Official – JavaFX To Be Released Dec 2
I wrote yesterday that it was time to get excited about JavaFX. Now, the official release date for JavaFX 1.0 for the Desktop has been confirmed. It will ship Dec 2, a little bit later than originally scheduled. Sun has taken the extra time to try to make it as polished as possible. For example, the FX team has done some work to optimize video playback performance that they wouldn’t have been able to do if the release had been in November. The aim seems to be make the version 1.0 release a “true” 1.0 release; so while it may not be ready to outdo Flash (which is at version ten) just yet, it should be ready for production use. JavaFX Mobile is set to follow the Desktop release in February 2009.
Now you know…
Asam Bashir wrote:
So when should we expect a JavaFX enabled mobile? I’m trying very hard to be excited about JavaFX, I’ve heard your reasons why it should allow app development with very high quality user interface elements, but apart from Sun and Java developers, I’ve not heard much else. What will a JavaFX mobile do that my iPhone can’t?
It’s all very well Sun executives banging on about JavaFX but there’s nothing to see here for end-users, at least with Google when they say something they deliver, as is the case with the Android HTC G1, whether it actually sells beyond the early adopters is yet to be seen. If it does sell, I’d imagine it to be at the expense of Windows Mobile rather then affecting iPhone sales in anyway. With Microsoft so pre-occupied with trying to rescue it’s desktop business with Windows 7 development, and nothing else coming out regarding mobile phones, Sun had a real chance here, but as I’ve said a million times, the problem with Sun is not it’s products, it is the executives that have no clue about consumer devices or how to sell and market them.
Posted 09 Nov 2008 at 4:37 pm ¶
Asam Bashir wrote:
Even on the Open Solaris front Sun is moving too slowly and not adding the details that would make it attractive to end-users. It’s only with the yet to be released OpenSolaris 11.08 release that it’s going to become useable on a laptop and that is too slow. There is still a whole year to go until Windows 7 is finally regurgitated out, if Sun had a viable OS product for laptops combined with an integrated method to manage a JavaFX Mobile, it would be onto a winner. However, with Sun, I’ve found, they’re all pipe-dreams and when they fail they’ll blame the economic downturn, when in reality it’s simply a lack of understanding of the consumer market.
Posted 09 Nov 2008 at 4:53 pm ¶
simon wrote:
Asam wrote:
What will a JavaFX mobile phone do that your iPhone can’t?
That’s a really good question. I don’t think anyone knows for sure what the answer to that is. Here is some of the potential, though:
- Allow what could be called “iPhone-class” apps to be developed to run on many hundreds of millions (and over the course of a couple of years, billions) of feature phones and smart phones.
- Allow the *same* rich apps that can run inside a browser on a desktop, to run on a phone.
- There will also be, I think, some all-new cloud services to go along with JavaFX Mobile. The idea is to provide a whole bunch of stuff to make it easy for JavaFX Mobile apps to interact with the Internet in interesting ways.
Of course, the devil is in the detail on all this. Think of JavaFX Mobile in terms of really raising the bar for app quality on low-cost feature phones (which is badly needed!).
The billion dollar question is: will the JavaFX Mobile end-user ecosystem be orders of magnitude bigger and more valuable than the iPhone user ecosystem? That will, in large part, depend on Sun’s hardware partners for JavaFX, and how robust and consistent their Java implementations are. What I will say is: the situation today in that regard, while far from perfect, is a lot better than it used to be.
Asam wrote:
So when should we expect a JavaFX enabled mobile?
I’m not sure. I’ve not seen a clear statement on this. There’s a chance that it will actually be a download for hundreds of millions of existing mobiles. I say this, because I’ve seen specifications that describe what Java ME apis will be required to run JavaFX Mobile apps. So, it may just need a phone with those specs to run. It may, though, require new phones.
The only date I have for JavaFX Mobile is “February 2009″. That could be when a developer SDK is released with phone simulators etc. It could be when JavaFX phones are released. Or it could be when future JavaFX phones are announced. I guess more details on this will emerge in the next couple of months.
Developers can, in principle, actually write JavaFX mobile apps today, because the JavaFX api is documented so people can see which parts are common to JavaFX Mobile and JavaFX Desktop; and which parts are exclusive to JavaFX Desktop. The best-case scenario answer to your question is: JavaFX Mobile apps can be shipped from February 2009. It could be a lot later than that though.
Re: Sun had no clue about consumer market
I think it’s hard to argue that Sun has “no clue” about the consumer market. For example, there are:
- 3.5 billion Java smart cards in the market (I’m not sure, but the SIM card in your iPhone might even be programmed using Java)
-2.1 billion Java-enabled mobile phones currently in the market
- 800 million desktop PCs have Java installed
- Every Blu-ray player can run Java, and all the content on Blu-ray discs is authored by using Java
- Android mobile apps are built using the Java language
- Plus, Java is widely used in the embedded electronics market e.g. in parking payments stations, in-car dashboard displays etc. etc.
So Java has been highly successful as a consumer technology in many ways. That doesn’t mean it’s perfect, though! Far from it. I’m just saying, it’s not really true to say they have “no clue”.
Obviously, JavaFX represents a renewed push for Java in terms end-user graphical stuff. Also, I think it’s true to say that Sun hasn’t benefited financially from Java’s success as much as it could/should have done.
Re: Sun being too slow to release products
I wish they were faster. However, they only have finite resources, and they’re doing a *lot* of stuff. I think the IT/software community really needs Sun to survive – they are doing so many things that are valuable to software developers (there are millions of Java developers, and millions of MySQL developers for example). I hope they can turn the ship around before they get acquired, and the true value gets destroyed.
Posted 09 Nov 2008 at 7:00 pm ¶
Asam Bashir wrote:
Re: Sun had no clue about consumer market
I don’t disagree that Sun has a lot of experience in the consumer space, but for most of that it didn’t have to deal with consumers directly, so it doesn’t have consumer marketing and human-user interface experience. If you want an ‘iPhone-like’ experience, those are the two main areas you would need to be good at.
Further you wrote,
2.1 billion Java-enabled mobile phones currently in the market
Only a very small fraction of those phones will ever have their firmware upgraded in their lifetime. Sun should concentrate on phone specifications for the 2008-on period, it’s wasting time trying to develop for older phones. So when you say, there’s a chance that it will actually be a download for hundreds of millions of existing mobiles”, that would be a mistake and distraction using up valuable time and resources.
At this stage Sun really needs to find hardware partners that can work closely with it to develop both OpenSolaris and JavaFX Mobile into a cheap alternative to Mac OS X and iPhone. Apple isn’t interested in the low end of the market, Jobs has already said he would never ship a sub-$500 laptop because they haven’t figured out how to make one that’s not crappy. Sun should move quickly whilst Microsoft is pre-occupied with Windows 7 and integrate it’s development for a complete laptop/phone package.
Re:The billion dollar question is: will the JavaFX Mobile end-user ecosystem be orders of magnitude bigger and more valuable than the iPhone user ecosystem?
Why compare to iPhone market when Sun is never going to have a product to match Apple? Again, this is a distraction for Sun, the product they need to compete with is the Windows eco-system since here open-source has revenue potential in the sub-$250 netbook/laptop and sub-$100 ’smart phone’ markets where Microsoft’s business model breaks down, and Apple isn’t interested in selling to those customers.
Posted 10 Nov 2008 at 3:29 am ¶
simon wrote:
Only a very small fraction of those phones will ever have their firmware upgraded in their lifetime.
Sure. I was actually meaning that there’s a chance that it could be bundled with an app, in a similar way to LWUIT
https://lwuit.dev.java.net/
If you develop an app with the LWUIT APIs, you bundled the LWUIT platform with your app. That is, no firmware upgrade required. I’m not saying it’s likely this is the case; just that I haven’t seen a clear statement on it one way or the other.
Why compare to iPhone market when Sun is never going to have a product to match Apple?
For the simple reason that software developers/companies wanting to develop mobile apps have to decide which mobile platforms to support; and how to prioritize development for those platforms. All sorts of factors come into play in making those decisions, including a realistic estimate of the size of the market opportunity.
(Sun) doesn’t have consumer marketing and human-user interface experience. If you want an ‘iPhone-like’ experience, those are the two main areas you would need to be good at.
Clearly, Apple is a leader in these areas.
Only time will tell if they maintain that lead, or if one or more of their many competitors catch up.
Posted 10 Nov 2008 at 12:09 pm ¶
Asam Bashir wrote:
I think my problem is that I don’t believe it’s possible to write-once run-everywhere and still have a decent user-interface experience. What developers want and what consumers want are different things, Steve Jobs understands this basic concept, Sun and Java don’t. If I buy hardware with the latest specifications and features then I want applications that take advantage of that technology now, not when Java has caught up several years later when the equipment has become obsolete.
Apple’s integrated model of software and hardware with a holistic approach means that it will always be several steps ahead, whilst Sun with it’s Utopian visions will make lives easy for developers writing once and running nowhere. But then, Sun doesn’t really care about it’s developers making money, as long as it can sell Java licenses to hardware makers. Sun doesn’t care about end-users, nor does it understand what kind of products they want, because at the end of the day, they’re not even aiming to get revenues directly from them, they just need to fool developers and hardware makers with Sun geek talk and sound like they know what they’re talking about…
Posted 10 Nov 2008 at 3:58 pm ¶
Asam Bashir wrote:
“The displacement of the RAZR by the iPhone 3G represents a watershed shift in handset design from fashion to fashionable functionality,” said Ross Rubin, director of industry analysis for NPD. “Four of the five best-selling handsets in the third quarter were optimized for messaging and other advanced Internet features.”
http://www.marketwatch.com/news/story/The-NPD-Group-iPhone-3G/story.aspx?guid=%7B7DDDC6C0-37E3-4534-B2D1-5F1399EBDB04%7D
So, end-users don’t buy a phone because it can run the same Java game as their mate, they choose fashionable functionality that is built-in rather then something available as download later.
Posted 10 Nov 2008 at 5:24 pm ¶
Asam Bashir wrote:
Looks like Sun has competition in the write-once run-nowhere market with VMware Mobile Virtualization Platform (MVP), will this be the final nail in the coffin for Java?
Posted 10 Nov 2008 at 5:55 pm ¶
simon wrote:
There is no doubt that there are challenges in making “write once, run anywhere” work well, especially on a diverse range of hardware/OS platforms. The situation isn’t as black and white as you make it out to be, though. I think that’s for a few reasons…
Firstly:
Do you use browser-based applications on your Mac e.g. apps like Facebook, or Google? I think you do. Do those applications use the latest features of your hardware? No they don’t. Not all applications need to support every aspect of a hardware/OS platform; and some of the most widely used applications don’t need to do that.
Secondly:
Write-once, run anywhere doesn’t have to be for the *entire* application. It’s still a huge win if you can re-use 90% of your code across devices, but have to write separate code, say, to make user interfaces that fit well with the look and feel of a particular hardware/OS platform.
Thirdly
One of the things that the success of native browser apps, and Flash-based browser RIAs, have done is change consumer expectations of graphical user interfaces. Prior to this, consumers tended to expect a consistent look and feel across apps. Today, consumers are used to figuring for themselves what things can be clicked on, what things are pull-down menus etc. This means the challenges for developing successful cross-platform apps can be fewer today than they used to be; and can make “write once, run anywhere” much more easily achievable than it used to be.
Apple clearly believes it needs to control both the hardware and software to be successful in the mobile phone market. RIM takes the same view. Microsoft, on the other hand, clearly believes it’s better to separate hardware innovation from software innovation; as does Google.
I’m not sure if anyone can really say for sure which is the winning strategy. Was the iPhone a game-changer for the mobile industry. Yes, it was. But how much of that was due to the fact that Apple has brilliant designers; how much was due to their “control hardware and software strategy”; and how much is due to the fact that, today, developing a new piece of electronics hardware is more about assembling off-the-shelf components, and programming the chips i.e. hardware innovation doesn’t really mean what it meant in the 1980s?
BTW – the VMWare mobile virtualization effort is about virtualizing the OS, whereas Java is about virtualizing the applications. It will be interesting to see how well VMWare does with this…
Posted 10 Nov 2008 at 6:12 pm ¶
Asam Bashir wrote:
I don’t get your first point, yes I do use browser based applications on my Mac, but on my iPhone I use dedicated apps made by both Google and Facebook which do use the latest features of the iPhone, and it works very well. Before the iPhone, I tried to use these same applications on Simbian S60 3rd edition Opera 4, but I gave up after a few seconds because the browser experience on a small screen just didn’t work. Ideas that work on a laptop screen don’t scale down to phone sizes. When phones do get screen sizes comparable to laptops, with future pico projection, they’re going to need a completely new way of human-interaction with them.
Regarding your second point, if you have to recode even 10% then you can’t really call it write-once, run-everywhere, that would be more like something being offered by the VMware Mobile Virtualization Platform.
Posted 10 Nov 2008 at 6:49 pm ¶
simon wrote:
Re: my first point
My point was – you were saying that you want applications that use the latest hardware features. However, on your laptop/desktop, you are happy to use apps that don’t use *any* particular hardware/OS-specific features of your computer. In other words, the hardware was irrelevant to the application. I’m not saying that you want to use browser-based apps on your mobile phone; I was just saying that there could be plenty of apps you might want to use on your mobile phone that could be written to run on the Java ME platform.
Re: second point
About re-coding the 10%. What you call it isn’t important. How long it takes to develop the app, and much effort it takes to maintain it over time is what matters. If you can re-use 90% of your code across platforms, that’s a really *massive* win. And, as I said in the third-point, the need to re-code that final 10% is going down over time as consumers stop expecting a consistent “look and feel” between apps.
Posted 10 Nov 2008 at 7:06 pm ¶
Asam Bashir wrote:
When I use Safari on my desktop, or on my iPhone, it is using the latest hardware/OS-specific feature, like touch gestures, multiple processors or the latest graphics cards accelerating aspects of browsing that are not dependent on what applications I run within that browser.
And when that 10% of coding, takes 90% of the time what happens is that applications only get developed for Windows whilst Mac and Linux users have to wait and hope the developer finds the time to think about them. Sun’s write-once run-anywhere utopian wet-dream is as big a myth as that of the clash of civilizations. As Mac users we’ve seen that time and again, and using that myth Sun helped Windows build the very monopoly that’s beginning to fall apart at the seams much the same as global banking.
Posted 10 Nov 2008 at 8:00 pm ¶
simon wrote:
I get where you’re coming from, but honestly, that 10% of coding takes 10% of the time. Java apps can work really well on Mac OS X. To quote from Apple’s own Java developer docs…
“Mac users are known for their high standards, so if you’re bringing your Java application over from another platform, you’ll find that adding support for Mac OS X is one of the easiest things you can do to make your program look and behave like a native Mac application. No matter what type of application you intend to deploy — server or command line, application bundle, or WebStart — by paying attention to the platform you can create a seamless Mac OS X experience. It’s no big deal to provide Mac users with a relocatable, easy-to-install application with a native menu bar and support for native events. Simply drop in your jars and resources and set properties with Jar Bundler, create an icon with Icon Composer, and you’re ready to go.
Because Java is integrated into the operating system and kept updated through regular software updates, applications don’t need to bundle a JRE. Java support is transparent to users, so when you distribute your Java application to your Mac OS X users, it includes all the flexibility, power, and charm of the native Mac OS X platform, allowing you to easily layer and customize your application to give your users the consistent user interface of the Aqua look and feel.”
There is some frustration that Apple doesn’t keep their implementation of Java as up-to-date as people would like. However, while they’re sometimes slow delivering, the Apple Java team does a really good job when they do ship their Java bits.
Posted 10 Nov 2008 at 11:25 pm ¶
Caleb Payne wrote:
So what’s the point of JavaFX? To be honest, I haven’t kept up with it, because I always thought it was going to flop. Just a kiddie play-thing. I mean, who would use something like this? I’ve been a java developer for about eight years now and I’ve been part of several different java development teams in several different environments (to include government, finance and micro-electronics). At no point have I seen a need for this type of technology.
So, is Sun’s reasoning behind this just to say, “Well, Microsoft and Adobe have an RIA technology, so we need to have one too” ?????
Don’t get me wrong; I love Sun’s products. I’ve used Solaris for years and do most of my development in Java. But sometime you guys just seem to be clueless when it comes to where you should go with technology. **STOP** trying to play catch-up with silly fad-technologies like these. Either innovate and come up with something that is “one-of-a-kind” (much like you did with Java) or improve the technology that you have now, making it that much better than the competition.
This is so disgustingly frustrating, it makes me want to get out of Java development. I can’t imagine what a Swing application will look like two years down the road. I’d say I’d switch to another technology, but there aren’t any others that come close to the power of Java that don’t already have this kind of crap built into it.
Maybe the OpenJDK will fork and keep this junk out of it? …. one can only hope, I suppose…….
Posted 11 Nov 2008 at 6:21 pm ¶
simon wrote:
Caleb,
First thing to say is: there’s nothing to stop you using Swing, if that’s what you want to do. It’s not going away!
To answer some of your points… as I understand it, there are a few drivers for JavaFX:
1) To enable developers to be more productive in developing compelling graphical user interfaces.
2) To allow project teams to increase the quality of their work by enabling interaction designers (experts in human-computer interaction, but not experts in programming) to design the user interfaces, while developers work on other parts system functionality.
3) To allow project teams to increase the quality of their work by enabling collaborations with graphic designers that can easily create high-quality digital assets for productive inclusion into user interfaces
4) Make the power of Swing easier to access for more developers. Swing is really an incredibly powerful GUI toolkit. However, few developers take advantage of most of that power; and those that do have to spend a long time on their GUI coding to access that power.
5) Extend the cross-platform capability of compelling Java applications from running on multiple desktop hardware/OS platforms (Windows, Mac, Linux, Solaris etc), to *also* running on large numbers of mobile phones, and televisions. That’s because JavaFX applications can be written to run not only on Java SE, but also on top of Java ME.
6) Taken together with enhancements to the Java platform (Java 6 Update 10), to enable media-rich (particularly audio and video) applications to be developed.
I’m don’t know why you’re so angry about JavaFX. It all seems pretty useful and worthwhile to me. I’m really looking forward to the 1.0 release.
Posted 11 Nov 2008 at 7:12 pm ¶
Caleb Payne wrote:
In response to your points:
1) What do you mean by more “compelling” graphical user interfaces? Do you mean rounded border, transparent buttons, flashy frames, animations and the like? If so, take a look at what the majority of developers write. Would you see that type of interface in a bank? An insurance company? How about for an inventory management system? Productivity tools (such as OpenOffice and the like)? IDEs? Professional environments have a tendency to keep flashy stuff like that out of their interfaces by design – it takes away from what the application is supposed to do. A good UI comes down to layout and accessibility, not flashiness. (Correct me if I’m wrong, but that’s what I interpret FX to be – a new-agey, flashy, look at my neat-o UI tool for kids.)
2) That’s in the ideal situation where you have the manpower and discipline to organize your development team in such a way as to make sure that your interface developers only touch the interface, etc. I’ve not been in a situation where that’s been applicable in the past … let’s see … well, never, actually. Having to learn a new scripting language in order to be able to maintain someone else’s interface is going to add a level of complexity that isn’t needed and will be rejected. (That’s right -I’m calling it. I’m saying the Java community is going to, in large part, reject JavaFX. And when we do, please don’t pull a Microsoft and stick your head in the sand and pretend like everything is hunky-dorey and proceed to shove it down our throats.)
3) Again – I don’t see that as something practical. How many developers would ever actually utilize this? Yeah, it sounds cool – but it’s impractical. Very few people are going to be working on this type of interface. Hiring graphic designers to write UI code?! That just sounds nuts. The graphics designers that I’ve worked with would be the last people I’d talk to about building a UI. (Maybe the individual elements, graphics, etc. of the UI, but they’d never actually touch the UI.)
4) Swing is perfectly accessible to all java developers who have an interest. It’s also great the way it is; it means that those who are capable of programming Swing now don’t have to go learn some new scripting language. Just because it takes a long time to code doesn’t mean that you’re doing something wrong!! If you took that approach, then we should just abandon Java and go with .NET. (For example: What’s the code necessary in Java to load an XML document, select a node, perform some arbitrary manipulations on that node, and save the document again? It’s worlds easier in .NET. How about a web service? I can whip up a web service in .NET, have it compiled and deployed in less than 10 minutes.) Just because they’re *easier* to work with, doesn’t mean that they’re better. .NET lacks the power and level of interaction and control that Java provides in at least the two examples that I provided.
I simply don’t understand what you mean by “make the power of Swing easier to access for more developers.” That just sounds like a hollow, empty, Obama-esque sales-pitch that carries no meaning. Could you please elaborate on that a bit?
6) The one true place where JavaFX would be necessary. Now take a step back, look at the java applications that are in existence today and tell me, as a rough estimate, what percent of those applications fit into the category of development that you suggest here? I’m willing to bet that your estimate, like mine, falls short of 10%.
I’m angry because it seems like Sun has put so much time, money and other resources into a useless technology, when you could have spent your time improving what you already have. Funnel some of that money to Solaris. Funnel some of that money and those resources to Java. If you think that the Swing API makes it difficult to utilize, then retool the API. I just don’t see a need to come up with a whole new technology (that isn’t new at all – it’s just a mimicking of what others have already done).
I know I’m coming off as an angry, irrational a-hole right now and for that I apologize. I tend to get sarcastic and stinging when I feel passionately about something and feel like those who have some sway “just don’t get it.”
Posted 11 Nov 2008 at 7:47 pm ¶
simon wrote:
Caleb wrote, Professional environments have a tendency to keep flashy stuff like that out of their interfaces by design – it takes away from what the application is supposed to do. A good UI comes down to layout and accessibility, not flashiness.
With respect, you’re simply not correct about what makes a good user interface; and neither are you correct about the drivers behind software design decisions in companies that have major requirements for custom software (banking, insurance, etc.).
I’ll resist the temptation to go off-topic though, and try to answer your question! “What do I mean by more compelling user interfaces?” I mean something very specific – that the user interface is both easier for the end-user to learn, and easier for them to use once they have learned it. There are lots of subtleties involved in this, which I won’t go into. If you’re interested, you might want to read some books on cognetics as a starting point. Suffice to say, things like animations, colour changes etc. really can help.
2) Re: interaction designers. I’m not surprised you haven’t been in that kind of environment before. Few people have. However, that’s a direction that software development is taking now. So, enabling that kind of workflow is important. Why? Because, to quote you – “take a look at what the majority of developers write” in terms of UIs. The record shows that most developers have no idea how to develop a good user interfaces. To make progress with software quality, then, new roles such as interaction design experts are needed.
3) I didn’t say graphic designers should *code* the UI. I said JavaFX aims to enable graphic designers to *collaborate* effectively with software teams to create high-quality graphic assets that can be used in the project. That kind of thing is a benefit, btw, whether the roles of designer vs developer are physical, or logical. It’s all about making it easier to manage creation and iteration of graphics as the project progresses.
4) re: “make the power of Swing easier to access”. I mean that:
- You have to write far fewer lines of code in JavaFX compared to Swing to achieve the same result
- The code you do write is easier to understand, and therefore easier to maintain
6) Yes, the use of Java in these kind of apps today is minimal. However, lots of developers would like to use Java to code these kind of applications.
In terms of investing to “re-tool” the Java itself for building user interfaces… Well, Sun has done precisely that. If you like the idea of declaratively defined UIs, then you can always use the all-new Java APIs that are complementary to JavaFX script.
The truth is, if you care about Java on the client (desktop, mobile, TV), JavaFX is good news. It’s making Java stronger, not weaker.
BTW, I’m nothing to do with Sun, so there’s no need to keep saying, “you” when talking about them
Posted 11 Nov 2008 at 8:59 pm ¶
Caleb Payne wrote:
Regarding interaction designers … you got my point with that. As I said and as you reaffirmed, very few people work in that type of environment. Just because you think that it “should” be that way, doesn’t mean that that’s reality. The reality of the fact is that it simply is *not* that way and I don’t foresee that shift any time in the near future (wish as you may).
I’ll concede that I’ve not spent countless hours researching UI design. What I – a developer and a software user – consider to be “good UI design” is an application that has an intuitive interface. None of what *I* personally find attractive and/or intuitive involve animations, that’s for sure.
3) How is JavaFX any different than swing in that graphic designers can collaborate with Java developers??????? Graphic designers collaborate with Swing developers on a daily basis…… without problems. JavaFX – a scripting language – is certainly not going to alleviate any “collaborative issues” that may arise between a software/UI developer and a graphic designer. Much to your (and my, I suppose) dismay, JavaFX isn’t going to be the Messianic Saviour of Java, who will bring the Hope, Change and Prosperity that the Java Nation needs. (Sorry for yet another cheap shot at BHO.)
6) JavaFX is *not* java. It may run in a JVM, it may interact with java, but it is not Java.
I don’t think it’s strengthening java. It’s trying to pander to a handful of people in a small, niche market that does not appeal to the vast majority of java developers. I, on the other hand, believe that it *is* weakening java, because the time, resourses, money, effort that have been put into JavaFX could have been placed on more worth-while investments.
As for the “you” remarks, I apologize. I linked to this blog from James Gosling’s blog and wrote what I wrote in hopes that some from Sun would follow the link. These remarks have more been structured as rants directed at Sun (and not so much attacks on your opinions). If you’ve taken them personally, I meant no disrespect. I did a poor job of conveying where I was laying the blame. It’s not on you, it’s on Sun. When I say “you”, I’m referring to the prospective Sun employee who happens to stumble upon this blog
Posted 11 Nov 2008 at 10:11 pm ¶
simon wrote:
Caleb wrote, I’ll concede that I’ve not spent countless hours researching UI design.
Maybe, then, should you at least consider the possibility that there might be people that know more than you about: a) what makes a good user interface; and b) what the future trends are in terms of how graphical user interfaces will get built.
How is JavaFX different from Swing in terms of how graphic designers collaborate with developers?
Part of the JavaFX platform is software that enables designers to create JavaFX assets directly from within tools such as Illustrator and Photoshop. It’s called “Project Nile”. Those will be part of the 1.0 release; and there are other JavaFX design tools in development.
You’re mistaken to say that JavaFX is *weakening* Java. It is an objective fact that improvements have been made to the Java platform that were made because of Sun’s investment in JavaFX.
It is perfectly valid, though, to hold the opinion (which I think you do) that the investment in JavaFX represents a big opportunity cost to Sun and to the Java platform.
I don’t happen to share that opinion. Clearly, though, Sun has made a big bet on JavaFX. I think it’s is a good idea; and I’m hoping that it will be well-executed. However, until we see whether JavaFX is a hit with developers, the jury on that must remain out…
As of the 1.0 release, JavaFX is going to be, by definition, for early adopters. If early adopters find success with it, then I think the ideas are good enough that it has a great chance of gaining widespread adoption.
Posted 11 Nov 2008 at 10:57 pm ¶
Asam Bashir wrote:
How much could you buy out Sun for right now with shares at $4? Apple has about $29 Billion stashed in the bank, Steve Jobs could buy Sun when they hit $3, put it out of it’s misery. I’m sure Apple could handle Sun big iron sales pretty well and Snow Leopard and ZFS will achieve nirvana. Standardise Java to use Cocoa/Webkit, put a layer of Cocoa and Webkit on OpenSolaris and offer it for free for the sub-$500 laptop/netbook market that’s too cheap for Mac OS X. Sell version of Cocoa/Webkit/Java to Nokia or something…..
Posted 12 Nov 2008 at 1:33 am ¶
Mr X wrote:
I’m not sure what the market is for RIA – there are somethings that are better in the web – frankly I would hate to see the BBC news web site as some awful RIA flash/flex/silverlight/DHTML/Java app – how often have you found yourself choosing the ‘classic’ version of something? For text information browsing hypertext is the best solution – perhaps argumented here and there, but essential hypertext.
In my view most of the web content should stay hypertext as most is the written word.
Then there is the other-side – applications – like word processing or managing photo’s – people have tried an all web technology approach because of the easy of deployment[1] and universal access – work/home/mobile/cybercafe etc. Here the development problem is already solved – desktop toolkits abound – some of them good – talking over the network is solved – the issue as I have mentioned above is really one of deployment – ie the *ONLY* difference between something written with Cocoa and with javascript/flash etc is the deployment model – client server is nothing new and not exclusive to web apps – with varying amount of effort you can write similar apps.
Deployment, deployment, deployment is why people are writing apps in, lets face it, generally slower and less feature rich enviroments for the web. ( That and to some extent they are more modern, making some aspects of development easier than in a language with direct memory access).
Deployment is what Sun are hoping to crack with Java Update 10 and JavaFX – easier and everywhere. The VMware approach is certainly interesting for the enterprise – it’s a way of taking back the hardware from the OS vendors.
Personally I don’t have such a deployment issue right now – we are fortunate to have sorted this long ago – we need to get IT just to install one thing and we have complete freedom to push out apps – however there may come a time when even that might be hard.
Apple and Microsoft have also sorted deployment for it’s own software ( OS, iLife, iWork ), app store for the iPhone for consumers, and there is Windows update, office update etc. Enterprises can use Windows update – but are still very reluctant to change anything – but don’t care which web apps you load – the problem in large organisations is you can’t find anyone who is willing to take a decision or risk ( they don’t control the web space so can’t be blamed ).
I do wonder however how many apps you actually need to access everywhere – email and….sales automation….casual games…. photo sharing – how many more consumer RIA are their to write? What matters is the enterprise were the trend is increasingly web because of the issues in[1] – the customers can bypass IT entirely – it’s trivial for them to buy access to a website or service and not involve or even tell IT.
So to be successful any platform has to take the deployment and updating aspects away from corporate IT – they just need to be bootstrapped and left alone to get on with it – that’s what we’ve done with our in-house deployment model – that’s what the web brings – that’s what browser plugins are on the edge of.
[1] I happen to be talking to people from really large companies at the moment as part of an interview process and believe me deploying something on the desktop in a number of large enterprises is nigh on impossible – people are pushed to the web and even outside suppliers by the inflexibility of their IT departments – it’s funny – back to the old IBM days of total lockdown – why even have a desktop PC if it’s as locked down as a thin client?
Posted 12 Nov 2008 at 12:12 pm ¶
simon wrote:
You make some great points about the issues surrounding deployment, particularly in terms of corporate IT. I think the trends are actually even more dramatic than those you’re painting.
Not only do regular people by-pass their corporate IT departments by going direct to services providers that can give them IT solutions via the browser; people are also starting to by-pass the whole corporate IT environment altogether.
I’m seeing people rejecting company desktops, laptops and phones, in favor of using hardware that they buy themselves with their own money; just because corporate IT doesn’t meet their needs. Don’t like corporate e-mail? Simple – forward your mail to your own e-mail provider. Don’t like the ban on installing apps? Simple – install the apps you want on your own desktop/laptop, and put the company data on that. Don’t like the fact your company won’t let you have an iPhone, or a Mac? Simple – buy them yourself, and use them for work. Don’t like the fact that your company bans you from using Facebook at work? Simple – plug in your own mobile broadband system, and access whatever web-sites you want.
So, the type of people that are the key IT decision-makers in companies is changing. It used to be IT professionals that made all the calls. Increasingly, it’s just regular employees that are effectively calling the shots.
Increasingly, to attract key IT decision-makers, the challenges are becoming equivalent to those involved in attracting users to consumer applications.
To succeed as a computing technology in this new era, JavaFX really needs to enable developers to build apps that will appeal to consumers – for both personal and business use.
As to the market size for RIA, I think asking “How many more RIAs do we need?” is a bit like asking, “How much new software do we need?” or “How many new web-sites do we need?”
The answer is – a huge amount more…
Posted 12 Nov 2008 at 3:01 pm ¶
Andrew John Hughes wrote:
Does this release include the source code under a FOSS license?
Posted 19 Nov 2008 at 7:22 pm ¶