QNX Press Release
Jim Gunnell
comes this:
"By combining PersonalJava or EmbeddedJava technologies with the QNX Realtime
OS, information appliance manufacturers gain a unique advantage: they no
longer have to be locked into a single processor platform," said Linda
Campbell, vice president of strategic alliances at QNX Software Systems.
"That's because we've designed QNX drivers, file systems, and other OS
services to be source-code identical across CPUs - just like Java technology-
based applications.
"As a result," continued Campbell, "manufacturers can now switch processor
platforms on demand. Or they can create an entire family of information
appliances - across multiple processors and at different price points - using
a single, platform-independent code base."
Now, couple this with the statements made by the esteemed people at
Amiga, and if half the rumours of the "MMC" chip are true, the NG machine is
really going to be one hellva beauty.
Jim Gunnell - Member - Team Amiga
A3k Desktop turned
tower/18&2Mb/Opalvision/CyberVision/PPC&060&80M/EZ230/E3/VidiPro24/...
Random taglines for today are.....
None Are So Blind As... Helen Keller.
If at first you don't suceed, destroy all evidence that you tried.
No News Is... Impossible.
Shawn
Jim Gunnell <jg...@icom.ca> wrote in article
<1297.850T198...@icom.ca>...
> From the QNX News area at http://www.qnx.com/hotnews/index.html
> comes this:
>
> "By combining PersonalJava or EmbeddedJava technologies with the QNX
Realtime
> OS, information appliance manufacturers gain a unique advantage: they no
> longer have to be locked into a single processor platform," said Linda
> Campbell, vice president of strategic alliances at QNX Software Systems.
> "That's because we've designed QNX drivers, file systems, and other OS
> services to be source-code identical across CPUs - just like Java
technology-
> based applications.
> "As a result," continued Campbell, "manufacturers can now switch
processor
> platforms on demand. Or they can create an entire family of information
> appliances - across multiple processors and at different price points -
using
> a single, platform-independent code base."
>
> Now, couple this with the statements made by the esteemed people at
> Amiga, and if half the rumours of the "MMC" chip are true, the NG machine
is
> really going to be one hellva beauty.
>
> Jim Gunnell - Member - Team Amiga
>
And coupled with another article from the QNX news area at
http://www.qnx.com/hotnews/pr/nov24_98-citrix.html
*** OTTAWA, November 24, 1998 - QNX Software Systems Ltd. (QSSL) today
introduced QNX RTOS support for the Independent Computing Architecture
(ICA) from Citrix Systems Inc. As a result, businesses can now use QNX and
Citrix ICA to distribute their existing Windows applications to virtually
any type of embedded system or thin client in the enterprise - be it a
network computer, handheld device, or realtime control system.
"To understand what this means," said Linda Campbell, vice president of
strategic alliances at QSSL, "imagine a diskless network terminal - or even
a wireless portable device with just 4M of RAM - accessing full-blown
enterprise solutions like PeopleSoft and SAP, as well as desktop
applications like Word and Excel. Better yet, these applications will look,
feel, and respond as if they were running locally on a desktop PC."
"With the combination of QNX and our Independent Computing Architecture,
there are virtually no limits to where a business can distribute its
enterprise applications," said Janice Roberts, director of client business
development at Citrix. "Devices that were previously too small, too slow,
or had rigorous realtime requirements can now access Windows-based
applications quickly and seamlessly, without regard to available
bandwidth." ***
***
It's an extention of what can be done between QNX powered devices but gives
QNX the abillity to apply the same methods of remote working to windows
implications for the amiga...
The stranglehold of the windows OS is further loosened, and if it's
available for QNX then the transfer to the new AmigaOE would seem an
inexorable and inevitable process.
The need for software to exclusively be written/compiled for the new
AmigaOE is ruduced but the access to the software is much increased
While we may not see the release of major software for the amiga we won't
need to, but amiga software made for amiga devices will shine much
brighter.
Shawn.
Paul E. Bell
I'm not so sure that either the Java or the Citrix code will be so
available to the Amiga. Remember, AmigaSoft OE is built on the Neutrino
_kernel_, not on QNX in general. They are as different as a truck and a
boat that just happen to use the same engine. If Amiga wants this
functionality in AmigaSoft OE, I am sure they will have to pay for it.
If we want it, I am sure we will have to clamor for it.
--
Paul E. Bell Email and AIM: wd0...@millcomm.com
ifMUD: Helios | IRC: PKodon, DrWho4, and Helios
(I'd put my webpage here, if it had anything on it.)
_____ Pen Name/Arts & Crafts signature:
| | _ \ _ _ |/ _ _(
| | (_X (_/`/\ (_) (_` |\(_) (_) (_|_) (/`
)
Philip Kaulfuss
> a single, platform-independent code base."
Wasn't this the idea behind C though? What makes Java better?
--
,- Philip Kaulfuss ---------------. Website: http://www.boehme.demon.co.uk :
| | IRC: PhilK in Undernet #Henry |
| Email: Change nospam to boehme | |
: ICQ: 21755556 `----------------------- Philip Kaulfuss -'
Vanni Torelli
As far as I remember, Jim Collas said something about Java and Jini being
certainly a strong feature of the new AmigaOS. We can't be sure that
would be an adaptation of QNX's Java, still I bet we are going to have a
full Java 1.2 (or higher...) support from day one.
> I'm not so sure that either the Java or the Citrix code will be so
> available to the Amiga. Remember, AmigaSoft OE is built on the Neutrino
> _kernel_, not on QNX in general. They are as different as a truck and a
> boat that just happen to use the same engine. If Amiga wants this
> functionality in AmigaSoft OE, I am sure they will have to pay for it.
> If we want it, I am sure we will have to clamor for it.
--
. ..................................................... ... .. .
\ _/__ vanni torelli mind soul whatever in between
\/ electronic engineering student - audio engineer
(_ va...@cleansky.force9.co.uk - van...@freeuk.com
. .. ... ..................................................... .
t...@ramjam.u-net.com
> Wasn't this the idea behind C though? What makes Java better?
C is platform independent (when written to be so) at the *source code*
level - i.e. C programs must be compiled for each platform that it is
to run on. Unless the source is available, you are limited to running
it on the platform(s) that the author compiled it for. Supporting
each additional platform is extra work.
Java is platform independent at the *byte code* level - i.e. the same
Java byte code will execute on *any* Java platform, even if its one
that the program's author has never even heard of. Supporting each
additional platform is no extra work at all.
Tim
--
======================================================================
Tim Corringham e-mail: t...@ramjam.u-net.com
Ramjam Consultants Ltd tel: +44 (0)118 946 5940
Reading, England fax: +44 (0)118 946 5941
John Millington
: *** OTTAWA, November 24, 1998 - QNX Software Systems Ltd. (QSSL) today
: introduced QNX RTOS support for the Independent Computing Architecture
: (ICA) from Citrix Systems Inc.
Heh. So _that's_ what ICA means now, huh? In earlier Citrix products it
meant Intelligent Console A[something]. I guess the word "console" wasn't
sexy enough. :-)
: implications for the amiga...
: The stranglehold of the windows OS is further loosened, and if it's
: available for QNX then the transfer to the new AmigaOE would seem an
: inexorable and inevitable process.
: The need for software to exclusively be written/compiled for the new
: AmigaOE is ruduced but the access to the software is much increased
: While we may not see the release of major software for the amiga we won't
: need to, but amiga software made for amiga devices will shine much
: brighter.
More support for more things is better, so I won't argue that getting an
ICA implementation would be good. But this isn't the same thing as being
able to run 'Doze apps. ICA is pretty much the same thing as X11,
PCAnywhere, etc. It's a way for the user to be at a different computer
than the one running the app.
But unlike X11, it is closed/secret/proprietary. You cannot get information
about the protocol without signing papers, paying money, etc.
<sour_grapes> I once called Citrix several years ago to try to get
information about it, because I wanted to implement an ICA client on -- you
guessed it -- the Amiga. They were very uninterested in talking to me.
(And yes, I was polite, and not in amiga.advocacy mode :-) </sour_grapes>
Yog-Sothoth Neblod Zin,
John Millington
Ed Dana
> In article <MPG.11e46f5c5...@usenet.force9.net>,
> va...@cleansky.force9.co.uk (Vanni Torelli) wrote:
> > Hi Paul E. Bell! In the mist of cyberspace, I spotted you writing...
> >
> > As far as I remember, Jim Collas said something about Java and Jini
> being
> > certainly a strong feature of the new AmigaOS. We can't be sure that
> > would be an adaptation of QNX's Java,
>
> AFAIK QNX' Java is simply standard Java. QNSS is not another Microsoft
> trying to personalize Java, HTML and all else to make sure that only
> their programs are compatible.
>
> But yes, Jim mentioned several times that Java was part of the OE. IIRC
> he mentioned Java2.
Hey, Luca! Welcome back to AmigaLand... : )
Havemose *definately* intends to have Java as part of the OS. He was even
considering the possibility of including a Java "Notepad" as part of the
OS. This implies the possibility of other Java based utilities.
It's a good idea, especially if your are planning an OS that is
multiplatform.
--
Sincerely, | "The problems of two little people don't amount
Ed Dana | to a hill of beans in this crazy mixed-up world!
Consultant/Developer | But this is OUR hill, and these are OUR beans!"
Amiga Enthusiast. | -- Naked Gun via Casablanca.
=========== http://OurWorld.CompuServe.com/Homepages/EDanaII ===========
luca_...@my-deja.com
va...@cleansky.force9.co.uk (Vanni Torelli) wrote:
> Hi Paul E. Bell! In the mist of cyberspace, I spotted you writing...
>
> As far as I remember, Jim Collas said something about Java and Jini
being
> certainly a strong feature of the new AmigaOS. We can't be sure that
> would be an adaptation of QNX's Java,
AFAIK QNX' Java is simply standard Java. QNSS is not another Microsoft
trying to personalize Java, HTML and all else to make sure that only
their programs are compatible.
But yes, Jim mentioned several times that Java was part of the OE. IIRC
he mentioned Java2.
Luca
------
To reply remove "nojunk" from gdi...@hpu.nojunk.edu
http://www.geocities.com/SouthBeach/Canal/3184 (Under Construction)
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
Paul May
> va...@cleansky.force9.co.uk (Vanni Torelli) wrote:
>> Hi Paul E. Bell! In the mist of cyberspace, I spotted you writing...
>>
>> As far as I remember, Jim Collas said something about Java and Jini
>being
>> certainly a strong feature of the new AmigaOS. We can't be sure that
>> would be an adaptation of QNX's Java,
>AFAIK QNX' Java is simply standard Java. QNSS is not another Microsoft
>trying to personalize Java, HTML and all else to make sure that only
>their programs are compatible.
>But yes, Jim mentioned several times that Java was part of the OE. IIRC
>he mentioned Java2.
Hi Luca.
the JAVA2 reference can be found in the Q&A onthe executive update page,
NTO2.0 as per the QSSL 3rd partys is still not ready for prime time
(still plenty of time before the consumer AmiQNX kit is due) but its
just been posted that some of the 3rd party QSSL Nto'ers have seen
the latest release and its got JAVA in it.
i'v been telling you guys/girls for ages that the QSSL Nto2.0 and the
AmiQNX are far more inter-related than you give it credit, its about time
everone here starts taking a far greater interest in QSSL 3rd party
partnerships and so on, then you might start to see just how vast this
AMIGA partnership could be in the near future.
i'm sure there are things/options that need to be voiced out in public
so that Jim, Rick, Bill and the rest can see our POV and perhaps help
them see options that have past them by.
i'v pointed out both privately and publicly on other lists about
such things as NCi (now www.liberate.com) a subsidury of ORACLE.
http://www.zdnet.com/pcmag/features/future/concept01.html concentrateing
on the future products from the other markets and how it might effect
our platform, how might the AmiQNX adapt to better these concepts.
'Piano' Motorola's research in wireless Tec, it can transmit information
at speeds of upto 500 magabits per second over a distance of 10 feet.
DOC (Disk on Chip) Tec is another interesting product line that could
be of use if it gets enough push into consumer space.
several weeks ago, i posted the personalJAVA QSSL announcment,
in it that there wereseveral things we should look at,
notably the realtime JAVAone conference and again the IBM connection,
QNX Takes Center Stage at JavaOne Conference
In his keynote address this week at JavaOne, Java
inventor James Gosling presented a demo that proves
realtime applications can be written in "pure Java."
The demo, based on the QNX RTOS and a modified JVM,
was jointly developed by IBM and QNX Software Systems,
both members of the Real Time Java Expert Group. The
group's mandate: to develop a specification for realtime
extensions to the Java API. The demo highlighted
extensions scheduled for public review later this year.
Details: http://www.qnx.com/hotnews/pr/jun15 99-realtimejava.html
_____________________________
http://www.zdnet.com/intweek/stories/news/0,4164,2274831,00.html
its basicly saying that Motorola, Cisco, and Sun will make a
new internet protocol high speed wireless network intigrating
Sun's JAVA SW.
http://www.real-time.org real-time CORBA, again expected in AmiQNX
and related to the QSSL connection.
going back through my posts to sep98, we have
http://www.theregister.co.uk/980916-000018.html
The secret project, codenamed Miramar, allows information to be
organised in such a way that URLs and other often used files and
addresses can be placed in a virtual 3D environment.
A source close to the project said that the software is intended to be
licensed to other companies and it will not be available for another
two years. The user interface does not compete with Microsoft
software, he said.
________________________________________
as its always nice to add something new , it might be an advantage
to take a look at WAP (wireless Application Protacol)
theres quite a bit of stuff and i ended up useing a search engine
to find the most interesting stuff, WAP along with several other
wireless Specs were also referenced off the JAVA/one sites.
all these things are what we should be looking at as a possible
option/future for our platform, not troting out the same old thing,
killfiled RUe threads slipping through ECT.
if you know of something that might be useful and werth pointing Jim
in its direction, just start a thread and got it down, we desperatly
need some good informative threads starting here PLEASE.
go to it.
--
Paul May, UK
Team *AMIGA* , *KOSH* , *OwlNet* , *REBOL*
KO...@reedsweb.net
--
http://www.KOSH.net
Kommunity Operating System and Hardware
http://www.owlnet.net
Open Platform Support Services, Home of Team *AMIGA*
http://www.rebol.com
Open Platform Scripting Language, Version 2 now available
Ancipital
Kaulfuss) wrote:
>A potato resembling Jim Gunnell miraculously spoke thus:
>
>> a single, platform-independent code base."
>
>Wasn't this the idea behind C though? What makes Java better?
Portable object code, and a consistent abstracted API that allows the
same calls to produce, for example, native gui elements on whatever
platform, or make network connections without any adaptation, for a
start?
I'm sure our resident java hackers could give a much better answer,
mind :)
Ancipital- Inedible Buddhas reality control #1
http://www.buddhas.org is currently tqt- back soon.
To unmung email addr, remove all instances of "aremadeoffish"
"I'm not crying victim, but I am stating that a lot of spammers
are genuine scumbags." -Sanford Wallace
Daithi O'Cuinn
Philip Kaulfuss wrote:
>
> A potato resembling Jim Gunnell miraculously spoke thus:
>
> > a single, platform-independent code base."
>
> Wasn't this the idea behind C though? What makes Java better?
Well, C code needs to re-compiled for every new platform, and if it has
any platform specific features, it has to be adapted as well.
Java programs are compiled into bytecode, which is then interpreted.
Only the interpreter (the Java Virtual Machine) has to be ported from
platform to platform, and that is a one off job.
--
Daithi O'Cuinn
luca_...@my-deja.com
Ed Dana <EDa...@CSI.com> wrote:
> Hey, Luca! Welcome back to AmigaLand... : )
Thanks Ed. I missed to elections to the council though. :-(
Damn HD broke and took me two weeks to get a new one and a Net
connections on my Ami in Italy.
Eelke Blok
luca_...@my-deja.com wrote:
>
> In article <377AB51D...@CSI.com>,
> Ed Dana <EDa...@CSI.com> wrote:
>
> > Hey, Luca! Welcome back to AmigaLand... : )
>
> Thanks Ed. I missed to elections to the council though. :-(
> Damn HD broke and took me two weeks to get a new one and a Net
> connections on my Ami in Italy.
Heyhey,
Maybe a bit late, but: welcome back! :)
Cheers,
Eelke
--
Eelke Blok, student Electrical Engineering, University of Twente
http://home.student.utwente.nl/e.blok, ICQ: 19514933
Amiga-page: http://home.student.utwente.nl/e.blok/amiga
"Our Lady of Blessed Acceleration, don't fail us now!" - Elwood Blues
luca_...@my-deja.com
Eelke Blok <e.b...@student.utwente.nl> wrote:
> Heyhey,
>
> Maybe a bit late, but: welcome back! :)
Thanks Eelke.
Hey, are you gonna make it to the Comdex?
Luca
PS: Aram was right, after eliminating the posts of you know who, it
didn't take me too long to catch up with the group :-)
Paul Hill
>A potato resembling Jim Gunnell miraculously spoke thus:
>> a single, platform-independent code base."
>Wasn't this the idea behind C though? What makes Java better?
It's trendier :-)
--
Paul
Glenn Saunders
>applications like Word and Excel. Better yet, these applications will look,
>feel, and respond as if they were running locally on a desktop PC."
Fancy "dumb terminals".
Michael Nielsen
>
> > > a single, platform-independent code base."
>
> > Wasn't this the idea behind C though? What makes Java better?
>
> level - i.e. C programs must be compiled for each platform that it is
> to run on. Unless the source is available, you are limited to running
> it on the platform(s) that the author compiled it for. Supporting
> each additional platform is extra work.
>
> Java is platform independent at the *byte code* level - i.e. the same
> Java byte code will execute on *any* Java platform, even if its one
> that the program's author has never even heard of. Supporting each
> additional platform is no extra work at all.
Nothing inherently wrong here, but, to be completely true, java is not
platform independant, it is highly dependant on the virtual platform
(also called java virtual machine), this machine being ported to other
architectures, does a mapping between the virtual machine operations and
the real machine operations, and thus giving the illusion of platform
independance. (You said it your self, *any* java platform, or basically
only on the JVM (java virtual machine)).
So in one respect it is not platform independent, any more than
scripts written in perl, csh, sh, BASIC, and so forth, are. It still requires
porting of the interpretor... What is new with java is the technology
that converts the script (byte code) to machine code, and on the next
iteration the machine code is used, and not the script any more, and
that a pseudo binary format is used for distribution so that you don't
need to distribute the source, (though byte code can be converted
to source, if the desire should be there).
But I guess it's like comparing a rose and a flower, basically
the rose is a flower right :-> ?.
mvh
mike.
--
#include <stddisclaimer.h>
-----------------------------------------------------------------------
(Ingeniør) Michael Nielsen BE(Hons)
telf: +45 9828 4611 x 26
USE THIS -> Private email : mi...@anarki.dk fax : +45 9828 4565
-----------------------------------------------------------------------
At arbejede er godt, så længe man husker at leve!
(To work is good as long as we remember to live!)
Dion Wooning
>
> t...@ramjam.u-net.com wrote:
> >
> > Java is platform independent at the *byte code* level - i.e. the same
> > Java byte code will execute on *any* Java platform, even if its one
> > that the program's author has never even heard of. Supporting each
> > additional platform is no extra work at all.
>
> Nothing inherently wrong here, but, to be completely true, java is not
> platform independant, it is highly dependant on the virtual platform
> (also called java virtual machine)
> What is new with java is the technology
> that converts the script (byte code) to machine code, and on the next
> iteration the machine code is used, and not the script any more, and
> that a pseudo binary format is used for distribution so that you don't
> need to distribute the source
Just FYI, actually even that idea is not new, and in the past even game
companies such as Infocom have used virtual machines (the Zork machine)
that execute byte code. By the way, the conversion of byte code into
native machine code in the fashion you describe is called JIT (Just In
Time) compilation/interpretation. I am not 100% sure if this technique
has been used before as well (but it seems a logical step to take).
---
Dion Wooning
Michael Nielsen
> > What is new with java is the technology
> > that converts the script (byte code) to machine code, and on the next
> > iteration the machine code is used, and not the script any more, and
> > that a pseudo binary format is used for distribution so that you don't
> > need to distribute the source
>
> Just FYI, actually even that idea is not new, and in the past even game
> companies such as Infocom have used virtual machines (the Zork machine)
> that execute byte code. By the way, the conversion of byte code into
> native machine code in the fashion you describe is called JIT (Just In
> Time) compilation/interpretation. I am not 100% sure if this technique
> has been used before as well (but it seems a logical step to take).
I was talking to one of they guys working on the teoretical jit's and
he "claimed" that it was new teknology, and I really think that this
is about the only really new thing java brings in.. Standardizing
the API has been attempted by Posix, X, Microsoft (sledge hammer approach
- use it or else - though not cross platform), as you said virtual
machines is nothing really new either. Must say though that java
as a language is a lot cleaner than C++ (objectoriented speaking),
but it stille needs a good compiler before it becomes useful for
large gui applications, there the JIT, cannot do a proper job, and
from what I've seen hotspot is a joke.
t...@ramjam.u-net.com
> Dion Wooning wrote:
> I was talking to one of they guys working on the teoretical jit's and
> he "claimed" that it was new teknology, and I really think that this
> is about the only really new thing java brings in.. Standardizing
> the API has been attempted by Posix, X, Microsoft (sledge hammer approach
> - use it or else - though not cross platform), as you said virtual
> machines is nothing really new either. Must say though that java
> as a language is a lot cleaner than C++ (objectoriented speaking),
> but it stille needs a good compiler before it becomes useful for
> large gui applications, there the JIT, cannot do a proper job, and
> from what I've seen hotspot is a joke.
JIT compilers are not new, they've been around for 20 years or more.
The Java language isn't exactly new either - it's pretty much smalltalk
with C syntax.
What Java does do is pull a lot of separate developments together, and
package them as one (marketable) technology. However, since Java is so
new (no other language has ever become popular so quickly) there are
weaknesses in current implementations. These will be resolved quite
quickly.
Tim
==========================================================================
Ramjam Consultants Ltd, 28 Queensway, Caversham, Reading, RG4 6SQ, England
e-mail: sup...@ramjam.u-net.com http://www.ramjam.u-net.com/
telephone: +44 (0)118 946 5940 fax: +44 (0)118 946 5941
Matthew James Scott
Machine too...
There are also java chips too... processors that run bytecode natively
Matt
Dion Wooning wrote:
> Michael Nielsen wrote:
> >
> > t...@ramjam.u-net.com wrote:
> > >
> > > Java is platform independent at the *byte code* level - i.e. the same
> > > Java byte code will execute on *any* Java platform, even if its one
> > > that the program's author has never even heard of. Supporting each
> > > additional platform is no extra work at all.
> >
> > Nothing inherently wrong here, but, to be completely true, java is not
> > platform independant, it is highly dependant on the virtual platform
> > (also called java virtual machine)
> > [....]
> > What is new with java is the technology
> > that converts the script (byte code) to machine code, and on the next
> > iteration the machine code is used, and not the script any more, and
> > that a pseudo binary format is used for distribution so that you don't
> > need to distribute the source
>
> Just FYI, actually even that idea is not new, and in the past even game
> companies such as Infocom have used virtual machines (the Zork machine)
> that execute byte code. By the way, the conversion of byte code into
> native machine code in the fashion you describe is called JIT (Just In
> Time) compilation/interpretation. I am not 100% sure if this technique
> has been used before as well (but it seems a logical step to take).
>
> ---
> Dion Wooning
--
Kewill Electronic Commerce
Name: Matthew Scott
Dept: Software Development
Tel : +44 (0)161 952 8027
Fax : +44 (0)161 952 8021
Mail: mailto:matthe...@kewill.com
W3 : http://www.kewill.com
Dion Wooning
>
> In comp.sys.amiga.misc Michael Nielsen <m...@dde.dk> wrote:
>
> > Must say though that java
> > as a language is a lot cleaner than C++ (objectoriented speaking),
> > but it stille needs a good compiler before it becomes useful for
I agree on Java being nicer in design in a few areas than C++. About
good compilers/JIT: I haven't kept up with the latest, but personally
I would not replace C++ by Java unless the difference in execution
speed becomes only a few % (dependent on project, of course).
> What Java does do is pull a lot of separate developments together, and
> package them as one (marketable) technology. However, since Java is so
> new (no other language has ever become popular so quickly) there are
> weaknesses in current implementations.
Maybe C got popular quite quickly in its days as well? But you are
right about Java.
P.S. Tim, your quoting was a bit lousy :) It made it seem as though
Michael's words were coming from me, upon first reading. Never
mind though.
---
Dion Wooning
ims...@my-deja.com
Stop dealing with Usenet for AmigaNG info and discussion.
Join into a community of professionals and multimedia aficionados
helping to advance the cause of a revolutionary new "Next Generation"
computing system!
The future is now!
To join just send email to: AmigaNG-...@onelist.com
Thats all that you need to do.
See ya on the list!
Ciao,
dupont
Joe Cosby
Dion Wooning hunched over his computer, typing feverishly;
thunder crashed, Dion Wooning laughed madly, then wrote:
> t...@ramjam.u-net.com wrote:
> >
> > In comp.sys.amiga.misc Michael Nielsen <m...@dde.dk> wrote:
> >
> > > Must say though that java
> > > as a language is a lot cleaner than C++ (objectoriented speaking),
> > > but it stille needs a good compiler before it becomes useful for
> > > large gui applications [...]
>
> I agree on Java being nicer in design in a few areas than C++. About
> good compilers/JIT: I haven't kept up with the latest, but personally
> I would not replace C++ by Java unless the difference in execution
> speed becomes only a few % (dependent on project, of course).
>
Well, I -think- that a highly 'MMC'-optimized Java implementation
is a large part of the plan. The system will suck otherwise.
That's what I've gotten out of the news of the last couple weeks,
am I hallucinating this?
--
Joe Cosby
Amiga Fanatic
Dion Wooning
>
> Well, I -think- that a highly 'MMC'-optimized Java implementation
> is a large part of the plan. The system will suck otherwise.
>
> That's what I've gotten out of the news of the last couple weeks,
> am I hallucinating this?
Well if you are, I'm hallucinating too. If the MCC wouldn't have
a speedy way to run Java bytecode, it would be madness to have the
OE base level consist of Java objects.
---
Dion Wooning
Eelke Blok
Dion Wooning wrote:
> Well if you are, I'm hallucinating too. If the MCC wouldn't have
> a speedy way to run Java bytecode, it would be madness to have the
> OE base level consist of Java objects.
Exactly. Should't we assume the guys at Amiga are competent enough
(incompetence is not exactly a term that comes to mind when looking at
the Amiga line-up) to recognize that Java is slow if you don't take
special measures and LAWKI is too unfriendly to be used as a consumer OS
without heavy work? And that they, consequently, are doing something
about it? Seems much more likely then those big names sitting there and
suddenly stop doing their jobs.
Michael Nielsen
>
> t...@ramjam.u-net.com wrote:
> >
> > In comp.sys.amiga.misc Michael Nielsen <m...@dde.dk> wrote:
> >
> > > Must say though that java
> > > as a language is a lot cleaner than C++ (objectoriented speaking),
> > > but it stille needs a good compiler before it becomes useful for
> > > large gui applications [...]
>
> I agree on Java being nicer in design in a few areas than C++. About
> good compilers/JIT: I haven't kept up with the latest, but personally
> I would not replace C++ by Java unless the difference in execution
> speed becomes only a few % (dependent on project, of course).
That was what I was saying - :->. I wasn't referring to jit, which goes
some of the way, but only works on really simple programs, such as servlets
and the like, can't handle gui work properly yet :-<.
regards
Michael Nielsen
>
> Just my 2pence worth... A DOS-Prompt box running in Windows is a Virtual
> Machine too...
If you really want to twist the virtual machine definition, then yes you could say
this, and so extend it to all OS's.
> There are also java chips too... processors that run bytecode natively
Sorry to provoke, where are these mythical processors currently used,
unfortunately I know of no commercial application of these as yet. The
idea is good, but they seem to have failed in the practical market.
mvh
mike.
Michael Nielsen
> >I was talking to one of they guys working on the teoretical jit's and
> >he "claimed" that it was new teknology, and I really think that this
> >is about the only really new thing java brings in..
>
> jit? New? Hardly. There used to be a pascal app that ran on a pdp-11
> that was supposed to be a 24/7/365 app. Sorry, no banana, that thing
> compiled to p-code on boot, and interpreted the p-code on the fly to
> run. Thats jit in my book. It also crashed at least daily, and took a
> 15 minutes just to boot. I miss it like a busted rib...
>
> jit? One of many 'buzzwords' to make somebody *look* as if *he* invented
> this particular wheel.
Yeah, well you did notice the "" around claimed, to emphasise, that I'm just
repeating what I was told (never bothered researching it though).
Erik Bruser
Michael Nielsen heeft geschreven in bericht <37B00E27...@dde.dk>...
hold your breath untill the bomb falls.........
Michael Nielsen
> >Sorry to provoke, where are these mythical processors currently used,
> >unfortunately I know of no commercial application of these as yet. The
> >idea is good, but they seem to have failed in the practical market.
> >
> >mvh
> > mike.
>
> hold your breath untill the bomb falls.........
Once again, I'm sorry to say, that one would suffocate :-<, I've been
waiting for that technology for a long time (it has application to what
I'm working on, but....).
mvh
mike.
Erik Bruser
Michael Nielsen heeft geschreven in bericht <37BBCED1...@dde.dk>...
>Erik Bruser wrote:
>
>> >Sorry to provoke, where are these mythical processors currently used,
>> >unfortunately I know of no commercial application of these as yet. The
>> >idea is good, but they seem to have failed in the practical market.
>> >
>> >mvh
>> > mike.
>>
>> hold your breath untill the bomb falls.........
>
>Once again, I'm sorry to say, that one would suffocate :-<, I've been
>waiting for that technology for a long time (it has application to what
>I'm working on, but....).
>
>mvh
> mike.
>
pessimistic approaches always seem more easy, try to believe.......