If this is true, I am more than a little peeved. This effectively
bricks the many 8 core/16 GB MacPros my employer purchased, expecting
that they would be able to run our Java 6-based application. Now we
are stuck.
On the other hand, all of these intel Macs would make truly wonderful
Ubuntu boxes.
-Ron
I'm not sure if a boycot is the right move. We might end up with less
Java support. For me the mac has more pros than cons (even if the
lack of Java 6 is a big fat stinky con). Is there a way to protest
that doesn't suggest "yeah, we should all just be using Windows"?
/Casper
Here is the bad news:
intosh:~ steven$ java -version
java version "1.5.0_13"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_13-
b05-237)
Java HotSpot(TM) Client VM (build 1.5.0_13-119, mixed mode, sharing)
Macintosh:~ steven$
--
Macintosh:Versions steven$ pwd
/System/Library/Frameworks/JavaVM.framework/Versions
Macintosh:Versions steven$ ls -al
total 48
drwxr-xr-x 12 root wheel 408 26 Oct 12:27 .
drwxr-xr-x 11 root wheel 374 26 Oct 12:27 ..
lrwxr-xr-x 1 root wheel 5 26 Oct 12:27 1.3 -> 1.3.1
drwxr-xr-x 3 root wheel 102 29 Sep 16:53 1.3.1
lrwxr-xr-x 1 root wheel 5 26 Oct 12:27 1.4 -> 1.4.2
lrwxr-xr-x 1 root wheel 3 26 Oct 12:27 1.4.1 -> 1.4
drwxr-xr-x 8 root wheel 272 21 Oct 2006 1.4.2
lrwxr-xr-x 1 root wheel 5 26 Oct 12:27 1.5 -> 1.5.0
drwxr-xr-x 8 root wheel 272 21 Oct 2006 1.5.0
drwxr-xr-x 7 root wheel 238 26 Oct 12:27 A
lrwxr-xr-x 1 root wheel 1 26 Oct 12:27 Current -> A
lrwxr-xr-x 1 root wheel 3 26 Oct 12:27 CurrentJDK -> 1.5
On Oct 26, 1:08 pm, "Mark Derricutt" <m...@talios.com> wrote:
> I was thinking before that if Sun want JavaFX to be successful and
> replace/complete with Air/Flash and continue to have JavaFX require Java6
> then we'll *need* a Java6 for Mac (unless theres a way to get JavaFX working
> on 1.5?)
>
> mark
>
On Oct 25, 8:09 pm, Jess Holle <je...@ptc.com> wrote:
> Yet without solid Java support Apple is burning their bridges in the CS
> education arena -- both at school and for student use in the home.
>
> Educators either want to teach cross-platform languages (and Objective C
> really does /not/ count here) and tooling or just pander 100% to the
> Also, throw all the stones you want at Java, but I get the feeling that
> Apple is turning antagonistic towards any cross-platform development
> tooling that is not guaranteed to stay solidly tucked away in the "geek"
> zone. Apple seems threatened by the notion that one might ship a
> cross-platform Java application instead of investing the time to produce
> a platform-specific Cocoa app. [If so, I see their fears are misguided
> -- better a Java app then no app on the platform at all.]
Really? Do you think Apple is "turning antagonistic" to Flash and AIR
too? Those are user-facing cross-platform development tools and
platforms. And Apple happily ships the latest version of Flash with OS
X.
Ugly truth: the big difference between Flash and Java on the Mac is
that Java is largely a write-only language on the Mac: developers may
write Java apps on the Mac, but few end-users run a meaningful amount
of Java code on the Mac, because there just isn't a lot of user-facing
Java out there. Name any, ANY meaningful user-facing Java app that
requires Java SE 6. Is anyone other than developers going to care if
it's not installed with Leopard?
OTOH, Flash is ubiquitous, popular with software developers, content
developers, and end-users. Apple has to guarantee a good Flash
experience to end-users, and they're fortunate that Adobe develops and
sells the CS3 tools for Mac (presumably, Adobe makes enough money off
Mac users to keep doing so).
"Better a Java app than no app on the platform at all" sounded goo
back in the 90's when Java first hit the scene. A lot of us believed
it. But pretty much nobody developed end-user Java apps of any
consequence, and most of the enterprise Java developers happily wrote
Windows-only Java (you would not care to know how many times I have
seen hard-coded path strings in Java code that start with "C:\"). The
idea of cross-platform user-facing Java simply hasn't panned out, for
whatever reasons. It's in Apple's self interest to expect developers
to make real Cocoa apps, which will differentiate the platform, and as
the Mac market-share increases, they can expect more and more
developers to do just that, out of their own self-interest (not
wanting to throw away 10% or more of the potential user-base).
I keep getting on the side of this issue that pisses off my friends
and colleagues, but I really see the Java-on-Mac story as a "canary in
a coal mine", a *severe* warning about Java's viability. Seems like
most Java developers would prefer to kill the messenger and write
posts about what a dick Steve Jobs is. OK, maybe he is. But he's
really smart, and the company is astonishingly successful right now,
and if he and his team think de-emphasizing, demoting, or even igoring
Java will make them more money, not less, then it is really incumbent
upon us to understand that reasoning.
--Chris
Ron, it sounds like you built an application for Java 6, betting that
Leopard would have Java 6. Apple never promised that it would, so you
have only yourself to blame for losing that bet.
As FSJ might say: Java is open source, if you frigtards want it on the
Mac then port it yourselves! Bokay? Why the heck should we care?
Neil
One thing is clear to me: I am really NOT getting myself a Mac :)
-Roman
Let me try another way. In which ways will a normal Mac user typically
come by a desktop Java application? I'd argue it is not a large list:
Azureus, online banking... hmmm... now I'm actually starting to run
out of examples. To that effect, it's down right reasonable why Apple
would not want to concentrate on this high-hanging fruit, since Steve
Jobs probably isn't particular in love with (nor rely on) Java anyway.
It might not be what Java enthusiasts with a Mac want to hear, but
from a market perspective I think it's fairly logic considering how
busy Apple are on many other fronts these days.
/Casper
I don't think Jobs is going to lose much sleep over your threat :-)
As for Java 7, the changes may well be big but it remains to be seen
whether they will be "important".
Neil
-Roman
Surely it will be important to many, to be able to leverage better
abstractions and smarter technologies. While Java has trouble getting
its head out of its own rear interface, other platforms have long time
ago moved on to explore components, binding, universal querying,
inference, parallism, DSL friendliness etc.
If it is important to me as a developer to remain enthusiastic about
what I do, it's certainly just as important to my employer to be able
to go from problem to solution as fast as possible.
/Casper
Anyway, this is a discussion for another thread. As I said in reply to
Roman, if Java support on Mac is important to Sun, then Sun needs to
do something about it. Or if it's important to the open source Java
community, then WE need to do something about it. It's clear that Java
is not a priority for Apple, and Apple are therefore, quite rightly,
not devoting much resource to it.
Neil
-Roman
[For my part I think there are some very notable and important
advancements in Java 6 and I don't see a critical need for every
possible language do-dad in Java for it to be remain compelling.]
> As I said in reply to
> Roman, if Java support on Mac is important to Sun, then Sun needs to
> do something about it.
Agreed.
> Or if it's important to the open source Java
> community, then WE need to do something about it. It's clear that Java
> is not a priority for Apple, and Apple are therefore, quite rightly,
> not devoting much resource to it.
>
Agreed to a point. If Apple is going to leave customers who purchased
Macs under the understanding that "Java on the Mac will be great" as
Jobs had said a year or two back (and that Java on the Mac would be
maintained/updated at least) high and dry the least they could do is
turn what they've got over to the open source community to develop from
there. If they can't do so legally in general at least they could turn
over hard bits they own (e.g. Mac-specific graphics pipeline code, etc)
for others to glue on to an OpenJDK port.
--
Jess Holle
On Oct 26, 3:17 am, Ron Bowers <ron1...@gmail.com> wrote:
> On the other hand, all of these intel Macs would make truly wonderful
> Ubuntu boxes.
Paying the Apple tax on Leopard: 129 EUR
Getting Ubuntu from https://shipit.ubuntu.com/ : 0 EUR
Being able to apt-get install sun-java6-jdk in Ubuntu for more than a
year and half: Priceless.
:)
cheers,
dalibor topic
- vineet
On Oct 27, 3:14 am, vineetb <vineet.vine...@gmail.com> wrote:
> They did have Java 6 sessions at WWDC. I guess Java 6 development is
> lagging behind and wasn't ready for GM?
> In any case, leave feedback here -http://www.apple.com/feedback/macosx.html/
On Oct 27, 4:23 am, "Mark Derricutt" <m...@talios.com> wrote:
> Leopard won't even recognise my partition :( So I can't even -start- the
> installer :(
>
Considering that they had a nearly-finished working version, based on
Java 6 b88, a year ago, it can't possibly be a technical issue. So
why the silence? The conspiracy theorists and whiny bloggers opt for
the easy "Steve Jobs hates Java", but that's stupid -- if that were
the case, the Java team would have been fired or reassigned long ago
and wouldn't be available to answer questions on the java-dev mailing
list.
There's another possible explanation for their silence. I wonder if
anyone else will figure it out.
--Chris
Alexey
2001 Honda CBR600F4i (CCS)
1992 Kawasaki EX500
http://azinger.blogspot.com
http://bsheet.sourceforge.net
http://wcollage.sourceforge.net
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Don't know many (outside Sun anyway) who'd have an idea for an end-
user desktop application and do it voluntarily in Java. It's certainly
the last thing I'd do... too much trouble moving ahead from idea to
satisfying solution. The sheer lack of these kind of applications
today would certainly suggest that sentiment to be shared among others
than I. Contrast that to i.e. Paint.NET, a full fledged image
manipulation suite for .NET (also running on Linux).
Do you think it'll get there? It's been 15 years with a highly
wobbling chartered course by Sun. Current trend is of course JavaFX
and the focus on tooling, the former is a lost battle (way too late
compared to competitors) and the latter is nice albeit also somewhat
late (very little in NetBeans hasn't been in Visual Studio for years).
It seems to me the desire to put a little Java everywhere, had the
effect of Java not really being there fully anywhere. Perhaps if Sun
had been a little better at picking their battles, then Java would be
a viable end-user desktop application platform and Apple would have a
higher interest in supporting it.
/Casper
But you're right.. one person acting alone can't make much
difference. But there are lots and lots of geeks using Macs and I
think a huge number of them are doing so because of Java. I know for
myself that was a large part of choosing to switch to Mac OS X with
10.0. I saw "Unix with a pretty face" and "Java" and was sold.
I think, though, that if Java were to have the same client side
significance as Flash then it would be a no-brainer for Apple to go
out of their way to make sure Java on Mac OS X was up-to-date.
Maybe JavaFX will change this (over time).
- David Herron
At home though I use Mac OS X.. despite having previously made fun of
Mac OS (cooperative multitasking? what a joke!) ... after 6 yrs using
this thing, Apple does a great wonderful job at making systems that
fit together extremely well. Ubuntu isn't quite there.. yeah, Ubuntu
is a great implementation of Linux and they've paid a lot of attention
to the same sort of usability questions which Apple is famous for.
But Ubuntu is hampered by the usability problems which hinder Linux/
GNOME overall.
At the moment I'm questioning whether I want to make further
investments with Apple. Maybe they are doing as they did in 10.4,
where the release of an up-to-date Java happened with 10.4.2 rather
than 10.4.0. It's a pain that they aren't doing everything that I
want them to do.
The suggestion to leave feedback with Apple ... http://www.apple.com/feedback/macosx.html/
... is a good one. I just did. As a JDK quality team member.. ah..
The suggestion elsewhere in this thread for Sun to take care of Java
on OS X is.. well.. adding another platform amplifies our cost in ways
you guys might not appreciate. Where would the money come from to do
this?
The suggestion to port the OpenJDK to Mac OS X is interesting and
ought to be doable. The way we at Sun would approach this (ramping up
a build infrastructure, test infrastructure, doing regular builds and
testing on another platform) is different than how y'all might do it
as a community project. The QA of it especially.
- David Herron
On Oct 26, 7:26 am, robilad <dalibor.to...@googlemail.com> wrote:
> On Oct 26, 3:17 am, Ron Bowers <ron1...@gmail.com> wrote:
>
> > On the other hand, all of these intel Macs would make truly wonderful
> > Ubuntu boxes.
>
> Paying the Apple tax on Leopard: 129 EUR
> Getting Ubuntu fromhttps://shipit.ubuntu.com/: 0 EUR
Obviously I believe there *can* be more kickass client Java apps or I
wouldn't have joined the JavaFX team (yet another thing I should
discuss in an interview on the posse). So my suggestion is for
people to be constructive and brainstorm awesome applications that
could be written in Java and would be hard or impossible to do in
Flash, Ajax, or Silverlight. We do have some advantages on our side,
such as a larger programmer base, excellent locale & i18n support, hw
accell 3d, real sockets, new NB support for the app framework & beans
binding & JNLP, and access to JNI. Surely we can come up with some
creative ideas. We're an awfully clever bunch. :)
- Josh
/Casper
Wasn't Netscape 5 or 6 an attempt on a large mainstream application in Java?
And how come only minute parts of Open Office are relying on Java? Have Sun ever studied why there are so remarkable few successful mainstream Java applications? After all, on paper a Java based solution should be superior in almost every way - unless of course the "write once run anywhere" involves too many compromises.
I think Corel's attempt was called Barista. I went Googling for it a
few years ago and found only a very few scraps attesting to its
existence.
There was an attempt to rewrite Netscape in Java around that time,
which I think was called Jazilla, and may or may not be related to a
Sourceforge project of the same name.
As far as I know, both projects were abandoned and considered
failures.
Thanks also for noting ThinkFree, which is probably the best applet in
existence. I hated it as a paying customer (it kept destroying my
data), but they've really done good stuff with it over the last couple
years.
That said, I still think overall that Java's been a horrendous
underachiever in the client-side/desktop/user-facing realm. And I'm
saying that after doing desktop Java for 10 years, and writing two
books on it.
Cooper has an interesting proposal on his ONJava blog -
http://www.oreillynet.com/onjava/blog/2007/10/dear_steve_or_fake_steve.html
- basically suggesting that Apple punt to OpenJDK and contribute their
stuff as the basis of an AWT implementation. Cooper also cites the
back-and-forth we've had on Screaming Penguin and [Time code]; about
all this.
--Chris
"I had posted to a "Where is Java 6" thread on the Apple Leopard
discussion forums; there were several Mac users and Java developers
expressing disappointment. However, the thread has mysteriously
disappeared. I can't say I'm shocked, but I'm very disappointed in
Apple's approach to the whole thing. "
http://www.javalobby.org/java/forums/t102924.html#92180758
-Roman
Jess Holle wrote:
> Casper Bang wrote:
>> Wasn't Netscape 5 or 6 an attempt on a large mainstream application in
>> Java?
> Nope. It was just a poorly written native app.
>> And how come only minute parts of Open Office are relying on
>> Java? Have Sun ever studied why there are so remarkable few successful
>> mainstream Java applications? After all, on paper a Java based
>> solution should be superior in almost every way - unless of course the
>> "write once run anywhere" involves too many compromises.
>>
> The most widely publicized attempt in this area that I'm aware of was
> Corel's office suite thing (what was that called again?) and Java was
> simply /not /ready yet then -- it was '97 after all! After that
> well-publicized flop few dared to try again.
>
> Unfortunately, Java was not /really /ready for this sort of thing
> until fairly recently by most accounts -- including mine. By that
> point the overwhelming prevailing wisdom was "it won't work". From
> NetBeans, IntelliJ, JEdit, and other Java desktop applications and
> experience with our own GUI apps, I'm pretty convinced Java 6 /is
> /ready (though still not perfect), but most of the market has not
> changed its stance since 1997.
>
> That's not to say there are no attempts in this space.
>
> See ThinkFree's desktop office suite at http://www.thinkfree.com/.
> It's written in Java and looks pretty compelling overall. I'm /much/
That said, I still think overall that Java's been a horrendous underachiever in the client-side/desktop/user-facing realm.
Java Web Start and that whole delivery mechanism would have *great*
potential if it wasn't so horribly obvious that its Java Web Start
(can only install shortcots with one single icon, can not uninstall
from the menu, everything you run as web start will get registred and
clutter up the install/remove applet, there's no way to interact with
a Java 1.6 splash screen set in the jnlp file etc.). So I guess my
pledge to Sun as a simple developer is, finish what you start and
spend a little time polishing it before moving on to, oh say arm-
wrestle with Adobe.
/Casper
On Oct 29, 12:13 am, Jess Holle <je...@ptc.com> wrote:
> Chris Adamson wrote:
> > That said, I still think overall that Java's been a horrendous
> > underachiever in the client-side/desktop/user-facing realm.
>
> Agreed.
>
> Part of this is due to the market (or most of it) failing to give
> desktop Java another chance after it improved technically to the point
> where it was useful. In other words, one can do really cool stuff with
> desktop Java today, but the market is largely not bothering to try.
>
> The other part of this is that desktop Java has taken a long time to get
> "up-to-snuff". It languished as the unloved backwater of Java for years
> before Sun started /really /treating it seriously in recent JDK
> releases. Sun has now given it front-and-center focus with the Consumer
> JRE and JavaFX -- and I believe Java will end up as a /great /desktop
http://sellmic.com/blog/2007/10/28/beware-of-rotten-apples/
If you crave for a Unix shell and you are not using Linux, cygwin is
always a good alternative, and won't burn you as bad as Apple likes to
do with developers.
The irony is, must of us just want our Java projects to compile so we
can deploy them to servers running Linux or Windows.
Apple's secrecy policy seems to be a long held philosophy which
creates buzz and protects the sales of their products as they head
into twilight. So while this might work for consumer goods, it doesnt
work so well in the developer space. But this is the bed we have
chosen to lie in.
For years many resented Bill Gates and Windows, with no viable
alternative other than open-source (which is not bad, just a
'different game' when you just want to go buy something and have it
work as well as a Mac). So now many, like me, feel rather happy being
in the slightly less evil, or evil in a different way, Mac camp.
On Oct 29, 5:52 pm, Augusto Sellhorn <augusto.sellh...@gmail.com>
wrote:
For years many resented Bill Gates and Windows, with no viable
alternative other than open-source (which is not bad, just a
'different game' when you just want to go buy something and have it
work as well as a Mac). So now many, like me, feel rather happy being
in the slightly less evil, or evil in a different way, Mac camp.
The kick-ass, must have, silver bullet java apps on mac os x are:
1. eclipse
2. netbeans
3. idea
4. java itself so you can test what you're writing.
Those are in no particular order.
That's enough. That's all. Getting developers is a -huge- benefit, as
they are invariably early adopters and 'tech leaders' - people who
influence businesses' computer decisions, where apps are going,
compatibility issues, and last but not least, tend to osmose their
preferences into the world at large simply by being technically savvy
and trustworthy. This isn't the microsoft twist on 'developers,
developers, developers' (which involves hooking developers into tools
that can only produce code that works on your own platform - those
days are mostly gone). It's just simply the fact that a developer is a
power user with tech skills, that's all.
Between the comments of Tor and the other Posse Gang and the sheer
amount of apple logos that show up at any JavaONE, it's obvious that
Sun itself has a boatload of 'down in the trenches' engineers that use
apple hardware. They'll surely care.
This leads to simple conclusion #1: Not having a java6 for leopard,
and not being open about apple's intentions on java in particular, is
a huge bleedin' mistake. It's a fine way to undermine the remarkable
uptake of apple hardware amongst developers in general and java
developers in particular.
Moving on to the practical stuff, what's this mean for us java
developers:
First of all, Sun is much more into the consumer experience than it
was, AND there are boatloads of sun engineers who have apple hardware,
as mentioned. I'm not aware of contractual obligations that prevent
sun from making a JVM for os x. The biggest roadblock right now seems
to be apple's silence. Not much point making your own version if
apple's going to release an update next week.
Secondly, the OpenJDK effort is chugging along, nay, flying like a
speeding bullet. The vaunted 'apple centric performance boost and GUI
upgrades' make little sense to me - it should be near trivial to take
an OpenJDK and make it work on macs. Here's what apple might add to
the pie:
- better PowerPC support. Who cares? We're on intel chips. We have
JVMs that blaze on intel chips. intel chips in macs are no different.
- Access to apple's own libraries. Apple actually stopped supporting
this a while ago and no one was doing it anyway. I used Quicktime 4
java and it wasn't great. I doubt anyone really cares about this one.
- a plain GUI in general. SWT has always been developed outside of
apple. With nimbus, how hard would it be to make nimbus run on a cocoa
canvas? It just can't be too difficult. No problems here.
- GUI acceleration. With due apologies to Romain Guy, I doubt this is
important. Right now the vast majority of java development is still a
plain GUI or web based. Having a slower GUI on macs will be an
annoyance to be sure but no show stopper.
- apple look-and-feel for GUI apps. There are nice documents over at
apple on how to make your java GUI apps like aqua-ish. Including
getting the preferences and about menu options in the right place and
having the resize thingie in the bottom right corner not be obscured
by your own GUI, and packaging it all up in a nice .app file. This
would have been a big issue except that no one uses java apps on apple
in that way. Netbeans/eclipse/IDEA have the resources to make custom
wrappers (and eclipse itself never even used the packaging option in
the first place!), and those are the only 3 apps that matter. Anything
else will look slightly off, maybe. Big whoop.
- 32 bit vs 64 bit. While having a java that integrates well with
leopard's 32/64 bit stuff would be great, the status quo on all other
platforms is that you have to choose one. Doing the same thing for the
hypothetical future Mac OpenJDK version shouldn't be difficult in the
least.
Long story short: Worst case scenario, apple dropped java like a
stone, which would mean a 6 to 12 month delay as a proper OpenJDK port
gets set up. Then everything is back to normal.
This is exactly what we are doing. Please try the Java 1.6 update N
preview release (on a *non-critical* system! :).
- J
Thanks, had some trouble seeing what exactly was changed in update N
apart from the VM preloading. Nice to hear Java Web Start is going to
be improved on as well. And last but not least, awesome that sites
using errornous Java no longer deadlocks Firefox 2.0 on Windows (as
i.e. java.sun.com/developer/technicalArticles/Programming/sprintf/
usually do). It will take some time to get used to, not to have to
hold your breath and cross your fingers when an applet is loading. :)
/Casper
On Oct 26, 4:30 am, Neil Bartlett <njbartl...@gmail.com> wrote:
> I can't understand what the fuss is all about. Where are all these
> amazing new features in Java 6 that you simply can't live without?
Are you serious?
http://java.sun.com/javase/6/docs/api/java/sql/package-summary.html
http://java.sun.com/javase/6/docs/api/java/util/LinkedList.html
There are few other great additions to the API that I didn't realize I
lived without.
and then there is the speed
http://www.jroller.com/dgilbert/entry/is_java_se_1_6
Fake Steve - Java developers finally realize the party's over -
http://fakesteve.blogspot.com/2007/10/java-developers-finally-realize-partys.html
"But I think they don't quite know which party they're talking about.
See this screed where some Javatard says Apple has been spitting in
his face because we didn't include Java 6 in Leopard. Or something. So
he says he's selling his Mac. His headline is, "So long, Apple. The
party's over." Well he's right about one thing. The party is over --
the Java party, that is. Glad to see the Javatards have finally
figured that out. Word is they're co-sponsoring a support group with
Lotus Notes developers. Fun bunch."
Daring Fireball linked to the newer of the Javalobby threads and said
"Java Developers Unhappy With Leopard. I fail to see why anyone
(other than Java developers themselves) would care."
Of course they are... as is everybody else outside the 'Java
community' :-)
...
Java6 will come eventually to OS X, but some enthusiasts seem to
grossly overestimate Java's importance outside of it's 'core market'..
Not releasing the latest and greatest version of someone's favorite
toy
from day one is a far cry from 'treating with disdain'.
Sorry, but sometimes the feelings of entitlement and self importance
within the 'Java community' can get a bit absurd.
The sad fact is despite suns belated efforts, client java has died.
Maybe it will rise from the grave, but I have learnt from bitter
experience not to expect that (always happy to be surprised).
Who knows, maybe apple have something great in store.
Irrespective of its importance Apple should not treat its customers with such clear disdain.Not releasing the latest and greatest version of someone's favorite toy from day one is a far cry from 'treating with disdain'.
Sorry, but sometimes the feelings of entitlement and self importance within the 'Java community' can get a bit absurd.
Evidently, in order to see this, you have to not be madly in love with
Java (denial). When I mentioned that Java on the desktop was dead in
an email that was read up by The Posse about a year ago, it was not
received very well (anger). Not that I expected it to, the Java
ecosystem has always felt very stubborn and pleased with itself (lack
of acceptance).
Acceptance is required in order to want to fix, and recent fresh
blowing breeze suggests this to have happened at least at some level
inside Sun. And it just might ignite the sparks into a full fledged
fire again - although I fear the breeze is too weak or too late.
So I can understand why people outside this ecosystem would be
laughing at a crowd worthshipping platforms which are now incapable of
operating together. It's a bucked of cold water in the face, but
perhaps it was needed.
/Casper
I mean, sorry, but holy shit, when does naiveté start to become a
liability? Who in his right mind makes relevant decisions based upon
old promises of some CEO?
...
A couple of thousand Java developers using Macs to write software for
a different platform is not a 'market segment' for Apple. They're
certainly not "opinion makers" or "thought leaders". They're a Solaris
refugees, no more, no less.
--
Jess Holle
The fact that java client side is dead is totally irrelevant. People
want their java to run eclipse, netbeans, IDEA, and tomcat/jetty/
struts/spring/rife/whatever.
What I don't get is this:
How hard could it be for sun to build a version of java6 for mac os x
that forces itself into headless mode? I can even boot up jconsole
using the apple 1.5 which does do GUIs. I bet that covers 95% of the
current complaints and I seriously doubt it would take much effort.
It's okay if the release is intel-chip only, even. It's got a posix
kernel and everything.
You're (probably) dead on.
The discussion revolves around some 'radical elements of the Java
community' thinking Java6 is important enough for Apple to justify
another delay of Leopard and everybody else considering that
ridiculous. :-)
To find out about the new changes to Objective-C you had to actually
pay cold hard cash to sign up to their developer program.
That is one thing I love about the way Sun handles Java, everything is
out in the open for _years_ in advance of changes being made. Whereas
Apple is ultra secretive about everything.
I think it is not entirely a coincidence that outside of Apple there
are ... thousands(?) of OS X developers, whereas outside of Sun there
are millions of Java developers.
OTOH, if the only applications that matter enough to ensure porting
are all IDEs, that is a sure sign that client side Java is dead and
pushing up the daisies*. Moreover, on the server side there is a vast
amount of work, but in my local area Java EE 5 isn't gaining
traction. It is so frustrating. Can someone please put a stake
through the heart of the unholy abomination that is hibernate (and 99%
of the other xml driven frameworks too) already? Java EE 5 is such a
huge leap in the right direction. Also, on the client side some of
these guys are doing incredible things, making beautiful software. So
the tools exist today for Java to be truly great, but there is this
vast inertia of the lowest denominator dragging it down and strangling
it.
*Back with 1.4 I worked on a client side program which was fairly
slick looking (nothing in the league of bling bling as the Romans,
Romains, Kirills and Joshes though), so I know it is possible. I play
Puzzle Pirates occasionally, so I know first hand that Java can do
pretty.
OTGH, I think the most exciting thing happening in client side Java
today is not JavaFX, but GWT. I just wish someone would put a bullet
in Struts (and the other web frameworks) so that we could use it!
And the most exciting thing on the server side is not Scala, it is
annotations.
Actually, from what I've read the number of Java developers worldwide
is an estimated 6 million. If only a third of those use Macs it would
be a couple million, not a "couple thousand"; certainly not a
demographic to pooh-pooh.
My question is: What about servers? If the Java 6 omission is a
symptom of a greater indifference to Java by Apple, then aren't they
effectively conceding the server market? Can anyone deny the
significance of Java on the server side?
Again, I feel this is not indifference but resources. Their (fairly
small) team can only do so much and they apparently decided that
getting a improved 1.5 onto Leopard was the highest priority. I get
the impression that there were a lot of changes under the hood (which
presumably will benefit Java 6 as well). Remember, Leopard itself
was late and Java 6 isn't the only feature that was cut from the
10.5.0. I prefer to always go with simplest explanation. It's just
late.
- J
>
>
>
> >
- Josh
> * Promising to make MacOS X the premiere Java platform...
... in the year 2000
> * Later saying "Java is dead"...
... in the year 2007. It is quite legitimate for a company to change
their strategy after seven years.
> * Doing a lot of promising looking Java stuff to be delivered in Leopard
It is not that Apple won't deliver Java 6. There is no anti-Java
conspiracy going on. When it is ready Apple will deliver. But
expecting Apple gives Java the same priority as their core products is
pretty weird.
> * Not delivering the Java stuff in Leopard and giving no information
> where they're really at with Java
This just Apple being Apple. They do this all the time, every time and
have been doing this for over thirty years. Everybody dealing with
Apple knows that. Sure it sucks, but it is the same treatment for
everyone.
...
> I think it is not entirely a coincidence that outside of Apple there
> are ... thousands(?) of OS X developers, whereas outside of Sun there
> are millions of Java developers.
It is also no coincidence that Apple is actually profitable and that
these thousands of developers write applications consumers like to
use, while millions of Java coders only talk about creating them.
....
> If only a third of those use Macs
Apple does not even have this kind of (world wide) market penetration
within its core markets.
I'm confused. That vastly contradicts the 3 most recent blog entried
by Ben Galbraith:
http://weblogs.java.net/blog/javaben/archive/2007/10/the_java_on_os.html
/Casper
Between Filty Rich Clients and similar focus by java developers, and
the consumer JRE, now is just the time when client side java might
finally be birthed. Perhaps we should retire the metaphor, actually.
That sounded horrible :-P
General note on the javalobby article: It was full of dingo droppings.
It jumped the gun more than a bit. Do read Ben Galbraith's counter
arguments.
Java6 will be released in at most a month or two by apple, and that'll
be the end of this storm in a teacup.
I'm guilty myself of overhyping this issue, for which I apologize.
As far as the pipeline goes, it should be plenty fast enough. I
suspect the reason certain particular apps are slowing down is
because they were counting on certain image formats being fast to
draw, and those formats changed when the pipeline changed. Properly
written code should convert images into particular formats that are
fastest on the underlying platform. There are utilities in SwingLabs
to do this (heavily used in Romain's demos, of course). BTW, When I
say format I mean the layout of bytes in the BufferedImage, not file
formats like gif & png. After updating code I'm sure those
applications will be fast again.
- Josh
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google
> Groups "The Java Posse" group.
> To post to this group, send email to java...@googlegroups.com
> To unsubscribe from this group, send email to javaposse-
> unsub...@googlegroups.com
> For more options, visit this group at http://groups.google.com/
> group/javaposse?hl=e
Holy crap, that's a lot of type-unsafe/hackish-feeling client
properties. How's does that fit write-once-run-anywere?
> The new features in Java 5 on Leopard are real. More importantly, I
> think the things they have been working on under the hood will make
> it easier to get Java 6 running sooner (things like the new Aqua L&F
> that better uses Cocoa).
Correct me if I am wrong, but didn't they drop the Cocoa bridge over a
year ago? Also, the sentence "...look and feel more like Cocoa
applications..." would indicate it has nothing to do with Cocoa peers.
/Casper
> Holy crap, that's a lot of type-unsafe/hackish-feeling client
> properties. How's does that fit write-once-run-anywere?
Quite nicely, actually. These are just client properties. Something
extra which you can add to components that you want to style. Other
Look and Feels will completely ignore them, as will older versions of
Java on the Mac. It means that your app will always do the right
thing. Also notice that they didn't call most of their properties
things like apple.searchbar=true; but instead used
JTextField.variant=search. This means that other Look and Feels
which have their own specific way to render a search box could
recognize the same client property. (in a future version of WinLaF,
perhaps?) And remember, these are all completely optional. Think of
them as css rules that older browsers would ignore. It's a great way
to add new features that degrade gracefully. Quaqua has been doing
this for years quite successfully.
>
>> The new features in Java 5 on Leopard are real. More importantly, I
>> think the things they have been working on under the hood will make
>> it easier to get Java 6 running sooner (things like the new Aqua L&F
>> that better uses Cocoa).
>
> Correct me if I am wrong, but didn't they drop the Cocoa bridge over a
> year ago? Also, the sentence "...look and feel more like Cocoa
> applications..." would indicate it has nothing to do with Cocoa peers.
Well, I don't know for sure because I've never seen the code to the
Aqua L&F, but I did spend a year and a half working on the Windows
Look and Feel. In WinLaF (starting with Java 5) we call to native
Win32 methods to get the actual bitmaps that Windows uses to draw
native widgets. This is why recent versions of WinLaF will work with
3rd party themes. Since we get whatever Windows itself is using, we
automatically pick up the theme changes. I imagine they are doing
something similar in Aqua.
- J
Could you be thinking about the Carbon bindings? Although in
researching this I did find some blurbs about "reduced" cocoa support.
Greg