I just happened upon an article by Matt Marshall on Venture Beat: “How HTML5 will kill the native app.”

Ugh.

This article reads like an HTML5 marketing document. There’s good reason to be excited about HTML5. But I believe there are a couple of key things missing from this discussion:

1. The value of cross-platform code to developers is a myth. — Yes, many people say they would love to standardize on one platform and write once and save “billions”. But in reality, developers like to learn new skills, platforms, and languages. And clearly having to rewrite code to a brand new platform hasn’t stopped hundreds of thousands of apps getting written for iOS. The best modern developers are well-versed in a variety of client and web-based technologies and platforms, and recognize that one solution doesn’t fit all. And ultimately they, and the businesses that employ them will flock to any platform that has a real promise of commercial success and novel functionality no matter how much new code they have to write. Do we really think iOS is the last time that a new platform will attract tens of thousands of developers to write hundreds of thousands of new apps from scratch? If that’s true the software industry is dead.

2. HTML5 has still not addressed a critical piece of the UX — responsiveness. – HTML5 and it’s predecessor Flash have are not focused on the degree of responsiveness we demand from really polished software. It’s true that in many cases, we don’t need instant responses. And with the advent of AJAX style development web-based apps have come a long way from needing to reload the page every time you make a state change. However, the fundamental value of an HTML page (and app) being able to load progressively is often counter to the type of rock-solid responsiveness that we need from many software experiences. I know that most user’s will live with little delays and not even be able to articulate that there’s a problem. But like the soft click of a door closing on a well-engineered luxury car, customers do know when something just “feels right” (and conversely… when it doesn’t). When I can load thousands of items in a list on a webpage without having to do pagination, when that loading feels instantaneous (even though there may be progressive loading of the data into memory), and when scrolling feels smooth as butter and super fast, then I’ll feel like web apps are getting closer. I don’t think there’s a technical limitation on this per se in HTML5, it’s just that it’s not optimized for these types of interactions. Responsiveness is one of the unsung heroes of a polished user experience, and even with all its innovations and AJAX goodness, GMail can still be frustrating to use for heavy mail users.

To be clear, I’m a fan of HTML5 and here at Jackson Fish Market we will use it as appropriate. It’s a tool, like many other tools in our toolbox. We’ll use it when it’s the right tool for the job. And we’ll use other tools when they are appropriate. The most rational and easy to work with developers I know share this philosophy. I’ve found that developers who like to spend lots of time arguing about which tool is the “end all be all” are doing me a favor by letting me know up front that I shouldn’t be working with them.