Weblogs Forum - Free As In Lawsuit

Articles |
News |
Weblogs |
Books |
Forums

Artima Forums | Articles | Weblogs | Java Answers | News

Sponsored Link •

Weblogs Forum
Free As In Lawsuit

39 replies on 3 pages. Most recent reply: Sep 30, 2010 2:32 PM by Standard IO

Welcome Guest
Sign In

Back to Topic List

Reply to this Topic

Search Forum

Threaded View


Previous Topic		Next Topic

Flat View: This topic has 39 replies on 3 pages [ « | 1 2 3 ]

John Zabroski

Posts: 272
Nickname: zbo
Registered: Jan, 2007

Re: Free As In Lawsuit

Posted: Aug 20, 2010 10:53 AM

java is already fragmented behind the firewall...

John Franey

Posts: 7
Nickname: jfraney
Registered: Dec, 2007

Re: Free As In Lawsuit

Posted: Aug 20, 2010 11:26 AM

> > > java has never been "open source", only fee free.
> >
> > These arguments sound like a WWF advertisement.
> >
> > Java has arguably been open source since day one. I
> > downloaded and read all of the source for JDK 1.0.2
> beta
> > back in '96. (Not "free as in GPL software", but open
> > source.)
>
> And, of course, since it's not GPL, both Harmony and
> Dalvik are in various degrees of trouble. If java were
> "open source", then forking in any way one wishes (which
> is what makes "open source" "open") would be a right. It
> isn't. Stallman was right. Oracle is proving it.

I think you are referring to the Java Trap: http://www.gnu.org/philosophy/java-trap.html
The proof to the Trap had come way long before Oracle's suit against Google. In fact, it is self-evident.
People who observe Stallman's ethics would not be writing Java code to begin with.

I believe what rankles people today has little to do with Stallan's definition of 'free'. None of the software here is free by his definition. By the way, funny to put Stallman on the same side with 'Open Source' on a patent suit. http://www.gnu.org/philosophy/open-source-misses-the-point.html. 'Open source' philosophy has bred this suit.

What is underscored by Oracle's suit, regardless of its rightness or outcome, is that open source is not free (as in freedom) and frankly, in solidarity with other developers, I am shocked, just shocked, that a corporation can willfully flaunt its profit motivation so... er..... openly. Things were better when we could shout with our hands cupped over ears while Stallman spoke to us of freedom. :)

Cameron Purdy

Posts: 186
Nickname: cpurdy
Registered: Dec, 2004

Re: Free As In Lawsuit

Posted: Aug 20, 2010 11:44 AM

> IIRC, in all previous releases, the spec was either before
> or contemporaneous with the reference implementation, i.e.
> the jdk. The fact that Sun/Oracle didn't do that, and the
> "control" of the JCP, leads many to be suspicious. This
> lawsuit simply confirms the suspicions.

You are right that this time around is a little different. That's because the JCP (including Oracle) was blocking Java 7 in an attempt to force Sun to the table on the licensing issue. Oracle apparently has inherited that issue, and has yet (to the best of my knowledge) to resolve it.

In the meantime, Sun pushed the items that they expected to be in Java 7 into the open source JDK, i.e. Open JDK (GPL licensed). This is the first time that the JDK was "out in front of" the specification, so you are correct that there is something different about how Java 7 is coming out thus far.

Nonetheless, as I said before, under the current process, it cannot be Java 7 until the JCP says it is. So JDK 7 is not (yet?) Java 7.

Previous iterations would have a beta of the JDK, along with specifications, etc. Normally, and in addition to any changes in the JVM or JLS references, the Java specification would simply reference existing, approved specifications from other JCP JSRs that had already been approved and released, i.e. a J2SE release would simply adopt them by reference.

Regarding suspicion, there is certainly a lot of room for Oracle to improve its communication over the direction of Java, the JCP, licensing, etc. I continue to encourage this internally, although to date largely unsuccessfully.

Lastly, both for prudent (company policy) and for legal reasons, I cannot discuss or comment on the lawsuit.

Peace,

Cameron Purdy | Oracle Coherence
http://coherence.oracle.com/

John Franey

Posts: 7
Nickname: jfraney
Registered: Dec, 2007

Re: Free As In Lawsuit

Posted: Aug 20, 2010 12:48 PM

> So, are there clear answers to these questions:
> * Is non-Android Java (e.g. server-side) in trouble, and
> why?
> * Are non-Java JVM languages, e.g. Scala or Clojure, in
> trouble, and why?

Here is why Google is being sued:
1) Android competes against Oracle's Java handheld JVM.
2) Allegedly, Android was developed with aid from former SUN JVM developers.
3) Allegedly, when the JVM Developers were SUN employees, they filed some JVM related patents which are now owned by Oracle.
4) Allegedly, the inventions protected by these patents were used in the Android implementation.
5) Google is projected to earn 10 Billion USD from Android.

Scala and Clojure does none of this. Even if Oracle wins, Scala, Clojure and every one of millions of Java applications are as safe as they had been before this suit started.

robert young

Posts: 361
Nickname: funbunny
Registered: Sep, 2003

Re: Free As In Lawsuit

Posted: Aug 20, 2010 12:50 PM

While I was out at Panera eating my French croissant and reading that Left Wing NY Times, I mind-mumbled to the history of WebSphere, which goes something like this (from memory):

- the servlet spec and RI (tomcat) are released; Sun explicitly says that the RI is not to be used for production web servers

- the jsp spec and RI (jasper) are released; again, with not for production

- WebSphere appears (or between?); I don't recall that the servlet part is explicitly tomcat (but I think so), but I do recall that the jsp implementation is jasper. both are later replaced with an IBM engine

So, what happens to these? Here's Colebourne from last Friday:

It would therefore be wise to consider there to be no guarantees that open source implementations of other JCP specifications can continue. This includes the whole of Java EE, including JPA, JMS, JSF, Servlets, JAXB, JAXRS and high profile projects including Tomcat, Geronimo, HornetQ and Hibernate.

Would Oracle be crazy to block these open source implementations? Yes!
But being crazy seems to be the order of the day.

John Franey

Posts: 7
Nickname: jfraney
Registered: Dec, 2007

Re: Free As In Lawsuit

Posted: Aug 20, 2010 1:02 PM

> > So, are there clear answers to these questions:
> > * Is non-Android Java (e.g. server-side) in trouble,
> and
> > why?
> > * Are non-Java JVM languages, e.g. Scala or Clojure, in
> > trouble, and why?
>
> That's what's causing the angst. No one really knows. If
> you're using all Sun/Oracle java bits, then likely not in
> danger of not being able to ship your application. BUT,
> if Oracle succeeds, then the next step is to say to all of
> those folks I just mentioned, "you're using our patents
> and copyrighted code, so you owe us 1% of your gross
> sales". That's the fear, uncertainty, and doubt. And
> precisely the way Larry behaves.

I don't see Larry's contribution to this discussion here at Artima. I guess I can wait to see him speak for himself.

You have glossed the details of a complex case in its earliest stage. You are speculating a future that has no basis in fact. You have half-attributed to Larry a quote he did not make. You should not be concerned so much about how Larry behaves. You should be concerned about what you write. You are propigating FUD yourself, perhaps inadvertently.

robert young

Posts: 361
Nickname: funbunny
Registered: Sep, 2003

Re: Free As In Lawsuit

Posted: Aug 20, 2010 1:51 PM

> I don't see Larry's contribution to this discussion here
> at Artima. I guess I can wait to see him speak for
> himself.

I didn't assign the futuristic statement to any one person, only to the corporation. You may believe that Larry wouldn't direct the corporation to squeeze others, if the suit is won; I disagree.

>
> You have glossed the details of a complex case in its
> earliest stage.

Not that complex. There are 2 cases that matter:
- in following the spec, Google had to implement a function to the spec; which function is covered by patent/copyright
- Google lifted some body of code from Oracle/Sun to do something unrelated to implementing the spec

If it's the first case, then we'll see how our court system works: does it support such patent traps? If the second, Google folks are blithering idiots.

The third case is that all the patents (we don't yet know what the copyright is all about) are ruled obvious/prior art. That outcome is really not complex.

> You are speculating a future that has no
> basis in fact. You have half-attributed to Larry a quote
> he did not make.

I didn't say he'd already said that. The suit (since Larry is CEO, he can reasonably be considered the speaker) calls for the impounding and destruction of Dalvik, et al. As to Oracle/Larry's behaviour toward the clients of acquired companies, there's a rich record of squeezing of gonads; facts from which a future fact may well be deduced.

> You should not be concerned so much
> about how Larry behaves. You should be concerned about
> what you write. You are propigating FUD yourself, perhaps
> inadvertently.

Cameron Purdy

Posts: 186
Nickname: cpurdy
Registered: Dec, 2004

Re: Free As In Lawsuit

Posted: Aug 20, 2010 2:27 PM

Hi Robert -

I suppose by some definition of an interpreted language, Java would fit.

However, I have never (to the best of my knowledge) referred to "interpreted languages". Rather, IMHO, the term "interpreted" refers to a runtime execution model that uses an interpreter. An interpreter is, as I choose to understand it, a software device that executes software in much the same way that a CPU is a hardware device that executes software.

Saying that Java is interpreted makes little sense to me. Consider:

* Java, the language, does not imply either interpreted execution or JITted execution or HotSpotted execution or pre-compiled-to-native execution. However, it does support all of the above.

* C, the language, was intended for a compile-link-execute model (i.e. native code), yet there are several C interpreters in widespread use. (For many years, MSVC++ had it as a check-box option to use pcode instead of generating the then much-larger x86 code.)

Java was obviously designed to support an execution model in which the Java-compiled application runs as native code. It was also obviously designed to support a simple stack-based interpreter execution model. To say that it is an "interpreted language" seems a specious argument at best, since the language is always compiled (to a .class), and it is the "linking" step that varies widely.

Peace,

Cameron Purdy | Oracle Coherence
http://coherence.oracle.com/

Max Lybbert

Posts: 314
Nickname: mlybbert
Registered: Apr, 2005

Re: Free As In Lawsuit

Posted: Aug 24, 2010 1:10 PM

> So, are there clear answers to these questions:
> * Is non-Android Java (e.g. server-side) in trouble, and why?

Only Android should be vulnerable to the copyright claim, if anybody is. You can't infringe copyright without first having access to the copyrighted material.

The patent claims could be leveled at anybody who isn't covered by the Sun/Java patent pledge -- which requires an implementation that does what the spec says and no more. OpenJDK does more than the spec, interestingly.

> * Are non-Java JVM languages, e.g. Scala or Clojure, in trouble, and why?

They should be safe because they don't implement any of the patents. Instead they run on JVMs which implement the patents. And the copyright claims sinply don't make any sense w.r.t. Scala or Clojure (or JRuby, or etc.).

Standard IO

Posts: 2
Nickname: stdio
Registered: Feb, 2006

Re: Free As In Lawsuit

Posted: Sep 30, 2010 2:32 PM

Dmitry Cheryasov:
"To tell the truth, MS Research does a lot of great innovative things, in C# and F# in particular."

if by innovative you mean copying-pasting more and more features from dynamic languages into the Delphi/Java bastard child and quite a few fine polish into M$ Ocaml, yeah, I agree.

As far as I understand, Microsoft Research is a bunch of old haskell developers hired to try to cram some of that goodness into VB users toolset... a very annoying task I guess...

Flat View: This topic has 39 replies on 3 pages [ « | 1 2 3 ]

Previous Topic

Next Topic