Re: Why MIDP, not PP? (was Re: MIDP 2.0 beta for palmOne devices now available)

What a great argument :-)

We have Palm Engineers, Sun Engineers, IBM and now a great author.

I have to admit I love a good row ;-)

Now to totally disagree with Eric...

Writing code and writing book about writing code are two completely
different things ;-)

Code on Nokia 6600 DOES run differently. Its a fact!

Code on different PJava VMs DOES run differently its a fact!

As Sun themselves said PJava code would not run in the same way on J9 VM as
on a pure CDC system due to the threading model.

As for the J2SE point. I seem to remember that J2SE 1.4 broke a whole lot of
J2SE 1.3 code. So thanks for proving my point Eric ;-) LOL

You try running a 3D engine on J2SE 1.4 and watch the graphical drop out as
you overload the triangle count from an ASE parser and tell me it runs the
same on J2SE 1.3.

It does not!

Furthermore yes there are coding techniques to reduce this but we all now
that Java is NOT portable. Only an absolute FOOL believes that.

It is quite clear that the leading developers target a subset of devices
like Nokia Series 60, Series 40, Series 30, Motorola or Sharp.

Then they develop explicitly for that subset of handsets. It is the only way
to develop highly performant games.

My point is that we INCREASE the lack of portability as we INCREASE the
differentiation in Virtual Machines.

My assertion is in no part a fallacy. Try and run a few of your examples out
of your book on different Virtual Machines. Try a Zaurus running CDC against
Internet Explorer 3.0 then come back to me and tell me I was wrong.

Sure you can REDUCE the damage by better coding techniques. For example you
can get round Garbage Collectors that are NOT compacting by coding in a
specific way so your memory get reallocated. Its an absolute pain in the ass
though and reduces turn around times. Great for those XP projects ;-)

I have read many book on Threading and I can tell you this it AINT portable
no matter how you code.

Peter



> From: Eric Giguere <ericgiguere@xxxxxxxxxxxxxxx>
> Reply-To: A mailing list for KVM discussion <KVM-INTEREST@xxxxxxxxxxxx>
> To: KVM-INTEREST@xxxxxxxxxxxx
> Subject: Re: Why MIDP, not PP? (was Re: MIDP 2.0 beta for palmOne devices
> now available)
> Date: Wed, 17 Dec 2003 13:23:51 -0500
>
> > I think IBMs virtual machine comment was interesting. If you are using
> > different Java Virtual Machines then this WILL result in different
> threading
> > models.
> >
> > For example even the Monty VM has a different threading model to the
> KVM.
> >
> > This leads unfortunately to UNPORTABLE code (see posting from Nokia 6600
> > developers)
>
> Code that depends on specific threading models is unportable, yes, but it's
> not the VM's fault. Everyone should spend some time reading Doug Lea's
> concurrency book to understand how to write portable concurrent code.
>
> > The future is ultimately in a totally portable CDC environment. Surely
> this
> > means locking down the Virtual Machine though doesn't it?
>
> Here's what I don't understand from that assertion... the threading model
> used by J2SE has changed over the years depending on the platform. What's
> different about that? If the J2SE threading model hasn't been locked down,
> why lock down the J2ME model?
>
> > I thought originally that CDC ran through a CVM?
> >
> > We really will loose portability if people all use different Virtual
> > Machines for their CDC ports.
>
> See J2ME Fallacy #3, which I've just updated to further clarify things:
>
> http://www.ericgiguere.com/j2me/fallacies.html
>
> Eric
>
> ===========================================================================
> To unsubscribe, send email to listserv@xxxxxxxxxxxx and include in the body
> of the message "signoff KVM-INTEREST".  For general help, send email to
> listserv@xxxxxxxxxxxx and include in the body of the message "help".

_________________________________________________________________
Hotmail messages direct to your mobile phone http://www.msn.co.uk/msnmobile

===========================================================================
To unsubscribe, send email to listserv@xxxxxxxxxxxx and include in the body
of the message "signoff KVM-INTEREST".  For general help, send email to
listserv@xxxxxxxxxxxx and include in the body of the message "help".