Straight from the horse’s mouth

By now you’re probably familiar with the 5by5 podcast network. If not, please do yourself a favor and visit their site. They have some great shows – The Critical Path by Horace Dediu is a house favorite. This week Mr. Dediu was a guest on The Web Ahead (another 5by5 show hosted by Jen Simmons) where he discussed how the forces of disruption apply to the web. It was an interesting discussion as usual, but if you’re pressed for time you should turn your attention to the bit after the sixty-five minute mark, when Horace volunteers some of his thoughts on the web vs. native application platforms.

His points are summarized in the notes below. Additional comments have been added in italic.

  • After the introduction of the iPhone, before the AppStore was born, there was this notion that people would get web apps on the iPhone. (Very true. The iPhone was launched in June 2007 and it wasn’t until mid-2008 that the iPhone SDK and the AppStore were announced. Upon launch, and for an year after that, the web was the official iPhone development platform)
  • There’s merit to the theory that the HTML5 technology will be disruptive to native apps, but nobody has solved the problem of where the money is going to come from. Usually the improvement of a technology comes from profits – the money tells you what’s the path the technology should take to improve itself. Horace calls this “the signal of pricing”. If you don’t have that signal, you’re dependent on intuition and committees (he probably means the W3C and he’s probably right, though HTML5 is more “market oriented”, having been born as a market reaction, in the shape of the WHATWG, to the slow-moving processes of the W3C ), and that doesn’t give you better information than the market does. The problem of HTML5 right now is that it doesn’t  have this “pricing signal” to tell it where to go. As a technology it probably has perfect or infinite potential, but it lacks that signalling. (This is an interesting claim, and while this correspondent is a believer in the efficiency of the markets, isn’t one of the characteristics of disruptive innovation the search for lower value markets that are ignored by an incumbent too busy with his cash cows? Isn’t Apple the incumbent and the AppStore the cash cow in this case?)
  • Native apps, on the other hand, are very well signaled by the marketplace (Signaled yes, very well signaled – well, that’s up for debate). And even if you consider the web technology superior to native technology, there are many examples of “less good” technologies with great market signals beating a better technology with less or no market signals. So HTML5 is an exciting technology with potential to disrupt native code, but we’ll need to see the revenue model.
  • Jen Simmons, the host, asks him to clarify what he means by a “market signal”. In case of apps, the market signal is the app store (you can probably assume he means application stores as a general concept encompassing the solutions from Apple, Google, Amazon and Microsoft and not the Apple App Store specifically). Apps didn’t come with the iPhone, Symbian had apps and Palm had apps. The word ecosystem as applied to an app market was coined by Palm’s marketing people.
  • The AppStore innovation (and here you can probably assume he means Apple’s, as it was the first to market with this model) is providing a central place, as prior to that you had dozens of stores all over the web for Symbian, Palm and Windows Mobile applications. When the AppStore came along and centralized it, it exploded like a supernova. The reason was improved discovery, quick publishing turnaround in a single storefront giving developers the ability to get metrics, actionable data to tweak application performance on the marketplace.
  • What is the HTML5 equivalent of this? You have the app, you have an advertising model, but that doesn’t resonate as well in this domain, so you haven’t solved that puzzle yet. The truth is we don’t know. Someone may figure it out at some point. ( There are web equivalents of some of the features mentioned above, and being centralized isn’t necessarily a boon for discovery. AOL wasn’t better than the web for content discovery, you just need the right tools – i.e. Google. Apple’s treasure trove is their database of hundreds of millions of credit cards and the AppStore billing engine with subscription and in-app purchase models. This is the killer feature of the native application distribution model that the web didn’t figure out yet).
  • On the subject of Google, he believes there are inherent conflicts. Here’s something Google should be nervous about: someone creates a killer app on Android, a content consumption app like Flipboard or Instapaper (call it Flippaper or Instaboard) and people start gravitating towards it, using the browser less and less. Android exists as a platform to enable an ad model and suddenly people are being pulled away from it, from their browsing behaviors that puts them in front of ads. How would Google react to this loss of eyeballs? Would they throttle back their support for apps? (Alternatively, you can ask what would happen if web apps start taking users away from the AppStore. Would apple throttle back their support for the web? Apple’s commitment is unquestionable. From the horse’s mouth: “We support two platforms at Apple. The first is HTML5, a fully-open, uncontrolled platform (…), and we are behind this 100%. It is fully open. The second platform we support is the AppStore.”  But would this level of support continue if the open web starts chipping away at AppStore revenues?)

Nothing that is said above is new to someone who has been following this closely, but it is overall an interesting overview of the Web vs. Native discussion. This is one of the favorite subjects around these parts, by the way, and one of the key battlegrounds on the upcoming web wars (FUD alert). Expect to hear more about it as this battle heats up.