Python finally succeeds in cross-platform areas where Java has beenfailing...

[Jonathan P.]

....that would be for desktop-based apps, games, 3d graphics,
and multimedia.

....thanks to APIs and bindings like Pygame, PyOpenGL, PyGtk
and PyGtkGLExt.

A summary of a lengthy post on the subject:

http://lists.free.net.ph/pipermail/compsci/2003-October/001510.html

[color=blue]
> - OpenGL for accelerated graphics.[/color]
(that's PyOpenGL)
[color=blue]
> - SDL for cross-platform sound, device input, etc...[/color]
(via Pygame)
[color=blue]
> - Gtk for excellent themable cross-platform widgets (make your
> Gtk apps look just like Win32 apps)[/color]
(via PyGtk)
[color=blue]
> - GtkGLExt for accelerated OpenGL display inside widgets. Yow![/color]
(via PyGtkGLExt)

One caveat is that while with .pyc files, Python has what are
essentially platform independent executables ala Java
class files, because the libraries mentioned are not yet a
part of the standard Python distribution, you have to
install the modules and dlls separately.

But if the Gtk and libSDL dlls were to one day come bundled
with Python (like Tcl/Tk is now), and with Psyco to provide JIT and
Python C extensions (being easier to deal with than JNI especially
with the help of SWIG) in the mix, who needs Java indeed?

[Jerome Alet]

Re: Python finally succeeds in cross-platform areas where Java has been failing...

Le Wed, 15 Oct 2003 05:09:25 +0800, Jonathan P. a écrit :
[color=blue]
> who needs Java indeed?[/color]

Sun.

;-)

Jerome

[Cameron Laird]

Re: Python finally succeeds in cross-platform areas where Java has been failing...

In article <TMcjb.219243$h E5.7385752@news 1.tin.it>,
Alex Martelli <aleax@aleax.it > wrote:

[Roy Smith]

Re: Python finally succeeds in cross-platform areas where Java has been failing...

How much of Python's cross-platform advantage over Java is due to the
single source of the Python interpreter and how much is due to the
language itself?

There's certainly nothing (that I'm aware of, other than time, money,
and skill) to stop me from writing my own Python interpreter, making
some subtle changes in behavior (either intentionally or by accident)
and promoting it in the field. If I could convince enough people to
install it, we'd suddenly have a cross-platform crisis in the Python
world.

This is really all that's happened in the Java world. The above
scenario describes what Microsoft did to Java. The only difference
between me doing my own Python and Microsoft doing their own Java is
that Microsoft has the resources and desire to pull it off.

What would happen if Microsoft saw Python as a threat and decided to
kill it by shipping their own incompatable Python interpreter with
Windows? Would we have any defense?

[Harry George]

Re: Python finally succeeds in cross-platform areas where Java has been failing...

Roy Smith <roy@panix.co m> writes:
[color=blue]
> How much of Python's cross-platform advantage over Java is due to the
> single source of the Python interpreter and how much is due to the
> language itself?
>
> There's certainly nothing (that I'm aware of, other than time, money,
> and skill) to stop me from writing my own Python interpreter, making
> some subtle changes in behavior (either intentionally or by accident)
> and promoting it in the field. If I could convince enough people to
> install it, we'd suddenly have a cross-platform crisis in the Python
> world.
>
> This is really all that's happened in the Java world. The above
> scenario describes what Microsoft did to Java. The only difference
> between me doing my own Python and Microsoft doing their own Java is
> that Microsoft has the resources and desire to pull it off.
>
> What would happen if Microsoft saw Python as a threat and decided to
> kill it by shipping their own incompatable Python interpreter with
> Windows? Would we have any defense?[/color]


Sure, the std open source software (OSS) defenses:

1. If the alternative is open source itself, then std OSS rules of
engagement apply. Thus, if the alternative has a few good ideas,
they will be absorbed by the main line. If the whole alternative
is better, the community may shift over en masse (a la gcc a few
years ago). If not, everyone will know it and will avoid the
alternative. Legal use of the Python name is (I think) up to PSF.

2. If the alternative implementation (including all libraries and all
extensions) was completely green-room, then it could perhaps be
kept closed source. In that case it might take a while to realize
it was incompatible. For a bad enough mismatch, the name "Python"
might be withheld, just as SUN fought MS's use of "Java" for
J++. When word got out, very likely there would be a backlash, and
people would avoid it if they could.

3. "avoid it if they could" is the crux of the issue. If MS
orchestrates DRM, Palladium, etc. so that only MS-owned languages
can play on a MS Win** box, then MS might offer something like
python functionality (e.g., that is the sales pitch for C#). Under
these circumstances, it is up to the buyer to beware of lockins.
So long as PSF doesn't authorize MS use of "Python" for that
purpose, there will still not be a split.


--
harry.g.george@ boeing.com
6-6M31 Knowledge Management
Phone: (425) 342-5601