Josh, Some random thoughts. Please note that I've not played with javafx at all; this is entirely from the perspective of someone evaluating the technical state of play at the most superficial level:
(a) the demos being semi-broken on OS X (and even on windows, sporadically, and with no reason I could fathom) means I never take the next step. (b) the language isn't *bad*, but it's definitely not keeping up in terms of concision and expressivity. I'd say it's within spitting distance of javascript with regard to brevity, perhaps with some additional cruft around the access controls and imports (and maybe other bits, but like I said, I've not programmed in it yet). But jeez, take a look around at the other *mature* languages on the jvm -- in comparison, javafxscript is decidedly painful, and just close enough to java that one would likely start bleeding into the other in one's head when working in both worlds. Obviously, there's a lot of nice primitives for binding, effects, etc., but it's surrounded by so much boilerplate. Having just hosted Ben Fry last night talk about Processing (http://blog.snowtide.com/2009/04/28/snowtide-informatics- welcomes-ben-fry-of-processing-fame-to-northampton), I'm reminded of how little progress javafxscript represents in terms of language design. Obviously, the use-cases and specific functionality are different, but I think my point stands. However, this may not matter if the next point were resolved... (c) the core of javafx is apparently (last I saw) unusable within swing, and from languages other than javafxscript. I've read and heard that this will change, but until it does, and does so comprehensively (e.g. using well-supported mechanisms, points of integration, etc), we can't use javafx. With apologies to those working in media and entertainment, we create UIs to do "real work", which means varying amounts of (mostly) typical swing componentry, with some key views doing "serious" visualization. javafx would be nice to use for the latter, but it doesn't appear to fit within a "normal" swing app, and it can't do *everything*, because it doesn't have a complete component model. The set of examples that are out there seem to imply that there's an assumption that javafx will be used only for completely new projects, and that it is meant to own those apps' entire UI. That will never work in our context. We use clojure for 95% of all new code, so you can imagine that we're loathe to start slinging around javafxscript just to get access to a scenegraph and some simple binding primitives. I suspect scala and jruby and jython and groovy folks would have the same perspective. I hope the above is concrete enough to be helpful. To move into the 'intangibles' category, I simply don't grok what Sun's aim is for javafx in general, which is a general concern for me. Since javafx isn't a new stack (e.g. SWT -- javafx is built on top of AWT and Swing), I'm completely puzzled as to why there wasn't a completely bulletproof swing integration story from day one. You've got tons of swing apps out there and a new applet deployment story that *sounds* great (remember, we're on OS X); why not make all those swing devs out there drool about what cool things they could bolt into their existing apps tomorrow (instead of showing us toy game examples and yet another twitter client)? It seems like doing so wouldn't impact the other priorities of mobile and media/entertainment one bit (aside from resource constraints). Given this, I have to presume that the intent is to support mobile and media first and foremost (which I am decidedly not involved in). That'd be fine, if it weren't for the fact that (at least AFAIK) 90% of Sun's "installed base" are in the business app "market". A possible corollary would be if Flex came out of the gate with a great component model for business app development, but *zero* support for media and entertainment. Cheers, - Chas On May 6, 1:40 pm, Joshua Marinacci <[email protected]> wrote: > I would really like to hear back from Ed. He is not the first person > to have a bad reaction to JavaFX initially. There seems to be some > sort of a mental hurdle that a lot of people face, including myself 2 > years ago. Once they play with it enough they suddenly 'get it' and > become happy and very productive, even with the earlier buggy > releases. I've heard from many smart people going through this, so > clearly there is more going on here than just "It's different, so I > don't like it". Perhaps we are presenting the language improperly, or > we are demonstrating the wrong features first. I'm not sure what's > going on but I'd like to dig down into it and find out. > > - Josh > > On May 6, 2009, at 7:45 AM, Reinier Zwitserloot wrote: > > > > > Summary of OP: > > > I looked at it, it sucked. I won't tell you why; instead I'll just > > whinge, because I'm a non-contributing zero. > > > Ed, post some constructive criticism, or go away. > > > NB: Michael, Gladwell is a gifted author, a real yarnspinner, but you > > shouldn't quote him with the presumption that his delusional ranting > > has any basis in fact, at least, not in polite company. However, in > > the vein of completing the argument in a proposed faulty logic frame > > being just as effective as proving the logic is false in the first > > place: People who have actually used JavaFX almost never complain. If > > you follow the posse, or read anything about java at all, you'd know > > that the update to get is 6u10, which you didn't have. No wonder stuff > > isn't working quite as well as it should; the fact that it does work > > in the first place is a small miracle. > > > Joshua, do you know when apple will roll out something with the > > flavour of 6u10 across all macs? My mac is still on 1.6.0_07-b06-153. > > Could be because I've been downloading releases from > > developer.apple.com. > > > On May 6, 3:13 am, Ed <[email protected]> wrote: > >> Why is JavaFX such an extreme departure from Java? At least Flex has > >> ActionScript and MXML--something that anyone reading this list can > >> figure out without having to look up and strain to grok. > > >> I am trying, really trying, to get into JavaFX but I just cannot > >> tolerate it's ugliness. JavaFX has to be the single most unintuitive > >> language to come out since COBOL. I mean really...can you be > >> serious? > > >> I know that on one of the recent podcasts the posse was in agreement > >> that JavaFX was going to be the future of desktop Java but I > >> respectfully disagree. There is just no possible way any sizable > >> group of critical mass will ever adopt JavaFX. As with any > >> technology, there will be 'pockets of users' but the whole reason we > >> came over to Java from C++ was for the elegance and safety of Java. > >> The write once redevelop everywhere fantasy has been painful for the > >> past 15 years; Java is just now coming of age where we can actually > >> write something once and get the rest for free. Why did Sun, now > >> Oracle, ever let JavaFX out of the lab? > > >> JavaFX will do more harm than good for Java; the most JavaFX will do > >> is make people consider Flex, and or Silverlight all that much more. > > >> What's so wrong with Swing anyway, why can't we just rev Swing and > >> Java3D? > > >> I can see Groovy (or substitute your favorite JVM language Scala, > >> Clojure...etc here) breaking out with an elegant/terse wrapper around > >> the Swing, Java2D, Java3D primitives long...long before JavaFX ever > >> gets past the demo experiment that it is. Oracle should bury JavaFX > >> as fast as it can. > > >> That said the JavaFX 'rendering engine' is awesome. Just awesome. > >> If > >> you haven't yet tried it you are missing something truly great. > >> Oracle should roll the JavaFX engine into a standard Java7 > >> library. I > >> think JavaFX is the right idea it just needs (come on guys) a > >> realistic scripting language behind it. > > >> I know I have been hard on JavaFX, I have I hopes for the future of > >> Java and I strongly believe Java needs something like JavaFX going > >> forward. Great job to those who worked on JavaFX--as a developer I > >> know how much work it must have taken--it was a necessary first step > >> in a much needed direction. > > >> Overall I give JavaFX a 'B-' > > >> - > >> ed --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/javaposse?hl=en -~----------~----~----~----~------~----~------~--~---
