I’m not in much of a blogging mood, but I wanted to get this down now, with a timestamp, so I can say “I told you so” in 9 months: app is going to become a dirty word.
At work I’ve been fielding a number of calls from tech companies who would like us to develop a version of one of our offerings customized to their upcoming platform — whether mobile, or web-based, or (most ridiculously) centered around a particularly power-efficient x86 processor. This enthusiasm is a testament to the quality of the work my colleagues do and to the excitement currently surrounding the open gov space. Both of those are wonderful things! And to be clear, I intend to take advantage of some of these opportunities. There is publicity to be had. In some cases I think that saying “yes” is the rational, self-interested thing to do.
But only insofar as it allows us to take advantage of an essentially irrational trend. It is increasingly clear to me that, in the fall and winter, exclusive” app stores will begin proliferating at a pace that is unjustifiable, and which will likely lead, appropriately, to the concept of “apps” and “app stores” being denigrated and then stigmatized.
Companies have looked at the runaway success of Apple’s mobile application distribution model and found themselves slavering. Nevermind that the mobile space, and Apple, and their first-mover advantage are all unique. These copycats are going to try to recreate that success. Everyone can have their own walled garden! Every platform will be exclusive and revolutionary! An no, of course we’re not trying to lock anybody in to anything. Why would you even think such a thing?
It’s heinous, it’s stupid, and it’s contrary to the norms that have made the internet as great and powerful as it is.
To get philosophical for a moment, all this is an iteration of one of a handful of archetypal technology debates — in this case, open versus closed. The thing is, this is a very boring debate: we know that the answer is “almost always open”, and we’ve known it for a while. I’d much prefer to go back to fat client versus thin client (aka “the cloud will change everything forever and I have the powerpoint to prove it”). That debate is at least deservedly cyclical, driven by the ebb and flow of processing power, storage, technology’s social ubiquity and, more recently, battery life and wireless bandwidth. We’ll probably arrive at an answer there, too, but not before we get a bit closer to the fundamental physical limits constraining our technologies and/or nervous systems.
Tim Bray just posted some thoughts on this topic, and I particularly like the direction he’s goes when mentioning Amazon.
http://www.tbray.org/ongoing/When/201x/2010/07/18/How-to-run-an-App-store
What if Amazon sold downloadable apps? They already successfully sell digital assets like ebooks and MP3s. Instead of going it alone, Google could have turned to Amazon for the Android Market.
I don’t disagree with anything Bray said — I think that post is pretty insightful, in fact. But it seems to carry the implicit question: “How are we going to make this work?”
I think the answer is just: we won’t. Bray’s right that Amazon benefits from selling classes of goods for which we already have rich sets of expectations. But it’s more than that: one piece of clothing really is like another. Books cover a vast variety of human experiences, but they’re used in only a couple of ways, really.
But software? To me, this is like saying you’re going to set up a store that’s your “one stop shop” (they always say this) for baby cribs and pneumatic drills and consumer electronics. That kind of variety might work for Walmart, but with mostly fixed prices and no geographic considerations, I don’t think that example applies to online stores.
It’s like setting up a “things that take electricity” store or a “things made of plastic” store. I just think it doesn’t make a ton of sense.
And all this is to say nothing of the difficulty of getting people to pay for a zero-marginal-cost product. Apple’s made it look easy because they’ve sold us great hardware that brought a cleverly designed nudge. It’s left people confused about how easy this task is.
And, to be clear: I agree that Amazon’s produced a great shopping experience. But a big part of this is them being first to market, and doing a ton to make their online shopping experience more convenient than anyone else’s. Will any of these app stores have competitors selling the same products? Right now it doesn’t seem like it, which makes me think the analogy is not that great.
I agree with your point that curated app stores, as the sole way of running code that’s outside of your browser, are terrible, and won’t last. 9 months seems way too quick for the paradigm shift to take hold, but I’m willing to be surprised.
Let’s look at a few things that are like the iPhone App Store, but different:
1) Android Market. No curation. Not the only way to get apps on your phone, just the one with the lowest barrier. The “least harmful” of the app stores, in my opinion, and represents the short-term way forward. Does this need to die?
2) Chrome Web Store. Your existing web app can participate by supplying a JSON file. Doesn’t need to be written for the store, it’s just an alternative way to get your website noticed. I have deeply ambivalent feelings about the Chrome Web Store, especially in how it makes web apps look more like a black box (like a native phone app) and less like something you expect that you can dissect by hitting View Source. But, there’s definitely something interesting being explored here. Does this need to die?
3) The apt-get repository system. Ubuntu has been trying to improve the GUI experience for browsing through available apps. These are all available through other mechanisms, but apt-get and Ubuntu make it easier. Does this need to die?
I guess my point is that yes, in the form that Apple has popularized the concept of an “app store”, as a walled, curated app garden that Palm and Microsoft both emulated faithfully, it represents everything the Internet as a network of devices should not be. But there are other models that hit along a spectrum of middle grounds that don’t seem as easily pilloried, and that we can learn from in building an open, discoverable system. We don’t need to go back to the days of download.com, or to keep depending on Best Buy to install everything they think you need on your computer.
I think it’s smart to point out that these things can exist on a continuum. But let’s pull apart two distinct functions represented by these examples: package managers and app stores. To me, app stores primarily provide advantages in marketing and categorization. I don’t really thing apt is an app store in any meaningful sense.
The Android Market is clearly an app store, and I believe you when you say it’s beneficial. I just don’t get why we need such stores for executables or websites. We already have really well-established (if decentralized) mechanisms for organizing these products and managing their installation (particularly on Windows/OS X). I think adding app stores to these distribution ecosystems is more about organizations wanting to become gatekeepers than it is anything else.
I am writing to ask for your permission to include your posts on
DCguide.com and include a link to your blog in our directory. We would
include a link back to your blog fully crediting you for your work
along with a profile about you listed on DCguide.com . Please let us
know as soon as possible.
Mike@dcguide.com
Mike Thomas
Editor-in-Chief
DCguide.com