stevenberlinjohnson.com

Apple's Ten-Year-Old Breakthrough

Taking their cues from Steve Jobs' presentation yesterday, a lot of commentators have described the new Spotlight search tools included in the upcoming OS X Tiger release as a broader application of the iTunes search functionality -- particularly the idea of Smart Folders that automatically update according to pre-defined criteria, the way Smart Playlists work with music.

But the Smart Folders idea goes back much deeper in Apple history -- in fact, it pre-dates Jobs' Second Coming. Smart Folders used to be called "Views" and they were a pre-announced feature of Copland, the ill-fated OS that was eventually replaced by the NeXT OS. (It drew on the V-Twin text indexing that did make it into OS 9, I believe, and may well still be a part of OS X.) I saw a demo of Views when I visited Apple in 1996, and it made such a strong impression that I wrote about it in my first book, Interface Culture:

Apple's V-Twin implementation lets you define the results of a search as a permanent element of the Mac desktop -- as durable and accessible as your disk icons and the sub-folders stacked beneath them. In Apple's language, these new items are called "Views," for reasons we will come to understand. At first glance, a View looks and behaves like your average folder or sub-directory: it is represented by an icon; clicking on that icon opens a window that contains other icons, representing assorted files; clicking on one of those icons opens the appropriate document. So far so good. Things get tricky, however, when we try to add a file to a View manually, by dragging an icon over the View's window... The user has only indirect control over the contents of a View. He or she specifies its general attributes, using the language of the V-Twin search engine: "find all documents on my hard drive that are likethis other document." The computer then decides which documents fulfill that request, and places them in the View window. (Technically speaking, it places copies -- or "aliases" -- of each document in the View; the originals remain in their previous locations.) Unlike the temporary results of a "find file" request, the View window has what programmers call "persistence." Like an ordinary folder, the View remains on your desktop until you throw it away. During that lifespan, the V-Twin software regularly updates the View's contents whenever new files arrive that match the original search request.

I think it's pretty ironic that the most highly-touted feature in Tiger is one they've been trying to get into a shipping OS for almost ten years. Sometimes information society isn't quite as fast as it's rumored to be.

One other thought on the keynote: I was quite disappointed by the new displays. I'm completely in the market for an insane new flatscreen extravaganza, having enjoyed my original Cinema Display for about four years now. But sadly, none of the options really do it for me. The 20 and 23-inch are the same old displays in a new case, for pretty much the same old price. The 30-inch is drool-inducing, to be sure, but with the required video card it's $4K! Trying to persuade the kitchen cabinet around here that I need to spend four grand on a few more pixels would be a negotiation up there with the Reykjavik summit -- and it would no doubt end with the same amount of success.

A Little Less Construction

Okay, things are stabilizing a little now. Still some aesthetic tweaks to come, but the main infrastructure is in place. You'll see that I'm now requiring comment registration through TypeKey. I'm sorry to have to do that, but the comment spam was just killing me. (Literally half of the email in my inbox was comment spam being automatically emailed to me by the blog -- and it had taken a very dark turn in the past month or so, from the old innocuous "hey, great site" messages to pretty much non-stop bestiality links.) If you notice anything obviously broken, drop me a line, or post here. Thanks!

Under Construction

I'm in the middle of upgrading to MovableType 3.0, and so things may be a little funky around here for a day or two. Those of you longing for that picture of me on the beach -- or the woefully out of date sidebar information -- will have to entertain yourselves some other way for a while.

The Rest Is Noizzz!

I've known Alex Ross for more than two decades, dating back to the high school we both went to in Washington, D.C. (One of my first memories of him is not being quite nerd-cool enough to get into a regular game of Diplomacy that he and a few friends would put together on weekends -- oh, the crushing blows of being fifteen!) Ironically, it took us going off to different colleges for our friendship to bloom, and since then we've been having ongoing debates about the cultural merits of Madonna, CSPAN's political coverage, Radiohead and Pavement and The Smiths, not to mention all the strange folks from high school years that we occasionally stumble across. He wrote a wonderful extended essay on New Zealand indie rock in the very first "issue" of FEED, but is probably better known for being the New Yorker's music critic -- or I should say, "a" New Yorker music critic, now that Sasha Frere-Jones has jumped into the fray.

At any rate, I say all this to alert you to the fact that Alex now has a blog, The Rest Is Noise. I love Alex's NYer writing, but as so often the case with blogs, I think his site captures a bit more of his whole personality -- the intellect and wonderful prose, sure, but also the eclecticism of his tastes, and his humor. Check it out.

When Is A Golf Course Unfair?

Greetings, blog readers. You may have wondered where I disappeared to. The truth is I've been on vacation for the past week on Martha's Vineyard, and before that it was a crazy rush of birthdays and pre-vacation packing. (It was my 36th on June 6, followed almost immediately by my wife's, and our now one-year-old son's.) But we're back in Brooklyn for a long stretch, and posting should now continue at its usual erratic pace.

A quick word about the controversy over the U.S. Open this past weekend. For those of you who follow this sort of thing, there's been a flareup over the difficult conditions created by the USGA at the brilliant Shinnecock Hills course where the Open was played -- mostly concerning the greens, which were by Sunday as slick as a ballroom floor apparently. As the Times reports today, the scoring average on Sunday was a full eight strokes over par, significantly higher than any major championship in recent memory -- though the overall average for the week was still lower than at Bethpage two years ago. This quote from Tiger Woods was representative of the players' complaints: "There's nothing wrong with it being hard and difficult. But just don't make it so it's out of control unfair."

I bring all this up because I think it raises an interesting question about what it means for a course to be "unfair." Assuming everyone is playing under the same conditions, shouldn't any level of difficulty be fair? It might be a little tedious to watch players hacking away on every hole, but there wouldn't be anything inequitable about the conditions. (Granted, there was some dispute about the watering of the 7th green, which apparently benefited some players and not others, but that's the only hole whether there was any question of giving some players breaks and not others.)

In my mind, the debate shouldn't be about unfairness. It should be about randomness. There is a certain point of difficulty where the course stops distinguishing between good shots and bad shots, and thus fails to reward players who are performing well relative to their peers. Imagine a course set up like a pinball machine, where balls land in the fairway and then bounce fifty feet in the air in random directions. Winning in those conditions would be a matter of pure luck -- the score would simply be a register of who got the luckiest bounces.

I think it's this randomizing effect that the players are intuitively talking about when they talk about the layout being unfair. The problem is, the evidence doesn't really suggest that the layout is any more randomizing than your average tournament. In fact, I'd suggest just the opposite. The easiest measure of this is whether the best players on average end up doing well in the tournament; tournaments where high-ranking players tend to cluster at the top are less randomizing than tournaments where you have low-ranking players disproportionately represented at the top of the leaderboard. In fact, you could easily create a statistic that would measure this: the extent to which the final leaderboard deviates from the world rankings. Call that number the R factor: layouts that possess high R factors -- where a lot of low-ranking players out-perform the top pros -- don't let quality play rise to the top, while low R factor layouts separate the wheat from the chaff. You could use this number to determine the ideal conditions for rewarding quality play; you'd analyze all the tournaments and whichever venue turns the lowest R factor -- that's your model for the ideal tournament.

From eyeballing the scores from the past five opens, it's hard to believe that the Open -- despite its allegedly "unfair" layout -- doesn't have an unusually low R factor, given the quality of the players in the top ten each year. The reason Woods, Mickleson, Singh, Els, Goosen, et al show up again and again in the top fives for the Open is because the layouts reward solid play and experience, despite their difficulty. There's nothing unfair about that.

TiVo's Weight Problem

Seeing this rumor item on the next generation iPod reminded me of a question that's been nagging at me for the past month or so: why is TiVo so big? I'm in the market to buy one of those Series Two TiVos for our main home theater setup, but I simply don't have the room for another box right now, given the amp, the two DVD players, the XBox, and the cable tuner already stacked on top of each other. I do, however, have room for another iPod. If Apple can make media-playing device with a 40 gig hard drive that's smaller than a pack of cards, why is a 40 gig Tivo still the size of two shoeboxes? Seriously, I'd like an answer. I know you need room for a bunch of different i/o options in the back (phone, USB, S-Video, RGB, audio, ethernet), but couldn't you do something clever to minimize that real estate -- perhaps create some kind of single universal plug and then let people buy the adapters they need to connect to it? My current Series One TiVo is bigger than the VCR I bought in 1990. What's up with that?

Bloggers Save The World!

My buddy Jeff Jarvis alerted me to the laudable Spirit of America site, which has already been widely linked to through the blogworld, but every link counts, so here's mine. It's a fascinating model for combining long-distance philanthropy with targeted interventions. I won't bother going through the details since they're nicely summarized here. But the site makes me wonder whether this isn't the beginning of a fascinating new chapter in the web's gift economy. Thanks to the passion of the bloggers themselves, and clustering technologies like Technorati and Blogdex, we've already mastered the art of locating and quickly swarming around the week's hot news item or thinkpiece. (You know the drill: Clay posts a provocative essay about power laws on Monday, and by Friday there are fifty in-depth responses, a dozen fact checks, ten suggestions for future research, and a handful of requests for the Lazy Web.) What Spirit Of America suggests is a version of that swarming directed towards Good Causes: someone halfway across the globe (or halfway across the country, or the county) puts out a call for help setting up a wi-fi network in an under-funded school, or repairing a sewage treatment facility, and within five days they're flooded with funds, spare parts, technical expertise, and good will. And when the network goes online, or the sewage starts getting processed again, we all get to see the results. (Maybe not so fun for sewage, but you get the idea.) And then we get to move on to the next cause.