How Gruber is Subtly Wrong About Cloud+Software

John Gruber is focusing on the wrong piece of the cloud+web vs cloud+software. It’s the ‘+’ that matters.

The ‘+’ that matters is that Google’s ‘+’ is web/http and Apple’s ‘+’ is objective c talking over some protocol that is inaccessible outside of Apple’s world. Yes, the user interface matters (a lot), but that is not the key differentiator. Over time, native and web will unify from a user experience perspective (of course standards will always lag innovation, but that is another story).

Traditionally, the term cloud has been associated with the web. Most large cloud implementations are all about the web and web protocols. Amazon (both books and AWS), Google, Flickr, etc all implemented their cloud services using open accessible web standards. They embraced and leveraged the spirit of the web by utilizing well known http protocols along with simple textual APIs. Call it the semantic web, web services, or web 2.0. In the end, it is about the ability of developers (with a wide skill gap) to mash together data and functional services to create completely new and interesting software. Think of it as the world wide middleware. The linkage between web 2.0 and these cloud based services is ingrained in anyone’s thoughts about what cloud is. At this point, cloud and web and simply inseparable cognitively.

I don’t think that it is a reach to suggest that the availability and simplicity of those public web services have been the primary reason for the success of many iphone/ipad applications. Those services, that were leveraged, provided http accessible formatted text (sometimes json, sometimes xml). There was very little structure and very little barrier to entry. There were no requirements about programming language or platform. You made a network connection and you parsed the text (see Eric Raymond for the advantages that text based integration have given unix). This enabled all kinds of platforms to succeed in ways that were unexpected. The iOS, Android, and web browser platforms has all benefitted greatly from this neutral, simple approach. This is what created the network effect that has allowed folks like twitter and facebook to succeed (see Tim O’Reilly and his web platform talks).

The reason folks say that the bigs guys like Microsoft, Oracle, and (now) Apple don’t get the cloud is because they are trying to redefine cloud without web. I don’t mean the web as in the web browser. I mean the web as in the web platform (http, json, xml). It’s all about the ability to access cloud services from any language, any platform, simply and without rigor. It’s the platform that matters. That platform in combination with Apple’s legions of designer/developers is incredibly compelling. However, by creating a webless cloud with moats and walls, Apple is preventing the network effect that occurs when built on web standards. As a result, they are restricting interoperability and creating a cloud that feels anti-web. It’s not bad. Webless clouds just don’t feel right. It’s like picking up your glass of diet coke and getting a mouthful of sweet tea. It’s good – even great, just wrong. Google, Flickr, Instagram, Yahoo, etc all have “Web Clouds”. Apple is implementing something more akin to their app store, something more controlled and more isolated, the Apple cloud.

This isn’t likely to change either. Apple, simply, has no business advantage in open cloud APIs. In fact, without selling $99 subscriptions is has not business interest in web applications. I assume they will ultimately justify them by charging for additional space/features. However, that will never be a big money maker for them Instead, for Apple, the cloud is a feature that enhances the value proposition of their hardware/software stack. That is how they can justify free. For Google and other web companies that has cloud services, their business models allows them to profit from both platforms and applications. Apple, like other traditional software companies) are all still trying to find ways to line up their business model with an environment where is is hard to protect intellectual property and copyrights while differentiating your platform with features and content.

That is the fundamental difference.

Of course, this all could change if apple comes out with a series of web based APIs for iCloud.