Reading time:

5 minutes

Word count:

1081

AudioBoo Updates: QCon London 2009: Enterprise JavaFX for the Web Platform

Hi All

What follows in an extract of an email conversation, which yours truly had with Wylliams Barbosa Santos. Santos lives in Brazil. He is a MSc Candidate in Computer Science, reading at the Federal University of Pernambuco (www.cin.ufpe.br/~wbs). Santos watched my presentation: Enterprise JavaFX for the Web Platform , which place at QCon London 2009, (https://www.infoq.com/presentations/Java-FX-Peter-Pilgrim )

First, I answered Wylliams’ email as an audioboo podcast, however I was told that is not accessible to people who do not understand a lot of English as a foreign language. So for the benefit of those listeners I have copied my response here so that they can get a friend, who can read English very well, to translate. NB: I have significantly expanded also on my answers in the audio podcast.

2009/10/21 Wylliams Barbosa <wylliamss at gmail dot com >:

Thanks very much Wylliams, I am very pleased and glad that you liked my presentation. I think I did a good job and it should have been a lot better looking back with hindsight.

To be honest, I have not yet used Flex in anger. May be I should try it? So I cannot really advise you on a direct comparision. However, I believe in terms of components Flex has a lot of them. Flex has more richer component than JavaFX 1.2 at the moment. However, if you give time a couple months or so, at least until February 2010, then I believe this story will change. JavaFX will have rich components for enterprise and consumer development.

JavaFX has a lot potential for Java like rich internet applications that already trust in the JVM. If they [companies/enterpises] can trust on the server side with the JVM, it is a lot less hassle to get them to trust with the client-side.

Probably not yet. If you are thinking that the RIA market consists of mostly desktop application, then the Flash platform probably will be dominant, at the time of writing, 21 October 2009, closely followed by Silverlight (Microsoft) for certain applications.

The issue, I think is, what exactly is the RIA market and how do you classify it?

Do you define it by the desktop?

If so then we must look at the story of the Java applet since 1995. Sun was very late to realise it error in business strategy to allow client-side Java to fall away in the early “noughtes” (2000s)

In order for JavaFX to win market share on the desktop then the new Sun Microsystems App Store must be successfully. It has to allow users to download and run applications securely, efficiently and with superb usability. It also needs businesses to get behind the Java AppStore and push there commercial applications to it. As I said in my presentation QCon London that I believe portals and aggregated websites will eventually be replaced by aggregated application stores and RIAs.

Do you define it by the web and browser applications?

Well known web developers, design agencies, pushed the boundaries and came up with AJAX/JavaScript. There was a sudden jump in the number of popular JavaScript libraries and much improved respect for JavaScript as a proper programming languages. Later on these web developers reexamined and rebuilt the JavaScript interpreter engines with runtimes like V8 and the other Just-In-Time interpreters. Now they are moving to diminish the effect of browser wars further and end their constant battle with web standardisation, bodies and legacy versions of Internet Explorer 6. Even some web advocates are considering getting rid of the desktop and making the whole operating system UI as a dedicated web platform, presumably no desktop applications at all, or the very minimum there of.

In order for JavaFX to win market share on the web platform, the modularity, performance and ubiquity of the runtime must score high values. Modularity is about reducing the payload when a Java[FX] application first starts. By breaking the runtime into modules, only those most critical modules are downloaded, when your application first starts on the user’s machine. Performance is all about increasing scenegraph rendering speed, talking advantage of hardware acceleration, thereby reducing the load on the CPU and perceived latency by the user. Ubiquity is all about getting the software out there to the market.

How about mobile market applications?

It’s well known that Java can run on mobile phones and other embedded devices. We know there is a market for Java running on mobile phones with J2ME. There is now another market on runtime platforms, other emdedded device including operating on phones. In my opinion, the mobile and embedded field is confused and slightly immature in the standards out there, because there are so many of them. A solutions architect really does need to act with due-diligence when choosing the requirements for a system that work on an embedded device. The business choices at the moment for serious mobile development are iPhoneOS, Google Android, Windows Mobile and Nokia. In October 2009, the mobile phone market for rich internet application definitely has a clear leader in the runtime and platform. Unfortunately, all you can say that it is a walled garden. In which, you can only exist in the kingdom if you decide to play to by the rules of the state. I think there is new royal family on the block, Apple iPhone. The king is now iPhone, the queen is the Apple AppStore there.

In order for JavaFX to win on the iPhoneOS platform, the JVM has to simple run on it. A good second choice is the Android platform. [Business-]Decisions, decisions, decision …

I think that it is all to play for. So we must set out our stall and see what business comes out of it …

>
> Thank you very much!
>

No problem. You are very welcome. ANd congratulations on Rio de Janeiro winning the Olympics 2016.

> ————————————————————

AudioBoo: QCon London Talk / Enterprise JavaFX for the Web Platform

Enjoy. Do responds if you do have comments. Thank you.

This is Peter Pilgrim. Out.

https://blogs.sun.com/SDNChannel/entry/this_ain_t_your_dad