decoration decoration

When you want to know more...
For layout only
Site Map
About Groklaw
Legal Research
ApplevSamsung p.2
Cast: Lawyers
Comes v. MS
Gordon v MS
IV v. Google
Legal Docs
MS Litigations
News Picks
Novell v. MS
Novell-MS Deal
OOXML Appeals
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v Novell
Sean Daly
Software Patents
Switch to Linux
Unix Books
Your contributions keep Groklaw going.
To donate to Groklaw 2.0:

Groklaw Gear

Click here to send an email to the editor of this weblog.

Contact PJ

Click here to email PJ. You won't find me on Facebook Donate Paypal

User Functions



Don't have an account yet? Sign up as a New User

No Legal Advice

The information on Groklaw is not intended to constitute legal advice. While Mark is a lawyer and he has asked other lawyers and law students to contribute articles, all of these articles are offered to help educate, not to provide specific legal advice. They are not your lawyers.

Here's Groklaw's comments policy.

What's New

No new stories

COMMENTS last 48 hrs
No new comments


hosted by ibiblio

On servers donated to ibiblio by AMD.

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Saturday, April 21 2012 @ 01:18 PM EDT

This is big news. Huge. The judge in the Oracle v. Google trial, the Honorable William Alsup, has reached an important decision. He has decided he will rule on whether or not APIs can be copyrighted, not the jury.

The news came with the final report from our reporter on Day 4's coverage. I've added it there in Update 5, but it's significant enough that I wanted to highlight it, to make sure you don't miss it. This is, in my view, an important win for Google, in that it ensures that Oracle will not be able to confuse a non-technical jury that also presumably doesn't know much about the law. I know a lot of you have been wondering when Judge Alsup was going to realize that the buck stopped with him, and it has happened. I'll show you the notes.

From our reporter finsko, this snippet from the very end of the day, after the jury has been dismissed:

Judge Alsup: The copyrightability of the 37 APIs is my call. But I want more briefing on it. I want you to take a firm position. Could you get a patent on structure, sequence, and organization (SSO)? Say "yes" or "no".

Also, do you know if the copyright office investigates SSO of copyrighted source code? I would guess not.

[Google answers "no", Oracle answers "maybe yes, maybe no"; I'm not sure which question this refers to.]

[Copyright office only gets first 50 lines and final 50 lines of the source for a SW copyright registration, so of course they don't look at SSO.]

Judge Alsup: OK, what about derivative works? Does the plaintiff have to prove that the source work was actually used? What if it was accidental overlap?

Oracle: Proof requires access, plus substantial similarity.

Google: Unauthorized work has to include copyrightable material.

Judge Alsup: When Google did their clean room implementation, did they have access to the English language comments [in Java source]?

Google: Yes, they had the English language prose descriptions of the APIs.

Judge Alsup: Does that make it a derivative work?

Google: No.

Google: The fully-qualified names are the organization.

Judge Alsup. I have to decide on copyrightability, not the jury.

You don't have to do exact copies, e.g., you could use science.sqrt() as your method rather than math.sqrt()

Google: It wouldn't work; existing code wouldn't find it.

Judge Alsup: Is fair use a question for judge or jury?

[1:30 or so, Judge leaves the courtroom]

You might want to read Mark's most recent article on the filings and an intriguing GPL possibility regarding the APIs. The briefs requested are due Sunday, so no rest for the weary. The bags under my eyes are carrying their own bags.

The other big news, which you'll find in update 5, the rest of the Joshua Bloch testimony is that the lines of code that Oracle claims are copied that Bloch wrote were donated to Sun and Java by Bloch himself. Sun incorporated it into Java and praised it for being faster than what they had already. Get a load of this:

Q. You left Sun and joined Google in 2004. What did you do at Google?

A. I ported existing Google infrastructure that was primarily accessible from C++ so that it was accessible to Java. I joined the Android team in December 2008 or January 2009. Android had already been released, and phones were in the market.

Q. When you joined, what was the status of Android's Java run-time libraries?

A. Not as fast as they could be, and there were some bugs.

Q. Did you have any influence on the choice of APIs?

A. No.

Q. Did you ever consult Sun source code or Sun materials?

A. No.

Q. Did you change any of the APIs?

A. No, you can't. It would make it incompatible.

Q. What is Timsort?

A. A way of putting a list in order [somewhere in here there is reference to compatible Timsort (?)].

Q. How big is Timsort (referring to Bloch's Java implementation).

A. 900 lines.

[The rangecheck() function gets mentioned in here.]

Q. Was there a sort API in Java before Timsort?

A. Yes, in the arrays (class? package?).

Q. How did Timsort differ?

A. It was much faster: as much as 20 times, typically somewhere between 2 and 20.

Q. How big is the rangecheck() method? Is it a private or public method?

A. It's a private method, about the same size as max() (so: small).

Q. What does rangecheck() do?

A. It's necessary when you have a sort method and you might be sorting a partial list or a whole list. You need to check that the indices are within range. It would be an error if an index was negative, or the indices are in the wrong order, or are not within range of the list.

Q. Is rangecheck() simple or complex?

A. Very simple. Any high school programmer could write it.

Q. Do you know of the existence of other rangecheck() functions?

A. Yes, there's one in I wrote it. [Timsort: from Tim Peters, and originally in Python. The Java implementation was a port.]

Q. Where did you get the Python version of Timsort? Was it open source [this was 2007, pre-Android]?

A. Yes, Guido [van Rossum] pointed me to it, it's under a permissive open-source license.

Q. What did you want to do with your Java Timsort?

A. Put it into OpenJDK (an open implentation of the SE platform).

Q. Who controlled OpenJDK?

A. Sun.

Q. How does someone contribute to OpenJDK, and had you done it before?

A. Yes. [Discussion about source repositories, and Doug Lee at Oswego, NY].

Q. If you worked for Google, why would you contribute to Sun's JDK?

A. Java is important to me; it's given me a lot.

Q. Why did you use the same rangecheck() function in Timsort as was in

A. It's good software engineering to reuse an existing function.

Q. But why use the exact same code?

A. I copied rangecheck() as a temporary measure, assuming this would be merged into and my version of rangecheck() would go away.

[Discussion of Timsort dates and Android work dates.]

Q. Was Timsort accepted and added into OpenJDK?

A. Yes.

[Something about Java 7.]

Q. Did Sun ever say anything about Timsort?

A. Yes, Mark Reinhold praised it for its speed on his blog.

Q. What else did you contribute to Java 7?

A. ARM Block: Automated Resource Management. Like automatically closing a file. This wasn't just an API, but a language feature. It added to the language itself and changed the grammar.

Q. Is it often that case that the language and the APIs change in tandem?

A. Yes.

Q. Are the 9 lines of rangecheck() currently in Android?

A. No, not in Ice Cream Sandwich; I checked.

Q. Are you still contributing to Java?

A. Lots. I feel a personal responsibility to give back to the Java ecosystem.

Q. Have other people at Google contributed to Java?

A. Yes.

Now Oracle is suing over that very code. That doesn't make Oracle look so good. In fact, they look a teensy bit petty.

And the other point that struck me is that he was asked whether you can use the Java language without APIs. This obviously matters, in that it would be pointless to tell the world that you can use the language but not the APIs if in reality you can't use the language without the APIs:

Q. Could you do anything in Java without APIs?

A. You could add two numbers, but couldn't do any I/O. E.g., the simplest program, the first program you write is "Hello world". This isn't doable without an I/O API.

Here's what Mark wrote about the GPL and how it works, or a part of it, but do read the whole article:
So the question is, given that the Java programming language is licensed under GPLv2, are the APIs and associated class libraries necessary to make programs run in the Java language "associated interface definition files?" If so, one could argue that those 37 APIs are a part of the "work as a whole" (the Java programming language) and, thus, also subject to the GPLv2. Interesting.
I'll say. Imagine if, in a truly karmic result of bringing this litigation, Oracle loses not only most of the asserted patents but has to GPLv2 the APIs.


Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj | 503 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: DBLR on Saturday, April 21 2012 @ 01:48 PM EDT
This is good news.


"Democracy is two wolves and a lamb voting on what to have for lunch. Liberty is
a well-armed lamb contesting the vote."
Benjamin Franklin.

[ Reply to This | # ]

Off Topic
Authored by: DBLR on Saturday, April 21 2012 @ 01:50 PM EDT
Place Off Topic subject maters here.


"Democracy is two wolves and a lamb voting on what to have for lunch. Liberty is
a well-armed lamb contesting the vote."
Benjamin Franklin.

[ Reply to This | # ]

Only if Oracle win their copyright definition
Authored by: Anonymous on Saturday, April 21 2012 @ 01:50 PM EDT

Otherwise APIs are just what they always were, stuff programmers use to
communicate with stuff that other programmers have already done.

[ Reply to This | # ]

Corrections here
Authored by: DBLR on Saturday, April 21 2012 @ 01:51 PM EDT
If needed


"Democracy is two wolves and a lamb voting on what to have for lunch. Liberty is
a well-armed lamb contesting the vote."
Benjamin Franklin.

[ Reply to This | # ]

News Picks Here
Authored by: DBLR on Saturday, April 21 2012 @ 01:52 PM EDT


"Democracy is two wolves and a lamb voting on what to have for lunch. Liberty is
a well-armed lamb contesting the vote."
Benjamin Franklin.

[ Reply to This | # ]

The APIs are already GPLed
Authored by: Anonymous on Saturday, April 21 2012 @ 01:58 PM EDT
They appear in the GPLed version of Java, don't they?

[ Reply to This | # ]

Oracle has already distributed the APIs as GPL
Authored by: mossc on Saturday, April 21 2012 @ 02:12 PM EDT
As part of "Oracle Unbreakable Linux" Oracle has released the JAVA
APIs with gnu classpath. The only permission they have to do this is the GPL

I am fairly certain that the GPL would not allow this legal action and remain in

Google should be able to rely on this as permission from ORACLE.

I think a issue in common with the TSG saga is The SCO Group and Oracle seem to
think that when they buy a product, company or distribution channel, that they
can disregard all obligations by either entity and assume all rights.


[ Reply to This | # ]

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: tknarr on Saturday, April 21 2012 @ 02:18 PM EDT
You don't have to do exact copies, e.g., you could use science.sqrt() as your method rather than math.sqrt()

I'd note that when the judge says this, he's wrong by definition. The name "science.sqrt" isn't "math.sqrt", and by the rules of Java refers to a different function. The unqualified names, "sqrt", may be the same, but they have to be fully-qualified to be usable and the fully-qualified names are different.

Go look up and see if it's the same site as to see an example of this. They're both "groklaw", right? No they aren't, because the TLD qualifies the name "groklaw" and is a different fully-qualified domain name than The two may both be called "groklaw" in part, but that doesn't make them the same name and you can't go to and find yourself at

[ Reply to This | # ]

COMES transcripts
Authored by: Anonymous on Saturday, April 21 2012 @ 02:19 PM EDT
Put your comes transcripts here! Please use Plain Old Text to make copying easier for PJ!

P.S. I know someone said anons should not make a canonical section, but I'm working on some transcripts for PJ the section does not exist!

[ Reply to This | # ]

Oracle must be concerned now
Authored by: Anonymous on Saturday, April 21 2012 @ 02:20 PM EDT
If indeed as indicated by Google that
The fully-qualified names are the organization.

Then what is "Structure" and "Sequence" in the context of APIs? An example would be very helpful.

From my limited programming (mainly Visual Basic), I do not see how one can code with an aim for compatibility and in the process, employ a different Structure or Organization of APIs. For "Organization", it may be the case but still, it's a stretch.

[ Reply to This | # ]

What a relief
Authored by: naka on Saturday, April 21 2012 @ 02:23 PM EDT

Well, relief doesn't really come until all the decisions and then appeals are finished I suppose, but I'm so glad that it's Judge Alsup deciding this and not the jury.

There are some fairly abstract and nuanced ideas and interpretations of the law in play and it would be tough to get a jury up to speed. JA will be able to ask questions and have a better idea of when the answers are evasive than a lay-person.

Kudos to Judge Alsup for owning the decision now that he is (hopefully) starting to understand what is happening here.

[ Reply to This | # ]

Channeling Marbux
Authored by: sproggit on Saturday, April 21 2012 @ 02:52 PM EDT
Way (way, *way*) back in February 2005, PJ wrote an article in analysis of orders from the Honorable Dale A. Kimball in the case between SCO and IBM. In her article, PJ quoted from a Groklaw contributor, Marbux - who, IIRC, is a retired lawyer.

Here's a snippet of that article:-
"Viewed through that lens, here's my own interpretation of what Judge Kimball is saying, stripped of legalese and the shroud of impartiality that judges have to wear:

"IBM, you got my attention and I'm all over this case now. But we're going to do it my way. Don't worry, IBM. The briefs have convinced me that you are going to win this case, so I'm going to help you win. Our main task is to work together to create a record that has the best possible chance of standing up on appeal. People appeal my rulings all the time, but I'm pretty good at predicting how the Tenth Circuit will rule and I don't get reversed very often. I'm in charge, so listen very carefully to what I say.

I'm not going to sit back and let the lawyers run the show. This is a case I'm going to actively manage from here on out. Here are my ground rules for the winner and the loser.

SCO, your case is absurd, and you're trying to swim up a waterfall. You made your bed; now you have to sleep in it. Your motion to get rid of the counterclaims is denied and off my desk. I'm not worried about getting reversed on that subject."

As it happens, we were all partially wrong on that one, and Judge Kimball's decisions were, in part, reversed. The panel of appeals judges decided that Judge Kimball had erred in deciding that one aspect of SCO's case could be decided as a matter of law, finding instead that it needed a jury trial. And as the full record shows, SCO lost in Court and in spectacular style.

I have no legal expertise on which to base this hunch - it's just a gut reaction really, but I echo PJ's sentiment when she states that this is a very significant development for the case.

That gut reaction suggests to me that Judge Alsup has watched the initial days of the trial and has realised that the BSF/Oracle strategy is to play fast and loose, to perform some fancy footwork and to create enough confusion and doubt in the minds of the jurors that they fall back on "Well, I suppose..."

As in, "Well, I suppose that Oracle must know what they're talking about and this is a huge trial and it all sounds very technical and important and it's their stuff after all and even though I may not understand this, surely they wouldn't sue Google unless there really was something wrong? Surely there's no smoke without fire?"

I wonder if the good Judge hasn't seen to the heart of this ploy and thought,

"Not so fast, Oracle. I see what you're up to here, and I won't have you bamboozle this jury just to get the win. They may not understand the niceties of copyrights and patents, but I do. We'll do this properly, if you don't mind..."

I confess I am very much looking forward to the transcripts of Monday's day in court, and the reaction of BSF/Oracle to this development...

[ Reply to This | # ]

What does the jury decide on copyright issues?
Authored by: Anonymous on Saturday, April 21 2012 @ 02:56 PM EDT
If the judge is doing this determination, what is left in the copyright phase? The 9 lines of code donated1 to sun/oracle?

1: This is my understanding of the TimSort() testimony, please correct me if I'm wrong.

[ Reply to This | # ]

API copyrightability is a matter of law
Authored by: kawabago on Saturday, April 21 2012 @ 03:40 PM EDT
Judges have already decided that API's are not eligible for
copyright protection because they are necessary elements to inter-operate with
the API. They become facts not expression
in this context.

[ Reply to This | # ]

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: Anonymous on Saturday, April 21 2012 @ 03:43 PM EDT
"I'll say. Imagine if, in a truly karmic result of bringing this
litigation, Oracle loses not only most of the asserted patents but has to GPLv2
the APIs."

That won't happen. The APIs are already freely usable by anybody, there is no
need for a license. (Unless the good Judge decides he wants to crash the U.S.
software industry into an iceberg worthy of the Titanic).

Besides, it already happened, in the sense that OpenJDK contains all of the APIs
in question and is already available under the GPL, with a linking exception for
the class libraries.

[ Reply to This | # ]

Saw this coming
Authored by: Anonymous on Saturday, April 21 2012 @ 03:46 PM EDT

I commented a couple of days ago that the judge would rule on this in a summary judgment. No way was he going to let the copyright go to the jury once he realized what was what.

As for Java and the GPL, the Java APIs are NOT under the GPL. If you look at the license you will see that Java is under the GPL but the APIs under under what is commonly known as the Classpath exception. What this means is that anyone can write code that uses the APIs and their code does not have to abide by the GPL. Without this provision almost no one would use Java.

The implications of the exception on the API are not straightforward since they are packaged together with the implementation. However, Joshua Bloch did a great job in illustrating why this isn't important since the APIs aren't copyrightable in the first place.

[ Reply to This | # ]

Wrong question: SSO of copyrighted source code?
Authored by: Ian Al on Saturday, April 21 2012 @ 03:52 PM EDT
The thing in question is the Java API Specification. This is an instruction
manual for writing Java software. It is not source code.

The names listed in the Specification are the names used in programs and in
function libraries so that the library code can be added to the class file along
with the program code.

As Oracle explain, providing function code in libraries means the programmer
does not have to reinvent regularly used code like the code for doing math.

The way that function code is selected from the library is that the programmer
uses the same text label for the function within the syntax of the computer
language that the library and compiler use to pick out the actual function code
from the library.

The SSO in the Specification is a factual listing of where in the various
library packages the relevant named function code can be found. It is a
directory which is precisely constrained by the way the libraries were written.
The SSO in the Specification cannot be patented because it is a book.

The SSO is the SSO of the libraries and is not new creative expression put into
the Specification. Even if the Specification came first as an abstract idea, it
is modified so that it reflects what is actually and finally coded into the

The SSO in the libraries cannot be patented because the libraries are math...
no, because the libraries are a collection of components and not a machine,
process, transformation or method. A selection or organisation is not patentable
subject matter under the act. SSO is not an invention.

Anyone writing replacement libraries intended for use by Java programmers *must*
duplicate the SSO of the original libraries or else the libraries will not
function when the programmer attempts to access them; i.e. it is a functional
requirement of replacement libraries.

The SSO creative expression is not written into the Specification: it is written
into the libraries and reported like a dictionary or a schedule in the
Specification. The Specification is entirely and completely derived from the
library expression.

As such, the SSO in the Specification is the recording in abstract idea terms of
the function and disposition of the actual code in the libraries.

Ian Al
Software Patents: It's the disclosed functions in the patent, stupid!

[ Reply to This | # ]

I don't think the GPL versions are relevant
Authored by: jbb on Saturday, April 21 2012 @ 03:57 PM EDT
Google is using Harmony which is under a permissive Apache-2 license. They choose not to use the GPL version of this code. Despite what some people might say, the GPL and Apache-2 don't mix well. If Oracle's novel theory about API copyrights prevails then the GPL versions of the APIs won't help Google.

If Google were forced to switch to a GPL version then all the downstream manufacturers would be forced to GPL their code which I assume they do not want to do. I was disappointed that Google went with a permissive open-source license rather than a free software license (the GPL) for this layer of Android but I think it was understandable. I'm sure Google discussed licensing issues with potential manufacturers. I imagine using a Free Software license was ruled out early on.

Perfect is the enemy of good.

Our job is to remind ourselves that there are more contexts than the one we’re in now — the one that we think is reality.
-- Alan Kay

[ Reply to This | # ]

GPLv2 and the real story here.
Authored by: Anonymous on Saturday, April 21 2012 @ 03:57 PM EDT
Oracle's Java implementation is based on the OpenJDK which is GPLv2, so
yes the Java API could be licensed as GPLv2. Oracle, of course, as the
copyright holders, may license Java any other way they want. Companies
that don't want to use the copyleft OpenJDK would need to license Java.

Google didn't like the GPL licensing because it would require handset
manufacturers to open-source their modification to the code. Google
didn't want to pay Sun for a commercial license, either (or, as Google
claim, Sun were dragging their feet during the negotiations). Google
couldn't use any clean-room open-source implementation of Java because
Sun wanted to specifically restrict implementations on mobile handsets,
and so had the field-of-use restriction on the TCK, which is required
if you want to call your implementation "Java".

So Google went ahead and found a loophole. They wouldn't call their
implementation Java (eschewing the need for the TCK), and they'd use an
inferior, yet passable virtual machine that could run Java (Dalvik).
Trying to do as little as possible while still circumventing Sun's
restrictions, Google did not provide their own implementation of the
Java language at all (it appears from the descriptions here as if
Google re-implemented the Java language - they didn't), i.e. they
didn't write a Java compiler. What they did do was have developers use
the standard Sun Java compiler (or any other), and then translate the
compiled code to a format that could run on their VM. In the process,
Google has fragmented Java.

Oracle, in their frustration, realized what Google had done: replaced
as few pieces of possible, but just enough to bypass their licensing or
the GPL, and tried to sue Google over whatever it is they did. Just as
Google found an apparent loophole, so is Oracle trying to find a crack
in the armor. The tactics Oracle is using may or may not be effective.
They may or may not have legal merit - IANAL. But it has to be clear:
Oracle isn't trying to make a philosophical or general argument here.
They're simply trying to nail Google for the schtick they've pulled.

The legal arguments are, of course, very interesting, but I don't
understand the Google bias here. Google pulled a schtick on Oracle, and
now Oracle fights back. It's simply two corporations using the court as
another battleground.

The legal ramifications that the ruling may have on copyrights,
patents, etc., is, of course fascinating. But there are no good guys
and bad guys here. Only two fierce warriors.

[ Reply to This | # ]

Authored by: Anonymous on Saturday, April 21 2012 @ 04:00 PM EDT

I really liked that Google came back to ask Joshua about the Java 7 vote. Java 7 was stalled for years over the dispute with the Apache Software Foundation regarding the Java TCK. Several JCP members were willing to support the ASF and vote "No" on the specification. However, once Oracle and IBM signed their deal with regard to IBM moving from Apache Harmony to OpenJDK the writing was on the wall. The ASF had no recourse except to resign from the JCP.

Several other JCP members who would have voted "No" also resigned and were replaced by Oracle, so it is no great surprise. Stephen Colebourne documented these events extremely well in his blog in 2011.

[ Reply to This | # ]

  • Apache - Authored by: Anonymous on Saturday, April 21 2012 @ 04:58 PM EDT
    • Shocked! - Authored by: Anonymous on Monday, April 23 2012 @ 12:28 AM EDT
Danger ahead Will Oracle
Authored by: Anonymous on Saturday, April 21 2012 @ 04:44 PM EDT
A. You could add two numbers, but couldn't do any I/O. E.g.,
the simplest program, the first program you write is "Hello
world". This isn't doable without an I/O API.

Firstly - if Oracle wins, then they will have shown that you
can't trust them. In just the way that we don't trust
Microsoft. And once you can't trust a company or a person,
every interaction with them carries that baggage, and
whenever you have the opportunity to avoid dependency on
them you think about it. The consequence is that Java will
only be used when it's necessary and unavoidable.

Secondly - if Oracle win, then the I/O library and all
others become unusable by FOSS advocates. And we'll have to
write our own set of APIs and split the language base
between those that use the Oracle standard and those that
use the FOSS libraries; there can be no dependency on any
Oracle libraries/APIs.

Neatly accomplishing exactly what Oracle claim to want to
avoid. Anyone would think they want the language to die.

[ Reply to This | # ]

"GPLv2 the APIs"
Authored by: Anonymous on Saturday, April 21 2012 @ 04:52 PM EDT
Be careful what you wish for.

Every time that I write a program in Linux I use, at least,
two APIs which were included in GPLed packages. One for the
system calls and one for calls to gcc run time library(s).

I don't release my programs under GPL and it is legal and
proper because the people who wrote gcc or Linux have never
claimed copyrights of the APIs; unlike Oracle they play fair.

If the judge will accept Oracle's idea of copyrights to API
then plenty of people will be very afraid to write programs
on Linux or use gcc. The big winner will be Microsoft.

Not Good.

(And I thank Google for fighting against this idea, hard.)

[ Reply to This | # ]

Didn't the 10th Circuit abolish summary judgments?
Authored by: Anonymous on Saturday, April 21 2012 @ 04:55 PM EDT
OK, this is the 9th Circuit. So, that's good. But I still get the feeling that
this just creates an appeal and won't be settled for a long time to come.

[ Reply to This | # ]

APIs can't be copyrighted because they aren't creative content.
Authored by: Anonymous on Saturday, April 21 2012 @ 05:45 PM EDT
As I understand it, copyright can only be used to protect
creative content. APIs and classpaths are not creative
content, they are a rigidly enforced referencing convention
which code cannot deviate from for compatibility reasons.
Hence they are not creative but rather part of the Java
language's structure.

In addition, fair use of copyrighted material allows the use
of limited copying of names for functions, variables,
classpaths and comments for reference purposes.

If this fair use wasn't permitted, it would be possible for
the Coca Cola company to sue the publisher of any negative
article using the name "Coca Cola" for copyright
infringement and so stifle free speech. Similarly the
publishers of a street map of New York, and the New York
telephone directory would be able to charge a copyright free
from anybody who used the name of a street or a person taken
from their publications in any written communications.

This would be total madness - even worse than the patent
madness we have now.

[ Reply to This | # ]

openJDK, the API, the cat, the bag
Authored by: Anonymous on Saturday, April 21 2012 @ 05:49 PM EDT
One of the topics getting discussed here is Sun releasing openJDK under GPL with
the classpath exception. The argument goes 'If only dalvik was under GPL and
derived from openJDK not Harmony then Oracle could not complain, since this is
expressly permitted under the GPL...'. That argument is probably a dead end.

Lets try another argument:

1. openJDK contains all the APIs being argued about.
2. Each method in a java language API has either public, private, protected,
package access restrictions. All non-private methods can be used in inheritance
– if a method is public or protected in a non-final regular class or a public
interface then it is volunteering itself for inheritance/implementation when it
is consumed by a programmer using the API.
3. In a java program during inheritance the full signature of the method is used
to identify the method being overridden.
4. The licence of openJDK explicitly says that programs making use of it don't
need to be GPL.

/*license gpl+cpe*/
public class APIclass {
public int add(int, a, int b) { //GPL+CPE
return a+b;

/*license creative-commons-do-whatever*/
package mypackage;
class MyClass extends APIclass {
public int add(int x, int y) { //API method signature reproduced but no-longer
under GPL
System.out.println("adding " x + ", " + y);
return super(x,y);

public int add(int, a, int b)

So in normal use of a publicly facing API covered by GPL+CPE it is possible to
extract the method signature and sidestep the GPL+CPE. There is nothing
nefarious about it. Repeat this step for every public/protected method and then
extract just the method signatures from your own subclasses and you get the
public/protected API as a list. This process will produce a big API that may
exactly match the ordering in the openJDK source – but is under no copyright –
certainly not GPL-CPE

Lets look at private API methods that are not intended/cannot be consumed by
normal regular use. Well first of all these are part of the implementation so
are not usually part of the public API that would be given to a programmer but
there are several corner cases in java where someone re-implementing an API for
comparability reasons would also clone private method signatures (for example
where the class might be interrogated at runtime using 'reflection' and
everything marked private becomes public). Well we could get the private APIs
using reflection but IMO that is not clear-cut legal. Instead there is a simple
way: simply take the openJDK source, locate the private methods you want to
copy, change private to public and save the changes – the code is still under
GPL, how can it not be? Now inherit the method in your own non-GPL class and
extract the method signature from your own non-GPL code. The same procedure can
be used to extract signatures from classes marked final which otherwise can't
be inherited.

An alternative weaker argument can be made that the the public/protected method
signatures in the openJDK source are directly analogous to the header files in
c. Google have previously been accused of taking the headers from GPLed c/c++
source-code and rewriting the code so that is no longer covered by the GPL but
still has the original header files – this is done for interoperability with
existing code. Now this might leave bad taste in the mouth but is it legal? Well
according to the people who drew up the GPL it is perfectly fine:

<i>In an email Richard Stallman says 'I've talked with our lawyer about
one specific issue that you raised: that of using simple material from header
files. Someone recently made the claim that including a header file always makes
a derivative work. That's not the FSF's view. Our view is that just using
structure definitions, typedefs, enumeration constants, macros with simple
bodies, etc., is NOT enough to make a derivative work. It would take a
substantial amount of code (coming from inline functions or macros with
substantial bodies) to do that.' This should help end the recent FUD about the
Android 'clean headers.'</i>

By publishing their APIs under GPL+CPE Sun/Oracle have placed any
public/protected API contained in openJDK non-copywritable when presented as a
list with no source-code attached. Since the API can derived legally in at least
three ways:

By simply inheriting the class into your own code, overriding the method and
then running grep over your own code.

Using reflection to list available packages and methods from your own program
running on openJDK – do Oracle claim to own the output of your programs when
running on their own fully free and open-source GPL+CPE openJDK ?!

By analogy – the GPL does not extend to header files in c – according to the
authors of the GPL. Method signatures declared public/protected may be
equivalent to header files in c.

Conclusion: It is trivial to write legally permitted non-GPL code, using no
non-GPL Oracle API documentation, only the openJDK source code for reference,
that results in a list of Oracles precious API as a document.


[ Reply to This | # ]

Nelson - HA-Haw!
Authored by: SilverWave on Saturday, April 21 2012 @ 06:03 PM EDT
Nelson Muntz is a fictional character and bully from the animated TV series The
Simpsons. ... Simpson. He is best identified by his signature laugh

RMS: The 4 Freedoms
0 run the program for any purpose
1 study the source code and change it
2 make copies and distribute them
3 publish modified versions

[ Reply to This | # ]

Judge Alsup Will Decide if API's are still not Copyrightable ~pj
Authored by: webster on Saturday, April 21 2012 @ 09:25 PM EDT

It's about time. Oracle ought to move for a mistrial. They have been trying to
convince the Jury that some aspect of API's is copyrightable AND that there is
infringement of such copyrightable API's. There is hardly any disagreement on
the rest of the "facts" of this case. Now they have to proceed
without knowing what the Judge will decide about this issue. It is a lot of
wasted time. Oracle: "If API's are copyrightable, consider this; if they
are not, never mind." Without this issue there is little for the jury to
do other than listen to stipulations. Since they have been listening to this
API-"design" mumbo-jumbo, Oracle can argue that the issue's
disappearance hurts their credibility.

Judge Alsup will decide more mindful of legal and industry precedence. The
presumption is against copyrighting API's. The GPL and the Harmony license may
also preclude the API copyright issue.

There is little for the jury to decide. There is no wilfull infringement if
this case sets a new precedent. How can Google infringe by a rule that didn't
exist yet. And then the licenses ...

API's appear to be clips of code that one can precisely invoke in a language to
do powerful and flexible things. The API's derive from ancestral coders passing
on and enhancing functionality. They are basic to making the circuits in the
computer behave in an anticipated matter. It is code below the source
programming level but above the brutal binary level. It keeps everyone from
having to reinvent the wheel in every language. Use the old stuff unless you
see an improvement. Copyrighting these clips is little more advanced than
copyrighting words or letters or yes/no, or ... numbers, like 0 or 1.



[ Reply to This | # ]

Authored by: Anonymous on Saturday, April 21 2012 @ 10:01 PM EDT
Do I understand this correctly...
While employed at Google:
Bloch copied code from Timsort
Bloch donated code to Java
Bloch used part of same code at Google

This is the ONLY code that Oracle is said was copied by Google?
...or am I missing something

[ Reply to This | # ]

Why Are Only 37API's Copyrightable?
Authored by: Anonymous on Saturday, April 21 2012 @ 10:06 PM EDT
The Judge declaring he would be the one to decide if API's are
indeed Copyrightable is the best thing to happen for Google
this week!
Here's a case that might have some important information for
Judge Alsop's Copyrightability issues:
Lexmark Int’l v. Static Control Components

Let's just hope Judge Alsop decides correctly, so Google
doesn't have to appeal his ruling, that it took this appellate
court to overturn!

Why aren't all the Java API's deemed copyrightable by Oracle.
Same function/purpose as any other API's in the Java
programming language? So why only these 37 API's published in
the Specifications. Is that the only place they remain under
copyright according to their claims or is it that it's the only
place these API's can be seen to be published with a copyright
notice anywhere in Java? Why didn't Google post the Oracle
Copyright Notices in Android then? Maybe because these 37 API's
are not so special or different from all the rest of the Java
API assets taken out of context by Boes from the Copyrightable
Specifications rather than in their implementable form?

Great information from Tim Bray (Sun/Oracle 2006) speaking to
InfoQ, when Java was Open Sourced:
""we have to ensure that users of Java don't have to worry
about getting claims that by using Java they're infringing a
patent held by any of the contributors to Java (including Sun);
so we have to prevent anyone, whether by accident or on
purpose, from contributing code to Java and then being able to
launch claims against people who use it.""

And another interesting Tim Bray quote for you:
""The GPL doesn't require that bytecode classes using a GPLd
java.lang.Object be licensed under the GPL as well. That's
because neither the bytecode nor the source code using it are
derivative works of java.lang.Object, as all that ever ends up
in them are interface names and constants, and those remain the
same, regardless of the license of the java.lang.Object class.
Those symbols are standardised through the JCP, and published
as specifications. They are necessary for interoperability.
Therefore, the symbol names and constant values can not be
claimed as original works by a GPLd java.lang.Object, and
accordingly don't meet the bar for copyrightability.""

The judge needs to run the tests for "copyrightability"
himself, without Oracle or Google's input. So it's decided
based on what's legally copyrightable instead of what Oracle
wishes was copyrightable!

If.... Oracle's "API's are Copyrightable" claims are court
stamped and approved, then Oracle can use this "Come On" pitch
sales ruse to force you to pay for what even they say is FREE
to Use... the Java Programming Language. It's like as if Oracle
is putting out some magnificently delectable Banquet dinner for
FREE. But... then make you pay the equivalent price of every
item on the table, for the plate and utensils to eat it with!

[ Reply to This | # ]

Authored by: Anonymous on Saturday, April 21 2012 @ 10:47 PM EDT
Bloch explained that regex was borrowed from Java. Now we just need some
explanation about how string formatting was borrowed from C and how a lot of the
math functions were borrowed from other languages as well.

[ Reply to This | # ]

Finder of law vs finder of fact
Authored by: Khym Chanur on Sunday, April 22 2012 @ 02:04 AM EDT

Doesn't it have to go this way? That is, isn't whether or not APIs are copyrightable a matter of law, which makes it something the judge has to decide?

Give a man a match, and he'll be warm for a minute, but set him on fire, and he'll be warm for the rest of his life. (Paraphrased from Terry Pratchett)

[ Reply to This | # ]

APIs across computer languages
Authored by: Anonymous on Sunday, April 22 2012 @ 03:42 AM EDT
Perhaps more a question for the techies, but what happens if Sun is granted a
controlling copyright on these APIs. Does they mean they would have a
controlling copyright on those data structures and algorithms, regardless of
the implementation language?

Let us consider the example of a book. I cannot write my Barry Cotter novel
about a teenager sorcerer if it is a simply variation of another well-known
series by Ms J K Rowling. Can I implement a C++ API that sorts a sequence of
numbers stored in a sequential container without violating the copyrighted
Java container API. Or rather, can I provide a complete package of data
structures and algorithms, much as every computer language supplies out-

It is bad enough that we are facing the current trend to patent the 'words in
our dictionary', but at least pattens expire after 'only' a few decades.
Copyright essentially locks these things away for ever, given the trend of
extending copyright lifetimes each time certain culturally sensitive works are
approaching the end of their existent copyright terms.

I cannot see victory for Oracle as anything less than the end of the ability to

practice software in USA. In practice I suspect software businesses would
continue, but trading in the knowledge that they could be shut down by
lawsuit at a moments notice, which is a pretty strong incentive to move
elsewhere, were there not so many paying customers over here. It would
become a thoroughly toxic market though, and probably lose many of its top

[ Reply to This | # ]

JDoes anyone have proof that Oracle GPL'd the java programming language?
Authored by: s65_sean on Sunday, April 22 2012 @ 07:56 AM EDT
I questioned Mark's statement in the previous article that oracle had released
the java programming language under the GPL. I never received a reply pointing
to any proof of this GPL'ing of the java language ever happening. Here in this
article PJ repeats that statement.

Does anyone have a link or citation to support the statement that the java
programming language was released under the GPL?

The only thing related to java that I am aware of that Oracle (Sun) has released
under the GPL is the OpenJDK, which is not the java language. it is an open
source version of a set of java related programming tools, but it is not the
java language itself.

[ Reply to This | # ]

Has Google Really Fragmented Java?
Authored by: Anonymous on Sunday, April 22 2012 @ 07:57 AM EDT
Oracle talks A lot about the fragmentation of Java Android/Google has created.

But if you think about it they have just created a new very large alternative
Java Programing language ecosystem. And it is big as there is probably more
lines written targeting Android each day than there are targeting Java ME/SE/EE.
So eventually it should overtake these other forms of Java in total lines
written to date etc. This is probably what scares Oracle the most.

But I would argue that it actually reduces fragmentation because there is likely
to be only one main Mobile platform version that uses the Java Language and this
is going to be Android/Delvik (note Java ME is just a toy version of Java and
you can't run a mobile platform on it).

So because Google has done such a good job so far on Android there is no need
for anyone to further fragment Java with every phone maker trying to make
different versions etc. The only real effect is that instead of Oracle holding
all the Java cards we have some real competition now. So right now it is only
Oracle that is fighting to further fragment Java and hurt the language in


[ Reply to This | # ]

the difference between a public method in a class and an API
Authored by: IMANAL_TOO on Sunday, April 22 2012 @ 08:15 AM EDT
Here is one example of interfaces in programming, public methods in programming.

Leading-Edge Java
Design Principles from Design Patterns
A Conversation with Erich Gamma, Part III
by Bill Venners
June 6, 2005 [...]

In this third installment, Gamma discusses two design principles highlighted in the GoF book: program to an interface, not an implementation, and favor object composition over class inheritance.

Program to an interface, not an implementation

Bill Venners: In the introduction of the GoF book, you mention two principles of reusable object-oriented design. The first principle is: "Program to an interface, not an implementation." What's that really mean, and why do it?

Erich Gamma: This principle is really about dependency relationships which have to be carefully managed in a large app. It's easy to add a dependency on a class. It's almost too easy; just add an import statement and modern Java development tools like Eclipse even write this statement for you. Interestingly the inverse isn't that easy and getting rid of an unwanted dependency can be real refactoring work or even worse, block you from reusing the code in another context. For this reason you have to develop with open eyes when it comes to introducing dependencies. This principle tells us that depending on an interface is often beneficial.

Bill Venners: Why?

Erich Gamma:Once you depend on interfaces only, you're decoupled from the implementation. That means the implementation can vary, and that's a healthy dependency relationship. For example, for testing purposes you can replace a heavy database implementation with a lighter-weight mock implementation. Fortunately, with today's refactoring support you no longer have to come up with an interface up front. You can distill an interface from a concrete class once you have the full insights into a problem. The intended interface is just one 'extract interface' refactoring away.

So this approach gives you flexibility, but it also separates the really valuable part, the design, from the implementation, which allows clients to be decoupled from the implementation. One question is whether you should always use a Java interfaces for that. An abstract class is good as well. In fact, an abstract class gives you more flexibility when it comes to evolution. You can add new behavior without breaking clients.
The article goes on, still, I have a difficulty seeing the difference between a public method in a class and an API as is discussed in the Oracle vs Google case.

Maybe there is one, but I don't know what it would be. Maybe someone can enlighten me.



[ Reply to This | # ]

So a Developer can donate Code to GPL'd Java and then later get sued for using his own code?
Authored by: Anonymous on Sunday, April 22 2012 @ 08:18 AM EDT
I am not sure if I actually understand this:
A developer can donate Code to Java (like J. Bloch
maybe)presuming Java is GPL, Free and open source. The
donation is taken into the Java code base, Oracle/Sun
whoever, says: "thank you great contribution". And then
later down the road that same developer gets sued by Oracle
for using his own code that he donated to GPL-Java?

Is that really the way GPL-Opensource Java works? Or am I
missing something?

How many Non-Oracle/Sun authors donated their own code to
the GPL'd Java codebase?

How much of Java was written by Authors outside of
Sun/Oracle? Are they all going to be coerced now by
Oracle/Sun to pay for and license back their own
contributions to the Java Codebase?

Is that really how Opensource and GPL-Java works?

Is it just me thinking that this is fraud or am I wrong
that, just missing some key point?

[ Reply to This | # ]

Stop using "API" jargon, just say "interface"
Authored by: Anonymous on Sunday, April 22 2012 @ 12:07 PM EDT
I think the real problem for the court here is that the term "API" is jargon more than meaning. Without the experience of being an experienced programmer, it's very hard to understand the variety of ways and levels at which the term can correctly apply.

Instead of using this meaningless-to-laymen jargon, expand the definition to focus on the important part. Application Programming Interface.

The point that needs to be made in court is that an API is simply an "interface" between software components.

"interface" is a word which in general usage has precisely the same flexibility as API does in programming usage.

Drop the jargon abbreviation - speak English and get to the core of the matter.

[ Reply to This | # ]

Authored by: mcinsand on Sunday, April 22 2012 @ 03:33 PM EDT
Although some wiser, calmer heads here have been cautiously optimistic, I'm a
bit more confident. Judge Alsup has been steadly, carefully avoiding any
decisions that would pave the way for an appeal. Although it's been
frustrating, things like giving Oracle upmteen chances with the damages
estimates would give Oracle a decidedly uphill battle in appealing any ruling.
Given the Judge's cautious nature, I don't think he would rule on API copyrights
without a solid, concrete, well-documented legal basis. If the API's were
copyrightable, then the way Sun/Oracle GPL'd the language itself would create
gray areas for a jury to be required.

If he's ruling, then I think we have a great chance of uncorking something in
celebration soon!

[ Reply to This | # ]

Leo Cizek
Authored by: sproggit on Sunday, April 22 2012 @ 04:12 PM EDT
Reading through the testimony from Leo Cizek, there's an interesting exchange
where he is being questioned by Counsel for Oracle:-

Oracle: What was discussed regarding Android licensing Java?

Leo Cizek: Which of the two JME technologies would be most appropriate.

Oracle: Where was this meeting held?

Leo Cizek: All 3 meetings were held at Google's headquarters.

Oracle: Who attended meeting two?

Leo Cizek: Me and Vineet Gupta.

Oracle: And for Google?

Leo Cizek: Andy Rubin, Brian Swetland, [missed one attendee for Google].

Oracle: What was discussed?

Leo Cizek: Further discussions of Open Source licensing, regarding whether Sun
would [ missed ], and did Google need a JCL.

Oracle: And the third meeting, in December 2005, who attended that?

Leo Cizek: For Sun: Vineet Gupta, Alan Brenner, and me. For Google: Andy Rubin,
Brian Swetland, and Borenstein.

Oracle: Why was Brenner there?

Leo Cizek: He led a large part of the discussion regarding [trying to] convince
Andy Rubin to 1) use Sun's technology (CDC), and 2) not use Open Source for

It's that bit right at the end of my excerpt that caught my eye: Mr Cizek
clearly states that one of the Oracle deputation to Google went along
specifically to try and persuade Google to "not use Open Source for

Well, duh... if *Oracle* felt that this was something that they needed to
*persuade* Google over, then doesn't it follow that *Oracle* felt that Google
could not be *compelled* to ignore an Open Source approach???

I am not sure how well this would have been picked up by the jury, but it reads
like another foot-gun moment for BSF.

[ Reply to This | # ]

The root of Marks confusion and resultant mis-direct on GPL?
Authored by: Anonymous on Sunday, April 22 2012 @ 04:46 PM EDT
"When determining "substantial similarity" what is Oracle
comparing?" is the question Mark asked.

He then started talking about APIs.

I think at this point Mark took a left turn and inadvertently fell through
Oracles looking glass.

I think the answer is actually much simpler.

They are comparing Androids Book to their Book.

That Book is the API Specification, that is the English Language Descriptions of
the APIs
(cf. Day 3 of the ..Challenged Item 6)

The confusion I think arises as to what's what and the question of which comes
first. Let's see if I can articulate that.

If I write a phone book, you expect that phone book to be arranged like all
phone books, alphabetically, by surname, forename, street address/zip code.

But I'm not like you, I'm an artist, I am very creative and I work very hard.

I instead order it by the second digit in your post/zip-code as the first order,
and the third letter in your surname as the second order, after that fall where
they may.

That is covered by copyright doctrine.

That is, my choice of selection, structure and arrangement is most definitely
creative (albeit useless as a phone book)

(I believe doctrine states it is Arrangement rather than Organisation, but
seeing as it has become such, at least in discussion here, I shall refer to it
as SSO from here on)

I am entitled to (thin) copyright protection in that SSO.

You may not, without my permission/license, produce a phone book with the same

Oracle is pointing at it's Book, and saying look, we 'designed this', it is
complex, it is interesting, it is creative, it is (unlike my phone book), a work
of art for programmers, it is creative, we are entitled to protection for the
SSO as well as the English Language expressive content.

They then say, if it is so, such that our SSO is protected, then it follows that
after Google read our specification, they copied our SSO into their source code
(the 37 Apis), and therefore their source code is a derivative work, regardless
of its being an independent implementation. Their declarations are an identical
copy of our SSO.

I've been getting irritated through this discourse about Oracles wilful misuse
of the terms "API", and "API Specification", frankly the
diversion into the GPL made me want to scream. So much that I could only find
myself writing odd humour (I did the "Groklaw to be taken down" and
law.corporate.due_diligence() API Implementation, posts)

But the GPL thing made me stop and ask how do you go about dispelling something
that is offered up as confusing and something that appears to be a struggle to
understand even for technical people? And this is it.

Oracle are saying that their Book has a unique and creative SSO and everything
follows from there.

All the discourse around here starts at the declarations and implementations and
looks out, this is a function, this is a class, this is an interface, this is a
package, the API Specification is just the documentation for the stuff we've
just said.

It leads to the raging discussion "what is an API?". I even fell into
the trap of trying to provide an example of an API for OddOrEven, my post went
wrong and bugstomper beat me to the repost with a much better articulation.

But if you look at it the other way up, from Oracles position, it doesn't
matter a jot what an API is, or even what's in it, or whether its free to use or
GPL or open, or where the boundary is between API and Language or what the
various licenses permit, say or don't say.

SSO is the only thing that counts. It is not the APIs and their SSO, it is only
the SSO, all else follows from there.

It explains why in the list of Challenged items (again cf. Day 3 of the ....)
Item 1 The 37 APIs, and Item 4 the Declarations in the 37APIs are identical
symbol for symbol are called out separately, and yet in the not in contention we
see 2) Names of any particular element, despite all of these being basically the
same thing from a technical point of view.

(IANAL) Google can try all the estoppel, de-minimis, functional, merger they
want, but that's only pandering to the illusion and nibbling at the edges, they
are talking about code and Oracle aren't (TBF we haven't heard Googles side

It seems to me that if instead Google show that the code came first, that the
Book comes from the code, that the SSO is a consequence of and falls out of the
code, then the Book is just a Book and the SSO is '(in/ac)cidental' and no more
than just a happy result of using javadoc, then the whole SSO argument goes out
the window, and we stay exactly where we are. Your source code is protected,
Your book is protected, your binary is protected. Doesn't it?

Is it that? Or did I just muddy the water even further?

[ Reply to This | # ]

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: Anonymous on Sunday, April 22 2012 @ 07:51 PM EDT
I hope Google's lawyers point out that API's have been around
for over half a century, and no one has tried to copyright
them before.

[ Reply to This | # ]

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: Anonymous on Sunday, April 22 2012 @ 08:07 PM EDT

For quite some time now, Pl/Sql has been the de-facto language for programming
using an Oracle database.

Oracle Pl/Sql is based on Ada, it says so in Appendix C of the Oracle 11g Pl/Sql
Reference Manual, and has said so in the version of the manual that has
accompanied each major database release back to Oracle 8 at least. If memory
serves me correctly, it goes back to Oracle 6. It might even go back further
than that.

The Ada 83 Language Reference Manual is copyright the US Government
(, which is not
surprising as they paid quite a lot to get the language developed in the first

So - does this mean that Oracle copied the Ada APIs when they created Pl/Sql,
and if so, was a licence necessary, and did they obtain one??

Are they guilty of themselves doing exactly what they accuse Google of??

I'm sure that plenty of your readers with better memories and legal minds than
mine will know the history and details.

Keep up the good work


[ Reply to This | # ]

An API cannot be copyrightable because it has no fixed form
Authored by: LouS on Sunday, April 22 2012 @ 09:34 PM EDT
In order for something to be copyrighted it must be fixed in some tangible form. But an API has no tangible form - it is an abstract idea. What confuses people is that there are several entities related to an API the do have fixed tangible forms.

There can be an implementation of an API, which is copyrightable source or binary (or byte) code. But there can be many different implementations of the same API, so the implementation is not the API

There can be documentation of an API which is copyrightable human- language text, aimed at either the user or the implementer of the API. (Documentation aimed at the implementer is often called a "specification".) But there can be many different documents describing the same API, so the documentation is not the API.

Consider the words "Gee" and "Haw" which were commands used to tell a plow horse to turn left and right respectively. (You might consider these a "Application Plowing Interface :) The fact that Gee means turn left and Haw means turn right are abstract facts, separate from the fact that these meanings are listed in some document or that horses are trained to respond that way. There is no way to patent the meanings of Gee and Haw because theses meanings have no fixed form,

[IANAL so take the above with a grain (at least) of salt.]

[ Reply to This | # ]

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: Anonymous on Sunday, April 22 2012 @ 11:24 PM EDT
The last comment is totally bogus. The aforementioned APis are already
made available under GPLv2 in OpenJDK. The question is why GOogle
have not derived their implementation from this??

[ Reply to This | # ]

Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability ~pj
Authored by: Anonymous on Monday, April 23 2012 @ 03:46 AM EDT
The API's might not be copyrightable, however the implementation of the API's can be copyrighted and hence can fall under the GPL. Given that OpenJDK also distribute an implementation of the API's, OpenJDK can therefore be GPL'ed.

[ Reply to This | # ]

Other side of the equation
Authored by: vadim on Monday, April 23 2012 @ 05:44 AM EDT
It is very disturbing to see that when discussing copyrightability of API people seems to forget about the other side of copyright equation : Promotion of Progress of Science and useful Arts.

The U.S. Constitution says: To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries

So the question is If we allow copyrights on API are we promoting Progress of Science and useful Arts?

IMO the obvious answer is NO. On the contrary this will open the potential of miriads of new lawsuits. In will have an enormous chilling effect on industry.

[ Reply to This | # ]

Is this definite? Seems like there was some doubt on day 5
Authored by: Anonymous on Monday, April 23 2012 @ 07:05 AM EDT

Considering this article about the judge taking the copyright issue was during day 4, and the next day (day 5), there was apparently some doubt on the issue, how certain are we that he will rule on this issue rather than allow the jury to?

It seems possibly dangerous to allow a jury to rule on an issue like this unless they are given very clear instructions and information on the current law. They are probably confused... story=20120420210807212

Judge Alsup: I am understanding that there is a cloud over Apache (Harmony). What is the nature of that (cloud)?

Google: The fight was over Apache wanting to say that Harmony is Java. They wanted that right. Sun didn't want to give Apache that. (It's the) same with Android. "If you don't want to call it Java, fine." Everyone knew who was using what APIs.

Google: "The copyright claim is a little crazy."

Google: "None of the engineers talked about that" [an API copyright].

Google: "None of the business people ever talked about that."

Judge Alsup: That's what you argue. Even if you are right… will let the decision on SSO go to the jury. [Not clear to me that Judge Alsup has made a decision on this or is just mulling it out loud.]

Judge Alsup: What is the comparison made against?

Judge Alsup: What is the copyrighted work?

[ Reply to This | # ]

Final rejection of patent 5,966,702 overturned
Authored by: Winter on Monday, April 23 2012 @ 07:42 AM EDT

German Heise reports the rejection of Oracle patent 5,966,702 is overturned (I have difficulty getting at the page). It is valid again: Reexamination of US5,966,702


Some say the sun rises in the east, some say it rises in the west; the truth lies probably somewhere in between.

[ Reply to This | # ]

Oracle is being stupid
Authored by: Anonymous on Monday, April 23 2012 @ 08:18 AM EDT
If they win, WINE loses.

[ Reply to This | # ]

Possible definition of "the API"
Authored by: lwoggardner on Monday, April 23 2012 @ 08:41 AM EDT

Is this a way to define the API?

Source code contains the API, Code (implementation) and Comments (both describing the API and the code)
API documentation (Javadoc) is the API + Comments (the ones describing the API)
Bytecode is the API + Code

We know that the source code is copyrightable, and the bytecode is copyrightable, but this question is whether the API on its own is copyrightable.
So if we look at what is common between the bytecode and the javadoc then the API as a whole would be...

  • the package namespace (being an unordered list of classes)
  • the class hierarchy (inheritance relationships)
  • the unordered list of instance and class variables (name and type) for a class
  • the unordered list of methods defined for a class, a method being a name, return type, and ordered list of parameters (name,type)
  • If Judge Alsup determines that as a whole is copyrightable then the Dalvik source and bytecode would likely be considered derivative works of that API. Oracle have given that API away under GPL+CE. The Classpath exception would not apply here (since not about linking other modules) so a remedy for Google would be to GPL2 Dalvik.

    What is the value to Google of the permissiveness of Apache 2 over GPL2?, or does Oracle's past actions in respect of Harmony mean that the chance to monetize that difference has come and gone?

    [ Reply to This | # ]

    FUD in the news picks
    Authored by: Anonymous on Monday, April 23 2012 @ 09:49 AM EDT
    This article in the news pick is simply wrong. In fact if your GPL+CE jdk is
    derived from openJDK you can get a special TCK. Red Hat have already done this
    with icedTea. Its what caused the Apache Harmony fallout since Harmony was not a
    derivative. Its not fair or pretty but java is 100% free under GPL+CP - stop
    spreading FUD. (Interestingly it is possible to use the 'Reflection Mechanism'
    to extract the full jav API copyright free from openJDK or icedTea - so please
    don't spread FUD on this issue it is important and could even win the case for
    Google - hopefully)

    "The Java IP Story

    2) If you don’t care about the “Java” label, you can forgo certification.
    However, without the certification, you don’t get a license to the software
    patents buried in the OpenJDK code. Thus, any patent owner who happens not to
    like you might decide to sue your customers who will turn back on you and will
    find you without patent defense"

    This story is mis-information and FUD please look here instead for a more
    accurate take:

    "the IcedTea Project reached an important milestone - The latest OpenJDK
    binary included in Fedora 9 (x86 and x86_64) passes the rigorous Java Test
    Compatibility Kit (TCK). This means that it provides all the required Java APIs
    and behaves like any other Java SE 6 implementation - in keeping with the
    portability goal of the Java platform.

    The IcedTea project is not subject to the same licensing issues as Apache
    Harmony, as Sun has provided a special version of the TCK license that is
    targeted to any Java implementation that is a derivative of OpenJDK; something
    that Apache Harmony cannot claim."

    Please do not be helping spread FUD that damages our own case against Oracle.

    [ Reply to This | # ]

    Judge Alsup Decides He, Not the Jury, Will Decide the Issue of API Copyrightability - oops
    Authored by: Anonymous on Monday, April 23 2012 @ 11:00 AM EDT
    Apparently saying that it can be for now

    but also says might change that after verdict?

    [ Reply to This | # ]

    Groklaw © Copyright 2003-2013 Pamela Jones.
    All trademarks and copyrights on this page are owned by their respective owners.
    Comments are owned by the individual posters.

    PJ's articles are licensed under a Creative Commons License. ( Details )