Free Java advocacy (was Re: Improving Java for Linux)

[Jim Pick]

Tony Kimball <alk at pobox.com> writes:

> No, they were not.  There are so far 6 votes in favor of the
> application to be a standard submitter, and one against.  There are 
> 8 countries remaining to vote.  

I spend a few hours reading the submissions to ISO JTC 1 TAG about 6 months
ago.  I don't think Sun's position has changed much since then.

Since JTC1 only represents one vote, you are probably right.  It's a
sad state of affairs if it goes through anyhow.  ISO will definitely be
tarnished, in my mind. 

> Really, people, (and this is not specifically addressed to Jim Pick)
> if you are going to get your news from trade magazines supported
> primarily by Microsoft, you have to be careful to separate the
> factual claims from the spin.

I did actually read some comments on the www.jtc1.org web site a few
months ago.  Very interesting.  I found myself agreeing with the
companies that were against the proposal.  The arguments for the
proposal were very weak indeed.

Read this:

http://www.news.com/News/Item/0,4,15759,00.html

(I don't know if I'd call News.com a 'trade magazine supported primarily
by Microsoft')

The article does say the number of votes isn't important.  It's just
a guideline to figure out if there is consensus.  I'd say there isn't --
but that's not my call.

> : The main problems were that Sun
> : wanted to retain some measure of control, including control over the
> : Java trademark.
> 
> "Including" meaning "specifically".  I think this was a very bad
> pretext for the US to vote against Sun's petition, since they don't
> have to call it Java.  They could call it "Ishmael" for all I care.

There were lots of other objections.
 
> : This is primarily due to the fact that ISO is controlled
> : by corporate entities - most of whom have cut deals with Sun.
> 
> Are you stating facts, or attempting to smear Sun?

Facts.  Don't get me wrong - I like Sun more than I like Microsoft,
but they're all playing the same game.

> : They have no plans to "give away" Java.
> 
> Java source code, not for commercial use or redistribution.  Java
> trademark, no.  Java the language they gave away years ago.  And they
> keep giving away API designs.  I don't see what there is to complain
> about, frankly.

That's not exactly giving away control, which is what I meant.
 
> : I think they may push the core Java spec and language out to a standards
> : body beyond their control.  
> 
> Very shortly, in my opinion, since I believe they can get a majority
> ISO vote.

But a majority of the vote does not represent a consensus - so that might
not be good enough.

> : Essentially, Sun will use the standards process to cement their monopoly.
> 
> What monopoly?  Kaffe exists.  JV-Lite exists.  Your boogeyman is not
> credible.  Sun can't create a standard -- they can only submit a
> standard proposal, whereupon the ISO can do what it likes with it.

That's not what Sun is proposing.  Sun is proposing that their JavaSoft
division should actually be certified as a standards body.  That's really,
really bizarre.  ISO has done this before with consortiums - but never
with a division of a private company.

> : I do think we should deviate from the extended APIs Sun is 
> : advocating beyond that, however.
> 
> Let's get that far first -- cross that bridge after the river gets built.
> Until we have a free implementation of the basic portability layers, 
> a free implementation is not generally useful.

We're pretty close.  I've seen reports of Kaffe running with kore.  There
should be another release of kore in a few days (or weeks).

> : [GnuStep, LessTif, Wine...] You'll notice that every
> : one of these projects, although successful, have only been able to 
> : clone APIs that are several years behind the state of the art. 
> 
> Like Linux or FreeBSD?  But Linux and FreeBSD are successful.

But Unix implentations have been around since the '70s.  It's taken
about 15 years and many, many hundreds of programmers for the GNU project to
get to this point.

> Hey, I have no objection to the design and implementation of purely
> freeware APIs.  When you produce something useful, I'm sure it will be
> used.  But until then, I think the focus should be on meeting the
> immediate, non-controversial, concrete barriers to basic minimal 
> java execution in a free environment.

That's what I'm trying to say.  I don't think JFC or JavaBeans are
necessary for "minimal java execution".

Take a look at the JFC and JavaBeans APIs.  They're HUGE.  At this
point, the 20 or so hackers who have done real work on free Java
implementations have only got a fraction of the base JDK cloned.
Overall, it's a small percentage of APIs that Sun has available
(or planned) for Java.  I'd like to see everything cloned - but
that's just fantasy.

> Who has Sun sued?  Microsoft.  Name one other party.  You really do
> seem to be intent on smearing Sun.

They have threatened to sue literally hundreds of people who were using
the Java name.  The lawyers seem to get first crack at everything they
put out - it's like the legal department runs the entire show there.
That's just my personal impression -- I have no facts to back that up.

> Microsoft signed a contract to implement JNI and RMI.  Microsoft never
> agreed to implement the media engine or 3d or any of the other
> non-core APIs, so they don't get sued for violating their agreement on
> those points.  I don't think that being in agreement with Microsoft
> is a point in favor of your views.

I'm not sure about the nitty-gritty details of their license.  I don't
really care.  I just find it funny (hilarious even) that Sun is trying
to bully Microsoft into pushing their APIs.  Microsoft has the market
share, and owns the OS, and mostly wanted to just use Java as an
excellent tool to integrate some of their desktop environment together.
It really is a natural fit with COM - much better than using C or C++.
Sun wants them to follow their agenda instead.  Good luck.

Anyways, we have the same problem with our free Java implementation.
Do we follow Sun's agenda to the ends of the earth?  Or do we say,
"Thanks for the clever language.  We'll start using it now.  We'll
 take a look at your other stuff, and maybe well use it later..."

> If you want agreement with your points, you should refrain from using
> specious arguments in their favor.  I agree with your concrete points,
> when I take the time to consider them in isolation from your
> arguments.  But the rhetoric of the arguments could easily have put me
> off.

The rhetoric is important.  Otherwise people will just blindly follow
Sun's lead, and trap themselves in a proprietary situation.  Fact is,
we almost have a totally free Java implementation that works.  But I'm
concerned that nobody will even use it - because it doesn't implement
everything Sun has delivered.
 
> :  3) We should work to build free class libraries that integrate with other
> :     software in the GNU platform.  These would be original, and may
> :     interface with native methods.  This represents a "divergence" from
> :     Sun's plan for Java.
> 
> That's what JNI is for.  Use JNI and the java code can run on any
> platform which supports the native code.  Use something else, and your
> java code won't run on any other VM.  (And foot-shooting such as
> this is just the sort of danger which an unreasonable anti-Sun bias 
> can cause.)

I agree we should support JNI.  That doesn't seem too hard to implement.
 
> :  4) We should advocate that free software projects should only use APIs
> :     which are free, or for which there is a substantial effort underway
> :     to clone them.  Currently, JFC or JavaBeans don't meet that spec.
> 
> JFC isn't free?  I downloaded it with no licensing agreement.

Of course it isn't free.  If there is no licensing agreement, that means
you aren't even really allowed to use it (or download it).  

I'd suggest stopping by the www.fsf.org website, and reading everything you
see.  It does a pretty good job of explaining the difference between
free as in "free beer" vs. free as in "freedom".

Anyways, thanks for replying.  I might not agree with all your points --
but I just wanted to argue anyways.  :-)

Cheers,

 - Jim
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 345 bytes
Desc: not available
Url : http://kaffe.org/pipermail/kaffe/attachments/19971107/ca35fd97/attachment-0003.pgp