The Oracle/Google lawsuits, and how it affects choice of language

[Screamin Lord Byron]

Not if they can get a judge to decide that it violates their patents, and
that all copies of it must be destroyed in consequence, which is what
they're trying with Android.

If that's true then they are more than stupid. I don't believe they are.
The reason why they attacked Google is because Google avoided paying
Java ME royalties by creating their own "Java ME", so to speak. Sun
wasn't happy about it either but decided to do nothing. But then again,
Sun is no more.

I wonder if the public reaction would have been so explosive if Sun
filed the same complaint.

 

[Eric Sosman]

]

And Oracle does not have any obligation to anyone except their
stockholders.

This is true. Are you prepared to bet your career on a language owned and
controlled by a company which, as you say, does not have an obligation to
you?

That's why nobody uses C#.

 

[Mike Schilling]

This is true. Are you prepared to bet your career on a language owned and
controlled by a company which, as you say, does not have an obligation to
you?

I'm not willing to bet my career on any single language. Anyone who does is
a fool.

 

[Lew]

I'm not willing to bet my career on any single language.
Anyone who does is a fool.

Vous avez raison.

 

[Arne Vajhøj]

This is true. Are you prepared to bet your career on a language owned and
controlled by a company which, as you say, does not have an obligation to
you?

Nonsense.

Java is language.

OpenJDK is an implementation of that language.

Java is healthy.

OpenJDK may not be healthy.

Not if they can get a judge to decide that it violates their patents, and
that all copies of it must be destroyed in consequence, which is what
they're trying with Android.

If there are something that requires a patent license, then that
would need to be replaced with something that does not.

Note that JCP rules prevent Oracle from using the patent weapon for
patents that any Java implementation needs.

Arne

 

[Arne Vajhøj]

If that's true then they are more than stupid. I don't believe they are.
The reason why they attacked Google is because Google avoided paying
Java ME royalties by creating their own "Java ME", so to speak. Sun
wasn't happy about it either but decided to do nothing. But then again,
Sun is no more.

I wonder if the public reaction would have been so explosive if Sun
filed the same complaint.

Maybe not quite as upset, but probably still some.

Speaking about Java ME. Then a lot of people think the evolution
in HW has made Java ME obsolete.

To me it would make a lot of sense if Google and Oracle made
a deal that would make Dalvik the latest Java ME!

Arne

 

[Jukka Lahtinen]

Not if they can get a judge to decide that it violates their patents, and
that all copies of it must be destroyed in consequence, which is what

Is that possible?
I'm not sure about the validity of software patents everywhere in the
world.

 

[Joshua Cranmer]

Is that possible?
I'm not sure about the validity of software patents everywhere in the
world.

In the U.S., the patentability of software is a somewhat open question.
The only relevant Supreme Court cases are Diamond v. Diehr (which says
that software per se is not patentable, but its presence does not render
a patent unpatentable) and Bilski v. Kappos (which says that the State
St. decision is a load of bullshit, the machine-or-transformation test
is on crack, software patentability is left explicitly unexplored in the
decision). A lot of software patents rested on the State St. decision
(which says that anything that "produces a useful, concrete, or tangible
result" is patentable)--which the SCOTUS struck down in Bilski.

I'm not a patent lawyer, but my understanding is that the boilerplate in
most current software patents is now invalid (resting on an untenable
decision). I think, however, that it is possible to amend the patents to
change that boilerplate or something, so current software patents are
not, in effect, blanket-invalidated by any of the recent decisions.

I am not aware of the binding case law in the EU or other jurisdictions.
I do know that the EU in theory prohibits software patents, although the
actual practical application of patent law does appear to admit software
patents.

Given all of the patent treaties and general legal atmosphere, I suspect
that if a software patent were actually successfully used against a
company in a major jurisdiction, the company would be required to
destroy the infringing code.

But, in the end, IANAL.

 

[Screamin Lord Byron]

If there are something that requires a patent license, then that
would need to be replaced with something that does not.

That's the tricky part. How does one do that? It seems to me that in the
USA everything is patented. It wouldn't surprise me to find a patent
which describes how to breathe.

On the other hand, as I understand, many of those patents are quite
vague and lack appropriate technical rigour, but still may seem rigorous
and precise enough for an average judge who makes the verdict at the end
of the day.

This James Gosling's words pretty much sum it up:
"IBM sued [us] over a RISC patent that asserted that 'if you make
something simpler, it'll go faster.'...and we lost."

Note that JCP rules prevent Oracle from using the patent weapon for
patents that any Java implementation needs.

Exactly. In the Google case, Oracle relies on the patent infringement
case as a legal loophole which may or may not work. They have nothing
else, since Google didn't violate any licence agreements.

 

[Screamin Lord Byron]

Screamin Lord Byron wrote:

I'm not really sure I think Android has been a net positive for the open
source community. It certainly hasn't been a net negative,

<...>

Yeah, OK. I could probably agree with that, but I wasn't speaking about
Android's contribution to the FLOSS community. I was just saying that
Google did a very good job with their embedded platform. It's easily
portable and very usable to say the least. And it is an open source.

 

[Screamin Lord Byron]

Speaking about Java ME. Then a lot of people think the evolution
in HW has made Java ME obsolete.
Yup.

To me it would make a lot of sense if Google and Oracle made
a deal that would make Dalvik the latest Java ME!

Good point. Who knows, maybe Oracle made such suggestion to Google which
they rejected. This lawsuit could be a way of forcing it.

 

[Arne Vajhøj]

Good point. Who knows, maybe Oracle made such suggestion to Google which
they rejected. This lawsuit could be a way of forcing it.

Unfortunately Larry does not consult me ...



Arne

 

[Arne Vajhøj]

If there are something that requires a patent license, then that
would need to be replaced with something that does not.

That's the tricky part. How does one do that? It seems to me that in the
USA everything is patented. It wouldn't surprise me to find a patent
which describes how to breathe.

On the other hand, as I understand, many of those patents are quite
vague and lack appropriate technical rigour, but still may seem rigorous
and precise enough for an average judge who makes the verdict at the end
of the day.

This James Gosling's words pretty much sum it up:
"IBM sued [us] over a RISC patent that asserted that 'if you make
something simpler, it'll go faster.'...and we lost."

Software patents are indeed a quagmire.

But it is not any different for Android than any other type
of software.

Arne

 

[Tom Anderson]

]

And Oracle does not have any obligation to anyone except their
stockholders.

This is true. Are you prepared to bet your career on a language owned and
controlled by a company which, as you say, does not have an obligation to
you?

That's why nobody uses C#.

Or COBOL. Or PL/SQL.

tom

 

[Martin Gregorie]

On Mon, 23 Aug 2010 09:49:27 -0400, Arne Vajh?j wrote: [...]
And Oracle does not have any obligation to anyone except their
stockholders.

This is true. Are you prepared to bet your career on a language owned
and controlled by a company which, as you say, does not have an
obligation to you?

That's why nobody uses C#.

Or COBOL. Or PL/SQL.

I agree about PL/SQL (SQL suffers badly from bastardisation by various
RDBMS vendors) but not COBOL. The COBOL language specification is now
controlled by the ANSI committee responsible for adopting changes to the
language standard rather than any company as the successor to the CODASYL
committee. Unfortunately it does have areas that are not totally
portable, e.g. the COMP-n field types, but last time I used it, it was
considerably more standardised than when I learnt it. However, that puts
it on the same footing as C and Fortran.

Sadly, the company-controlled languages include Java, Delphi and RPG as
well as C#.

I'd like to see Java adopted by ANSI or an equivalent body but I'm not
holding my breath.

 

[Mike Schilling]

On 8/23/2010 2:18 PM, Simon Brooke wrote:
On Mon, 23 Aug 2010 09:49:27 -0400, Arne Vajh?j wrote: [...]
And Oracle does not have any obligation to anyone except their
stockholders.

This is true. Are you prepared to bet your career on a language owned
and controlled by a company which, as you say, does not have an
obligation to you?

That's why nobody uses C#.

Or COBOL. Or PL/SQL.

I agree about PL/SQL (SQL suffers badly from bastardisation by various
RDBMS vendors) but not COBOL. The COBOL language specification is now
controlled by the ANSI committee responsible for adopting changes to the
language standard rather than any company as the successor to the CODASYL
committee. Unfortunately it does have areas that are not totally
portable, e.g. the COMP-n field types, but last time I used it, it was
considerably more standardised than when I learnt it. However, that puts
it on the same footing as C and Fortran.

Sadly, the company-controlled languages include Java, Delphi and RPG as
well as C#.

C# is standardized by ECMA, and there are conforming non-Microsoft
implementations of it. What makes C# proprietary is the .NET framework, the
most interesting parts of which are not in the public domain.

 

[Tom Anderson]

On Mon, 23 Aug 2010, Eric Sosman wrote:

On 8/23/2010 2:18 PM, Simon Brooke wrote:
On Mon, 23 Aug 2010 09:49:27 -0400, Arne Vajh?j wrote: [...]
And Oracle does not have any obligation to anyone except their
stockholders.

This is true. Are you prepared to bet your career on a language owned
and controlled by a company which, as you say, does not have an
obligation to you?

That's why nobody uses C#.

Or COBOL. Or PL/SQL.

I agree about PL/SQL (SQL suffers badly from bastardisation by various
RDBMS vendors) but not COBOL. The COBOL language specification is now
controlled by the ANSI committee responsible for adopting changes to the
language standard rather than any company as the successor to the CODASYL
committee.

C# is standardized by ECMA, and there are conforming non-Microsoft
implementations of it. What makes C# proprietary is the .NET framework,
the most interesting parts of which are not in the public domain.

The same applies for COBOL: i'd be surprised if there were any significant
COBOL applications built entirely using the portable API, and using no
CICS or other IBM goodies.

tom

 

[Lew]

Martin Gregorie wrote

The same applies for COBOL: i'd be surprised if there were any
significant COBOL applications built entirely using the portable API,
and using no CICS or other IBM goodies.

Even where standards exist, and even where free versions exists, e.g., C, Java
and SQL, customers occasionally purchase quite expensive implementations
(Oracle's, IBM's), and use the proprietary extensions (Oracle Streaming).
Counterbalancing the risk of vendor lockin is vendor support and the power of
enhanced features.

I don't recall everyone going into this kind of tizzy over Sun's suit against
Microsoft for misusing the Java intellectual property. Everyone cheered Sun
then. Now when the new steward of Java does pretty much the same thing we all
get our knickers in a twist.

 

[Martin Gregorie]

The same applies for COBOL: i'd be surprised if there were any
significant COBOL applications built entirely using the portable API,
and using no CICS or other IBM goodies.

What about all those using Microfocus COBOL on Unix boxes? Not a lot of
CICS or DB2 there...

As for the CICS folks, when you app runs on big iron, you use the
ironmongery. Mumble, mumble ancient, undocumented, systems that nobody
understands well enough to port elsewhere or even to an RDBMS...

 

[Martin Gregorie]

I don't recall everyone going into this kind of tizzy over Sun's suit
against Microsoft for misusing the Java intellectual property. Everyone
cheered Sun then. Now when the new steward of Java does pretty much the
same thing we all get our knickers in a twist.

True enough, but then again Sun had a much more innovative, friendly
image than Oracle ever has. IME that sort of thing has influence.