I've been discussing HTML5 for some time now. In July of 2010, I mentioned that I wasn't particularly concerned about PowerBuilder supporting HTML5 in the initial PowerBuilder.NET release (12.5) because:

• "HTML5 is largely still in its infancy, and there appears to be too much opportunity for it to fragment as previous HTML standards have done." [1]

In December of 2010 I devoted an entire editorial to HTML5 [2], noting that:

• "I have some basic concerns that make me reluctant to recommend using HTML5 as the basis for any line of business application development in the near future."

But also recommending that as far as a future version of PowerBuilder, Sybase should:

• "Still focus on Silverlight, still work on HTML5 as well, and give us the capability of generating applications that implement both. If the Silverlight player is available, use that, and if not then downgrade to HTML5. That will ensure that we have the capability to deploy apps until such time as HTML5 is mature and capable of being handled by a majority of machines, but also allows the app to work on non-Silverlight enabled devices provided there is a HTML5 browser available to it."

In July of 2011 [3], I again looked at HTML5 and concluded:

• "I expect to see a significant increase in the market penetration of tablet devices in the near future, and that the majority of application development done for those devices will be the development of native apps, not web [e.d. HTML5] apps."

HTML5 also got a passing reference in my editorial for November of 2011 [4], in which I noted:

• "I still believe it's an immature technology."

But also that since Windows 8 is supposed to support the use of HTML5 to generate desktop applications that

• "It's beginning to look like HTML5 may become not only the best long-term bet for web deployment, but for desk-top deployment as well."

Why do I bring all this up? Because in January of 2012 I penned yet another article where I mentioned HTML5 [5], and some people seem to think that last article represented some sort of conversion experience and that I was now a HTML5 fan boy.

So, just to make sure everything is clear, I'd like to review what I said there, compare that to what I said in these earlier articles, and then elaborate a bit further to make sure there's no confusion.

What I said in the January 2012 article, was:

• "While HTML5 may be the future, the future isn't here yet. [...] Nonetheless, the lesson we might take away is that HTML5 currently isn't suitable for line of business application development, though it may be in a few years."

I've highlighted a couple of terms in that quote as well as from the previous quotes, because I think they're crucial in understanding what I've been saying for some time. I do not believe that HTML5 is currently a mature technology suitable for the generation of enterprise line of business applications. It may be in a few years, it currently isn't. I have to agree with the assessment of Mike James in his iProgrammer article [6] that HTML5 currently is "...another one of those false marketing ideas with very little substance" and that with reference to Microsoft's shift away from Silverlight to HTML5 that it was "...perhaps the most reckless abandonment of a technology in the history of technology." Don't get the wrong idea from those quotes, Mike goes on to explain why the eventual adoption of HTML5 could be a good thing. He's just noting that somebody needs to "work on the underpinnings that we need to turn it into a usable technology."

As Mike mentions in his article, and Richard Holdsworth goes into some detail about in his own piece in TechCruch Europe [7], when you visit an HTML5 website what is really doing the heavy lifting, not HTML5 itself, but JavaScript and CSS [8]. How much of the heavy lifting? Let's take the recently released HTML5 version of Cut the Rope as an example. Of course, Cut the Rope started out as an iPhone application, and it runs quite well on such mobile devices (it was later released for Android and just recently for the Sony Playbook). They recently made it available for HTML5 browsers by porting the OpenGL code over to the HTML5 canvas API and the Objective C code over to JavaScript. How much JavaScript? 15,000 lines running in the browser. [9] And what do I get when I run it in Safari on my iPad? This screenshot shown in Figure 1.

"A little too slow?" You've taken an application that used to run fine on the device, ported it to HTML5, and now it's essentially unusable there. I guess I'm not impressed. I actually get that same message with most other browsers than IE 9 running on a laptop. The primary reason that it even runs well in IE (and only IE 9) is because IE 9 pre-compiles the JavaScript on a background thread.

Don't get me wrong. I do look forward to the day when we can create and run applications that will run on multiple devices in different browsers and will perform adequately on them all and do that without having to use plugins. HTML5 offers the promise of making that happen. However, it hasn't delivered on that promise yet. When it has, I'll endorse it enthusiastically. Until then, I can only recommend it as something to watch and learn for the future, but not to use for production applications today.

References:

1. http://pbdj.sys-con.com/node/1464997
2. http://pbdj.sys-con.com/node/1658103
3. http://pbdj.sys-con.com/node/1884245
4. http://pbdj.sys-con.com/node/2070202
5. http://pbdj.sys-con.com/node/2120922
6. http://www.i-programmer.info/professional-programmer/i-programmer/3539-2011-the-year-html5-won.html
7. http://eu.techcrunch.com/2011/12/30/html5-idiots-are-confusing-meatballs-with-spaghetti/
8. When most people refer to HMTL5, they are most likely referring to the combination of HTML5, JavaScript, CSS and other web standards (e.g., WebSocket). HTML5 alone would give us very little functionality.
9. http://www.cuttherope.ie/dev/