Is Google Getting Dumber?
Much digital ink has been spilled about Google’s new initiative to integrate its various platforms to profile users and ‘provide them with a better experience.’ Basically, what they’re doing is compiling the profile information it once kept on separate platforms, so that it can get a better idea of who you are. Here is one good article that looks at the implications .
I will leave the privacy issues in more capable hands for the moment. What I want to talk about is a more concrete difference that has emerged recently. Basically, Google, as a tool, seems to be getting dumber, at least at the tasks I use it for.
As a translator, Google has been a godsend. My art translation work requires me to track down a lot of obscure quotes, not only from art history tombs, but from every facet of the Chinese and Western history of ideas. Being based in Beijing, I can’t simply go down to the library and check out the English version of whatever book or treatise is being quoted. Google books has made that task so much easier in recent years.
That is, until now. I don’t understand the technical side of what is going on, but Google seems to have figured out that I’m in China, surfing the internet from behind a VPN. This VPN allows me to circumvent the censorship on the Chinese internet, accessing blocked sites such as Facebook and Youtube, and, in certain cases, present myself to foreign websites as a US user. This is a very important feature. If your IP address is in China, Google will route you to its Hong Kong search engine, and China’s firewall will censor certain searches on the way back in. Sometimes the service is inaccessible altogether. If you try to download Skype from a Chinese IP address, it automatically redirects you to its Chinese partner site, where you can only download a compromised version that has a built-in back door for Chinese security agencies. Some people use VPNs so that they can access internet TV from their home countries, or to be able to buy books or newspapers that aren’t licensed to foreign markets.
So here’s my problem: Google Books seems to have figured out that I’m in China, and has decided to block access to one of my most important translation tools. Though my Google searches are not (yet) rerouted to their Hong Kong server, whenever I click on a search result to a Google Book, I am brought to a screen that says I must perform this search through Google Books Hong Kong, which has a fraction of the selection and almost no previews of the texts I need to find.
This is one case in which Google’s attempts to get smarter have made it a much dumber tool for my purposes. I am currently seeking a workaround.
That seems to be a simple technical or legal glitch, and not actually an example of Google’s search algorithms actually getting dumber. The other case, I think, is a very direct example of its waxing intelligence.
Even after years of translating, I still come across words I don’t know on a regular basis. Some of these words represent new vocabulary. Some represent technical terms or names for Western schools of thought, theories, historical events or Chinese transliterations of foreign names. In any event, most of the words I have to look up aren’t conveniently listed in your average Chinese-English dictionary. For the first type of term, I usually just need to see it in more contexts, so I will run a Google search to see how it is being used in Chinese. The second type of term requires a more complex approach. I will usually type it into the search bar along with certain related key words in Chinese and English to see if I can bring up a bilingual text or an article that lists these terms in English (such as putting someone’s proper English name in parenthesis).
Recently, however, this latter technique has appeared to get a lot less useful. I don’t know if Google has tweaked its algorithms or if it’s just a coincidence, but these searches are turning up less and less useful material these days. It appears that Google’s search engine is more willing to discard certain key words as long as it can find two of them together. For instance, a translation I was working on yesterday mentioned something called the 丐派 (Gai Pai) in relation to the Notre Dame and the French Revolution. Gai Pai looks to mean something like “band of beggars,” but I’m guessing there’s a specific term. I typed in丐派, beggars and French Revolution. No dice. It turned up English articles on the French Revolution, and Chinese articles on the other Gai Pai, a band of warrior beggars in Chinese fantasy Kung Fu novels. I decided to add Notre Dame, so now I had three distinct sets of articles showing up in my search: the first two, plus some articles about the Hunchback of Notre Dame.
I guess I’m just going to have to polish up on my Boolean, or whatever they’re using these days. I’m no technical guru, but I think what’s happening is that Google’s attempts to become smarter are actually making it dumber. While more intelligent profiles may make sense from a marketing perspective, it doesn’t always make sense for us, the users. Sometimes I need the search engines to stop making assumptions about who I am in order to turn up what I need. And sometimes, I need to be able to trick that search engine, and not have it ‘figure out’ what it thinks I’m trying to do. It reminds me of our good old friend Clippy from Microsoft Office. In one of those ancient emails with a subject line “Fwd: Fwd: Re: Fwd: Fwd: Fwd: Re: Fwd: Really Funny!” I got a picture of Clippy saying something along the lines of “It looks like you’re writing a letter. Would you like me to… – fuck it up for you? – screw around with all your formatting? – convert all the text to wingdings and send it to the wrong address?”
It just goes to show that sometimes smarter isn’t better, or even smarter for that matter.