Three Key Business Objectives For JavaFX
The technical side of the JavaFX platform is progressing well since its launch in December 2008, with technical innovations being delivered rapidly. There’s still a long way still to go of course. For example: more API features; optimized “user experiences”; better IDE support; new design tools; and improved video support are all needed. However what’s in place now is a great start, and the platform is quite usable. Developers, amateur and professional alike, are going to find JavaFX Script a super-fun, easy-to-learn language to code in.
With the technical side of the platform beginning to work well; a hundred million JavaFX runtimes and a hundred thousand SDKs shipped, I think some business issues have now come to the fore. I thought I’d blog what I believe to be three important business objectives for the FX team. If you like, you can file this under, “What I want from JavaFX”. In no particular order:
- Do what’s necessary to ship the JavaFX SDK for Linux (minus the mobile emulator if that’s too much work). Early adopter developer mind share is important for the platform while it’s new. While Linux has only a small market share, it has a disproportionately high share of early adopters.
- Work with Apple to get an up-to-date version of Java SE 6 shipped on Mac OS X. In particular, there’s a business need for Apple to ship Java Plug-In 2 on the Mac, and to fix Java SE 6 Web Start bugs. JavaFX is being held back while it still has to target Java SE 5 VMs. When Java SE 6 can run in the browser on the Mac, then it’s time to flip the switch and allow JavaFX to use Java SE 6 features.
- Persuade RIM to adopt JavaFX Mobile on Blackberry and announce it, including back-porting to OS platforms of current devices. Blackberry is one of the most important mobile developer platforms – it has an importance that is disproportionate to its market share; and RIM could really benefit from supporting JavaFX Mobile as core part of its platform. To make this happen might need more work than on many other mobile platforms. So be it. It’s important (as much for RIM’s benefit as anyone else’s) that this happens and is announced as soon as possible.
If the FX team can achieve these three objectives, I think it would move JavaFX to really a rather interesting place in the RIA world, and significantly enhance developer adoption as a result. Of course there are other objectives too. For example, there’s a need to get Nokia, Samsung and Motorola to join Sony Ericsson and LG Electronics as partners for JavaFX Mobile. The three objectives above would do me for now, though. NB I’m not suggesting any of the above are easy (I’ve done enough biz dev in my time to know that, in the end, you can only do the deals you can do!) – I am saying they’re important though…
Carl Antaki wrote:
Simon don’t you think the Plugin 2 team has to work hard on improving the startup time and the performance of video. It took me 18s in order to see the video of the JavaFX launch, 3s to load the prime factor example. I have a 7Mbps connection and a decent machine (Dual Core AMD, 2GB RAM, 128MB VRAM on Windows XP). I feel that no one worked on the performance yet, there are loading too many classes and Dlls. Flash on the other hand takes less than 0.3s. What also need to be done is improve the user experience, removing the ugly security dialogs
Posted 16 Feb 2009 at 12:30 am ¶
simon wrote:
Carl, absolutely – I wrote in the post, “There’s still a long way still to go of course. For example: … optimized “user experiences”… and improved video support are all needed.”
I don’t think it’s true to say no-one worked on FX performance. Start-up times of sub three seconds are a huge improvement on what’s gone before, and are, I think just about “good enough” – provided the whole user experience is properly addressed. Faster is better though. On the other hand, the c. 18 second start-up time for playing videos on JavaFX.com, is not even close to being “good enough” (I haven’t explored what the cause of that slow start-up might be, it’s very odd). Anyway, more work clearly needed on video. Having said that – it at least works – which is more than it used to.
Those kind of things, though, are more what I would call “technical” issues. Sun has a roadmap for FX. I’d expect those to be addressed. In this post, my focus was not on technical issues, but on *business* issues that need to be addressed. When the platform wasn’t ready to be used for real projects, the business issues weren’t so important. Now though, FX really *is* good enough technically to be used for many types of RIA project, and so these business issues will become a factor in slowing adoption in the very near future if they’re not addressed.
In the case of Linux support, it’s at least *partly* about priorities because the community has figured out by itself how to get the SDK “kind of” working at on Linux and Solaris. In the case of Apple and RIM, it’s about relationships with partners, and really getting them on board with FX and what’s needed to make it great. For sure these are not the easiest of objectives and in many ways are much more difficult to tackle than the purely technical objectives.
Posted 16 Feb 2009 at 1:16 am ¶
Chui Tey wrote:
Sun will need to make some quick moves in the developer tools arena and beef up it’s consulting division if it wishes to avoid going the way of Silicon Graphics.
Unfortunately, both of these are difficult to execute in a tough economy.
To get sufficient buy-in, Sun has to be sufficiently transparent about how far are they planning to commit resources towards making Java a viable client-side technology. Adobe is betting the farm on AIR. Anything more wishy washy from Sun will see it’s RIA lunch eaten by somebody else.
Posted 16 Feb 2009 at 10:08 am ¶
simon wrote:
Maybe. I’m not so sure the challenge is quite as dramatic as that. Silicon Graphics made some serious strategic mistakes: they started pushing Windows NT as their future, rather than IRIX (despite the fact their customers *loved* IRIX); and they failed to realize that the technology for high-performance graphics was going to be driven by the mass market for video games on home computers, and didn’t adapt.
Sure, Sun is under pressure commercially – they need to transform their business. The difference between Sun and SGI, though, it that the changes Sun are making aren’t dumb.
As for JavaFX, the Java community is pretty big, and on the whole it’s a smart bunch. As Java developers start to understand JavaFX, I’d wager large numbers will prefer FX to Air. Sure Sun needs new JavaFX tools; but it’s more important that these are awesome tools than that they’re delivered “next week” IMHO. Yes, the economy might affect that – right now, though, the Java division is growing revenues and profits. If they can keep it that way, there’s no reason they should face cuts in their strategic Java offerings like FX.
Also, I really don’t think the RIA battle is a zero sum game. That is, I don’t believe that for JavaFX to win, Air and Silverlight have to lose. Yes, Sun has to make JavaFX work amazing well technically; that’s a “hygiene factor” for any modern RIA tech. However, if a RIA platform is widely deployed, and works well, the barriers to adoption aren’t going to be technical, because there won’t *be* any technical barriers to adoption.
So, the barriers to RIA adoption are likely to be much more about the business side of things – how developers can make money and/or get their apps in front of large audiences in a cost-effective way. For example, what’s the reason the iPhone App Store is so popular? Yes, the iPhone is fun platform to code for; but it’s the business side that’s so killer – developers know they’re on a level playing field with everyone else; the playing field is one where customers have great experiences; developers don’t feel they’re getting ripped off financially; and it’s really easy to sign up; it costs nothing to ship a piece of free software to millions of users etc.
Posted 16 Feb 2009 at 11:18 am ¶