Joho the BlogIs the iPhone generative? - Joho the Blog

Is the iPhone generative?

Steven Johnson makes one of his typically brilliantly insightful points in his recent NY Times op-ed: The iPhone is a locked-down device, but it has been the site of arguably the greatest burst of software generativity in the computing era, much of it by small developers. This has led Steve to re-evaluate his adherence to the “unifying creed” that “Open platforms promote innovation and diversity more effectively than proprietary ones.” When Dan Gillmor challenged this in a tweet, Steve responded with a terrific blog post, further considering the point.

The argument is over the issue framed by Jonathan Zittrain in The Future of the Internet and How to Stop It. JZ defines “generativity” as “a system’s capacity to produce unanticipated change through unfiltered contributions from broad and varied audiences.” (p. 70) Steve suggests that we instead judge generativity by the type of results we see, not by the nature of the software or hardware environment on which they run; a generative platform is “a platform that is constantly being re-invented in surprising new ways by a diverse group of creators, where individuals, hobbyists, small startups, and amateurs compete on a level playing field with large incumbents.” So, while JZ assumes that a system’s capacity to produce generative results depends on the system’s openness, the burgeoning of software for the iPhone shows that closed systems can produce wildly generative results.

I think a few things are going on here.

First, Steve is right about the fecundity of the iPhone as a platform, and about its openness to amateur and small developers. But, he’s right because the iPhone is not purely locked down. Apple could exert any control it wants, to any degree, any time it wants, but so far it’s been pretty open. So, the iPhone and the iPad are generative because in practice they generally meet JZ’s criteria. They are, in JZ’s taxonomy, hybrid animals.

But, the fact that there are over 150,000 apps for the iPhone is not the only measure of generativity. Apple has announced it will exclude unruly guests from its party (and Apple gets to define “unruly”), so the unruly don’t even bother to ask for admittance. The AppStore is a ruly environment. Now, there are obviously advantages to the user (as well as to Apple, but we’ll leave that aside for now) in having a device that cannot be disrupted. (“Disruptive” figures large in JZ’s book, but not in Steve’s definition.) For one thing, a ruly device is less likely to melt into a puddle of palm-sized uselessness. But, that’s to say that the iPhone’s limits on generativity are desirable. Steve’s argument is different. He’s saying that the iPhone is generative.

In any case, I think Steve is wrong in his causality. The iPhone has generated 150,000 apps because it’s a cool piece of hardware with a preternaturally appealing UI, useful software affordances built in, and an appealing SDK. Not to mention, it’s got a gazillion users. And the App Store is well-designed for marketing small programs. The iPhone is not wildly generative (in Steve’s sense) because it’s a walled garden; the iPhone could allow other marketplaces for apps to exist without losing its generativity (as Steve notes in an aside).

But, the most important issue is not whether the iPhone is generative. The question is whether Steve is right to renounce the “unifying creed” that generativity depends on open platforms. The argument should not be over whether a particular hybrid device is generative — although it’s helpful to have the case raised — but over the future of the Internet. That’s why JZ raises the issue of generativity in the first place.

JZ defines generativity as part of a polarity. Here’s what he says at the beginning of his book:

…the pieces are in place for a wholesale shift away from the original chaotic design that has given rise to the modern information revolution. This counterrevolution would push mainstream users away from a generative Internet that fosters innovation and disruption, to an appliancized network that incorporates some of the most powerful features of today’s Internet while greatly limiting its innovative capacity—and, for better or worse, heightening its regulability. A seductive and more powerful generation of proprietary networks and information appliances is waiting for round two. If the problems associated with the Internet and PC are not addressed, a set of blunt solutions will likely be applied to solve the problems at the expense of much of what we love about today’s information ecosystem. (p. 8)

The danger is that as cellphones become mobile Internet devices, and as iPods become mobile computing platforms, our new generation of computing devices will be appliances open only at the forbearance of their creators. Those creators may be relatively benevolent, but the question isn’t whether this device or that creator is open. It’s what the future of the Internet and of computers will look like. If appliances become the dominant way of interacting with the Net (and thus how we interact with one another), then no matter how loosely the device creators hold the reins, we are accepting the bit in our mouths. If appliances become the default, then the market for challenging, risky, disruptive, subversive app development is in danger of drying up.

From that point of view, the generativity of the iPhone and the iPad is — to use JZ’s word — seductive. Steve Johnson is right that they have unleashed a torrent of creativity. But it is creativity within bounds. The very success of these devices, driven by the generativity that Steve Jobs allows us and to which Steve Johnson astutely points us, can lead us to the future that JZ fears.

17 Responses to “Is the iPhone generative?”

  1. I think part of the issue has to do with properly framing the term, “generativity.” Simplistically, Steve Johnson is correct: the iPhone, together with its captured market and marketplace, generates lots of creativity based on the age-old economic constructs of potential fame and money.

    However, if we understand “generativity” as the creative potential that could, under the right circumstances, obsolesce the environment that spawned the supposedly generative force in the first place, iPhone is definitely not, whereas the (more open) Internet is. In the latter sense, very few innovations have been truly generative throughout history.

  2. For me, a lot of what seems on the surface to be confusing falls into place when you consider that many of us look at the new devices as smaller versions of general-purpose computers, whereas Apple treats them like larger versions of proprietary embedded personal electronic devices.

  3. This actually makes me think of chiptune/8bit music. It never would have existed without the closed ecosystem of limited instruments (technical impositions, but impositions none the less) and continues to thrive even w/o those circumstances still being imposed.

    I think it can be tough to define a healthy ecosystem, and to make a unifying theory even tougher. I also think that markets will bear certain things, and if we really were in a state w/o any 99% user modifiable computers, that part of the market would recognize said need and create it.

  4. Dave, great post. Thanks for all the nice things you say, and for your typically nuanced parsing of this. Clearly we need someone with your philosophical background to make sense of a concept like generativity.

    I would say two things. First, while I agree that there are important “open” elements to the App Store, the platform is closed in ways that go beyond just the approval process, and those closed elements are crucial to why it’s such a delightful platform to create on: the tight hardware/software integration, the unified payment mechanism, etc.

    The more important point, which I didn’t have room to address in the Times piece, is that I don’t feel that Apple’s platform is necessarily in conflict with the open values of the internet, in the way that Jonathan, for one, seems to feel. If Apple were out there trying to route people away from the Web, or force its proprietary standards onto the Web, I would be totally opposed to it. But I think Apple has shown an unflagging commitment to open Web standards, and indeed the most “open” part of their entire company is probably WebKit. The ads for the iPad (and the iPhone before it) are just as likely to show a Web page as they are an app. One of the main taglines for the iPad has been “it’s like holding the Internet in your hands.”

    It seems pretty clear to me that closed architectures facilitate certain kinds of creativity, and open ones facilitate other kinds. Have you seen Star Walk for the iPad? Point your iPad to the cosmos and you get an augmented reality annotation of all the constellations and planets. It’s insane and magical and I’m sure it never occurred to Steve Jobs or anyone else at Apple when they were building the original iPhone platform. Good luck building that in a web app, even in HTML5. But at the same time, the Web is vastly better at building something like Wikipedia, or Facebook.

    What Apple is trying to do, as I see it (and some of this I’ve learned from reading John Gruber), is to try create a space where both kinds of innovation can flourish: the controlled, design-rich innovation of an Apple platform, and the creative anarchy of the Web. Why isn’t that an admirable goal?

  5. Brent, I think that’s right. It’s all about expectations. I don’t expect my Kindle to be open, because it’s a dedicated appliance. I do want the iPad to be more open because I think of it as a computing device. But, beyond expectations, there’s the question of what we (most of us, perhaps all of us) will be using to do what we currently use computers for. Those devices are likely to come from the small, portable, cheap-ish side of the spectrum where our expectations are for dedicated-use appliances. That’s why it’s important to be having this discussion now: we don’t want to carry the wrong expectations (assumptions, defaults, DNA) over into the future devices that replace the older generation of computers. (Not that you disagree.)

  6. Steve, thanks for the as-ever generous and kind reply. Yes, it’d be a much easier discussion if Apple were the epitome of evil and closedness. But it’s a mixed bag (as we used to say). And it always has been: Apple has always exerted tigeter control over its computers than Microsoft did, which is a huge part of why Macs are elegant to use and Windows has been like dragging a bag of pitchforks. On the other hand, it’s an important reason why Windows machines have generatived more cool stuff than the Mac. Obviously market share also has a lot to do with it, but so does the fact that Microsoft didn´t try to own the hardware, which meant we got driver Hell but more competition and innovation on Windows, and more beauty and grace on the Mac. (For the record, I used Windows for 20 yrs, switched to the Mac 3 yrs ago, and am very happy with the change.)

    The key question you raise in your comment (and you do have this habit of raising key questions) is whether closed and open architectures facilitate different kinds of of creativity. I’m not sure what I think about that, but my hypothesis is (apparently) that closed architectures enable elegance within a more limited range, and open architectures enable disruptive innovation. I don’t count StarWalk as disruptive, although it certainly sounds cool. (It also sounds like a Google app I have on my open Droid.) Isn’t it the case that Apple’s control over what can be loaded onto the iPhone has resulted in 150,000 safe and commoditized apps, some of which are cool, but none of which are disturbing? (Well, except for a couple of cases that Apple then pulled.)

    It may turn out, though, that the Cloud saves generativity: If we ultimately are able to do just about anything we want via the cloud and in a browser, then client app stores will be of decreasing importance.

    The Net may yet save the Net’s generativity.

    (PS: Sorry for whatever vagueness and incoherence has crept in. I’m jetlagged and using a computer in the hotel’s business center for my 30 allotted minutes.)

  7. Like Amazon or eBay or Threadless, Apple has created a closed platform that allows buyers and sellers of a particular class of products, in this case software, to discover each other and transact efficiently. Does anyone fret about whether Amazon, eBay or Threadless are open or not?

    They’re each (basically) closed and very good at what they do. But they haven’t gone on to take over the world. Likewise, the scope of Apple’s franchise, building great Walkman-cum-star-finders, will prove more limited than Jobs acolytes would like to think. For example, will Jobs thrive as an iAd peddler? He sure hopes so. App-sellers, like website operators in the nineties, will increasingly abandon upfront fees to focus on growing their user bases and earning ad revenues.

    Looking out five years, my guess is Apple’s closed iAd model won’t port well into the existing ad industry ecosystem. More thoughts on iAd’s future here:

    Yes, Apple’s on a roll, but it won’t roll far from the tree it dropped from.

  8. […] 4: David Weinberger sums up a debate over whether the iPhone and Apple’s overall ecosystem is “generative” — in […]

  9. Interesting talk… FWIW I think that confusion over the locus of “generativity” may stem precisely from, as DW puts it, the fairly polarized way in which it was originally defined — or at least as it has been widely interpreted/applied in the years since it was coined. I believe that the phenomenon that JZ captured with this snappy neologism is in fact very old and commonplace, although before JZ it was never named or referred to directly. If I’m right, Zittrain is using “generativity” is refer to that universal (albeit highly variable) feature of economic systems that causes them to produce emergent properties like “the division of labor” — or to use the more modern term “specialization,” the most ambitious/surprising and welcome instances of which are distinguished with their own special term of praise, “innovation.”

    In Adam Smith’s early exploration of this phenomenon, he associated its origins with the progressive elimination of natural and artificial impediments to the (frequency and widest possible scope of) voluntary, self-directed exchanges of value between independent decision making entities. Once the paleolithic requirements for such interactions are sorted out (e.g., existence of fixed population groupings larger than family, possession of means to move about and produce basic subsistence requirements), and barter transactions become commonplace, the most important development that contributed to basic generativity was the emergence/discovery (in this case probably best characterized as post-facto recognition) of a workable “liquidity mechanism” — i.e., some naturally occurring or manmade artifact that most people are generally willing to accept in trade for more tangible goods, because they are confident that they too will be able to use that liquidity token in exactly the same way, to obtain what they want at some other time/place thereafter.

    In the conventional economy, the divisible elements of this liquidity (or “medium of exchange”) function are generally called “money” — and in order to sustain liquidity amongst the scarce, rival, physical goods and real-time-consuming services that overwhelmingly dominate the conventional economy, the monetary class of liquidity mechanisms has to be similarly finite, rival, and time-bound. These conditions are achieved through the mechanism of circulation, which in turn is facilitated by the other functions that money quite predictably/necessarily came to embody (e.g., circulation = “means of payment,” which is made possible by money’s other onboard functions, c.f., “store of value” and “unit of account”). By contrast, in the world of network-based, packet-mediated, and thus inherently nonrival exchanges, the same basic liquidity function is much simpler to sustain; in fact it requires nothing more than an basic “attachment” mechanism, i.e., something that creates a durable/inter-temporal, non-directional, and transactionally-nonspecific association between each potential exchange participant and exchangeable factor and all of the others. On the Internet this function is embodied in individual IP addresses.

    The most important/relevant implication of this insight is that while the Internet is (at present) the largest and most all-encompassing nonrivalrous domain that might be characterized as “generative,” any subset of the Internet — or indeed, any completely separate packet-switched network (or any other system composed of independent, economically motivated decision makers) also possesses some nonzero (but probably less) potential to produce greater specialization and occasional innovations over time, i.e., to “be generative.” In effect, comparing the iPhone system to the Internet is like asking whether a specific firm in a specific economy is more or less innovative than the encompassing macro-economy, or maybe the economy next door. Even assuming that a coherent answer is possible, it’s not entirely clear what practical implications any answer might have.

    Of course, in this particular comparison, both entities are beneficiaries of the same basic structural endowments (e.g., packet-switching > nonrivalrous production, many distributed decision makers), but the Internet in general lacks an inline payment mechanism, as well as an externally imposed set of value relations that invariably/unavoidably dictate who shall pay whom for what. By contrast, the iPhone system provides that convenient payment facility, but also shoehorns all participants into a relatively inflexible, externally imposed commercial ontology. The marginal product that distinguishes these two systems from each other is the long, long tail of productive activity and exchange that the Internet fuels and sustains based solely on non-monetary inputs (i.e., individual time and attention), most of which would not exist within any current-iPhone-like system. Perhaps a future iPhone system *could* support those kind of activities one day, but probably only by selectively relaxing some elements of that imposed commercial ontology and payment system… i.e., by becoming (or maybe internalizing) its own Internet-like sub-domain.

  10. To further clarify the relevance of the above observations about liquidity mechanisms and their role in fostering generativity, the whole point of Adam Smith’s argument against mercantilism was that by (further) skewing the exchange landscape to favor of (more tractable, better understood, seemingly more locally net-beneficial) intra-national value exchanges over competing inter-national alternatives, national sovereigns were effectively creating/exacerbating national-scale “walled gardens” that hurt themselves as much if not more than they hurt their sovereign counterparts. Although the original argument was largely geared to convey the importance of that insight to national-level decision makers, the same dynamic is equally relevant in cases where distinct liquidity mechanisms coexist any other level of analysis — subnational (e.g., private banknotes vs. national-standard currencies, bank loans and credit cards vs. merchant-specific credit accounts) or super-national/extraterritorial (e.g., the TCP/IP based Internet and its 35k+ independent but transparently interoperable routing domains vs. the more private, more isolated variants like the iPhone network).

  11. […] Weinberger, one of the first theorists of the web, recently took up the issue (his use of “appliances” comes from another original theorist, Jonathan […]

  12. […] Weinberger, one of the first theorists of the web, recently took up the issue (his use of “appliances” comes from another original theorist, Jonathan […]

  13. […] for us.  Cory Doctorow of BoingBoing voiced his concerns about the locked nature of the iPad and so did David Weinberg.  Harry McCracken, posted a really thorough comparison of iPad to the various other […]

  14. […] David Weinberger Previous in series This entry was posted in All. Bookmark the permalink. ← Generative […]

  15. […] mountain of failures. Steven Johnson describes the “Star Walk” astronomy app as “magical” (which is completely is, by the […]

  16. […] (e.g., iTunes vs Google Play, Amazon App Store, SideME) — Shouldn’t there be something already written (researched) on how app stores are generative? (cv: there is a dissertation to be written on the […]

  17. […] […]

Web Joho only

Comments (RSS).  RSS icon