JavaFX 1.3 soon?

[zammbi]

By looking at http://javafx-jira.kenai.com/browse/RT/fixforversion/10237?selected=com.atlassian.jira.plugin.system.project:version-summary-panel

479/500 issues done. Does that mean we will soon be seeing JavaFX 1.3? is there a set release date?

[kappa]

JavaFX in its current state is horrid and really shouldn't have been released until it was done.

I'm really hoping they get this version right especially as its been in the works for a while. If they just nail two things

1) the performance
2) the user experience

Then it might be cool.

[princec]

A shame the JRE team are so intent on breaking the user experience at the moment.

Cas

Games published by our own members! Check 'em out!

[trembovetski]

Come on, you know it's not true. The JRE team is working within the restrictions of the current framework.

Are you saying you didn't see any improvement in deployment (I assume this is what you're alluding to) in the last few 6 update releases? Or is this just a knee-jerk reaction anytime you hear "JavaFX"?

[Orangy Tang]

Are you implying that the recent changes in webstart deployment provide an improved user experience? 

[princec]

JRE6.0u19 broke everyone's applets. In a point release. Sack whoever it was thought that was a good idea and then quietly patch it back so that it doesn't do that any more. Save that sort of change for 7.0 eh?

Cas

[Riven]

Backwards-compatibility is often an argument not to fix obscure Swing bugs.

Recently (both in 6u14 and 6u19), the JRE team has showed they don't give a damn about backwards-compatibility of basic functionality.

[princec]

Reminds me, I should have moaned about that byte order change in images in u17.

Cas

[CommanderKeith]

But all that changed was really just the timing of when the security cert was shown. Not really a big deal. The technique of delaying the security cert was just a dodgy workaround for the real problem which is that java security is not fine-grained enough - it's all or nothing (like windows vista security which MS have fixed in windows 7 and Sun should fix in java 7!)

Sun probably only just realised the existence of the ingenious technique of delaying the security cert dreamed up by DzzD and thijs. Doesn't seem like a feature Sun planned to have in the API.

I hope I haven't fanned the flames lol 

[Riven]

But all that changed was really just the timing of when the security cert was shown.

Not really. Class.forName(String) from one JAR to another stopped working. You have to modify the manifest to make it work again. There are zillions of applets / webstart apps broken now.

Games published by our own members! Check 'em out!

[CommanderKeith]

Gee, didn't know about that one. I wonder why that didn't come through on Sun's pre-release tests

[kappa]

yup Class.forName(String) was the reason LWJGL applets broke too. As a signed jar used that to access resources in an unsigned jar. Really odd thing to change.

[Riven]

It was allegedly a massive security hole. Somehow. Maybe I should read up on that...

[bobjob]

I dont understand why java is trying to self distruct the end user experience. is it to impress the server side users, and up the security rating of java? seriously alot of bad choices.

On a side note, flash is accessing resources like webcam with no problems. Yet QuickTime for java is now depricated, and no decent bindings have yet been developed for Mac's QTKit (even though it should be released as part of the JRE).

edit: to think if it, JMF was just to ahead of its time, before people had the internet speeds, or enough webcams to go around. And even then, there was no Mac binding.

[Mr_Light]

JavaFX in its current state is horrid and really shouldn't have been released until it was done.

I'm really hoping they get this version right especially as its been in the works for a while. If they just nail two things

1) the performance
2) the user experience

Then it might be cool.

Given you keep the list of things-to-not-use(like in the early(er) days of java) performace hasn't been bad to me.

Sure developing using netbeans misses polish to put it gently, the bit of code that handles code formatting was it programmed by a first year student or something? I mean at the end of the day you might get used to just about anything..

As far as user experience, so stuff broke... can't care that much in the light of the very much improved startup time. I mean I've visited pages and I actually had to look at the source to figure out that there was actually an applet on the page. As sad as it is, it still kinda surprises me.

No all in all Javafx has been pleasant

That all being said it does seem that there no drive from management to actually polish the user experience or developers with that as a primary role/responsibility, just some developers doing the good work in some spare time. - if I'm off base here please tell me, otherwise I'm going to presume that I'm right.

[jojoh]

Not really. Class.forName(String) from one JAR to another stopped working. You have to modify the manifest to make it work again. There are zillions of applets / webstart apps broken now.

WOW!!! 
That is truly unusually amazingly idiotic 
I use a framework for all my games, and want to use the very same jar for all the different game applets. I guess that is no longer possible, and already broken everywhere! I even get a version of this where one applet simply doesn't start, no threatening pop-up, and nothing in the console!!!  Just tested it from a work laptop, that had some MS security patch update, so I have to test with other HW to find out what the culprit is...

Do you guys have links on how to fix this? (just add "lazy" jars to boot jar manifest?)
Is there an issue to vote on, to fix this?
Who do I send angry mails to about this?

Sorry for derail...

[kappa]

WOW!!! 
That is truly unusually amazingly idiotic 

..

Do you guys have links on how to fix this? (just add "lazy" jars to boot jar manifest?)

apart from breaking tons of existing applets and making some things now impossible (even for signed jars), its not particularly hard to fix, just add a

Trusted-Library: true

or a

Trusted-Only: true

to the manifest file of your jars

see following page for full explanation
http://java.sun.com/javase/6/docs/technotes/guides/jweb/mixed_code.html

[kappa]

It was allegedly a massive security hole. Somehow. Maybe I should read up on that...

might be related to http://www.theregister.co.uk/2010/04/09/critical_java_vulnerability/
probably why a fix was rushed out.

[Riven]

might be related to http://www.theregister.co.uk/2010/04/09/critical_java_vulnerability/
probably why a fix was rushed out.

I love that article. Typical that Sun/Oracle responded with "we don't give a damn" and then wrote some dirty patch that screwed over everybody.

"I uninstalled Java more than a year ago and haven't had a single problem with any website," he wrote. "Why are people still running Java in the browser?"

[kappa]

ah lovely, its hit slashdot too now

http://developers.slashdot.org/story/10/04/09/2017232/Serious-New-Java-Flaw-Affects-All-Browsers



however some more details on the actual bug in the comments and when it was fixed.

[bobjob]

http://javafx-jira.kenai.com/browse/RT/fixforversion/10237?selected=com.atlassian.jira.plugin.system.project:version-summary-panel

Tavis Ormandy, a security researcher who prefers his employer not be named

with a  little googling...
aparantly he works for google

Of course, there's another mitigation that was tweeted Friday by security researcher Alex Sotirov that's looking more and more viable.
"I uninstalled Java more than a year ago and haven't had a single problem with any website," he wrote. "Why are people still running Java in the browser?"
Good question. ®

sigh

Does Java ever get any GOOD press?

[bienator]

Does Java ever get any GOOD press?

(nothing gets good press on The Register.)

[princec]

Indeed not They are famously cynical. But in this case they really can't be blamed. It's been rubbish for years, too little too late, and now this. Grrr. Still, Oracle may have grandiose plans to compete with Flash and Silverlight so perhaps they'll throw money at it and we'll see Java 3.0 rise like a phoenix from the ashes etc etc.

Cas

[kappa]

and it gets better

http://www.theregister.co.uk/2010/04/14/critical_java_vulnerability_exploited/

[TheAnalogKid]

1.3 is now out. http://blogs.sun.com/javafx/entry/javafx_1_3_released_delivers

[kappa]

cool, nice to see the massive focus on speed.

[Riven]

I see JavaFX content 7 seconds after the Java Console popups up. (which popped up 2 seconds after the page was loaded) (www.javafx.com)

IIRC, it's factor 10 better than it used to be. But we seriously need yet another factor 10 improvement to be on par with Flash.

[Eli Delventhal]

How is it overall?

[kappa]

yes still pretty slow compared to the competition (at least in perception). But pretty big improvement on 1.2 which was just unusabely slow. Feels a lot better now but again could do with more speed, hope we don't have to wait another year to get from 1.3 to 1.4.

[trembovetski]

yes still pretty slow compared to the competition (at least in perception). But pretty big improvement on 1.2 which was just unusabely slow. Feels a lot better now but again could do with more speed, hope we don't have to wait another year to get from 1.3 to 1.4.

You're talking about cold start, right? Warm start (at least on my box) is very fast. (not that I disagree that cold start needs a lot of work. Hopefully Prism and jdk7's modules will help with that).