You can’t make this stuff up. Someone sent me a couple of great graphics supporting my assertion. So please respond with anything that supports the contrary.

Update:

After getting a little splashback from some of my friends and colleagues, I thought that I’d add a little supporting evidence to my assertions.

“Moving” to HTML5 at this point is senseless gambling.

I’m not just blowing smoke up your pipe here either. Open your browsers to HTML5test.com and see for yourself. Come on… go ahead and do it!

The HTML5 support score on latest Windows 7 browsers that I have: IE 138; FF 330; Chrome 400. And then on the Mac OSX 10.6 that I have; Safari 319; FF 340; Chrome 400. That makes Chrome our top student with a whopping 80%. And I’m what many would call an “Advanced User.” Expect less from the general public.

Last I checked, 80% was a low “B-”. And with over 50% of the browser usage coming from browsers that have failing scores, you can see why I would not recommend HTML5 except in specific edge cases. And you thought IE6 was bad! (Browser stats from w3schools.com)

As a mobile or web developer, adding HTML5 to your list of skills is imperative. But with the severe lack of consistent support, moving an enterprise development project to HTML5 now is purely experimental or an exercise in ego.

If you think I’m wrong, please provide supporting evidence and I’ll gladly educate myself.

Original Post:

The fear, uncertainty and doubt (FUD) is being slung around by profiteers like folding chairs at a WWE event. The haters are still being haters. Nothing new there. But now I see JavaScript companies’ desperate pleas for Flex developers to start using their HTML5 software.
The context is all wrong here.  Very, very few Flex developers have shifted, moved, changed over, or whatever you want to call it… to HTML5 (or anything else JS-based.)

There is not a move to HTML5

I will go as far as to say that there is not a move to HTML5. The simple fact is that, developers are being developers. No matter the background, we are always trying to broadening our skill sets. This includes HTML5 since it started showing up a few years ago. For anyone to imply, or state outright, that there is some mass exodus from Flex is completely false!

The reality is still the same, Flash Player is still the most consistent cross-browser, cross-OS, and cross-device platform for software development.

It doesn’t matter if you are building business software or games, with one technology you are able to build for the desktop, Internet Explorer, Chrome, Firefox, Safari, Android, iPad, iPhone, Blackberry and Internet TVs.

What most haters fail to realize is how nice it is to go through your bug list and not find ANY bugs that are browser/platform specific. The only bugs I have, are actual bugs that I can fix. Not browser support related issues that you have no control over. For the first time in years I had to deal with browser specific issue when our On3 client embedded the application in a JSF, JSTL, ADF container. It reminded me of how good I have it. I don’t have to deal with this headache on a daily basis. In fact, it was one of the complete joys that drove me from building DHMTL development to Flex development.

So, the next time someone says, “Flex is dead” or “Everyone is moving to <insert tech here>,” take it with a grain of salt. In all likely hood, they have a hidden agenda.

Thank you Moai for this great graphic on the HTML5 Hype vs. Reality.

That is until the other day.

Adobe announced that they were donating the Flex SDK to the Apache Software Foundation (ASF). It’s about effen time!

I give the Adobe Flex team a lot of credit for creating a really great product, but in the end… they could only put so many resources on it for it to make financial sense. Going ASF is the right thing to do.

Adobe, go make more great tools for us… so that we can continue to inspire and engage our customers with great software.

Oh, and I’m hiring Flex developers. Just thought I’d throw that out there in case you happened to have been recently laid off.

The core of Flash Player is the Tamarin Virtual Machine, which is an open source project under Mozilla. While the SWF file format is not fully open, it is documented by the community on osflash.org. Additionally, there are numerous open source products that read and write SWF files.

The Flash Player’s product direction has traditionally been heavily influenced by the community and their needs. The core language for Flash Player is an implementation of ECMAScript 262, which is the same specification for JavaScript. Flex also uses CSS for styling of components/applications.

There are also several libraries included with Flash Player that are licensed through other parties (i.e. h.264) that are not open. Thus, preventing Adobe from making the whole thing open source if they wanted to. Not sure that they would, but this definitely kills the idea.

Come save us HTML5 in 2022 AD

HTML5 has been in the works since 2004 and is still in “draft”. Its primary intent is to reduce the need for proprietary plug-ins (like Flash Player and Silverlight).

I can definitely see the benefit of not relying on a plug-in for multiple reasons. There is a concern if users will have the plug-in, but the bigger concern is vendor dependence. I think Adobe has the install base issue covered fairly well, yet it should still be a concern for locked down environments. To the bigger concern, I’d say that we already depend on companies like Apple and Microsoft quite heavily and that Adobe is far from a fledgling start-up that would be considered very risky. Naturally, I understand to the concern and will help my clients choose the appropriate technology.

The reality is that HTML5 is not coming anytime soon

Steve Jobs, CEO of Apple, claims that “the world is moving to HTML5″. How is that going to happen Steve when Ian Hickson, editor of the HTML5 specification, expects the specification to reach the W3C Candidate Recommendation stage during 2012, and W3C Recommendation in the year 2022 or later?[http://en.wikipedia.org/wiki/HTML5]

Should we hold off development for a few years while Google (Ian works at Google) finishes the specifications?

Finally, how many different implementations of HTML5 do you think there will be?. There will most likely still be cross browser compatibility issues to deal with.

Flash Player and Silverlight

I’ll sticking with vendor dependence that I can use now over incomplete technology with potential compatibility nightmares any day!

A good friend of mine, Cliff Hall, is most known for the PureMVC framework which was inspired by his his lack of love for the Cairngorm framework. Well, Cliff apparently was irritated by the inability to create licensed products with Flex and AIR too, so he did something about it. Zarqon. He has created very useful tool for creating a licensed Flex and/or AIR applications. How you implement your app is totally up to you, Zarqon just helps you easily issue and validate licenses.

Are you a software skeptic, me too! No worries, there a free 30 day trial license that will allow you to fully evaluate the product and make a decision about whether it’s right for you.

It is now in “beta”, but if you begin implementation of a Zarqon-licensed product within the trial period that you plan to launch, let him know and you just may get a free license. Even if you do have to buy a license, it won’t set you back but $25.

I’m going to start my testing immediately and I’ll let you know my  thoughts later.

You can read all about it and sign up for the Beta at http://zarqon.net

The hidden reason is cost. Cost for setting up each developer environment. Cost of licenses.

The bigger reason is, or should be, security.

Enter Platform as a Service (PaaS) to the conversation.

Although the idea of using the cloud for development is good in theory, it can be cost prohibitive. Here is why:

• The cost of using the cloud. In reality, you’ll end up spending around $1k per computing instance per year. i.e. per developer

• You still have to license any development tools for each developer seat. Check your EULA, they typically address ways of circumventing license requirements via virtualization.

You do get the benefit of quickly setting up an environment for a team. But lets face it, the only real benefit is the level of security in that your code base is never on someone’s laptop. Don’t get me wrong… this is a really big benefit!

From my research, I would recommend VMware ACE instead of PaaS. VMware ACE allows you to provision standardized client PC environments inside secure, centrally managed virtual machines called ACEs. Each ACE contains a complete client PC—including the operating system and all applications. Administrators use the dynamic policy configuration capabilities in ACE to lock down endpoints with device and network access control, there by protecting confidential company data and ensuring compliance with IT polices. You will get security through a managed virtual machine and the ability to quickly deploy a “ready to go” environment for your developers.
No, its not cheap. But it is way less expensive than the cloud with the equal benefits.

I know this because I’m looking into these tools in order to manage our training facility where we teach ColdFusion, Flex and LiveCycle.

Most components have a toolTip property that you can set (typically bound to some data). But the List-based components are a little different. They use dataTipField and the dataTipFunction. The “data” version of toolTips is based on the list iterating through it’s dataProvider and then adding toolTip for each row.

So, while adding toolTips to a List with an itemRenderer, I ran into a problem. My toolTips weren’t showing up when I set the dataTipField. The dataTipFunction didn’t work either. What gives?!

Well, I had an inline itemRenderer for my list. And apparently the List doesn’t like adding toolTips, via the dataTipField, to a component that is inline.

The solution: Simply add a toolTip to the inline component. And in my case, bind it to the data property that is passed in by the list.

Note: You don’t even need to set showDataTips to true.

Date / Time
November 16, 2009
Event: 9am – 4pm
Registration: 8:30am

Main Website / Registration
http://www.adobe.com/go/flextrainingforcfdevelopers

Monday, November 16
The Curtis Hotel, 1405 Curtis Street, Denver, CO 80202

The Flex PMD tool is locate on the Adobe Open Source Projects site.

http://opensource.adobe.com/wiki/display/flexpmd/FlexPMD

It is available as an ANT task, command line and Mac OSX Automator.

And before you ask, no, there isn’t an Eclipse plugin… yet.