Oracle suing Google over Java use in Android

You've got to laugh at this. How many times have we heard free software weenies telling people to avoid using Mono because Microsoft might sue for patent infringement (despite the core being ECMA standardised and with a free patent grant attached)?

In fact I remember not so long about on this very forum a well-known troll proclaiming the benefits of Java and how great it was that Android used it and I replied that Andoid doesn't use Java - it uses a custom VM and bytecode to get around the Java licence costs for mobile use. Well it looks like that's come back to bite Google. They should have gone with a .NET based solution.

Oh, another interesting fact is that the licence agreement for Andoid specifically excludes any liability on Google for patent claims, so it's not like they didn't see this coming. All those free Andoid licences will become rather less free if Oracle win (or more likely Google settles) as all the phone manufacturers using Android will have to pay too.

When Java went open source, the license it went under (like most FOSS licenses) includes a patent grant. Oracle will have to have some good arguments on how this is invalid in the case of Google.

Java pretty much belongs to the "community". And by "community", I mean major multinational supercompanies like IBM, Google, VMware, etc. Oracle is going to have a difficult time putting this cat back into the bag if that's their goal. It's almost like AT&T trying to assert rights over C-like languages at this point.

Bass said:

When Java went open source, the license it went under (like most FOSS licenses) includes a patent grant. Oracle will have to have some good arguments on how this is invalid in the case of Google.

 

Java pretty much belongs to the "community". And by "community", I mean major multinational supercompanies like IBM, Google, VMware, etc. Oracle is going to have a difficult time putting this cat back into the bag if that's their goal. It's almost like AT&T trying to assert rights over C-like languages at this point.

If Sun  made a patent grant, it'll only apply to implementations that support one of Sun's 'standards'. You remember how obsessed Sun were with implementation complying with their standards? Android's implementation of Java is highly non-standard and in any case, the free licencing didn't cover the one profitable bit of Java: the Java mobile stack.

Bass said:

When Java went open source, the license it went under (like most FOSS licenses) includes a patent grant. Oracle will have to have some good arguments on how this is invalid in the case of Google.

 

Java pretty much belongs to the "community". And by "community", I mean major multinational supercompanies like IBM, Google, VMware, etc. Oracle is going to have a difficult time putting this cat back into the bag if that's their goal. It's almost like AT&T trying to assert rights over C-like languages at this point.

The way I understand this is, if you use the standard JVM, it's free to use it.  If you modify the JVM (as in not byte code compatible) you either have to release your code as GPL in which case you get the patent grant, else you want to stay closed source,  you pay a license fee to Oracle.  As the Davlik VM has dome neither of these, it's open to patent attack.
 
I quite surprised that there aren't more people siding with Google over this, based on the comments.

rhm said:

Bass said:

*snip*

If Sun  made a patent grant, it'll only apply to implementations that support one of Sun's 'standards'. You remember how obsessed Sun were with implementation complying with their standards? Android's implementation of Java is highly non-standard and in any case, the free licencing didn't cover the one profitable bit of Java: the Java mobile stack.

Oracle might have a case, but I don't think this will be a slam dunk. In the case of Sun vs Microsoft, AFIAK Microsoft and Sun had some kind of formal contract regarding the use of Java. Plus, the rights on Java where far more well defined in the 90s then they are now. So Oracle has their work cut out for them.

At the same time, they are risking their PR. As you said, the "Java troll" was promoting the opennness of Java. That is not uncommon. The alleged openness of Java is often touted as a major advantage of Java. By big companies (IBM, VMware) even. This will negatively affect the perception of Java, which is not a good thing for Oracle. Java shops even before the Sun acquisition where major consumer's of Oracles database, application servers and middleware. This might make some Java shops look elsewere.

They aren't a monopoly in enterprise software / tools, they need to remember this. Not just Microsoft, but IBM and VMware too are happy to accept the massive license fees Oracle gets.

Bass said:

rhm said:

*snip*

Oracle might have a case, but I don't think this will be a slam dunk. In the case of Sun vs Microsoft, AFIAK Microsoft and Sun had some kind of formal contract regarding the use of Java. Plus, the rights on Java where far more well defined in the 90s then they are now. So Oracle has their work cut out for them.

 

At the same time, they are risking their PR. As you said, the "Java troll" was promoting the opennness of Java. That is not uncommon. The alleged openness of Java is often touted as a major advantage of Java. By big companies (IBM, VMware) even. This will negatively affect the perception of Java, which is not a good thing for Oracle. Java shops even before the Sun acquisition where major consumer's of Oracles database, application servers and middleware. This might make some Java shops look elsewere.

 

They aren't a monopoly in enterprise software / tools, they need to remember this. Not just Microsoft, but IBM and VMware too are happy to accept the massive license fees Oracle gets.

I don't think many Oracle shops will jump ship to SQL Server. There are a lot of differences in the setup and maintenance and in the usage. Even the terminology is different, e.g. schemas versus databases, databases versus servers, etc.

I think the public support for Google is waning as more people realize that they're just another big corporation.

spivonious said:

Bass said:

*snip*

I don't think many Oracle shops will jump ship to SQL Server. There are a lot of differences in the setup and maintenance and in the usage. Even the terminology is different, e.g. schemas versus databases, databases versus servers, etc.

 

I think the public support for Google is waning as more people realize that they're just another big corporation.

I'm not talking about MS-SQL. MS-SQL doesn't even run on Linux or AIX/Solaris. Do people actually run Java EE on Windows?

Anyway much to Oracle's dismay, wildly popular ORMs like Hibernate support many, many DBMSes. PostgresSQL is my perferred DBMS. WebLogic has competition in WebSphere, JBoss, SpringSource tC, Apache Geramino, etc. So it's pretty easy to have a Java EE infrastructure without ever doing business with Oracle. Ellison is cocky, but he got no leverge here. Oracle software is among the most expensive and widely used enterprise software, but none of it is a monopoly.

First, Microsoft.  Now, Google.  Who's going to be sued next by Sun/Oracle over Java?  Apple?

Bass said:

When Java went open source, the license it went under (like most FOSS licenses) includes a patent grant. Oracle will have to have some good arguments on how this is invalid in the case of Google.

 

Java pretty much belongs to the "community". And by "community", I mean major multinational supercompanies like IBM, Google, VMware, etc. Oracle is going to have a difficult time putting this cat back into the bag if that's their goal. It's almost like AT&T trying to assert rights over C-like languages at this point.

It's my understanding that Sun specifically excluded patent rights, and only handed over copyright.

Bass said:

Do people actually run Java EE on Windows?

That's a good question, and the answer is a most definite 'yes'. I used to contract for a finance company that used a number of Websphere servers running on Windows. And I think you'll find that even if they're deployed on Linux, most J2EE applications are developed and tested on Windows machines (in fact, I have only worked for one company that tried to develop J2EE applications on Linux - a worthwhile experiment that failed, unfortunately, due to problems with display drivers). 

Bass said:

Anyway much to Oracle's dismay, wildly popular ORMs like Hibernate support many, many DBMSes. PostgresSQL is my perferred DBMS. WebLogic has competition in WebSphere, JBoss, SpringSource tC, Apache Geramino, etc. So it's pretty easy to have a Java EE infrastructure without ever doing business with Oracle. Ellison is cocky, but he got no leverge here. Oracle software is among the most expensive and widely used enterprise software, but none of it is a monopoly.

Java is open source-ish but has never been free. Hard to believe, I know. Sun has always charged a license fee to commercial server vendors (IBM, Oracle etc.) for use of the J2EE spec. They have also charged for the compatibility tests which is why there are not so many implementations as there should be. It also explains why folk are cagey about calling their Java-like languages 'Java'. Now bear in mind this has only applied to commercial vendors; open source implementations are not subject to a charge (this wasn't the case for the compatibility suite, but I believe that has now changed).

The other thing is that Sun has always charged for is the implementation of Java on mobile platforms and I think this is where Google has tripped up. They may have believed that since they are giving away the implementation then they are not subject to a license fee.  

Here's a piece on the 'free java' fight dated from 2002. The interesting part concerns Oracle's early actions.

ZDnet said:

But despite the JCP's democratic look-and-feel, Sun, which owns the patents to Java, still retains veto power. Also, Sun's intellectual property stake in the technology allows the company to connect other requirements to Java and to those who want to profit from it. First and foremost of those are licensing fees that Sun collects from vendors who attach the Java brand to their products (i.e.: IBM's WebSphere or BEA's WebLogic). The actual fees that Sun charges to those vendors are one of the IT industry's best-kept secrets. Sun refuses to disclose what it charges Java licensees, and the licensees never tell.

But evidence suggests that not everyone pays the same price. Sources familiar with the licensing schedule say the players involved with the mobile JSRs pay more than vendors in the other JSRs, perhaps due to the prevalence of devices like cell phones. Oracle is rumored to have cut a sweetheart deal in exchange for pulling out of an IBM-led organization (openserver.org) that Sun perceived to be a threat to the Java brand. (Oracle's withdrawal from that effort precipitated that group's collapse.)

As I said, Java has never been free. The language doesn't even have an ECMA spec. Sort of ironic, isn't it?

Now before folk start screaming about me being a '.NET troll', I'd like to point out that I develop J2EE apps for a living and that I'm as sorry as any other Java developer that this has happened. Now that Oracle has shown its true colours, we're going to see a lot of worried vendors looking for a truly free alternative to Java. (And it won't be any development tool, such as Grails and Scala, that requires a JVM under the hood)

Sorry, yes. Surprised? Not at all.

Ray7 said:

Bass said:

*snip*

As I said, Java has never been free. The language doesn't even have an ECMA spec. Sort of ironic, isn't it?

Now before folk start screaming about me being a '.NET troll', I'd like to point out that I develop J2EE apps for a living and that I'm as sorry as any other Java developer that this has happened. Now that Oracle has shown its true colours, we're going to see a lot of worried vendors looking for a truly free alternative to Java. (And it won't be any development tool, such as Grails and Scala, that requires a JVM under the hood)

Sorry, yes. Surprised? Not at all.

ray7 said:

The other thing is that Sun has always charged for is the implementation of Java on mobile platforms and I think this is where Google has tripped up. They may have believed that since they are giving away the implementation then they are not subject to a license fee.

I'm not so sure, it always seemed that they were relying on the fact their VM wasn't a Java VM to get away with it. The only odd thing at the time was that Sun didn't seem bothered, although with hindsight they may have seen the potential for a lawsuit as a more saleable asset than anything else.

AndyC said:

Ray7 said:

*snip*

I'm not so sure, it always seemed that they were relying on the fact their VM wasn't a Java VM to get away with it. The only odd thing at the time was that Sun didn't seem bothered, although with hindsight they may have seen the potential for a lawsuit as a more saleable asset than anything else.

Well, Oracle seems to believe that some the JVM code is being used in the Google implementation. If Google can (and I hope they can) prove otherwise then Oracle shouldn't have a leg to stand on. There is nothing preventing anyone from creating a clean room version of the JVM, but I do remember hearing one Sun executive saying that this would be impossible (though of course, he never said why).

An interesting take on this whole sorry business.

elmer said:

Bass said:

*snip*

It's my understanding that Sun specifically excluded patent rights, and only handed over copyright.

They licensed it under the GPL, which includes a patent grant. The patent grant may or may not apply to Google. The courts will have to decide.

Regardless of the outcome of this case it will be very damaging to Java.

Bass said:

elmer said:

*snip*

They licensed it under the GPL, which includes a patent grant. The patent grant may or may not apply to Google. The courts will have to decide.

 

Regardless of the outcome of this case it will be very damaging to Java.

Google Dalvik VM is an independently developed project under the Apache License (and uses a similarly licensed class library). It has nothing to do with the GPLd Java SE and therefore is not covered by its patent grant.

Ray7 said:

AndyC said:

*snip*

Well, Oracle seems to believe that some the JVM code is being used in the Google implementation. If Google can (and I hope they can) prove otherwise then Oracle shouldn't have a leg to stand on. There is nothing preventing anyone from creating a clean room version of the JVM, but I do remember hearing one Sun executive saying that this would be impossible (though of course, he never said why).

This isn't about actual code. Its about patented programming techniques and algorithms.

Which is why I think software patents suck.

PS: There's also a copyright aspect to this case. Maybe Google has been calling thier implementation "Java" when its not really according to Sun/Oracle.

PPS: Ignore that PS

DCMonkey said:

Bass said:

*snip*

Google Dalvik VM is an independently developed project under the Apache License (and uses a similarly licensed class library). It has nothing to do with the GPLd Java SE and therefore is not covered by its patent grant.

You got to look at the facts of the case. Oracle is suing Google for "copyright" and "patent" infringement. They do not specifics on what copyright infringement they think Google is violating. They are suing Google using seven software patents, a few of which are ridiculously broad (like this) and probably have some pretty visible prior art. They might get thrown out in court.

I think it all boils down to if Oracle files in Texas or not and how good Oracle/Google lawyers are in demonizing each other to a jury.