Moving on (forked from Re: JavaOne roundup?)

Daniel Zwolenski zonski at
Mon Sep 30 01:39:01 PDT 2013

The lack of information on iOS/Android is a major bummer, but this also
highlights a deeper problem here.

We have a situation where Oracle won't talk to this community because the
topic is important, it's too big a game changer for them to comment on.
It's tied in with share prices, and market strategies.

So won't that be the case for anything *important* going forward? We
"community" members are outsiders and very lowly ranked, well below "real"
customers and even below random punters from the media. There's not even a
way for us to rank bugs and get them attention (even if we provide fixes!).

What kind of community can this ever be if anything important can't be
discussed here before it's locked in, because it risks Oracle giving up a
commercial edge? Is this then a community only for discussing our favourite
method names for the API and pointing out that an enum constant is missing?

I can't see any way that this forum provides any significant contributions
back to the platform - the occasional bug fix at best. JIRA is fine for
discussing bugs, method names and little things like that. Any of the real
community initiatives are run completely separate to this forum because
Oracle doesn't want anything to do with them, and all the significant
platform work takes place behind Oracle's closed doors and we only hear
about it after it's a done deal.

>From where I'm standing, the Oracle community concept is fundamentally
flawed, and the root cause is that Oracle just don't get how to interact
with a community. You want to use us but you're not very good at it, you're
not trying to improve (you don't think there's a problem) and ultimately
Oracle's culture won't let you do it properly anyway. The current approach
is a little like a car salesman trying to be your Facebook "friend".

All the initiatives I got involved with through this forum have gone
nowhere - deployment (auto updating), the early Maven deployment work
(which Richard asked for), the tower defender game (which Richard asked
for), the jfx browser (which Richard asked for), even stuff as simple as
JIRA dashboards (which Richard again asked for).

All these hit points where they needed Oracle to do their part of it and
then just stalled and then died. This community could have fostered a lot
of tools and efforts, and really propelled JFX into the bigger dev
community, but instead, for me, it has been a constant source of stress and
dissatisfaction, a hinderance and a hurdle. All pain, no gain.

The only initiatives I actually made work were the JavaFX Maven plugin and
the RoboVM Maven plugin. With both of these I made a conscious decision to
not involve this forum or Oracle. I decided to cludge around platform
shortcomings, rather than work with Oracle to fix it (5 minute fixes would
have saved me days of work).

That was the only way I could make these initiatives succeed since this
forum is a hinderance to contributing. It gives a false sense that Oracle
is listening and actively supporting the community. To anyone out there
wanting to do something in JFX tool space, I'd say start by leaving this
forum and working out what you can do without any access to the Oracle
guys, even if you make your own code contributions to the platform. Assume
you're an outsider - the cavalry is not coming, you're on your own.

Given all that I'm walking away from this forum. I was waiting to hear
about the iOS/Android stuff first, but really even if they did announce
anything, it would be a long shot at best (untested, low resources, lack of
solid direction and most likely tied in with some Oracle ADF garbage or
similar). The uncertainty created by Oracle's mixed messages also killed
all momentum on the community RoboVM work. Meanwhile web based stuff is
getting stronger, cleaner and better tool support at an exponential rate,
including in the mobile space.

If JavaFX one day actually provides a usable platform for non-Oracle
entrenched customers, and the developer world notices, I'll certainly
consider it. I reckon I'll hear about that through the usual tech media
channels first, rather than through here though. As Oracle themselves
pointed out at the 2012 JavaOne session the smart money is on web based
stuff (check out backbone.js and marionette.js for a desktop-like coding
experience, not bad and will get better faster than JFX improves).

On that note, the JavaFX Maven plugin is about to go into decay mode. It
needs to be updated to work on Maven 3.1 (some libraries have changed from
3.0) and there are a number of bugs and feature requests building up that
I've been ignoring. I have no incentive to do any of this so it will
unfortunately just rot. If anyone wants to pick it up, let me know (you
need a few free hours a week just to maintain it). I'm picking up stumps
and moving on.

I also have the access rights for the openjfx Maven repo on Sonatype
(needed to deploy to Maven central). I imagine Sonatype would grant this
access to others if you apply and make a case for it, but if anyone wants
to do this let me know and I can notify sonatype to give you access and
save you some hassles.

I think Niklas has the RoboVM Maven Plugin sorted now and can do
enhancements on that but I'm sure if anyone wanted to help him out he
wouldn't say no.


On Mon, Sep 30, 2013 at 5:04 PM, Tobias Bley <tobi at> wrote:

> I don’t understand why „all“ this people who needs JavaFX on iOS/Android
> does not tell it Oracles management. And I don’t understand why all this
> people use their time to develop all this demos and Rasp.PI stuff. Who
> needs it? Why don’t we develop base stuff like iOS skins, Android skins,
> iOS/Android widgets, RoboVM for Android, RoboVM using OpenJDK, … I really
> love useful stuff like the „JavaFX maven plugin“ or the „AquaFX“ project.
> That kind of development we need!
> Best,
> Tobi
> Am 30.09.2013 um 08:50 schrieb Felix Bembrick <felix.bembrick at>:
> > No, you are *not* the only one. We *all* need it.  In fact, without it
>  happening soon, JavaFX is already dead.
> >
> > But let's not give up yet.  Perhaps it's closer than we know. I am a
> glass half full kinda guy :-)
> >
> >> On 30 Sep 2013, at 16:40, Tobias Bley <tobi at> wrote:
> >>
> >> I suppose „legal reasons“….
> >>
> >> For me it’s very frustrating to see every year the same procedure:
> JavaFX-iOS/Android related tracks were canceled - „nerd“ stuff like
> Rasp.PI, DukePad & Co were announced. Maybe I’m really the only one who
> needs JavaFX on mobile to use JavaFX on desktop as well… :(
> >>
> >>
> >>
> >>> Am 29.09.2013 um 18:13 schrieb Jeff Martin <jeff at>:
> >>>
> >>> It seems the JFX on iOS/Android were cancelled at the last moment. I
> tried to keep expectations low this year, but I admit I harbored secret
> hopes based on those sessions (a few embarrassingly optimistic
> conversations with clients notwithstanding).
> >>>
> >>> Last week Tomas offered this:
> >>>
> >>>> about cancelled sessions please contact Mr. JavaOne
> stephen.chin at I believe he will give satisfactory answer.
> >>>
> >>> I'd like to take him up on that satisfactory offer. Also, can we run
> the name "DukePad" by marketing again?
> >>>
> >>> :-)
> >>>
> >>> jeff
> >>>
> >>>
> >>>> On Sep 29, 2013, at 12:12 AM, Daniel Zwolenski <zonski at>
> wrote:
> >>>>
> >>>> The sessions aren't up yet from the looks of it. It would be great to
> get an overall roundup of any new announcements or directions in any case.
> Given this is the developer community network it would make sense in my
> mind to highlight stuff like that in here.
> >>>>
> >>>> For me, I'd love it if someone could quickly sum up any announcements
> or sessions made about JavaFX for iOS, Android or in the deployment space?
> >>>>
> >>>> What happened at the sessions Tobi highlighted before (
>, did anyone go to these and able to
> give us some info?
> >>>>
> >>>>
> >>>>
> >>>>> On 27/09/2013, at 7:07 AM, Richard Bair <richard.bair at>
> wrote:
> >>>>>
> >>>>> The sessions, I think, are all being uploaded to Parley's (
>, although I don't see any content there yet (not
> sure how long it will take them to post-process, but usually it is pretty
> fast).
> >>>>>
> >>>>> Richard
> >>>>>
> >>>>>> On Sep 26, 2013, at 2:00 PM, Daniel Zwolenski <zonski at>
> wrote:
> >>>>>>
> >>>>>> Has anyone done or seen any good roundups (text or video) of the
> JavaOne sessions relating to javafx?
> >>

More information about the openjfx-dev mailing list