Sunday, May 23, 2010

Will There Be an App for Everything?

Sue Halpern's article on the iPad is now available at the Web site for The New York Review of Books, even if the print edition has not yet arrived in my mailbox. The fact that I should be reading this piece at all by virtue of computer mediation is saying something, regardless of whether I am doing so on an iPad, with a smart phone, on my networked laptop, or even from my printer. This virtual approach still has its disadvantages. I cannot flip back to the Table of Contents to remind myself of why I know Halpern's name as readily as I can with my physical subscription copy. On the other hand there seemed to be something appropriate in reading about the "next generation brave new world" through some connection to the current generation.

This may be the first time that The New York Review has released a product review, rather than a book review or an extended essay; but it was a relief to encounter finally a serious effort to put this particular product in perspective, free from the gee-whiz rhetoric that Steve Jobs seems to engage each time he talks about yet another feature. On the other hand, given that the iPad is primarily Apple's big gun in the war that has begun over the future of electronic books, I would have expected that Halpern, who is no stranger to cognitive science, would have devoted a bit more column space to the relation of this technology of the act of reading itself. She does not ignore this issue, but she never goes deep over the question of how well the iPad fits her own reading habits. (On the basis of what I know from my own experience, I take her to be a very serious reader.)

Ultimately, however, neither her article nor the iPad itself is primarily about reading. I did not use the word "war" lightly in the preceding paragraph. These days almost everything that happens in cyberspace is aggressive. Put another way the Internet has become one vast zero-sum game; and those of us who naively think only in terms of utility tend to overlook the extent to which major corporations like Apple, Google, and Microsoft plan their every move with the long-range goal of controlling all the marbles. Thus, even Halpern cannot resist serving as a minor foot-soldier in the latest skirmish by using language such as "game-changing." Indeed, her concluding sentence is not about the utility of the device but of the future of this ongoing war:

This is what is really revolutionary and game-changing about the iPad: once there is an app for everything, it’s Apple’s Web, not the wide world’s.

So in her conclusion Halpern leaves the world of devices and utility and moves into one of the oldest oppositions in software development, the conflicting interests of generality and specificity. Back when using a computer meant programming it, there were major factions that aspired for a single programming language general enough to address all software needs. The most popular of these was probably PL/I (as in "programming language one," a curious name for something that was supposed to be the last word in programming languages, since it presumed the future existence of a "two"); and it probably spawned as many jokes as it supported software projects. The best of the jokes involved comparing it to the ultimate Swiss Army Knife: There was a blade that could do anything you wanted; the problem was that, in trying to get at that blade, you would cut yourself on all the others!

This bit of history repeated itself with the rise of Smalltalk. Smalltalk was even more like a Swiss Army Knife than PL/I ever was. It was driven by a vast library of "methods," each of which was highly specialized and effective at its job, just like any single knife blade. Using any method was a snap, but finding the right one could be a real bummer. In many ways Smalltalk first awakened us to the idea that search itself was a major intellectual task that required effective technical support.

In this historical context the "app" is nothing more than the latest incarnation of a Smalltalk method. The good news is that we now know a lot more about indexing and search than we did when a Smalltalk programmer could burn an entire day in hunting for just the right method. The question, however, is whether or not what we know will be able to serve us should it ever come to pass that there really is "an app for everything."

Actually, there is more to that historical context than one might imagine. The Smalltalk development team recognized that finding methods would be a problem, so they built a tool to facilitate the search process. The tool was called the Smalltalk Browser, providing an interesting foretaste of the future semantics of the world of the Internet. Negotiating an inconceivably large number of methods evolved into negotiating an inconceivably large number of Web pages, so there may be some inevitability in our now moving into a world of apps too numerous to mention. In this world the device will be a side show, the only real question being whether lots of apps multitask on a single device or whether lots of specialized devices are tightly networked. On the other hand finding the app you need without cutting yourself metaphorically on lots of other apps is likely to be the real problem that identifies how the game has really changed.

No comments: