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.

Day 2 of Oracle v. Google Trial ~pj - Updated 9Xs - Ellison, Kurian, Page and Google's Opening Statement Slides
Tuesday, April 17 2012 @ 02:46 PM EDT

Today was the opening statement by Google's lawyer Robert Van Nest, and then the first live witness for Oracle, namely Larry Ellison, the very one. We have a quick report sent during the break from our reporter in the courtroom, with more to follow, so stop back by.

Update 6: We have Google's slides [PDF] from the opening statement now too. And James Temple at the San Francisco Chronicle does an excellent job of explaining what this case is all about.

Jump To Comments

[Update 1, Update 2, Update 3,
Update 4, Update 5, Update 6,
Update 7, Update 8, Update 9]

From our reporter, mirror_slap:

Taking a break now. First objection raised by Van Nest over whether Ellison has the technical expertise to answer, "Is it necessary to use the Sun API to run the Java programming Language?", which seems to me to be a badly-phrased question, but it is Boies asking it, so I assume it's meant to confuse the jury :-)

Judge Alsup asked Ellison if he knew the answer from direct personal experience, to which Ellison replied "yes", and the objection was overruled.

Elllison said no, and cited the Springboard Java implementation. [PJ: I see a comment that may help us comprehend what he is talking about.]

I hope that Van Nest revisits this on cross.

Interesting that Judge Alsup called a recess part-way into Ellison's testimony, leaving him cooling his heels. :-)


Update 7: Internet Archive has the 2008 Sun Microsystems' description of all the Core Libraries in OpenJDK. Here's the opening sentences of the description of java.lang:"Package java.lang Description - Provides classes that are fundamental to the design of the Java programming language. The most important classes are Object, which is the root of the class hierarchy, and Class, instances of which represent classes at run time." Here's my question: is OpenJDK not under the GPL? If so are these core libraries not freely available to the world? And given the requirements of the GPL, can you add further restrictions on top of the GPL? And if Oracle ships GCC/GCJ with its Unbreakable Linux, doesn't that contain all these APIs? How, then can it sue anybody over something it freely distributes to the world?]

All the details will follow at the end of the day's session.

Live tweeting from the courtroom, once again:

  • James Niccolai: Ellison says writing APis is 'arguably one of the most difficult things we do at Oracle' - building the case that APIs can be patented
  • Dan Levine: Ellison didn't talk about importance of Java in $7.4 bln price tag. Alsup had told Google could rip him to shreds on cross if he did
  • Brandon Bailey: Google atty Van Nest cross examining Ellison. Q: Nobody owns the Java language? A: I'm not sure. // Then shows video where Ellison agreed.
  • Ginny LaRoe:Oh, Wow. Google lawyer Van Nest has one low-tech prop to explain "APIs" in Java source code: A filing cabinet.
  • James Temple
Ellison is in a suit and tie.

Yesterday, there was some talk about Apache Project's Project Harmony. In case Oracle thinks it can escape its involvement or isn't "sure" if it was involved or not, here's something to refresh their memory, from 2008, on Oracle's website, showing their direct involvement, some might say encouragement of, Project Harmony. Oh, before I forget, it's about working on class libraries, with a link to where to get them and even how to install them.

I mention this because if you look at Oracle's slides [PDF] from yesterday's opening statement, which Oracle put on its website, you'll see that Project Harmony seems to have disappeared from the narrative. For example, look at slide 34. It says in bright red letters:

Google took Java API designs
But wasn't it, in fact, from Project Harmony?

By the way, I don't think you get copyrights based on how hard something is. Sweat of the brow and all that. Groklaw member nsomos points out in a comment that if difficulty was the standard for getting a copyright, then no one should be able to copyright a photograph, because the camera does most of the work.

I see the Mercury News has Ellison saying this:

If people could copy our software and create cheap knockoffs of our products, we wouldn't get paid for our engineering and wouldn't be able to invest what we invest," Ellison testified.
Umm. You mean like taking Red Hat's software and offering it as a cheap knock off called Oracle's "Unbreakable" Linux? You mean like that, Larry? He's done for the day on the witness stand, I've just heard. But he'll be back tomorrow for more wonderfully inventive quotable quotes. Maybe he can ask around the office and find out if anyone knows if everyone is allowed to run Java freely or not. The report adds this detail:
Ellison's testimony came after Google attorney Robert Van Nest argued to the jury that Oracle sued Google because Oracle wanted a piece of the growing smartphone business after failing to develop its own product.

But under further questioning by Oracle attorney David Boies, Ellison said Oracle's interest in smartphone software was unrelated to the Google lawsuit.

Uh huh. Got it. Unrelated. Is it related to money, though? Maybe? Google's money from Android? I'm not "sure", but here's what Google's lead attorney, Robert Van Nest just told the jury:
It was only after Oracle failed in its own efforts to build a smartphone, according to Van Nest, that Oracle challenged Google's use of Java in the Android software.

"It's not about them protecting their intellectual property. It's not about protecting the Java community," Van Nest told jurors. Referring to Oracle, he said, "They want to share in Google's success with Android even though they had nothing to do with Android."

Can I hear an Amen?

It's Larry Page up next, with Oracle's David Boies trying to turn him into catfood in his deposition video. He later took the stand too.

Information Week's Thomas Claburn interviewed an attorney who asks some interesting questions:

David Long, an attorney at Washington, D.C., law firm Dow Lohnes, said ... that a key issue will be what happens when you have critical intellectual property in a standard like Java that companies have elected to follow. "Is presence in a standards body enough to mean that you will allow people to take a license?" he asked.

Long sees a parallel between FRAND in the patent world--an obligation to offer a patent license on fair, reasonable, and non-discriminatory terms--and the resolution options Oracle and Google will have to consider. He questions whether Oracle can give away Java as a language but still require licensing for Java tools and API access. A possible outcome, he suggests, is that Oracle's ability to seek damages and to obtain injunctions may be limited.

I mean, the language is free to use, except if you do, you get sued?

Update: The Guardian reports this fascinating bit:

Although Oracle has spent more money on other companies, Ellison depicted Java as the company's most cherished prize. "Of all the things we have ever purchased, by far the most important we ever purchased was Java," he crowed in typically bombastic style during his 80-minute appearance in the witness stand.

There were other times during his testimony, though, when he looked bamboozled. At one point, Google lawyer Robert Van Nest reminded Ellison of all the nice things that he had had to say about Android and Google during an on-stage appearance in 2009, when Oracle was still awaiting regulatory approval to buy Sun.

With Sun co-founder Scott McNealy standing by his side, Ellison had assured the crowd that he was "excited" and "flattered" about Android's reliance on Java. Ellison hailed his "friends from Google" and said he looked forward to many more mobile devices running on Android.

Before Ellison took the stand, Van Nest also tried to persuade the jury that Sun Microsystems had encouraged and endorsed Google's use of Java in Android. That contrasted with opening statements on Monday by Oracle lawyer Michael Jacobs, who focused on emails indicating that Google's top executives knew for years that they should be paying to license some parts of the Java technology that helped create Android.

"Tried to persuade the jury"? That makes it sound like he was spinning a tale. But, it's simply true. A fact. That's why then-CEO of Sun Jonathan Schwartz is on tap to so testify on Google's behalf, as the article goes on to report. Here's a link to that video of Scott McNealy and Ellison talking about Java and Android. He clearly says he saw no reason why some of the Java devices couldn't come not just from Google but also from Sun. Another one, in which he says Oracle's entire product line, above the database, is based on Java, and they loved it for the same reason the audience of Java developers did, because it is "open and extensible".

And Google's lawyer Robert Van Nest did explain to the jury about the emails, that they were in a context, from back in 2005 and 2006, when the two companies were trying to work out a work partnership to co-develop Android, which fell apart when Sun insisted on charging for Android, which Google wasn't interested in doing.

Update 2: I *finally* found the Sun announcement about releasing Java under the GPL in 2006. Maybe this will refresh Mr. Ellison's mind about Java being free, but I must say, if Oracle would stop making documents disappear or finding new corners to stick them in, they might find it easier to remember things:

If anyone can get that ogg download to work, send it to me, please. I love rehashing historic events. Also, you might want to see what other goodies lurk in the far corners of the Archive. Do share. Happy hunting.

Thank you, Internet Archive, for keeping the world's players more honest.

Update 4: You can find notes taken from the video in this comment. But it leaves out my favorite part: the final graphic, which reads:

Java. Free and Open
Paging Mr. Ellison. Paging Mr. Ellison. Here, see for yourself, from this screenshot:

Update 3:

And here is mirror_slap's report on the day's events. Most of it. He'll send the final bit after he rests up a bit:

Oracle v. Google, Day 2
Opening statement from Google
First Witness: Larry Ellison
Second Witness: Larry Page
Third Witness:

By the time I got to the courtroom at 7:30AM, they had already started. The jury was not seated.

Judge Alsup: It has boiled down to a handful of copyrights. It is not a case of Java versus Android. If and when Mr. Ellison tries to work in larger dollar amounts, Mr. Van Nest can bring the subject back to this trial. To Oracle: the letter with the offer of $2B is too self-serving. You are not allowed to bring the letter to the attention of the jury.

Regarding questions from lawyers (regarding the introduction of evidence via witnesses or not)" [Goggle asked a question, but I missed the question]
Judge Alsup: Don't give it to me in mismash form. Other judges will make you designate weeks ahead of a trial. I am trying to save you time and effort by not doing that.

Google: Having some trouble with Oracle's counsel [regarding the amount of documents they might cite]
Judge Alsup: They [Oracle] will have to use it. Oracle will have to read every word of it, and them will burn up their time.

Google: Issue with trial exhibit TX431. It has proprietary Google information.
Judge Alsup: I will allow it to be used. This is a public trial.
Google: But this has strategic information to 2013 in it.
Judge Alsup: This is a public trial. Request denied.

Oracle: Can exhibits come in other than via witnesses?
Judge: Here's how that part works: If the lawyers agree to have an exhibit placed into evidence, then they will be published to the jury without explanation (or with explanation from a witness). If there is an objection, the exhibit will not come in by Court order. A foundation must be laid.
Oracle: We'd need time.
Google: We've gotten a long list of exhibits from Oracle where they say that they will be bringing them into evidence.
Oracle: We sent a list of 14 over.
Judge: That seems reasonable.
Lawyers for the parties agree to confer on exhibits becoming evidence.

Judge: Before we start with the Google opening statement, I want to say to the two teams and the members of the public that no hacking, coughing, or shuffling of papers. I do not want anything to distract the jury's attention from what is being presented to them.

7:50. Jury seated.

Handouts are given to the jury. One is a timeline and the other is a glossary of terms. Both have been stipulated. Each is a single page.

Judge: One-page documents are good. To the jury: you can fold these and put them in the back of your notepad for handy reference. You can write on them. They are yours for the trial. When the trial is over, they and your notes will be shredded.

Google's Opening Statement, Robert Van Nest presenting.

The phenomenal success of Android is the result of the hard work of Google engineers. The Java language is free and open to the public. It was put into the public domain by Sun. As you heard yesterday, there is nothing wrong about using Java.

Google didn't need a license to use Java in Android. The source for Android was written by Google engineers or it came from other Open Source projects. Yes, it is true that Google negotiated with Sun to build Android together. Google put 3 years, 100K hours and millions of dollars into it. Android source was publicly posted (the source code for Android). Sun could see it. They could download the source code for it. Sun didn't complain. Jonathan Schwartz the CEO of Sun, posted on his blog on Sun's web site, congratulated Google for Android in 2007, saying "We support Android's use of Java".

The problem came in 2009 when Oracle came onto the scene. Oracle wanted to go into the mobile market. They tried to build their own mobile phone and failed. They tried a second approach and that failed. Lastly, they tried to partner with Google on Android, and that failed.

There are 15M lines of code in Android. Sun was fully aware of what was in Android in 2007. Nobody else had successfuly built a smartphone using Java.

The Java language is open and free for anyone to use. Mr. Ellison knows this. He said so in his deposition.

The names of the disputed API package specification are all free to use as are the class names and method names. The Java language APIs? Sun said that they are fundamental to the language. Sun's idea was to say, "Let's get everybody hooked on Java". What Oracle is claiming is akin to saying that Java is English, the APIs are nouns and the libraries are verbs. You can speak English for free, but don't use any nouns or verbs.

GNU Classpath, the Open Source implementation of the Java APIs and libraries, was written in 1998 using all the APIs that Oracle is complaining about.

How Android came about: Google's approach for products is to give them to consumers for free and to charge advertisers for access. Google has a huge R&D investment. Of the 13,000 employees, 4 in 10 are R&D engineers. Google gives Android to manufacturers for free. It gives Android to developers for free…some 100K+ Android developers. [Puts up his own timeline, which has some additional elements in it than the stipulated one the jury has, but it is at an angle where I cannot see it]. We start in 2005, but before that, in the 90's was the GNU classpath implementation. Google's key decision regarding the smartphone was: buy or build? Google talked with a lot of companies about that. In 2005, Eric Schmidt talked with Scott McNealy about a partnership, with Google offering to buy the actual code in the libraries rather than Google building them from scratch. Sun wanted a lot for it, and they couldn't reach an agreement. So Google built it themselves, and they did not use Sun's technology.

[Van Nest shows a functional block-diagram of the Android software -- Apps
app framework
libraries (not Java) and Android runtime (the disputed part)
Linux kernel

[This demonstrates the proportion of Android that is in dispute. Of 15M LOC and over 57K files, the visual is good.]

Oracle's claim is about "structure, selection, organization". What is an API? Words a programmer would use while writing a program. Example is given of a website (Amazon) where the programmer wants to be able to show products ordered by price, using the max() method as defined by the API and implemented in the library source code.

Oracle said (yesterday in their opening statement) that the API is the blueprint. They are not a blueprint. java.lang.math.max() is a typical usage of the Java language, and this is taught in colleges to students of programming everywhere.

"I am going to approach the cabinet" [a file cabinet labeled "Java.lang" with drawers labeled "math" and other similar names. He steps through the organizational aspect of invoking the method, how the packages are organized, pointing out the natural alphabetic sorting.]

The source code inside Android is different from Java because Google wrote it from scratch. Mr. Jacobs said yesterday that there was copying, but that there was not a lot of it. 9 lines out of 15,000,000. These lines came from a developer that Google hired from Sun late in the development of Android, and it should not have happened.

Sun publicly approved Android's use of Java. Google published Android in 2007. Sun didn't object to it. On his blog, on a Sun-sponsored site, Sun CEO Jonathan Schwartz added a post, congratulating Google on their Java/Linux smartphone, saying, "It is pleased to add Android to the list". Sun said this publicly, and privately (in 2007), Sun told Google, "Let us know how we can support your announcement. We are happy to do so".

In 2008, Sun tried to build their own Java-based phone. Enter Oracle. They too wanted to enter into the smartphone market.

Larry Ellison spoke at the April 2009 Java One conference after the announcement of the agreement for Oracle to buy Sun. Ellison said that "Sun has opened up Java and has given it away" and "We are flattered by Android. It is great to see Java devices coming from our friends at Google, and I expect more devices to come… why not from Sun/Oracle?".

Google transformed the APIs and did something that nobody else had done.

Google is a participant in the JCP, and Oracle's opening argument slide left them off. This is odd because Google contributed a lot to Java.

Examples of Sun trying to build a smartphone using Java: They launched 3 major projects and none succeeded. Oracle wanted to get into smartphones and compete directly against the Apple iPhone. They intended to base the phone on Java FX, and launched "Project Java Phone". Oracle looked at building it on top of Android, but found that (in Oracle's words), "they had very limited expertise to be able to succeed". Oracle contemplated buying RIM, but RIM was too expensive. They considered buying Palm, but considered them not to be competitive. Finally, they tried approaching Google CEO, Eric Schmidt with a business proposition that would allow Android phones to boot faster, run faster and use less battery power, but the two parties failed to agree on terms.

The Lindholm emails had nothing to do with the mid-late 2000's negotiations. The Lindholm email was written in 2010, which was after Oracle had threatened to sue. Oracle was effectively asking Google to change the floor plan of a house that had been completed and was occupied.

Google has made fair use of the Java language APIs.

Request to the jury: remember that we go second, Oracle first. So please keep an open mind.

Google expects the following 3 findings to be reached:

1) there was no copyright infringement; the language is free and the APIs are necessary to use it.
2) Sun approved its use.
3) Android is a fair use of the Java APIs.
[I have to hold myself back from applauding.]

Judge to jury: We are only dealing with copyright in this phase of the trial.

First witness: Larry Page via video deposition, Aug. 24, 2011

Trial exhibit 1 (TX1), deposition 517.
[Lots of starting and stopping of the video. Oracle's counsel has messed up their exhibit numbering scheme, so it's confusing to everyone. The judge is not at all happy about it, because as each email that Oracle is asking Larry Page about comes up, its exhibit number needs to be translated to the *correct* exhibit number on the fly.]

Boies appears to be trying to box Larry Page in while asking about documents from 2005. Asks what "TCK" is. Page doesn't know. Page is obviously thinking about the questions before answering them, ensuring that he understands their scope and is answering correctly. Working on fragmentation, Page said that Google open-sourced Android, and that this implies fragmentation.

Second witness, Larry Ellison live

Larry enters the courtroom in a 3-piece gray suit. Walking a little stiffly, he takes the stand and is sworn in.

Mr. Boies asks Mr. Ellison to tell about his background, upbringing, schooling, founding Oracle, how big Oracle is today.

Ellison: Oracle was founded in 1977 with 4 people. Over 100K people employed there now. Still president… no promotion since 1977.

Oracle: What is Oracle's business?

Ellison: Designing, building, and selling computer hardware and software. [He hammers design aspect.]

Oracle: Asks about copyright protection of designs.

Ellison: If we didn't protect our designs, then anyone could create a cheap knockoff and undercut our prices, leaving us no revenue to fund more R&D.

Oracle: Let me focus very broadly on Java. [Oxymoron.]

Ellison: Basically, a programmer writes a Java program and then runs that program in the environment.

Oracle: Java programming...

Ellison: Programmers use pre-written code. This makes them faster and more productive. The APIs are a command structure, used as building blocks.

Oracle: Is it necessary to use the Sun API to run the Java programming language?

Google: Objection, calls for expert testimony.

Judge to Ellison: Do you know the answer to the question, from your own personal knowledge?

Ellison: Yes.

Judge: Overruled.

Ellison cites the European company Spring, which produced their own version of Java.

Oracle: Tell me about the licenses.

Ellison: Oracle has a variety of licenses. Open Source code is covered by the GPL, free so long as any changes and additions are published. Also gives a license to see and use the source code and any patents.

Oracle: Was a license available to Google?

Ellison: Yes.

Oracle: What is the Java Specification License (JSL)?

Ellison: The JSL lets you look at all the design specifications. It allows you to build your own version of Java. After building your own version, you have to run and pass a compatibility test, called a TCK. Oracle charges money for the TCK. Once the TCK is passed (and accepted by the JCP), you are granted rights to the Java copyrights and patents. The JSL is free, the TCK is not.

Oracle: Why is it important to have compatibility?

Ellison: Uses the example of Apple Macintosh versus Microsoft Windows software. Java is different in that it must be able to be "Write Once, Run Anywhere" (WORA). This only works if there is compatibility.


Discussion of Oracle's mismatch of deposition numbers.

Judge: Will there be any other instances where you lawyers don't follow my instructions? [to Oracle]

Ellison stands and looks a little lost after the judge leaves the bench. Eventually he gravitates to the plaintiff's table. This seems odd to me, but the judge later said that it was okay until after Ellison's testimony of today concluded.

[Break over]

Oracle: Compatibility issues… please explain fragmenting versus forking.

Ellison: In creating incompatible versions of Java, 1) developers have to learn 2 versions, and 2) it ends the WORA promise.

Oracle: JSL requires that the engineer has to create a compatible version of Java. What other kind of licenses is available for Java?

Ellison: A Java Commercial License (JCL). Oracle charges for this.

Oracle: Do other companies take this kind of license?

Ellison: Yes.

Oracle: Please give us some examples.

Ellison: RIM, Amazon, Nokia, LG, Samsung.

Oracle: Was a JCL available for Google?

Ellison: Yes.

Oracle: Did Google take a JCL?

Ellison: No.

Clean room. Ellison explains that a requirement for accepting a JSL is that the implementor agrees that the implementation must be done in a cleanroom, independent of having seen Sun's (now Oracle's) source code.

Ellison: The JSL implementors cannot look at our IP.

Oracle: Explain the Java Community Process (JCP).

Ellison: IBM, Red Hat, SAP and others all contribute to Java. IBM, Oracle, SAP, HP, and Red Hat are all board members of the JCP. Google is on the board. Requests for additional features to be added to Java go before the board. Example, the latest version of Java, Java 7, was just approved before the board. Google was the only company that voted against approval.

Oracle: When you became aware that Google was copying the APIs, what steps did you take?

Ellison: [I missed, but effectively they asked Google to stop.]

Cross-examination of Larry Ellison by Mr. Van Nest:

[This went by very fast. I missed a lot, but Mr. Van Nest was able to show on at least two occasions direct conflicts in Mr. Ellison's video depositions versus his testimony before the Court today. For me, it spoke to credibility of the witness. It was really blatant.]

Google: The Java Programming Language (JPL) -- nobody owns the Java Programming Language, right?

Ellison: I am not sure.

Google: Anyone can use the JPL without paying royalties, yes?

Ellison: Not sure.

Google: Sun created the APIs, yes?

Ellison: Yes.

Google: You distinguished between the APIs and the libraries. You said that, "API's are a command structure."

Ellison: [I think he said, Not sure.]

Google: Didn't you just testify under oath ten minutes ago that the API is a command structure?

Ellison: [missed the response. Mr. Ellison is not speaking up, but Mr. Van Nest really is.]

Google: You didn't participate in the negotiations with Sun in 2005-06?

Ellison: No.

Google: As far as you know, Google has never branded its phone as being a Java phone?

Ellison: No.

Google: [Established that Jonathan Schwartz was CEO and was in charge of making decisions for Sun.]

Ellison: Agrees.

Discussion of Ellison's separate meetings with Eric Schmidt and Larry Page in 2010, when trying to come to terms.

Google: Do you know that Android was released in 2007 and the first Android phone came out in 2008?

Ellison: Knew when the first Android phone came out.

Google: Asking about the April 2009 Java One speech with Scott McNealy [TX 2939 accepted with no objection]. Played for the jury. McNeally expresses that a lot of the Java One attendees are concerned that "we (Oracle/Sun) are going to close the technology up." Ellison says, "Don't expect a lot of change".

Google: Is that you in the video?

Ellison: Yes.

Google: Were the remarks you made recorded and given to the government?

Ellison: Don't know.

Google: [Requested permission to approach the witness with exhibit 2041, a transcript of the words spoken at the Java One conference.]

Google: Confirm that the words are the same as what was spoken.

Ellison: They are.

Google: From Page 7 of TX 2041, items sent to the SEC (Securities Exchange Commission)-- [again, a blur of questions, getting Ellison's quotes into evidence]:

"James … has done a great job of opening Java"
"Oracle is flattered that Android used Java"
Google: The press was there (at Java One)?

Ellison: Yes.

Google: Lots of developers?

Ellison: Thousands.

Google: Hadn't you tried to build your own Java phone?

Ellison: No. We explored that idea and decided it would be a bad idea.

Google: When you acquired Sun, wasn't the primary idea to enter into the smartphone market?

Ellison: No.

[Comparing now with the video deposition-- two apparent direct contradictions]

Ellison: Java was a factor in the decision, not a primary motivation.

Google: TX 2042-- asks Ellison to identify it.

Ellison: It's an email that I wrote to Scott McNealy.

Google: After Oracle agreed to buy Sun, proposing building a smartphone based on Java.

Ellison: [Equivocates some.]

Google: Did you learn that Sun had tried and failed to build a smartphone?

[Much back-and-forth. It turns out that McNealy had sent Ellison's email to James Gosling, and Gosling replied to Ellison. TX 2043: May 2009 Gosling to Ellison.]

Ellison: Sun never funded the [smartphone] project.

Google: What was Oracle's approach to building a smartphone?

Ellison: Looking at all options and making a decision.

Google: One of the potential starting points that Oracle evaluated was using Android?

Ellison: We did a detailed study of Android.

Google: This was in 2009, before you met with Eric Schmidt?

Ellison: Yes.

Google: You asked your engineers to look at the top 3 manufacturers who were building Android?

Ellison: Yes.

Google: Enter TX 2044 exhibit of "Project Smartphone at Oracle". What was "Project Smartphone"?

Ellison: An analysis of what it would take (resources) for Oracle to enter into the smartphone market.

Google: How long did it take to produce? Weeks? Months?

Ellison: I think weeks, not months.

Google: Look at the agenda item, "Proposed M&A associated with the project."

Ellison: If we didn't have the expertise inside Oracle, we'd have to either hire a bunch of people with the expertise or buy a company.

Google: You considered buying RIM or Palm?

Ellison: We looked at everything. RIM was too expensive. Palm wouldn't work well.

Google: To page 6, "Handsets and HW Options, Recommendations for V1.0"
"talk with 3 Android handset vendors, sign up with one"

Google: The idea was to save time?

Ellison: It would be convenient and would save time to market.

Google: Page 11, "Handset/HW"
"Target a single OS only- Android Linux"

Ellison: [Answer lost. Lots of fast words. Van Nest jumps in.]

Google: Excuse me, your engineers said that this was Android Linux.

Ellison: [?]

Google: Page 12, "Next steps for development"
"Run JavaME (mobile edition?) on Android".
But (in the doc) you (Oracle) had problems… you didn't have the internal expertise to be able to make smart decisions.

Google: I take it that you didn't develop a smart phone, based on this evaluation?

Ellison: Yes.

Google: After this evaluation (build), you went to to evaluate buying Palm?

Ellison: Yes.

Google: [Moving onto the meeting with Eric Schmidt, CEO of Google at the time.] What was the business proposal to Google?

Ellison: Google would take (a license for) Java from Oracle and put Java into Android.

[Discussion about whether this was a joint project, Google and Oracle engineers working together. Ellison denies that it was a joint project. The playback of Larry Ellison's video deposition appears to directly contradict his testimony today, again.]

Google: You never got an agreement with Mr. Schmidt?

Ellison: Yes (we never got an agreement).

Google: You met with Mr. Page (CEO) later and made the same pitch?

Ellison: With fewer options than were offered to Eric.

[Referencing page 90, lines 23-191 of deposition… about the pitch to Larry Page]

Google: Mr. Ellison, that never happened [was not agreed to, either].

Ellison: Yes [no agreement].

Google: And the next thing that happened was this suit?

Ellison: Yes.


Oracle: Can you copyright a computer programming language?

Ellison: I don't know if you can copyright a language. I just don't know that.

Oracle: Question about the quote that Sun/Oracle was flattered by Android using Java.

Ellison: I always thought that we'd be able to make Google make their Java compatible.

Oracle: [Some questions about Oracle building smartphones. Also established that LE couldn't do any deals w.r.t. Sun's IP prior to the acquisition closing.]


Google: [TX 2362, email from Jonathan Schwartz… getting the timeline of the email… Sun One conference… into evidence]


Judge: Encourages Oracle to get their act together on their exhibit numbering.

[End break]


Third witness, Thomas Kurian (sp?) Executive VP for Software development, Oracle

[to be continued]

Update 5: And here is part two:

Third witness, Thomas Kurian, Executive VP for Software Development, Oracle

Reports to Larry Ellison.

Oracle: How does Java fit into your job?

Kurian: Java is used throughout Oracle's product families, including the Oracle Fusion Middleware (OFM)

Oracle: Please describe Oracle's use of Java before its purchase of Sun. How did it happen that Oracle started using Java internally?

Kurian: We integrated Java into our database products, into creating development environments (JDeveloper), tools (OFM), and packaged applications.

Oracle: What made Java so attractive?

Kurian: 1) availability of Java developers; 2) Java enforced a set of standards that allowed multiple vendors to produce competitive products that would satisfy large customers seeking to avoid technological lock-in;

3) eliminates cost of porting, WORA.

Oracle: How do Java APIs and class libraries make Java attractive?

Kurian: Class libraries are packaged program objects that at are pre-packaged utilities that are extracted through the API. Class libraries improve productivity, cut time to market. The Java specification describes the set of standards to which the language adheres.

Oracle: Is there a particular spec that Oracle made its products work on?

Kurian: JSE vs. JEE vs JME (Standard, Enterprise, and Micro).

Oracle: Did Oracle take a license from Sun prior to the acquisition?

Kurian: Yes, we took all 3 varieties of licenses: JSL, TCK, JCL. The TCK, once passed [ and approved by the JCP ] gives trademark and copyright rights.

Oracle: Are there other types of licenses available?

Kurian: Yes. There is a royalty-free one for non-profit organizations and schools. And there's the GPL.

Oracle: How does the GPL work?

Kurian: OpenJDK is the reference version, contributed to largely by IBM and RedHat. Source code that you use for the JVM, if you modify it, you have to contribute back the changes. This helps to keep the language from fragmenting.

Oracle: References exhibit TX610.1 - J2SE Development Kit License statement.

Kurian: Grants a perpetual, non-exclusive worldwide, limited license.

Oracle: How does the specification relate to the implementation of a class library?

Kurian: For a developer implementing class libraries, there are additional requirements:

1) Implements all functionality fully. Cannot just implement a part. Without this requirement, standard programs might not work on the specific implementation, and programs written for the new implementation may not work on a standard VM.

2) Prevents Java from vendors who want to add or subtract from Java. Cites Microsoft as an example of this.

Oracle: What did Sun do about Microsoft?

Google: Objection.

Judge: Sustained- answer would be hearsay.


3) TCK - if you are a class library developer, you have to run the new class library through the TCK. The JSL is free. The TCK costs (a nominal charge).

4) Implementation has to be independent, not derived from proprietary work from Sun/Oracle. [gets a little into "cleanroom" here]

5) Trademarks- once the new class library has passed the TCK [and has been approved by the JCP], the developer can use the Java trademarks.

Oracle: Are there companies that have made an independent implementation?

Kurian: Yes. Many.

Oracle: Did Google take a license?

Kurian: No.

Oracle: Can you cite some companies who have taken a license?

Kurian: Sony, for their Blu-Ray drives; Cisco for their soft phone,; GE for home appliances; Nokia for Symbian [ooops!]; eBay, for Java Server Pages; VISA, for Java card; Amazon, for Kindle.

Oracle: All of these companies took a Commercial license (JCL)?

Kurian: Yes.

Oracle: How about TCK licenses? [what companies took TCK licenses]

Kurian: IBM for Websphere has JSL, TCK, and TCL.

Oracle: Open Source licenses, how does that work?

Kurian: OpenJDK is the reference version. The main issue with the GPL is having to contribute back.

Oracle: is Android under any of these licenses with regard to Java?

Kurian: No.

Oracle: Since the acquisition, did Oracle's role w.r.t. the Java Community Process (JCP) change?

Kurian: Yes. In January, 2010, Oracle's goals changed. Oracle

1) became the steward of the JCP

2) took over engineering of the reference implementation

3) became sponsor for user groups and the development community.

Oracle: Any idea how much Oracle spends on Java each year?

Kurian: Hundreds of millions.

Oracle: Will Oracle's expenditures on Java change?

Kurian: No.

Oracle: Did you have a meeting with Andy Rubin?

Kurian: Yes, in May 2010.

Oracle: Did you have a meeting with Alan Eustice (sp?)?

Kurian: Yes, in July 2010.

Oracle: What did you discuss in these meetings?

Kurian: Two topics:
1) Proposed mechanism by which Google could make Android compliant, which consisted of 3 sub-topics:

1a) Make Dalvik compliant, or eliminate it and replace it with [sounded like CVC?]
1b) Make the Java class libraries compliant
1c) keep the current apps runnable under the (new) compliant JVM.
2) [ missed this]
Oracle: With regard to 1b), what did you say to Google?

Kurian: Google must make the class libraries compatible. Within Android, there is an implementation of the class libraries from the Apache Harmony project. Google would have to replace these libraries. Google rejected this.

In the next meeting, in July 2010, the same topics were discussed in a different meeting: remove Dalvek, remove Apache Harmony libraries.


Google: You mentioned that you spoke with Mr. Rubin about Apache Harmony, something that wasn't owned by Oracle.

Kurian: Yes. Apache Harmony was never granted a license.

Google: It was first published in 2005. Did you ever do any investigation into it?

Kurian: Intel and IBM helped with Apache Harmony.

Google: All 37 of the impeached API's are in Apache Harmony?

Kurian: Yes.

Google: Apache was out there using these libraries since 2005… Apache was using the same libraries and API's.

Kurian: Sun was constantly trying to get Apache to take a license.

Google: Please describe the API.

Kurian: The API consists of 3 pieces:

1) blueprint
2) specification
3) implementation
Google: Apache used all of these with no license from Sun, at the same time that Google (Andy Rubin) said that Google didn't need to take a license.

Kurian: [I think that he said yes? Sorry, cross-examination is fast. ]

Google: The Dalvik VM is different than the Java VM?

Kurian: Yes.

Google: [Manages to establish that Oracle wanted Google to buy its JVM to replace Dalvik.]


Oracle: What was the problem with Apache Harmony not having a license?

Kurian: Apache Harmony couldn't guaranty that downstream developers would remain compliant.


Google: During the period before Oracle bought Sun, wasn't the Sun CEO, Jonathan Schwartz, responsible for the relationship with Apache Harmony and Android?

Kurian: [something grudging, like, "I suppose so"]

Larry Page

Oracle: In 2005, Google was considering acquiring Android from a startup?

Larry Page: Yes.

Oracle: Who was responsible for Android at Google?

Larry Page: Andy Rubin.

Oracle: Produces exhibit 1, July 26, 2005 Android presentation, "Key Strategic Decisions around Open Source".

Oracle: Who prepared this document?

Larry Page: Andy Rubin and the Android Team.

Oracle: Referring to page 8, "Why Java"?
carriers (telcos) require it.
MSFT will never do it
Elegant tools story
Safe sandbox for 3rd party
[Oracle: starts going after LP regarding Googles cleanroom practices, to the point where Judge Alsup admonishes Bois that he is overstepping.]

Oracle: In common industry practice, what is a cleanroom?

Larry Page: [Describes the essential elements of a cleanroom ]

Oracle: In terms of the cleanroom implementation that Google had, what steps were taken to prevent the use of proprietary source?

Larry Page: [Explains that he had no direct influence over how the cleanroom was implemented, that this would have been the responsibility of Andy Rubin's team.]

Oracle: [Further question on cleanroom implementation directions to Andy Rubin.]

Larry Page: I don't recall ever talking with Andy Rubin about cleanroom implementation.

[Larry Page is holding fast that he knows nothing about code copying.]

Oracle: If you discovered that lines of code had been copied from someone else's IP, would this be a violation of Google policy?

Larry Page: The question is a hypothetical one, but yes, it would be a problem.

Judge: This witness says he has no direct knowledge. Move on.

Oracle: Back to the exhibit,.. "must take a license from Sun". "Proposals: G/A with support from Tim Lindholm, negotiates the first OSS, J2ME, JVM license with Sun"

Oracle: Do you know Tim Lindholm?

Larry Page: [I don't know him well.

Oracle: Did you ever give him any direct instructions?

Larry Page: I don't recall.

[On cross-examination, Google got the point made that the Lindholm emails are happening in 2010, *after* Oracle starts saber-rattling.]

[Larry Page is dismissed.]

Judge: How many methods are in the 37 API's?

Oracle: 8,693 methods.

Judge: Are these the same as are in the Oracle package [I assume that Judge Alsup is asking about the namespace?] I want to know what differences are there. Are there any additions/subtractions?

Judge: Did Sun develop the methods themselves, or were they developed by some 3rd party?

Judge: I need answers from both sides regarding the "Declaration of API elements". What is this?

Oracle: The first line of the API and the corresponding class libraries.

Judge: When something is considered to be a derivative work…

Google: The derivative work question depends on what is being claimed. The declaration (or method signature) is

a) the name of the method
b) variable type returned
c) order and type of arguments
[Further exploration of nomenclature, and the Court adjourns at 1PM.]
I know you join me in wanting to say thank you to mirror_slap for his extraordinary note-taking abilities and for his willingness to take time from his busy life to be our eyes and ears this day.

Update 8: I just found this on Internet Archive, an interview with James Gosling, the inventor of Java, on the day that Sun released Java under the GPL:

Q: What kinds of things can a developer do with the open-source Java SE platform pieces right away?

Gosling: Probably the most useful thing you can do with it is look at it and learn from it. It is somewhat traditional, but I always say that the source is the documentation of last resort. We have tried really hard to make sure that we have real documentation.

Of course, you've been able to read the code for a very long time as part of the JDK 6 project. There will certainly be people who will be locally tweaking things just for themselves.

In the past, there have been a lot of researchers in universities using the code to do research projects. And people who are making contributions will be able to build their contribution and to make sure that all the contributions fit.

Q: How do you think this move will affect other open-source implementations of the Java programming language -- for example, Apache Harmony or GNU Classpath? Gosling: It's hard to know. They'll certainly be able to mine our source for stuff to incorporate into their projects.

In other words, Gosling told the world that Apache Harmony and GNU Classpath could "mine" the Java source code and "incorporate" that code "into their projects".

That's what Oracle is suing Google for using, the Apache Harmony APIs.

Ladies and gentlemen of the jury, I rest my case.

: )

Update 9: All the trial exhibits are now available as PDFs here. Some are also done as text. Look for the date nearest the day, as they are listed by the date they were entered, which could be a day or so after the date of their use in the courtroom.


Day 2 of Oracle v. Google Trial ~pj - Updated 9Xs - Ellison, Kurian, Page and Google's Opening Statement Slides | 342 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
Authored by: Kilz on Tuesday, April 17 2012 @ 02:51 PM EDT
Please mention the mistake in the title of your post.

[ Reply to This | # ]

Off Topic
Authored by: Kilz on Tuesday, April 17 2012 @ 02:52 PM EDT
For All posts that are not on topic.

[ Reply to This | # ]

Authored by: Kilz on Tuesday, April 17 2012 @ 02:53 PM EDT
Please mention the news story's name in the title of the top

[ Reply to This | # ]

Authored by: Kilz on Tuesday, April 17 2012 @ 02:54 PM EDT
Please leave all transcriptions of Comes exhibits here for
PJ. Please post them as HTML in plain text for easy copying.

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj
Authored by: Anonymous on Tuesday, April 17 2012 @ 03:08 PM EDT
Who decides if API's are copyrightable?

What happens to the statements from Oracle and Google Debate: Can You Copyright Computer Languages, APIs?

I don't really understand why this question (are API's copyrightable) isn't answered first. Anything until now was (more or less) useless if they weren't. Or am I missing something?

[ Reply to This | # ]

Owning Java
Authored by: The Cornishman on Tuesday, April 17 2012 @ 03:19 PM EDT
Q: Nobody owns the Java language? A: I'm not sure.

Trick question. I imagine Mr Ellison is confident that his company owns the relevant copyrights in the Java softwares; as we've said a hundred times, the GPL depends on copyright. "Nobody owns" seems to me to be a fuzzy sort of definition of 'public domain', which the Java software clearly is not.

That ownership, though, isn't equivalent to owning the language, which the jury might think meant having a right to control anything and everything expressed in that language. In the case of that interpretation, Mr Ellison indeed mightn't be sure (but wishes Oracle did).

(c) assigned to PJ

[ Reply to This | # ]

Larry - Ellison says writing APis is 'arguably one of the most difficult things we do at Oracle.
Authored by: Anonymous on Tuesday, April 17 2012 @ 03:22 PM EDT
When Larry said:
"Ellison says writing APis is 'arguably one of the most
difficult things we do at Oracle".

Being difficult does not make it special. A farmer plows
many acres of farmland, harvest many bales of hay (often
having to move the bales by hand), gets up
early in the AM to milk cows... all difficult, and most
likely more difficult than Larry's APIs.

So, He deserves protection just because it is hard to do?
Or, because he his who he is, and needs protection because
he want's it (as life is understood in one's alternate

[ Reply to This | # ]

When do friend of the court briefs get submitted
Authored by: Anonymous on Tuesday, April 17 2012 @ 03:50 PM EDT
Are these mostly used in the appeal process, or should this have been done
during the original proceedings? And do courts care about implications of their
rulings, or do they leave that problem to the legislature? One would hope that
when making new law (such as whether API "arrangement" is
protectable), that the effect it has on industry would be relevant.

Whichever way this case is decided, there will be 3rd parties who disagree with
the findings, and will hope to get the appeal decided in their direction.

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj
Authored by: tknarr on Tuesday, April 17 2012 @ 03:59 PM EDT

I think Oracle's trying to confuse the jury. In Sun's (and Oracle's) Java, there's a sun.* namespace. That's the vendor-private API namespace for all the Sun-specific functions that Sun's implementation of the java.* API uses. It isn't neccesary to use the sun.* APIs to run Java, in fact it was assume that other vendors would place their own implementations in their own namespaces (to the point that the Java language spec defines how to create your own namespace: take your DNS domain name and reverse the order of elements, so that Groklaw could use the net.groklaw.* namespace and know it wouldn't conflict with anything or anyone else). But that doesn't change the fact that it's necessary to use Sun's java.* APIs to run Java.

I can see setting up a shell game here, talking about the sun.* namespace and then switching to talking about all the APIs Sun defined for Java without clarifying that you're talking about two different things.

[ Reply to This | # ]

Springboard doesn't need APIs
Authored by: Anonymous on Tuesday, April 17 2012 @ 04:06 PM EDT
I'm interested to hear of this Springboard platform that
apparently proves it isn't necessary to use Java APIs to
"run the java... whatever"

I think the badly phrased question was probably intentional,
not to confuse the Jury, but to allow Ellison to say it is
possible without lying. If the question were phrased in a
straightforward way, I think the answer would have had to be

I'm familiar with Java, Android, and some web frameworks
using different languages, but am not familiar with
Springboard at all. Googling it suggests it's a web
framework, in which case perhaps it dodges the "output"
problem by handing it off to a webserver & browser. Whether
they coded around all the APIs, including java.lang /
object etc is another issue, and frankly I doubt it.

If anyone has specialist knowledge about this framework,
perhaps insight may be had.

It is absurd to claim that Java can be used for anything
meaningful with no APIs, and the fact that APIs are used
even in a "Hello World" app means that there is clearly no
bright line between the language and APIs.

[ Reply to This | # ]

Picture Camera Metaphor Is Recent
Authored by: Anonymous on Tuesday, April 17 2012 @ 04:26 PM EDT

Back in the day, taking a picture with a camera was only
the first step.

Next came developing and then printing.

Both developing and printing could be done different ways,
different solutions, different time lapses, etc. that could
materially affect the picture.

In those cases, the camera did a minimal amount of work --
it opened a shutter for a time period and exposed a piece
of film. Focusing, lighting, etc. were all done by the
photographer (or assistants).

Just Saying.

jcjodoin ... not logged in ...

[ Reply to This | # ]

LarryCodeSpeak: “I’m not sure" == “That’s correct”
Authored by: SpaceLifeForm on Tuesday, April 17 2012 @ 04:28 PM EDT

On cross-examination, Google came out firing and the room got tense quickly. "Do you understand that no one owns the Java programming language?" lead counsel Robert Van Nest asked.

Ellison began a longer answer, but Judge William Alsup interrupted him and said it was a "yes or no" question. Finally Ellison said, "I’m not sure."

"And anyone can use it without royalty?" Van Nest followed up.

"I’m not sure," Ellison said again.

Then Van Nest showed a video of Ellison receiving the same question on a deposition video and answering "That’s correct" to both.

And the Footgun keeps on working.


You are being MICROattacked, from various angles, in a SOFT manner.

[ Reply to This | # ]

Oh, Video Please
Authored by: Anonymous on Tuesday, April 17 2012 @ 05:06 PM EDT
I would just love to see Mr Van Nest demonstrating the relationship
between language, APIs and application source code, using his
metal filing cabinet, like an old fashioned chemistry professor...

[ Reply to This | # ]

Did Oracle violate Sun's copyright before their acquiring Sun?
Authored by: mossc on Tuesday, April 17 2012 @ 05:15 PM EDT

I was just pondering Oracle's Unbreakable Linux (OUL) and thought of something that had not occurred to me before. Perhaps Oracle is in the same boat as Caldera/The SCO Group was in their case as far as distributing what they claim is being infringed under the GPL.

As a company they have distributed a recompiled version of RHEL for a while. I think there have been various java options in the distribution since they launched OUL. Currently I show libgcj and java-1.4.2-gcj-compat-javadoc packages available for RHEL 5. I would bet they are on OUL as well.

From the gnu website

GCJ is a portable, optimizing, ahead-of-time compiler for the Java Programming Language. It can compile Java source code to Java bytecode (class files) or directly to native machine code, and Java bytecode to native machine code. Compiled applications are linked with the GCJ runtime, libgcj, which provides the core class libraries, a garbage collector, and a bytecode interpreter. libgcj can dynamically load and interpret class files, resulting in mixed compiled/interpreted applications. It has been merged with GNU Classpath and supports most of the 1.4 libraries plus some 1.5 additions.

interesting quote from the FAQ

2.8 What features of the Java language are/aren't supported. GCJ supports all Java language constructs as per the Java language Specification. Recent GCJ snapshots have added support for most JDK1.1 (and beyond) language features, including inner classes.

I would think this includes APIs and is licensed under the GPL

A few questions for Larry Ellison:
1. Was Oracle in violation of Sun's copyrights before they were acquired?
2. What license did they rely on to distribute libgcj and gcj?
3. Did that include the APIs?


[ Reply to This | # ]

Hard work & copyrights
Authored by: Anonymous on Tuesday, April 17 2012 @ 05:51 PM EDT
To further pj's argument I would add that the Supreme Court has already
addressed the issue of hard work and copyrights.

In particular, the ruling in which they addressed the matter related to an item
in the public domain. The item had been digitized through a significant amount
of work and copyright was claimed on the digital copy.

However, the Supreme Court ruled that the digital copy was still that... a
copy... and not a creative work.

In the ruling they acknowledged the difficultly involved in digitization, but
very clearly stated that level of effort had absolutely nothing to do with

As a result, they ruled that the digital copy of a public domain item remains in
the public domain.

--- nyarlathotep

[ Reply to This | # ]

critical intellectual property in a standard
Authored by: SpaceLifeForm on Tuesday, April 17 2012 @ 06:22 PM EDT
Like Rambus ?


You are being MICROattacked, from various angles, in a SOFT manner.

[ Reply to This | # ]

On The Value of Sun
Authored by: sproggit on Tuesday, April 17 2012 @ 06:34 PM EDT
I find myself particularly annoyed at the specious mis-representation, by Oracle, of the value JAVA in their transaction to purchase Sun Microsystems. This is a blatant attempt to wow the jury with big numbers in the hope of a substantial damages award.

It is extremely frustrating to see the difference between what the jury are being told and what was reported by Oracle at the time of the purchase.

For example, if you look hereyou will see a press release from April 2009. In this article, you will find comments such as,

“The acquisition of Sun transforms the IT industry, combining best-in-class enterprise software and mission-critical computing systems,” said Oracle CEO Larry Ellison. “Oracle will be the only company that can engineer an integrated system – applications to disk – where all the pieces fit and work together so customers do not have to do it themselves. Our customers benefit as their systems integration costs go down while system performance, reliability and security go up.”

Well, he's wrong, of course. IBM have been doing that for years... So did Honeywell-Bull for that matter.

In fairness, that same article then goes on to describe Solaris and JAVA as Sun's most important assets, but the truth is that they were not the reason that Oracle made the purchase. The reason, quite simply, was that Sun was at risk of going bankrupt, and if that happened Oracle would have taken a big hit. Whilst Oracle is available on many OS platforms, it sells more licenses on Solaris than all the others combined. The failure of Sun would have seriously hurt Oracle's bottom line. In that sense the purchase was protective/defensive rather than a traditional "growth through opposition" deal.

There is lots of evidence to suggest that Oracle really didn't understand what they were getting into, especially with the loss of many of their best hardware people shortly thereafter.

There is also evidence to support my statement that Sun were in a bad way. Witness their 10Q filing for Q1, 2009, which includes news of the proposed Oracle deal, but adds:

"Other key financial metrics for the quarter ended March 29, 2009, as compared to the quarter ended March 30, 2008, include the following:
• Total revenue decreased by $652 million, or 20.0%.
• North American revenue decreased by $211 million, or 16.8%.
• European revenue decreased by $265 million, or 23.8%.
• Gross margin as a percentage of net revenue decreased by 2.2 percentage points.
• Research and development expenses decreased by $64 million, or 14.0%. • Selling, general, and administrative expenses decreased by $146 million, or 14.8%. "

Does anyone see any good news in those 10Q statements? Other than the reduction in R&D expenses [which were cut because they simply could not afford to spend money they weren't earning, not when revenues were down 20%] everything was on the slide.

At the time, Sun probably had 2 more quarters, at best, before they tanked. IBM had looked and decided that there was nothing of value there and had walked away. Ellison did the deal in order to get Sun hardware and MySQL.

I went hunting the 10Q because I was interested to see if the revenue split out that generated for JAVA. Whilst it doesn't do that, exactly, here's what it does say:-

Net Revenues and Gross Margins(dollars in millions)
Server Products: $1,094
Storage Products: 425
Systems Net: $1,519

Support Services: $ 853
Professional Svcs: 242
Services Net: $1,095

Total Net: $2,614

Obvious question: can anyone show me, in here, where it lists out all the vast sums of money that Sun were making from JAVA? No. Because they weren't making vast sums from it, plain and simple. If you look at the total numbers, the only place that JAVA licenses could be accounted for in the above 10Q statement would be in the detail line, "Support Services". That's an amount of $853 Million in a Gross revenue of $2.6 Billion.

The problem is, of course, that Sun licenses lots of other things in addition to JAVA. Every Sun Server sold ships with the Solaris Operating System, for which Sun charged a license. In addition to the license, Sun charged a maintenance and support charge for the software that covered things like bug fixes, second and third tier support and so on. This would have been included in "Support Services". They would also, of course, charged hardware maintenance fees on all their servers.

All of those things appear to have been bundled in this one number for 10Q purposes.

I haven't followed discovery closely enough to see if Google discovery questions went to the topic of "JAVA Revenues", but I'll go out on a limb here and suggest that it's probably significantly less than 5% of the annual revenues being realised by the company.

Quite how Larry Ellison can turn around and try and claim that JAVA is the reason that Oracle bought Sun is beyond me. Especially when surveys at the time were suggesting that JAVA development was on the *decline* in the big development shops...

[ Reply to This | # ]

Anyone else think that Larry would make a great Lex Luther?
Authored by: Anonymous on Tuesday, April 17 2012 @ 07:34 PM EDT
Ellison Flustered as Google and Oracle Argue Over Java.

Anyone else think that Larry would make a great Lex Luthor?

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj
Authored by: Anonymous on Tuesday, April 17 2012 @ 07:42 PM EDT
Where is the list of these 7 APIs that Oracle says Google
stole? I've not seen them listen anywhere. Inquiring minds
want to know.

[ Reply to This | # ]

Ars: Ellison, Page both take the stand as Google argues "Java language is free and open"
Authored by: SilverWave on Tuesday, April 17 2012 @ 08:59 PM EDT
Ellison, Page both take the stand as Google argues "Java language is free and open"

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 | # ]

Boies Career Goals
Authored by: Anonymous on Tuesday, April 17 2012 @ 08:59 PM EDT
Does anybody else get the feeling that David Boies grew up watching "Perry
Mason", but idolized DA Hamilton Burger? I can just hear him telling his
mom "When I grow up I'm going to be an attorney who only takes cases where
I lose and end up looking like a fool!".

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj - Updated
Authored by: Anonymous on Tuesday, April 17 2012 @ 10:02 PM EDT
PJ (in the first(?) update):
And Google's lawyer Robert Van Nest did explain to the jury about the emails, that they were in a context, from back in 2005 and 2006, when the two companies were trying to work out a work partnership to co-develop Android, which fell apart when Sun insisted on charging for Android, which Google wasn't interested in doing.
That was also mentioned in the Guardian article, immediately after the paragraphs PJ quoted from it.

It is certainly true that Van Nest "tried to persuade the jury." No one should read that as just "spinning a tale" as PJ put it.

[ Reply to This | # ]

Found this while poking around
Authored by: Anonymous on Tuesday, April 17 2012 @ 10:37 PM EDT

[ Reply to This | # ]

ogg file
Authored by: Anonymous on Tuesday, April 17 2012 @ 10:39 PM EDT
Re the ogg file that you wanted to get your hands on, is this
the one you meant?

[ Reply to This | # ]

dealing with elephants in the room
Authored by: mcinsand on Tuesday, April 17 2012 @ 10:50 PM EDT
If I was a juror, how BSA/Oracle handles certain elephants
in the room would be critical to me, with one key example
being Project Harmony. Harmony is a lynchpin to this case,
particularly since Google is arguing that they used Harmony
as a source, rather than Java. If I were a juror, the way
Oracle handled the opening slides would be a very bad start.
Oracle's slides seem very clear, to say that there is Java,
Google copied Java, and that is that. When Harmony comes
into the discussion, I don't see how a juror could help but
notice the omission.

When there is an elephant in the room, how it's handled is
very important. In a case like this, the handling makes a
huge difference when it comes to credibility, and Team
Oracle's behavior isn't helping. The topic may be
important, but, by not taking the time to explain their
position of it not being relevant, they made a mistake.
Whether it is relevant in their view or not, I would view it
as lying by omission. If they honestly see Harmony as
irrelevant, then they blew a key opportunity for presenting
their side... in addition to sacrificing credibility.

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial BSF being BSF?
Authored by: rsteinmetz70112 on Tuesday, April 17 2012 @ 11:30 PM EDT
I have to wonder how much of the exhibit "confusion" was to purposely
confuse witnesses who had been prepared for different exhibits..

Rsteinmetz - IANAL therefore my opinions are illegal.

"I could be wrong now, but I don't think so."
Randy Newman - The Title Theme from Monk

[ Reply to This | # ]

Open Sourced Java SE?!! That is the 37 APIs right there...
Authored by: mjscud on Tuesday, April 17 2012 @ 11:42 PM EDT
The difference between Java ME, Java SE, and Java EE is exactly what APIs they
include! And the 37 APIs at issue are in the Java SE package.

If you open source the API implementation then you have open sourced the API!

It is exactly what the good judge warned against, Oracle which succeeded Sun is
trying to take back what Sun gave away.

Even a fool, when he keeps silent, is considered wise. Proverbs 17:28

[ Reply to This | # ]

Major thanks to the groklaw reporters!
Authored by: Anonymous on Tuesday, April 17 2012 @ 11:49 PM EDT
No, seriously. Thank you!. The reports coming out from Groklaw are the best out there. Very detailed and cover the whole thing in a way the others don't quite reach. I love the near word for word transcript and how the breaks are included as well!

Notice how Ellison gets caught off guard repeatedly? I counted at least 3 statements of his that google was able to practically "impeach".

Also amusing how Oracle's lawyers had a hard time with case numbering. Why would they mess that up, incompetence? Sure annoyed the judge (reading between the lines). No mention of this in other news on the case. A fine example of the quality here!

Thank you!

[ Reply to This | # ]

Larry Ellison 2010 talk
Authored by: kds on Tuesday, April 17 2012 @ 11:54 PM EDT
If you listen to the highlights of his January 27, 2010 talk about Sun and Oracle being one company, it is interesting to hear where Java is placed amoungst the list of things they got from Sun:

  • That huge patent portfolio
  • Fabulous Sparc chip
  • High-end Unix Operating System Solaris
  • Java - the industry standard middleware
  • A very popular database - MySql
  • ZFS / Sun 7000
Later during the highlights the only other mention about Java is:
The Java business is gonna grow.
Doesn't sound to me like he thinks Java was worth the entire $7.4 billion, but that may just be me.

I haven't had time to listen to the entire 59 minute webcast so feel free to post any other tidbits. My main point is if Java was what they bought Sun for, why is it only 4th on the list of 6 items mentioned and why do they only give a footnote of one sentence about it in the "Highlights" video?

(I'm still trying to locate the .ogg clip mentioned in the update but found this instead.)

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj - Updated 3Xs
Authored by: Rubberman on Wednesday, April 18 2012 @ 12:09 AM EDT
Great recap - wish there was a video of it! As for the
Wayback postings of the Java open-sourcing event, from the
links I found, the Wayback Machine didn't save/cache the
videos... :-( There's no there, there it seems.

[ Reply to This | # ]

Judge: Encourages Oracle to get their act together on their exhibit numbering. Oh Dear.
Authored by: SilverWave on Wednesday, April 18 2012 @ 12:18 AM EDT
How much are these guys getting paid?

This made them look bad..

Not as bad as Ellison did but, still :-O

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 | # ]

Does it take a smart company to make a smart phone?
Authored by: Anonymous on Wednesday, April 18 2012 @ 01:07 AM EDT
But (in the doc) you (Oracle) had problems… you didn't have the internal expertise to be able to make smart decisions.

If this is a correct quote, I vote for this as the Barb Of The Day...

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj - Updated 4Xs
Authored by: Anonymous on Wednesday, April 18 2012 @ 02:37 AM EDT

Video of Larry Ellison and Scott McNealy where they talk about Java being free - giving Java to the world. m/watch?v=_Dtqe1e0tXg&

[ Reply to This | # ]

Authored by: Anonymous on Wednesday, April 18 2012 @ 02:49 AM EDT
The analogy of an API comparing to nouns may be helpful for a non-tech juror.
This is in a way correct. But it lacks a critical point. No one, even Oracle,
can successfully launch a programming language that doesn't offer a broad help
to programmers to deal with their tasks.

It would be like baking bread by starting to invent a new well, mill and stove.

In fact, a programming language is successful if there is a good environment for
programming, to ease the repetitive tasks of a programmer. And this environment
has to offer a library, where the programmer can use predefined functions and
procedure. No one, even at Oracle, starts from scratch. Everyone starts from a
base that was build by a myriad of programmers and mathematics that done basic
and hard stuff before. In research and development. Without anyone paying for
it. Like using an algorithm for sorting. This isn't astounding, it is business
as usual. If R&D has to guaranty to be cost effective, it is production, not

The success of a programming environment depends on the library. That has to
offer all that basic stuff that other programming environments already offer.
And some things that go further. That doesn't mean they copy it. You can get
coffee everywhere, but all is different, it isn't the same at all. You don't
steal from Starbucks when you make your own coffee at home.

If Oracle succeeds, doesn't that imply that every author needs to define and use
his own set of figure of speech?

Wolves and Lamas moisten your face. Because I'm not allowed to use "It's
raining cats and dogs." I invent my new and set of figures of speech. Get
used to it.

[ Reply to This | # ]

Is the judge being a little cavalier here?
Authored by: Anonymous on Wednesday, April 18 2012 @ 03:09 AM EDT
Google: Issue with trial exhibit TX431. It has proprietary Google information.
Judge Alsup: I will allow it to be used. This is a public trial.
Google: But this has strategic information to 2013 in it.
Judge Alsup: This is a public trial. Request denied.
Google is in this situation as an involuntary defendant, certainly not by any choices they made. Just because it is a public trial, it does not mean that any of their private and confidential data can be exposed to the general public willy nilly.

Whilst he seems very competent and in control, asking for clarification where necessary, I find that occasionally some of the decisions made by this judge to be abrupt and lacking in deeper analysis. It is like a hot button has been pushed with him and he sometimes reacts automatically without looking at the merits of the argument.

Does anyone else agree?

[ Reply to This | # ]

Wayback has the APIs!
Authored by: Anonymous on Wednesday, April 18 2012 @ 03:11 AM EDT
Using the Wayback link in Update 2 to the Sun opening announcment, on the live page, follow the Open JDK link in the right side box.

On that page, follow link Core Libraries link on lefthand side column.

On that page you'll find lots of java library links, for example: java.lang [Described as

Provides classes that are fundamental to the design of the Java programming language such as String, Math, Enum, and the wrapper classes for primitive types (e.g. Boolean, Float, Character).]

If you click the java.lang link, you'll get into all the definitions of what it does, and deeper links that describe the interfaces/APIs, and what they all do down to excruciating detail.

[ Reply to This | # ]

Reminiscent of SCO trial reports
Authored by: Jimbob0i0 on Wednesday, April 18 2012 @ 03:30 AM EDT
This reminds me so much of the SCO/Novell trial....

Staying up late to catch the first reports... waking up in the morning and
hurrying over to groklaw to read the later reports on the train to the

Thank you reporters for providing the view into the court room....

[ Reply to This | # ]

Am I missing something here?
Authored by: Anonymous on Wednesday, April 18 2012 @ 03:53 AM EDT
Oracle touts their nine/thirteen lines of code found in
java/util/ rangeCheck() as proof positive that
Google misappropriated their code.

Leaving aside the fact that that's a rather small proportion
of the Java classes (and that there's not that many ways to
write it efficiently), the top of that file in the OpenJDK
(as released by Oracle) states that it's covered by GPL2.

Why do they then think that Google is not allowed to use
that file?

[ Reply to This | # ]

Big Honking Smoking Gun
Authored by: s65_sean on Wednesday, April 18 2012 @ 07:41 AM EDT
I'm sorry if someone has already noticed this and posted a comment, but I didn't see it in the comments section and when I read it in the article it just blew me away:
Google: [Moving onto the meeting with Eric Schmidt, CEO of Google at the time.] What was the business proposal to Google?

Ellison: Google would take (a license for) Java from Oracle and put Java into Android.
Emphasis added.

Oracle wanted to PUT JAVA INTO ANDROID!

Doesn't this imply that Oracle was admitting to Google that up to that point that they felt that JAVA WAS NOT IN ANDROID?????

Why would they need to put java into Android if java was already in Android? And if java was not already in Android, then why are they suing Google?

Why didn't Mr. Van Ness jump all over that one? Can they recall Mr. Ellison and wipe the floor with him over this?

[ Reply to This | # ]

What's Google's Plan B if they Loose?
Authored by: Anonymous on Wednesday, April 18 2012 @ 09:37 AM EDT
Why doesn't Google just rewrite Android apps in Ruby?

Look, morally I think Google's right. But if the Google looses at trial, can't
they switch languages?

The cost is large but Google has plenty of financial resources. It seems to me
that they could port the entire ecosystem to something else with $2 to $3
billion -- 10,000 developers for two years.

In fact, it seems to me that Google should already have two competing hidden
projects to do just this.

I've heard that Ruby is the best 5th generation language.


[ Reply to This | # ]

OpenJDK Legal Docs
Authored by: hAckz0r on Wednesday, April 18 2012 @ 10:37 AM EDT
The Legal Docs are here, such as the GPLV2 with classpath exception and the assembly exception.
GNU General Public License, version 2, with the Classpath Exception

To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it.

OpenJDK Assembly Exception

The OpenJDK source code made available by Sun at and ("OpenJDK Code") is distributed under the terms of the GNU General Public License version 2 only ("GPL2"), with the following clarification and special exception. Linking this OpenJDK Code statically or dynamically with other code is making a combined work based on this library. Thus, the terms and conditions of GPL2 cover the whole combination. As a special exception, Sun gives you permission to link this OpenJDK Code with certain code licensed by Sun as indicated at ("Designated Exception Modules") to produce an executable, regardless of the license terms of the Designated Exception Modules, and to copy and distribute the resulting executable under GPL2, provided that the Designated Exception Modules continue to be governed by the licenses under which they were offered by Sun. As such, it allows licensees and sublicensees of Sun's GPL2 OpenJDK Code to build an executable that includes those portions of necessary code that Sun could not provide under GPL2 (or that Sun has provided under GPL2 with the Classpath exception). If you modify or add to the OpenJDK code, that new GPL2 code may still be combined with Designated Exception Modules if the new code is made subject to this exception by its copyright holder.

DRM - As a "solution", it solves the wrong problem; As a "technology" its only 'logically' infeasible.

[ Reply to This | # ]

Judge's question "Did Sun develop the methods themselves, or were they developed by some 3rd pa"
Authored by: Anonymous on Wednesday, April 18 2012 @ 11:16 AM EDT

A few weeks ago, someone anonymous posted here stating,

From a programmer's perspective, syntax is integral to a language, most APIs are not. A few are: it's hard to imagine C without printf() and malloc(), [...]. I tend to think of those things as a single API - "the standard library." In the case of Java, the trademark requires a whole bunch of libraries, some of which I'd argue aren't crucial. I can't remember my Java package names at the moment, but generally I'd say that the system stuff is crucial [...] as are basic programming utilities (regular expressions, sorting, data collections like vectors

An aside, there certainly are situations where someone uses C without malloc() or printf() ... Linux kernel development, 16-bit Windows development, Postgres user-defined functions... although whatever the environment is, it generally provides a memory-allocator and a varargs-based formatted-output function or macro.

But as one answer the judge's question, look at java.util.logging and java.util.regex, and compare them to Apache Log4j and Apache Jakarta Regex. Sun probably did not use any third-party code for their logging implementation (although there are adapters to run Log4j on top of JUL, or vice versa), but they took the overall library design that Apache had already made popular and defined an API for it in the java.util hierarchy.

Log4J is still popular even after the introduction of JUL, and still runs just fine on Java 1.7; it would probably run fine on Android, haven't tested it myself. If Android couldn't use j.u.r for some reason, an Android app could probably switch to Jakarta Regex without too much rewriting; but Jakarta Regex depends on the following classes from two of the other 36 accused packages based only on its own API Javadocs:

  • java.lang.Object
  • java.lang.String
  • < tt>java.lang.StringBuffer (private field, perhaps that doesn't count)
  • java.lang.Throwable
  • java.lang.Exception
  • java.lang.RuntimeException

[ Reply to This | # ]

Kurian: Sun was constantly trying to get Apache to take a license.
Authored by: Ian Al on Wednesday, April 18 2012 @ 02:43 PM EDT
Just a reminder, Sun contracted with JCP member, Harmony, to allow them to
create their own clean room implementation of the entire Java API Specification
and to allow them to run the TCK which, if passed, would allow Harmony to call
their implementation, Java.

Sun breached their contract term to allow Harmony to run the TCK.

As long as Sun remain in breach of contract by refusing to allow the running of
the TCK they are refusing to allow Apache to take a license. Only Sun are
preventing the Harmony project from being licensed.

How would you characterise Kurian's statement to the court and the jury that
'Sun was constantly trying to get Apache to take a license.'?

Yes, so would I.

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

[ Reply to This | # ]

"Mine our source" line still on Oracle website
Authored by: Anonymous on Wednesday, April 18 2012 @ 07:06 PM EDT
Nice find in Update 8. Helpfully e/gosling-os2-qa-136546.h tml is still available and includes the "They'll certainly be able to mine our source for stuff to incorporate into their projects" line.

[ Reply to This | # ]

Day 2 of Oracle v. Google Trial ~pj - Updated 8Xs - Google's Opening Statement Slides
Authored by: Anonymous on Wednesday, April 18 2012 @ 08:09 PM EDT
That Final Update - the one with the comment by James Gosling ¨Q: How do you
think this move will affect other open-source implementations of the Java
programming language -- for example, Apache Harmony or GNU Classpath? Gosling:
It's hard to know. They'll certainly be able to mine our source for stuff to
incorporate into their projects.¨ really drives home to me HOW SIMILAR this is
to the SCO case

New Management comes along and changes the rules and (tries to) rewrite history
with a wild dream for billions of dollars - and then slowly reality kicks in
SCO is now synonymous for many bad things in IT. DO ORACLE really want to keep
going down the same path and the thought of as kind of company as SCO ended up
as ???

[ Reply to This | # ]

"Thank you, Internet Archive, for keeping the world's players more honest."
Authored by: Anonymous on Friday, April 20 2012 @ 08:04 AM EDT
Did you type that with a straight face?

[ Reply to This | # ]

"mine" our source for stuff
Authored by: Rook on Tuesday, April 24 2012 @ 04:58 PM EDT
Gosling: "They'll certainly be able to mine our source stuff
to incorporate into their projects."

PJ: "In other words, Gosling told the world that Apache
Harmony and GNU Classpath could "mine" the Java source code
and "incorporate" that code "into their projects"."

Gosling's "stuff" doesn't necessarily mean "code." What
else could it be? Ideas, functional methods, etc. --- all
that stuff that copyright does not cover (and Patents may or
may not; if it's something that others can incorporate, that
suggests that Gosling thinks that that "stuff" isn't patent

If Gosling generally understands the GPL (I'd guess yes),
then he'd know that GPL Java code released by Sun cannot be
incorporated into an Apache-licensed code base because of
the incompatibility of the licenses. For that reason, I'd
guess that he didn't mean copyrightable-code when he said

[ 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 )