search is my life blood.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: