Innovation again (was Re: Text classes)
neugens.limasoftware at gmail.com
Wed Dec 6 08:11:14 UTC 2017
I think Phil said that, the way to propose such changes is to file a Jep
and discuss it here.
On Wed 6. Dec 2017 at 09:07, Markus KARG <markus at headcrashing.eu> wrote:
> I think what John actually asked for is whom to send his design upfront at
> the JFX team to get an initial judgement whether it is worth programming
> it, or whether it bears such flaws that it makes not much sense to invest
> any more time. Whether or not that decision is done by an Oracle employee
> or not, he simply needs to know whom to sent his proposal for early review.
> -----Original Message-----
> From: openjfx-dev [mailto:openjfx-dev-bounces at openjdk.java.net] On Behalf
> Of Philip Race
> Sent: Mittwoch, 6. Dezember 2017 06:50
> To: John-Val Rose
> Cc: openjfx-dev at openjdk.java.net
> Subject: Re: Innovation again (was Re: Text classes)
> There needs to be a viable community that is not just Oracle to support
> you here ..
> I think everyone has come to be dependent on Oracle to "be there".
> But if there is a specific community need that Oracle doesn't see as
> essential, then the community should help out.
> On 12/5/17, 9:27 PM, John-Val Rose wrote:
> > Well, that’s all fine but you didn’t address the issue of working with
> someone within Oracle to get these innovations done.
> > Sure, I could just toil away by myself but clearly it would be better
> all around if there was someone with much more extensive knowledge of
> JavaFX and its internals who was accessible when required.
> > I would assume that a member of the Oracle JavaFX team would be such a
> person. If not, then who?
> >> On 6 Dec 2017, at 15:53, Philip Race<philip.race at oracle.com> wrote:
> >> I think looking at it as an Oracle-owned and controlled project maybe
> the first mistake here.
> >> Yes it was closed source and then Oracle controlled, but not any more,
> OCA requirements aside.
> >> It is not even a "java specification". It can be evolved at an API
> level without a JSR.
> >> The JEP process is the main thing to be followed, although we also use
> CSRs too to track API.
> >> Consider it that anyone who is a contributor owns (not the right word
> ?) a piece of it too.
> >> So standing on the project is what matters. Not the company who pays
> you to work on it.
> >> -phil.
> >>> On 12/5/17, 8:21 PM, John-Val Rose wrote:
> >>> Phil et. al.,
> >>> Whilst I’m not going to be quite as “passionate” as some on this issue
> (although I do understand the frustration), I would like to point out again
> that this is indeed a huge gap and it is critical that it is filled ASAP.
> >>> Obviously a solution where every word in a text document is a Node
> would be unworkable so it would need to be architected from the ground up.
> >>> I would be happy to work on such as feature, just as I was happy to
> work on implementing WebGL, but my hesitation is concern over the
> assistance and involvement from Oracle.
> >>> If I am going to have to spend months working on something without any
> or only minimal involvement from Oracle, only to find at the end that
> Oracle either doesn’t like the design, implementation or something else
> then it is wasted time I’ll never get back.
> >>> There are lots of other innovations too that I would like to see in
> JavaFX but I just don’t “feel the enthusiasm” from Oracle.
> >>> If there is someone on the JavaFX team who would be willing to work
> with me (at least in some capacity), please have them contact me privately
> via email.
> >>> The innovations I could work on and contribute include:
> >>> 1. WebGL support in WebView
> >>> 2. Better text support including text documents& rich text editors
> >>> 3. Significant improvements in scene graph rendering speed using
> >>> modern game-engine style structures and algorithms
> >>> JavaFX cannot survive without innovation and I am keen to see it
> happen and contribute as much as possible.
> >>> Graciously,
> >>> John-Val Rose
> >>> Rosethorn Technology
> >>>> On 6 Dec 2017, at 11:36, javafx at use.startmail.com wrote:
> >>>> Sorry about all the typos previously.
> >>>> Question- why not use the code in awt ? I am not totally up on what's
> going on with the platforms' native rendering engines ( meaning, I have no
> idea whatsoever) or how they have changed, but golly it sure does still
> work pretty well.
> >>>> At least it seems to me looking at awt that a smallish number of
> things are 1) well defined by the native platofrm and 2) would more or less
> translate directly to an Java API and 3) from those small number of
> building blocks, (Font and Glyph metrics and this kind of thing) text
> line layout algorithms can be written by ordinary civilians along with all
> the other stuff that goes into a text editor.
> >>>> And yes, everything does look easy when someone else is going to do
More information about the openjfx-dev