Monday, January 21, 2008

Focus? Focus? We don't need no stinking focus!

This is the most wonderful thing I've seen recently.

The most wonderful thing I've read recently is Murakami's Blind Willow, Sleeping Woman. The quality of the stories stayed high throughout. I love his voice, and it's amazing to see all the variations it takes through these stories, which span his career.

I've found myself thinking a lot about voice while reading this. "Voice" is a term that gets thrown around a lot by readers, and even more by would-be writers (myself included). What does this mean?

To me, an author's voice is that quality which makes you identify a work as being uniquely theirs. This is most obvious with writers such as Hemingway. You can pick up a Hemingway book, and even if you didn't know the author, within a few pages you would peg it as either being Hemingway or an imitator. Now, this does not mean that an author can't have a lot of variation within their works. "The Sun Also Rises" is a markedly different work from "Old Man and the Sea", in their scope, their narrators, their philosophies and aims. Still, there's an essential Hemingway-esque element to both that is hard to define but is nonetheless there.

One of the wonderful things about a short story collection is the way you can get more acquainted with an author's voice by seeing what remains constant as he changes between stories, narrators, tones, style and topics. To pick two dramatic examples: "The Rise and Fall of Sharpie Cakes" may be the one story I enjoyed the most. It is silly, pointed, arch and taut. I found myself thinking of George Saunders' best work when I read the dialog. On another extreme, the story "Crabs," while also "about" food, is austere, haunting, sinister and melancholy. What could possibly link the ironic description of a food convention with a man quavering over a toilet bowl? Their differences help underline what remains the same: the undefined mystery at the edge of the story, the author's reluctance to give pat explanations, the interest in how little people can actually know one another, a sense of individualism hopelessly adrift in a sea of uncaring chaos. You can laugh at that, as you do in "Sharpie Cakes", or it can make you sad, or even give you nightmares. (Last night I had a creepy Murakami-inspired dream that at one point included a cat burrowing into my chest.)

After finishing the book, I'm more inclined than ever to agree with Murakami's foreword, where he talks about how novels and short stories are very different things. I was expecting to summarize BWSW by saying, "It's good, but Kafka on the Shore is still my favorite," but I don't think I can even compare the two. To pick a crude analogy, it's like comparing a box of fine assorted chocolates with a pristine chocolate cake. They are similar and wonderful, but essentially different and trying to accomplish different things.

Other random thoughts:

It's only a matter of time until one or more of these stories gets turned into a Hollywood movie. The result will probably be horrible, but will lead to more people reading his stuff, which is all for the good.

I was amazed to get all the way through a Murakami book without encountering a single reference to incest! Wow!

I really like the attitude Murakami, and the Japanese in general, have towards animals. They give them a lot more personality and weight than you tend to encounter in American novels (at least, based on my experience).

Abruptly shifting gears:

I read an interesting article here recently about the California primary. Clinton has a lead here, though it is shrinking. There a couple of factors. First, Clinton's support is mainly coming from "traditional" Democrats: blue-collar, high school educated, low income voters, as well as members of ethnic minorities. (California is a minority majority state, so that's a very powerful bloc.) Obama supporters, on the other hand, tend to be college-educated, upper-income, and predominantly white. Anyways, this article distilled the issue down to, "If Obama is going to win the state, he needs to dominate the Bay Area." It sounds like this region is the strongest in the state for Obama, and mobilizing voters here will be key to overcoming Hillary's strength downstate. I love the fact that my prejudices against Southern California continue to be reinforced despite me spending almost no time there. On the other hand, though, California has an interesting primary system, where independent voters can vote in the Democratic primary but not in the Republican. Since independents (and some conservatives) have been strongly backing Obama, it's not impossible to imagine that we could end up with an Iowa-style upset here.

There was an interesting piece on Google in the New Yorker this week. I tend to really enjoy Ken Auletta's reporting, but he seemed a bit... off this time. I'm generally frustrated when I read media coverage about technology, so some of that may be in play, but I also was just not sure what the point of the piece was. It felt like a forum where a lot of people aired criticisms about Google, and then Google replied to them, all of which was dutifully recorded by Mr. Auletta. This kind of "journalism" is common in political coverage, but I guess I've come to expect more from the New Yorker.

That's neither here nor there... maybe he had a bad week, or maybe I was in a bad mood. I do, though, want to talk a bit about Google, cookies, and privacy. There is a lot of chatter out there, and little solid information about what's going on. Hopefully this will clear the waters a bit.

First: What is a "cookie"?
You can see for yourself! If you're using Firefox (which you should be), take a quick detour from this post. Click "Tools" in the top menu, then "Options..." Look for the "Privacy" tab and click it, then click "Show Cookies." You should see a bunch of site names pop up. You can search for specific cookies you're interested in - try "google". I'll wait here for you.

As you can see, each cookie has several types of information. First, there is the site (or "host") that the cookie came from, such as,,, etc.
Next, there is the name of the cookie, which is usually something short and cryptic.
Third, there is the "content" of the cookie. This is usually a long series of letters and numbers, and is the meat of the cookie.
Finally, there is an expiration date.

So, let's take a step back: what are cookies, again?
What you're viewing here are small text files stored on your computer. When you connect to a site like google, it can pass back a "cookie" that gets saved. The next time you connect to google, your browser will pass that cookie back to google.

This is important and useful for several reasons. Most significantly, it lets the site "remember" who you are. When you sign in to Amazon, it uses a cookie to remember you, and because it knows who you are, it can recommend products, show you what is in your shopping cart, etc.

What Google does is a little more complex. First, if you use any Google services, like gmail or personalized search, it acts like Amazon: it looks up who you are, and logs you into the right account. Even if you don't have a google account, though, the cookie is still very useful for Google, because even if it doesn't know who you are, it will know how you search, and can use that information to improve its products.

Let me run through an example. (I should emphasize here that I am not a Google employee and don't have access to their algorithms, so the details may be a bit off, but I'm confident in the overall design.) Suppose that I am searching for information on how to fix a flat bike tire. I search for "flat tire", don't like any of the results it gives, so I hit the "Next" button to go to the second page of results. I continue this several times before finding something on the fifth page that I click on. I read it for a bit, then come back to Google, and try another search. I type in "bkie flat tire". As you can imagine, I only get a handful of results back. Realizing my mistake, I correct this and do another search for "bike flat tire." This time I get many more (and better) results, click on the first three, and then stop once I find what I'm looking for.

If I had disabled cookies on my browser (which you can do), Google would have only seen a random sequence of searches, with no way of knowing that they came from the same person. Because of cookies, Google can look at the way a user is interacting with the site, and use that information to improve.

In this specific example, Google would realize that, at least in this particular case, the results for "flat tire" didn't seem to match what the user was looking for. If many users did the same sort of thing (not click on anything until the fifth page), Google could automatically re-sort the results, pushing the helpful ones higher up so people more quickly find the useful information.

Also, by seeing that "bkie flat tire" was immediately followed by "bike flat tire", Google can realize that the first query was probably a typo, and the second one was correct. If a lot of users do the same thing, then Google can start prompting people automatically when they type "bkie flat tire," asking them if they meant to type "bike flat tire".

There are a few points worth summarizing about Google and cookies:
1. Cookies are only sent to the site that originated them. That means that, for example, Yahoo will never see your Google cookies. When talking about "privacy," therefore, the only question that matters is how much you trust Google with your data.
2. Cookies are stored on your computer, but don't contain personally identifiable information. Only Google can interpret the information stored in the cookies. Even if someone stole your computer and looked at your cookies, they would only see gibberish.
3. Cookies have a legitimate role to play in improving a user's browser experience by keeping them logged into an account.
4. Even absent an account, cookies have a legitimate role to play by allowing companies such as Google to see how people are using their sites. Without this data, companies are forced to guess at what people are doing rather than actually seeing it.

As I see it, therefore, everything boils down to a simple question: do you trust Google? What people are concerned about is that the semi-anonymous information, such as people's searches, will be linked together with personal information, such as a credit card if you've created a Google Checkout account. If Google tore down the barriers between different kinds of information that they collect, AND made all that information available to advertisers, then theoretically you could start getting phone calls about products you'd searched for, text ads about your credit report, and similarly invasive, too-personal openings. (Current Google ads are semi-personal, only in the respect that they are displayed in connection with what you are doing RIGHT NOW, as opposed to what you've done before.)

I guess that individuals need to make a decision about how they feel about Google and offset the immediate convenience of a better browsing experience against the risk of a future invasion of privacy. Personally, I think that Google is well aware of the responsibility it has; as Brin says at one point in the article, they have to respect privacy, because if they don't, they'll lose everyone's trust and business. Google has become a phenomenally successful company while carefully maintaining individual privacy, and I just don't see the incentive for them to turn away from this in the future.

No comments:

Post a Comment