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.

To read comments to this article, go here
Day 9 at the Oracle v. Google Trial ~ pj - Updated 9Xs (Schwartz, McNealy, Gering, Rizdi, Astrachan)
Thursday, April 26 2012 @ 12:39 PM EDT

Our man in the courtroom today at the Oracle v. Google trial has been having email issues. But Jonathan Schwartz is on the stand now. Also today Craig Gering, Hassan Rizdi, Scott McNealy, and Google's expert Dr. Owen Astrachan. More from mirror_slap later. Meanwhile, enjoy the tweets from the stream Feldegast set up for us or this one. Wired's Caleb Garling tweeted:
"Has the Java programming language always been free to use?" -- "Absolutely" - Schwartz
And Ginny LaRoe tweeted:
straight from the Java man's mouth --> Schwartz says APIs always free along with open source Java language. Never proprietary.
And that's the ballgame, if the jury believes him. That's if the issue even goes to a jury. Google has a Rule 50 motion before the judge right now, asking for a ruling as a matter of law that, as Mark Webbink's article puts it, could blow the door off Oracle's copyright claims. What I noticed yesterday about it turned out to be accurate, that Google claims the Java copyright registration doesn't match Oracle's claims, and that really could be the ballgame.

And let me just say, I've never seen a chess move like this before. Whatever the judge rules, it's been an honor and a privilege just to watch Robert Van Nest, Esq. in action. Just stunning work. Someday, when I have grandchildren, I'll be telling them about it, I'll bet.

Jump To Comments

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

A few more details from ZDNet on other witnesses of the day. Scott McNealy is expected to take the stand once Oracle gets its rebuttal time. That should be fun. It could be today.

And Garling has a link to a photograph he says Judge William Alsup took but Caleb uploaded. Hopefully, Judge Alsup isn't a suing kind of guy.

: )

There be sharks in the waters. And if Oracle wins, there will be more. And the most innocent actions can land you in a court of law.

Here's more from ZDNet:

Schwartz also confirmed to Google counsel Robert Van Nest that Sun promoted the open use of Java APIs as well. “You had to if you wanted to see that language broadly accepted,” Schwartz explained. “Those APIs enabled people to write full, complete applications that leverage all the technology underlying the platform.” He added that the distribution of those APIs across the world enabled the adoption levels that Sun was seeking.

To further boost those adoption levels, Sun had to find a way to be bigger than Microsoft, and Schwartz said that was made possible with the open Java community. Partners included Oracle, SAP, Sybase, and many other Silicon Valley giants. “It would give us something to pull together that was bigger than the monopoly itself,” Schwartz remarked....

However, even though Schwartz said that Sun wanted to get revenue from Google if a partnership could be hammered out, Schwartz said that the deal did not fall apart for money. “We probably would have paid them to work with us on a Java phone,” Schwartz admitted.

Ahead of the announcement of Android in 2007, Schwartz told Van Nest that Sun was aware of a few things about Android, including that Google would be using the Java languages and APIs.

I wonder if that remark could affect damages, in the event Oracle prevails on anything. I mean, if Sun would have paid Google to do the deal, how much was the deal worth? How much can you ask Google to pay now to allegedly make Sun, now Oracle, "whole"? Maybe Oracle should pay Google.

Hey. There's an idea. How about Oracle pays Google for bringing them into court on these stupid API claims? Now *that* would be justice, or at least karma.

Update: And the email issue is resolved, so here is mirror_slap's first report, on the discussion before the jury came in, and of course it was all about the Google Rule 50 motion:

Oracle v. Google, Day 9, part 1

[ 7:30 AM ]

Judge: Any issues?

Oracle: [Michael Jacobs] Depositions… Agrawal, next morning.

Oracle: Coming: Gupta. We got a brief from Google last night on ownership and their Rule 50 motion.

Oracle: Background: when Mr. Baber (Google) page 4, March 9, Oracle's copyrights TX6-066-538 -- "as a protected work…" same copyright form as Sun's other versions".

Oracle: Then Mr. Reinhold's description of thee JSR's… In a break, Oracle brought up the issue previously, percentage of 3rd party ownership.

Oracle: [quoting Google] "We are not disputing the right of ownership. We are disputing the… [ lost ]

Oracle: I specifically said, "Is ownership in contest?"

Oracle: And they said, "ownership is not in contest".

Google: [Mr. Baber] We do not dispute that they own the copyright registration. The question, the question of law, is what the copyrights cover. This is not a new issue in the least. The parties identified jointly issues of laws to be satisfied, whether some pretrial order, docket 525, page 10, covered this issue. Whether these 103(b) issues cover the individual files.

[PJ: This is referencing US Copyright Law Section 103(b). And you can find more details about this motion here and here.]

Judge: The first time the word "collectively" was used was a week ago.

Google: Mr. Jacobs used that word a week ago, but they said that it was not a compilation, and they'll prove ownership of the underlying components.. Now, in the middle of the trial, they are saying that they are a collective work. This is similar tactic that they have used all along.

Google: The issue, to be clear, is that they own the copyright registrations, but what they cover is unclear.

Oracle: Google's counsel is confusing things, accidentally.

Judge: They are saying that the copyright exists for the collective work. You might not know what.

Oracle: In the context of Mr. Reinhold being examined in the context of the JSR's...

Oracle: Only addressing the effect of the registration on our ability to sue. The question of what the copyrights cover is tangential. The question is whether we can sue under the collective rights.

Judge: Reinhold was the 6th witness. Was it at that point that ownership was given away?

Judge: How can you say to Google that they cannot bring in their reliance on lack of pursuit by Sun?

Oracle: We don't need the registration to assert our infringement claims here. We can sue on the collective work, and for components of the work.

Judge: In 103(b), "...extends only to the work contributed by the author."

Judge hypothetical: 3rd party develops an API package, and donates it. Author is not an employee of Sun or Oracle. It could be that there is an assignment document that the person signed that says, "I give you all my rights". But we haven't seen that yet. But if this is a collection of API packages assembled in a clever way, but some chapters are written by somebody else, I can see you suing over the SSO for all 167 packages, but I cannot see you suing over a smaller subset of the API packages.

Judge: How far does the copyright extend?

Oracle: Restates at 5th witness from Mr. Purcell. [PJ: I'm not clear what this means, but will ask.]

Judge: It says right here that ownership is in contest, Page 15, number 9.

Judge: It's in your joint statement. Maybe it meant something else, but that is what it says.

Judge: I am not going to decide this right now.

Judge: My guess is that somewhere there is a document (assignment) at Sun.

Judge: Mr. Baber, would this cover that?

Google: It would, if they exist. But we started with 51 and now we showed that they are only 37 that Oracle owns. The other 14 were donated by Dr. Lee, and were donated to the public domain. Developers don't have to assign to Sun/Oracle. Author can retain ownership.

Judge: When someone contributes, what do they contribute? Method, class, package?

Google: Sometimes a whole package (Professor Lee contributed 3 packages). Via a JSR, any or all of the three can be the case.

Judge: Different or related question to Mr. Jacobs: I confess that I could easily miss things in the record in this gigantic case. But it is related to the question as to whether Oracle can sue individually on each of the 37 packages. You went on and on regarding how they were interconnected. I didn't realize that you were also going to sue over the individual SSO's. Scope issue in no way effects Sun/Oracle's ability to sue on the overall SSO.

Judge: In the special verdict form, question 1 is going to go to the jury. We are arguing over question 2.

Judge: Where is it that you said that you wanted to put to the jury that each of the 37 SSO's was in contest? I have a suspicion that this case was launched to consider the 37 APIs as a whole, but now we are looking at individual API packages. Is the individual part in the pretrial [filings]?

Oracle: I don't recall, Your Honor.

Judge: If we were to revert back to the 37, I will say it's 37 versus all the APIs in Java, and not Android. I don''t think that you can surgically remove the 37 and try to address the work as a whole. If we get down to the individual API level, then the proper comparison is a package-by-package analysis. But I don't think that this is the way we agreed to try this case. If I am right, we will never get to the collective issue.

Google: [Baber] Are the April 12, Oracle statement, 7 different items, docs for all 37 API packages, all together.

Judge: Doc is going to be a one-to-one. SSO is only in the compilable code concept. You are going to lose on the big fat manual.

Google: I understand. Collective work issue is framed to. It is fundamental, not a technicality or tangential

2) SSO of all names of source code for the 37 packages.
3) declarations in the source code
4) source code and object code that implements the packages
5) copied files (individually)
Judge: Did anything you said bring this in?

Google: No. We didn't have any claims for them to assert.

Judge: Not making any ruling now. want to see brief from Oracle on this point. If Oracle wants to add evidence on this, I will allow it, but it's coming out of your time.

Google: A problem with our filing last night… "SSO should be before the jury" should read "should NOT" be before the jury. [ laughter ]

Judge: I think I am going to hold you to that [ joking ] [laughter]

Update 2: And here is the next section, the testimony of Craig Gering:

[Witness brought in.]

[Jury brought in.]

Google calls Craig Gering.

Testimony of Craig Gering

Google: Introductions -- worked for Sun [Sun's senior director of embedded Java software] and then Oracle, leaving 2011.

Google: In 2006, you took over engineering service and Java licensing for mobile devices?

Craig Gering: Yes.

Google: You were in that job until Oracle took over from Sun?

Google: TX 2053, in evidence October 4, 2006 email.

Craig Gering: Yes.

Google: Related to Sabadje [sp?] trying to build a phone?

Craig Gering: Yes.

Google: Sun bought Sabadje?

Craig Gering: Yes.

Google: when did Sun acquire Sabadje?

Craig Gering: Do not recall.

Google: Ever heard the term "full stack"?

Craig Gering: Building a complete mobile platform, vertical platform.

Google: Multiple layers, as compared to Java, which was horizontal stack across multiple platforms.

Google: So in contrast to Java, Android was a full stack?

Craig Gering: Yes.

Google: Apple's iPhone is a full stack?

Craig Gering: Yes.

Google: When Sun bought Sabadje, Sun didn't have a full stack?

Craig Gering: Yes, they did not have a full stack.

Google: You do know that Sun tried to turn the Sabadje platform into a full stack?

Craig Gering: Yes.

Google: Project Acadia was the internal name?

Craig Gering: Acadia at one point in time, changed name as it moved between internal organizations in Sun.

Google: Sun never got any product to market from its Sabadje purchase, right?

Craig Gering: Customer engagement products (Cisco phone).

Google: Deposition, Page 134, lines 19-25

Q: Did you perceive any value in purchase by Sun?

Craig Gering: So there were technology and people but only if the technology got to market.

Oracle: Line 7-23.

Google: Read form line 1 for completeness.

[ end of readback ]

Google: You are not aware of any phone that Sun brought to market?

Craig Gering: No.

Google: Sun never brought a mobile product to market?

Craig Gering: No.

Google: After Google released Android, Sun tried to develop products based on Android?

Craig Gering: Researched the possibilities.

Google: TX2052, presentation in Sun, Java in Wireless, Business review, March 16, 2009: page 20 of the doc… page discusses project Deneal(?), and Sundroid.

Craig Gering: Project existed, with similar goals.

Craig Gering: Deneal had 2 goals: 1) to put the SunVM in beside Android VM, 2) replace Android VM with Sun's.

Google: And phase 2, which is a full Linux platform.

Craig Gering: Phase 2 was not as well defined.

Craig Gering: Don't recall specifics.

Google: Sun got as far has having a prototype of a phone with Android stack?

Craig Gering: Correct.

Google: Never got to market?

Craig Gering: Correct.

Google: TX2061, in evidence, email exchange to/from Mr. Gupta. Email to bottom of page is an email from Mr. Gupta?

Craig Gering: Yes.

Google: In January 2009, what was Mr. Gupta's job?

Craig Gering: CTO of embedded sales force, involved in licensing.

Google: Mr. Gupta says, "I've had several requests regarding partnering with us to provide a Dalvik/JavaME combined platform. Samsung is really pushing for a partnership asap"

Craig Gering: Yes.

Google: TX missed. Email October 2009, attaches a couple of presentations.

Craig Gering: I see it.

Google: Look at Java landscape, "Sun's leadership is considered stagnant and software is legacy."

Craig Gering: I do [ see that ]

Google: "fragmentation between Java SE and ME"

Craig Gering: I see that.

Google: Java SE and ME are different versions?

Craig Gering: I don't think that they are that different.

Google: Page 4, "Modernize unified Java."

Craig Gering: Yes.

Google: Bullet point: Define one dev model across all device types, allowing for industry-specific customization without fragmentation of features.

Cross examination of Craig Gering
Oracle: What do you do now?

Craig Gering: Consultant for FAIR-ISAAC.

Oracle: Continue answer regarding fragmentation?

Craig Gering: Hardware fragmentation: different devices. Software fragmentation: different implementations that had bugs or performance issues.

Oracle: Jave ME have a presence on smartphones?

Craig Gering: Yes. Nokia, RIM blackberry

The witness is dismissed.

Update 3: Here's the testimony of Hassan Rizdi:

Testimony of Hassan Rizdi
[Sworn in at 8:25 AM ]

Google: You work for Oracle?

Hassan Rizdi: Yes.

Google: Since 1990?

Hassan Rizdi: Yes.

Google: Left in 1998?

Hassan Rizdi: Yes.

Google: Returned in 2001?

Hassan Rizdi: Yes.

Google: Report to Mr. Kurian?

Hassan Rizdi: Yes.

Google: Manage Java business?

Hassan Rizdi: Yes.

Google: Java language is free to use without paying royalties to Oracle?

Hassan Rizdi: Yes.

Google: June 25, 2010 email to you form (?}, Java Business model overview.

Google: Java clients is a $296M business, growing at 13% a year. includes Java SE, ME, Card.

Is Oracle's [Java] revenue still growing at this rate?

Hassan Rizdi: No. Almost flat now.

Google: Playback page 229, lines 13-21. July 28, 2011.

Google: Figure of 10%. Is the Oracle Java business still growing at this rate?

Hassan Rizdi: cannot say for sure the growth rate.
--end of playback

Google: Oracle VP of development left?

Hassan Rizdi: Yes.

Google: Oracle employee, Jeet Kol?

Hassan Rizdi: Yes.

Google: Jeet was in charge of Sun's Java SE, ME and Card?

Hassan Rizdi: Yes.

Google: [Going to fragmentation, Oracle wondering whether Sun was deliberately fragmenting Java.]

Oracle: Objection, hearsay.

Judge: Mr. Messenger was where?

Google: Oracle.

Oracle: Line of questions is intended to elicit hearsay.

Judge: Overruled.

Google: To what extent did Mr. Messenger express concern over deliberate fragmentation.

Hassan Rizdi: It was a big concern.

Google: TX2110, not in evidence, email between the witness and Mr. Messenger re: Sun's deliberate fragmentation of Java,12/9/2008.

Google: Mr. Kol running a large part of Sun's Java operation?

Google: Ask to be placed into evidence.

Oracle: Objection, hearsay, 403b.

[ sidebar ]

Judge: [Exhibit is entered into evidence.]

Google: "Steve Harris and I met with Sun, and they are a mess over there. Jeet Kol is forking Java, regardless of what it does to Java, just to make money."

Hassan Rizdi: Yes.

Google: You recall being told that Jeet Kol, running Sun's Java division, was fragmenting Java to make money. Did you do anything to address this?

Hassan Rizdi: No. We saw no direct evidence of this.

Google: You didn't investigate and conclude that he was wrong?

Hassan Rizdi: No.

Google: "if you continue down this path, you will either need to take one of 3 separate forks of Java, Apache Harmony, JDK 1.6, OpenJDK"

Hassan Rizdi: Yes.

[ Argument. Judge allows witness to respond fully.]

Hassan Rizdi: Sun's attempt to do development while the standards process wasn't working well.

Google: OpenJDK is still available?

Hassan Rizdi: Yes.

Google: Available under GPL?

Hassan Rizdi: Yes.

Google: When someone uses the OpenJDK, there is no need to run the TCK?

Hassan Rizdi: Don't know.

Google: Familiar with the term Full Stack?

Hassan Rizdi: Yes.

Google: Google's Android is a full stack?

Hassan Rizdi: Yes.

Google: iPhone?

Hassan Rizdi: Yes.

Google: Symbian? Yes.

Google: Sun's Java platforms are not full stacks, are they?

Hassan Rizdi: No. Java platform is OS-agnostic.

Google: Sun hasn't had a full stack on the market?

Hassan Rizdi: Not aware.

Google: Oracle?

Hassan Rizdi: No.

Cross Examination of Hassan Rizdi, by Oracle's Fred Norton
Oracle: Familiar with the term API spec?

Hassan Rizdi: Yes.

Oracle: When you say that the Java language is free to use, do you include the APIs?

Hassan Rizdi: No.

Oracle: When you got the email from Mr. Messenger, did you see any evidence that Mr. Kol was taking steps to deliberately fork Java?

Hassan Rizdi: No.

Oracle: Sun?

Hassan Rizdi: No.

Oracle: To the extent that Android forks Java, does this make the situation better or worse?

Hassan Rizdi: Worse.

Google: Objection.

Judge: Sustained, outside scope.

Oracle: Have three questions that are out of scope.

Judge; Okay, no leading.

Oracle: Once Oracle acquired Sun in January 2010, to what extent did you discuss pursuing a license with Android?

Hassan Rizdi: I had several meetings with Google, including with Andy Rubin.

Oracle: What was the purpose?

Google: Objection.

Judge: Overruled.

Hassan Rizdi: [ missed ]

Oracle: In your discussions with Google over a license for Java, did anyone ever say that Google didn't need a license based on what Jonathan Schwartz said?"?

Google: Objection.

Judge: Overruled.

Hassan Rizdi: No.

[ break, jury leaves ]

Judge and Lawyers Discussion:

Judge: Any issues?

[ Big fight over getting the copyright registrations for Java at Library of Congress (LoC) into evidence ]

Google: Issue with getting into evidence, the copyright filings from Library of Congress. No stipulation. TX 3529, 3530

Judge: What are all these redactions?

Google: The Library of Congress allows redaction to protect IP.

Judge: What is the objection to this?

Oracle: There is a disk, the Copyright office has the disk, but cannot produce the disk. We will allow it if the disk is produced.

Google: TX3530, the other copyright registration.

Google: No requirement to send in the disk. Library of Congress now says that there is no data on the disk.

Judge: "The attached photocopies is a true representation of what is on the front of the disk".

Google: Behind that are the redacted 50 pages of the source.

Judge: Legal significance?

Google: To complete the record, materials are not complete, not certified. We cannot get from the Copyright Office a copy of the entire work. Plaintiff needs to prove that they have a registration and what that registration covers.

Judge: On 3529, there is no reference to a disk.

Google: Correct.

Judge: What is the legal objection to these documents?

Oracle: No objection on one, but objection on the other.

Judge: Objection overruled. These are entered into evidence This is a certified copy from a federal agency.

Judge: One item for you: please ask counsel to file handouts that were given to the jury so they can be entered into the record ( timeline, glossary, etc).

Judge: Juror 2 doesn't feel well, is coming down with a cold. Does she feel she cannot go on? Can she pay attention and continue on today? [ to clerk ]

[ 15-minute break ]

Update 4: Next up is former CEO of Sun Microsystems Jonathan Schwartz, and aside from testifying that Java was free and so were the APIs, he reveals that the alternatives to Java, like Gnu Classpath and Apache Harmony, didn't have a license or need one, and when asked who used Apache Harmony, he said IBM and Oracle. Yes. Oracle. And he says Oracle helped fund it too:

Judge: Will allow in the exhibits that Oracle objected to.

Jonathan Schwartz

[ jury comes in ]

Judge: Is that you?

Jonathan Schwartz: That is, in fact, me. [ smiles ]

[ sworn in ]

Judge: Welcome, sir.

Google: [Van Nest] Introduce yourself.

Jonathan Schwartz: CEO of Sun 2006-2010, live in SF. CEO of startup, CareZone. [ describes CareZone ]

Google: Background? Jonathan Schwartz: Grew up on both coasts… UConn Math and [?] Started WhiteHouse Designs, bought by Sun.

Google: Career at Sun, 1996.

Jonathan Schwartz: Left intact, general manager, become responsible for product marketing for Java technologies.

Google: 1998 became product marketing person for Java.

Google: Java programming Lang, free to use for everyone?

Jonathan Schwartz: Yes. Microsoft was defining what was available, Java was all about being able to write an app that didn't require Windows. Strategically very important to open market and opportunities. Google: Promote teaching of Java in colleges and universities?

Jonathan Schwartz: Yes, High Schools, colleges, universities. Sun went around the world to help educational institutions.

Google: What is the Java API?

Jonathan Schwartz: Language is the semicolons, operations. You have to have specifications that deliver what you want to get done. Example: make your computer beep. Use the API from the sound library… allows you to communicate with the computer [ hardware ]. Need API's.

Judge: In your example, what would be the API?

Jonathan Schwartz: 1) identify the musical instrument, 2) identify the note, 3) …

Judge: What words would you put down on the line to call the API?

Jonathan Schwartz: Would have to refer to the specification.

Judge: From memory, would you know what to use?

Jonathan Schwartz: I'd have to look it up.

Google: Sun promoted the APIs as well?

Jonathan Schwartz: You have to promote the APIs, the distribution of the language and the APIs that would allow us to bypass Microsoft.

Google: APIs were free and available to everyone?

Jonathan Schwartz: Yes.

Google: Were the APIs ever sold or licensed separately from the language?

Jonathan Schwartz: No, of course not.

Google: You mentioned separate ( competing) implementations, what do you mean by that?

Jonathan Schwartz: You need to…

Google: Implementation is written in source code?

Jonathan Schwartz: All Java technology is written in code. The implementation is the thing that gets compiled.

Google: How was Sun going to make money off of Java?

Jonathan Schwartz: One company was monopolizing the desktop. Sun was able to get together with Oracle, Sybase, etc, getting together to form the Java community, so it can run on any computer. That was our way of bypassing the monopoly. Allowed us to pull together all these companies and be bigger than the monopoly. This was our key strategy.

Google: Explain open specification, open APIs.

Jonathan Schwartz: Scope… builds trust. equal access, equal opportunity, no tilted playing field. We made a lot of noise about open APIs.

Google: Was there ever a time at your tenure at Sun where the Java APIs were considered to be protected by Sun?

Jonathan Schwartz: No, and if anyone did, we'd have worked very hard to [ shut them down ].

Google: When did you become COO?

Jonathan Schwartz: 2004.

Google: And CEO?

Jonathan Schwartz: 2006.

Google: What did you do?

Jonathan Schwartz: Responsible for everything.

Google: Licensing and IP?

Jonathan Schwartz: Yes, including strategies around these.

Google: What is Jonathan's Blog?

Jonathan Schwartz: Sun seems like a big company, but our competitors are ten times bigger than we were. Always faced with a human scale problem, they had more resources (advertising, sales rep).

Google: Was the blog posted on Sun's web site?

Jonathan Schwartz: Yes, it was our way of telling our customers what we thought was important. [ the equivalent of calling a press conference, but without having to call the press.]

Google: Describe the concept of Open Source.

Jonathan Schwartz: Wikipedia is an example… an open source encyclopedia.

That all changed with an individual named Linus Torvalds.

Google: Did Sun participate in the OSS movement?

Jonathan Schwartz: We were slow to do this. We didn't see how to make money off of it. RedHat was doing it brilliantly. Make money off of support. widely-adopted. As an enterprise, complexity is your enemy.

Google: What is GNU Classpath?

Jonathan Schwartz: Yes I know it. Dynamics of Open Source community. Linux was a clone of Unix… GNU Classpath was an OSS implementation of Java. There was little that we could do to stop them; they didn't call it Java.

Google: They used the Java Programming Language?

Judge: And the APIs?

Jonathan Schwartz: Yes.

Google: Did they ever call themselves Java?

Jonathan Schwartz: Not to the best of my knowledge.

Google: As CEO of Sun, were you satisfied that what GNU Classpath was doing was fine?

Jonathan Schwartz: There was nothing that we could or did do to stop it.

Google: GNU Classpath had a license?

Jonathan Schwartz: They had no license.

Google: What is Apache Harmony?

Jonathan Schwartz: Similar to GNU Classpath. Example of on a browser, looking at a stock ticker tracker… a competitor to Sun's Java. Differences in approach to philosophy

Google: Was Apache independent of Sun?

Jonathan Schwartz: Funded largely by IBM and Oracle.

Judge: Did Apache have any kind of license?

Jonathan Schwartz: They were a part of the JCP… the trademark and specifications were tied. We couldn't keep people from [ rolling their own ]. Understanding the business model: companies producing Apache were paying Sun fees… would then allow, companies helping Apache could say, "we'll just run Apache".

Google: Example of companies using and funding Apache Harmony?

Jonathan Schwartz: Oracle, IBM, etc.

Google: Did Sun and Google start discussions over a mobile phone?

Jonathan Schwartz: Yes, we wanted them to take a license so that they could call their phone a Java phone.

Google: What did you want to get from the relationship with Google?

Jonathan Schwartz: Revenue. Get a big license fee for them to call their phone a Jvaa phone. Then, 2, go to major marketplace (dominated by Nokia).

Google: Was part of it that you would form a partnership?

Jonathan Schwartz: We wanted Google to feel comfortable.

Google: TX 435, you recognize it?

Jonathan Schwartz: I do, indeed.

Google: What is it?

Jonathan Schwartz: 27 April, 2006 email from me to Eric Schmidt.

Google: You wrote: "I believe this effort is an important project for both of our companies. We're at a critical stage in the industry where we still have a chance to successfully create an open platform that can target consumer devices".

Google: Were you able to reach an agreement with Google?

Jonathan Schwartz: We did not.

Google: Why not?

Jonathan Schwartz: It was a bit opaque to me… it turned out that we didn't have anything that they wanted.

Google: When working with Google, did you learn that Google was going to be using Java for Android?

Jonathan Schwartz: We learned that they were going to be using Linux and Java.

Oracle: Objection.

Judge: Did you have a belief with regard to what Google was going to be doing, using the Java APIs?

Jonathan Schwartz: Yes. Google said that they were going to be creating a Java Linux phone. [ Google didn't try to hide it .]

Google: ... [ alarm goes off ]

Judge: Do not panic. At various times, at 10 o'clock, that thing goes off.

Google: Is that coming out of my time your honor?

[ laughter ]

Google: You know Eric Schmidt?

Jonathan Schwartz: Very well. He was my first boss at Sun.

Google: Google became a customer of Sun?

Jonathan Schwartz: Yes. For the toolbar. We had the ability to do updates. This was valuable to Microsoft and Google. We called them both.

Google: Timeline. Sometime in the fall of 2007, did you get a heads-up about the upcoming announcement of Android?

Jonathan Schwartz: I did.

Google: And you wrote a message to Eric Schmidt, November 9, 2007. What did you mean by that?

Jonathan Schwartz: Please let us get in on the announcement so it wold be good for us.

Google: When you wrote this email, you knew that Google was using Java in their phone?

Jonathan Schwartz: We did.

Google: Why did you publish a blog post about Android?

Jonathan Schwartz: To try to be relevant and to get Java developers excited.

Google: TX2352. Authenticate?

Jonathan Schwartz: That's my blog.

[PJ: The post in question is this one in November of 2007. Here's a screenshot:]

Google: The post you made in 2007?

Jonathan Schwartz: Yes.

Google: What is NetBeans, and what are you communicating?

Jonathan Schwartz: Example of Linux, when it first arose. Suing Linux would just make Linux more popular. You can embrace it or litigate. We could sue, or grit our teeth and make the best of the situation. "Hey, here's a new phone on the market. Come use NetBeans to develop for it."

Google: Did you say that Android was helping Java?

Jonathan Schwartz: They could have picked Microsoft Windows, or they could pick Java. Which would you prefer?

Jonathan Schwartz: University professors would say, "Great, we can keep teaching Java".

Google: Interview win Engadget Mobile, TX 2853.

Oracle: Objection. Hearsay and [?].

Judge: Sustained on latter ground.

Google: Some time after release of Android, did you meet with Eric Schmidt?

Jonathan Schwartz: Yes.

Google: Anyone else?

Jonathan Schwartz: No.

Google: At the time, Sun was trying to build your own phone, JavaFX and Android.

Jonathan Schwartz: Yes.

Google: The meeting you had, talked about a variety of topics. Main one was how Google and Sun could work together?

Jonathan Schwartz: Yes.

Google: Talked about Android?

Jonathan Schwartz: Yes. I was reflecting to Eric Schmidt that the carriers were suspicious that the phone wasn't a Java phone, encourage Google to take a license and call Android a Java phone. Customers said that they were suspicious of Google. "That's fine, but people are suspicious of your technology."

Google: Did Sun demo at JavaOne, a JavaFX phone on Android?

Jonathan Schwartz: Yes.

Google: Were you there? Jonathan Schwartz: I think.

Google: Remember it?

Jonathan Schwartz: Vaguely.

Google: TX3103, playback of video presentation at JavaOne:

Eric Klein, VP of Java marketing.
[ holds up phone… JavaFX phone. Shows an Android emulator running on a Mac (pink mac)
ConnectItLite, a JavaFX app running on the Android emulator.
Shows scalability of JavaFX.
Google: Remember that demo?

Jonathan Schwartz: Absolutely.

Google: You were there that day?

Jonathan Schwartz: I was.

Google: Remember complaints about compatibility?

Jonathan Schwartz: Yes.

Google: Remember the phrase "compatibility is optional"

Jonathan Schwartz: No.

Google: TX2707- do you recognize it?

Jonathan Schwartz: I do.

Google: What is it?

Jonathan Schwartz: Document about how we deal with Java and Open Source.

[ G works to get TX 2707 into evidence ]

Judge: Can use as a demonstrative. Cannot be put into evidence.

Google: Happy to use as a demonstrative.

Google: Explain cover to the Jury? what was the issue?

Jonathan Schwartz: The issue is, if someone takes a bunch of code and builds Apache Harmony or GNU Classpath, these are potentially incompatible versions. Should we be more lenient in our licensing? Should we litigate?

Google: First page, what is the problem? -- "Our compatibility rules are complex, intimidating, and sometimes difficult to understand"

Jonathan Schwartz: Highlight paragraph 2.

Google: Did you consider litigation against Google over Android?

Jonathan Schwartz: We did, but we didn't have any grounds.

Cross Examination of Jonathan Schwartz
[ Mr. Schwartz has back problems ]

Oracle: [Michael Jacobs] CEO of Sun?

Jonathan Schwartz: Yes.

Oracle: Sun did not claim that Java specifications are protectable?

Jonathan Schwartz: Yes. Oracle: That included dev's relying on APIs and competitors.

[ JS made clear that the calling of a product Java was the nub of the problem. ]

Oracle: TX18. in evidence.

Oracle: See about OS J2ME. 2nd from top. "Wish them luck", says Andy Rubin. "Sun APIs are copyrighted and branded."

Jonathan Schwartz: Referring to getting access to the TCK, and that conferring getting access to the brand.

Oracle: Back to TX2707, what Mr. Van Nest was asking about. He asked about page 1. Let me ask about page 2: "For independent implementations, we grant rights for implementation that satisfy compatibility requirements."

Jonathan Schwartz: It's a problem if they go out and call their independent implementation Java.

Jonathan Schwartz: At the beginning of the slide. Conflating two different things.

Oracle: If Apache wished to release Apache Harmony and not call it Java, they were wrong?

Jonathan Schwartz: Conflating brand and the specification.. No way of separating the brand from the specification.

Oracle: Did Apache just miss that?

Jonathan Schwartz: I spent a lot of time on this issue, and did a lot of interviews on this issue. They can ship their code. They cannot call it Java.

Oracle: TX917

Judge: In evidence?

Oracle: Recognize this letter?

Jonathan Schwartz: I got a lot of open letters. Yes, I recognize it.

Judge: Into evidence.

Oracle: Apache has been trying to get access to the TCK…

Oracle: Including on mobile devices, sir?

Jonathan Schwartz: Absolutely.

Oracle: From Apache Harmony FAQ: Why doesn't Apache Harmony ignore this and avoid passing the TCK?

Oracle: Users wouldn't be assured that Apache Harmony was compatible.

Jonathan Schwartz: He put that very eloquently.

Jonathan Schwartz: We were addressing one simple point: [ We wanted them not to call their product Java.]

Oracle: TX 1045, in evidence.

Jonathan Schwartz: When did they write this?

Oracle: Dec. 2010.

Jonathan Schwartz: That was after my time.

Oracle: Did you follow this issue after you left?

Jonathan Schwartz: Blissfully, no.

Google: Objection, foundation.

Judge: Overruled.

Oracle: "Oohhhhh, I thought that you were making a legal" [point] when you were testifying about the position on the brand.

Jonathan Schwartz: I can tell you the business agenda. I didn't write the contracts. I directed the business direction that was implemented in the contracts.

Oracle: [ field-of-use ]

[ trying to get Schwartz to say that field of use]

Oracle: TX563- email March 2007 between Jonathan Schwartz and Scott McNealy.

Jonathan Schwartz: They were effective negotiators. They had options.

Judge: What does "opaque" mean to the witness?

Jonathan Schwartz: They don't show their cards.

[ Characterization of G's negotiating acumen.]

TX565, email from Jonathan Schwartz 9/19/2007 to Vineet Gupta.

Oracle: Gupta says: continuing on Gphone. " Heads up, Mr. CEO, we may need to tackle this issue around licensing."

Jonathan Schwartz: Yes.

Oracle: TX1056. Looking at it?

Jonathan Schwartz: Yes.

Oracle: March 26, 2008, after first SDK?

Jonathan Schwartz: Yes.

Oracle: I was with my Google buddy over the weekend. They hate GPL, like Apache Harmony. Jonathan Schwartz: It is a "give and force back" license.

Oracle: Copyright holder can enforce compliance with GPL.

Jonathan Schwartz: Yes.

Oracle: You thought that the GPL was okay for you?

Jonathan Schwartz: For achieving our business purposes.

Oracle: [?]

Jonathan Schwartz: Wanted to integrate well with the GNU community so we chose the GPL.

Oracle: Google ultimately chose the Apache License, right?

Prior to the release of Android, you thought that they were going to be released under the GPL.

Jonathan Schwartz: No, I think that you misunderstood.

[ establishes that the blog post went up that Google was going to not be using our code. Warns that Oracle has some emails on this ].

Oracle: did you know that Android was going to be released under the Apache license.

Oracle: Scroll up in TX1056 "Google takes without paying".

Oracle: You also say, "I totally agree with this"… taking.

Jonathan Schwartz: Yes.


Oracle: TX 1057, in evidence, email covering 'Java and You' presentation.

Oracle: Are you on one of these email lists?

Jonathan Schwartz: No.

Oracle: Did you review before JavaOne?

Jonathan Schwartz: No.

Oracle: Review what Rich Green was going to say about Android?

Jonathan Schwartz: I don't believe so.

Oracle: TX530, not in evidence - "refers to you". You see that 530 refers to a Jonathan?

Jonathan Schwartz: That would be me.

Google: Objection. Foundation.

Judge: Sustained.

Oracle: Did you suggest a licence to Google that would make Dalvik compatible, and would require TCK? Did you have that in mind as a possible solution for Android?

Jonathan Schwartz: Yes, if they wanted to use the name "Java".

Oracle: TX2070, email Oct 23, 2008

Google: Not on list of disclosed… apologies, on the list.

Oracle: Exchange between Jonathan Schwartz, Vineet Gupta, Brian Swetland?

Jonathan Schwartz: Yes.

Oracle: Discussion of Android?

Jonathan Schwartz: Yes.

Oracle: Next page, 2nd paragraph, Vineet writing to you: "Either they become [ or they become a competitor].

Oracle: Sun was considering using the IP hammer to bring Google into compliance?

Jonathan Schwartz: No, Vineet was considering it.

Oracle: TX205.copy of email from McNealy, October 2006.

Oracle: Questioning of "submarining revenue"

Jonathan Schwartz: No.

Oracle: You invested a lot into Java at your tenure at Sun?

Jonathan Schwartz: Yes.

[Tries to say that Oracle's opponent invited him, so that Google thought that he would help Google. ]

Jonathan Schwartz: That's fair.

Oracle: Feb. 2010, that's the day you were no longer CEO?

Jonathan Schwartz: I resigned, they already had a CEO.

Jonathan Schwartz: Mr. Sukind led the integration activities between Oracle and Sun.

Oracle: Sun's business was suffering before the acquisition?

Jonathan Schwartz: The financial crisis made it quite difficult.

Redirect of Jonathan Schwartz
Google: TX2707, move into evidence.

Judge: In for all purposes.

Google: You mentioned that there were companies funding Apache Harmony?

Jonathan Schwartz: Yes.

Google: Who?

Jonathan Schwartz: Oracle, SAP, IBM.

Google: you said the same about them?

Jonathan Schwartz: Yes.

Google: TX2341, not yet in evidence.

Google: Accurate report of comment you made in 2007?

Jonathan Schwartz: Yes.

Google: What does it say?

Jonathan Schwartz: There is no reason for Apache Harmony not to ship their code. What they cannot do is to call their product "Java".

Judge: [ argument over TX2341, didn't quite say the same thing ]

TX 2195. Recognize?

Jonathan Schwartz: Yes.

Google: What is it?

Jonathan Schwartz: March 2008, email from Jonathan Schwartz to Mr. Fowler.

Google: What's the Apache issue, hardware support?

Jonathan Schwartz: Yes.

Google: "But the code is available, to be clear, just not the brand"?

Jonathan Schwartz: [Explains what this means.]

Google: Did you consider Android to be a fragment of Java?

Jonathan Schwartz: Hard question. "It was simultaneously a threat and an opportunity."

Google: TX1055, 2007, reflects views at the time?

Google: First page, please? "A separate implementation isn't a fork, so long as Google agrees to certify that their platform is compliant with the Java specification".

Re-direct of Jonathan Schwartz
Oracle: TX2371- email exchange between Jonathan Schwartz and John Markoff. NYT reporter Nov. 6, 2007, post blog post.

Oracle: How come they can use?

Oracle: "Subject: I don't get it" -- see that?

Jonathan Schwartz: Yes.

Oracle: Jonathan Schwartz response: "Off the record, how they avoid those licenses, they have shown a stunning naivite on the licenses".

Jonathan Schwartz: Was trying to get him to write a story about Java.

[The witness is dismissed.]

[ jury goes on break ]

Judge: In all my 25 years, I have never heard of a hand signal…[ laughter]. Be careful on that. If you raise your hand indicating that you've got impeaching materials, but you don't use it, that could be a problem .

[ break ]

[PJ: I was curious what the hand signal thing meant. Here's the explanation, from Ginny LaRoe at
When introducing an email he believed would undermine Schwartz's testimony, Jacobs permitted himself another flourish: "I'm going to raise my hand when I know something you don't," he told Schwartz, his hand in the air. "I have some emails on this."

Displayed on each juror's flat-panel screen was an email from Schwartz that said of Google: "They also take Java for Android without attribution or contribution."

A grinning U.S. District Judge William Alsup didn't interrupt Jacobs's performance. But later, when the jurors weren't around he said he couldn't help but bring up "a small point."

"I did practice for 25 years," he told the lawyers, "and I'd never heard of a hand signal."

Belly laughs all around. Then Alsup warned Jacobs not to inappropriately signal impeachment to the jury. "No hand signals."

Update 5: While we wait for the final part of the day to be transcribed, I see the media is reporting that Scott McNealy told the courtroom that Jonathan Schwartz's blog wasn't official, just a personal blog. I beg to differ. I don't see how a CEO can ever have a "personal" blog that doesn't impact the company, in that there are SEC rules about what a CEO can say to the public and how he has to say it. But let's go back in time, shall we, and take a look at the archives of Schwartz's blog, back to October 2, 2006, a couple of years after he began to blog, and after you read this entry, you tell me if Mr. McNealy was remembering accurately or plausibly:

One Small Step for the Blogosphere…

I’ve been an officer of a public company for a while, and I’ve had access to confidential information for a good while longer. And I’m used to holding my tongue on issues that’d be deemed material to Sun’s financial performance. Like a pending acquisition or big sale, or data related to how our quarter’s going. In a public company, there are very strict laws surrounding how information’s disclosed.

So a couple years ago, when I first started blogging, I and our illustrious general counsel were far less worried about what I was saying, than where I was saying it. For example, I couldn’t use my blog to announce our quarterly performance, or disclose a material transaction. I had to use a press release, or a conference call (with a telephone operator, no less!).


A regulation known as “Reg FD,” or Regulation Fair Disclosure – which attempts to ensure no one audience gets preferential access to material non-public information. It’s a great concept, designed to prevent selective disclosure, or actions that might advantage one investor over another.

Unfortunately, Reg FD doesn’t recognize the internet, or a blog, as the exclusive vehicle through which the public can be fairly informed. In order to be deemed compliant, if we have material news to disclose, we have to hold an anachronistic telephonic conference call, or issue an equivalently anachronistic press release, so that the (not so anachronistic) Wall Street Journal can disseminate the news. I would argue that none of those routes are as accessible to the general public as a this blog, or Sun’s web site. Our blogs don’t require a subscription, or even registration, and are available to anyone, across the globe, with an internet connection. Simultaneously.

Now we happen to have a like-minded Chairman at the United States Securities and Exchange Commission (the ‘SEC’), Christopher Cox. So Mike and I sent along a rather formal note last week, requesting a clarification to Reg FD, one that would permit our (and everyone else) using the internet (eg, a company blog or web site) to release material information. Without a press release or operator assisted conference call. We are, after all, the primary source of such material information – there’s no point in going through an intermediary if what we’re after is fair disclosure and full transparency. Let the light shine in, don’t buy a flashlight.

We’ve had enough interaction with the Chairman (and read enough of his writings) to know he understands the utility of the internet to inform investors – but until we see a formal revision or clarification to FD, we’ll still be limiting what we disclose via blogs and the internet. And consuming trees with press releases. Which can’t, in the long run, be all that desirable.

But we’ll take it one step at a time…

I’ve attached the letter below (and yes, before you ask, we did fax it, and send by overnight mail).

Via Fax and Federal Express

September 25, 2006

The Honorable Christopher Cox
United States Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549

Dear Chairman Cox:

You have been a leader in the drive toward bringing greater transparency and access to information for individual investors. Indeed, you recently challenged us to the capitalize on the full potential of the Internet for the benefit of the American investor: “So the question for us today is how do we put the current communications technology to the service of the American investor. How do we harness the Internet which is serving so many customers in so many other ways to deliver the maximum benefit to those in our regulated capital markets.” (Chairman Cox Remarks to Interactive Data Roundtable, June, 12, 2006).

Sun Microsystems fully supports and applauds your recognition that the Internet is a “great instrument of national and international communication… [and] also a critical engine of American productivity.” (Id.) We have been on the same side of the issue from a technology perspective – from the evolution of open standards for document interchange (such as the Open Document Format), identity interchange (Project Liberty) to the drive toward open source (OpenSolaris and OpenOffice). Our view is that now is the time to fully exploit the innovation that only the Internet can yield in creating the most transparent environment possible for keeping all investors promptly and equivalently informed.

As adopted, Regulation Fair Disclosure’s requirement of widespread dissemination can be met through the filing of a Form 8-K or “through another method (or combination of methods) of disclosure that is reasonably designed to provide broad, non-exclusionary distribution of the information to the public.” (17 C.F.R Sec. 243.101(e)(2)) To date, the SEC has not taken the position that the Regulation’s “widespread dissemination” requirement can be satisfied through disclosure through the web-postings alone. While that may have been a pragmatic approach in 2000, we believe that the proliferation of the Internet supports a new policy that online communications fully satisfy Regulation FD’s broad distribution requirement.

Our corporate website ( currently receives on average of nearly a million hits per day. This website includes a blog that I write as CEO of Sun Microsystems (, as well as the blogs of over 4,000 of our other employees. My blog is syndicated across the Internet by use of RSS technology; thus, its content is “pushed” to subscribers. This website is a tremendous vehicle for the broad delivery of timely and robust information about our company. It is our view that proprietary news outlets are insufficiently accessible to the broad majority of Internet users and individual shareholders. It is certainly the case that the Internet represents a broader user base than those able to afford subscriptions to traditional forms of media and thus usage of this or any other freely available company blog or web site should be considered sufficient in satisfying the objectives of Regulation Fair Disclosure.

In 2001, upon the one year anniversary of the effective date of Regulation FD, Commissioner Unger requested a study designed to assess the implementation of the regulation. (“Special Study, Regulation Fair Disclosure Revisited,” U.S. Securities and Exchange Commission, December 2001, modified November 29, 2003.) Even then, the roundtable group recommended that “the Commission should embrace technology to expand opportunities for issuers to disseminate information online. The Commission should make clear that options such as adequately noticed website postings, fully accessible webcasts and electronic mail alerts would satisfy Regulation FD.” (Id.) The evolution of the Internet makes these recommendations even more compelling today.

We truly believe in the utility of the Internet – as a means of driving transparency throughout all governmental and corporate processes, as well as greater accessibility of health care, education and social services. Indeed, it is a principle on which Sun Microsystems was founded. We encourage you to look to the Internet to achieve the Commission’s objectives of greater investor access to information and would welcome the opportunity to further discuss with you our views in this area.


Jonathan Schwartz
Chief Executive Officer
Sun Microystems, Inc.

cc. Michael Dillon, General Counsel, Sun Microsystems, Inc

So, what do you conclude? Was this blog a just a personal blog? Me too. That's what I think, as well, in that it's described to the SEC, no less, as a corporate blog, hosted on Sun's corporate website. And he describes discussing things regarding it with the in-house lawyer, things a CEO has to think about that other employees don't have to worry about so much.

Update 7: Look at this entry in the Schwartz blog. By 2008, you can see that he included the Safe Harbor Statement. And notice the open sourcing of the Sun xVM:

xVM is our free, open source virtualization platform, which we unveiled at Oracle Open World, alongside our management platform, xVM Ops Center. xVM will virtualize Windows, Linux or Solaris, on either Dell, HP, IBM or Sun hardware. We've seen broad interest from across the world, especially from customers that want to avoid putting a proprietary virtualization technology at the base of large scale open source datacenters (“why go back?” one said to me). Interest in our virtualization story (from xVM to Solaris containers) expands to every industry, and nearly every customer – it's just about the number one item on the agenda.

Now, here's my question. Is it possible, do you think, that a letter like this would be sent to the then-head of the SEC without Scott McNealy knowing about it in advance? If so, maybe he didn't really hold the reins at the time? But that would work against his message in the courtroom today, which was that he was in charge, not Schwartz. But which one wrote to the SEC? If it was without McNealy's knowledge and he didn't give approval in advance, who was actually running the company in 2006?

Of course, over the years, our memories can fade, and as we age, sometimes we lose some memories. But when reminded, we can refresh our recollection. In that spirit, and hoping it will help him to refresh his memory, we offer this walk down memory lane.

Update 6: And here's the final piece, the testimony of Scott McNealy, the initial part of the testimony of Professor Owen Astrachan, and the end of day discussions between the judge and the lawyers:

Testimony of Scott McNealy, examined by Oracle's David Boies
Oracle: TCK, you know what that is?

Scott McNealy: Yes.

Oracle: It has a role in maintaining compatibility?

Scott McNealy: It ensures compatibility with the Java specification.

Oracle: About the positive comments about Android from Sun ( after Google's announcement)… Did you read the blog of Jonathan Schwartz?

Scott McNealy: Well, don't tell him, but I didn't read his blog.

Oracle: Was [his blog] a representation of any formal policy of Sun?

Scott McNealy: No. Blogs were personal, not corporate.

Oracle: Were you aware of any deals that were being negotiated with Google in 2006/2007?

Scott McNealy: Of course I was aware.

Oracle: TX205, already in evidence- it's an email Feb. 8, 2006, in response to an email from Jonathan Schwartz to you?

Scott McNealy: Yes.

Oracle: In it, you say, "I am worried about how we are going to replace the revenue that would be submarined by this deal". What did you mean by "submarining?

Scott McNealy: Google wanted a relationship [ with Sun ] that would minimize the revenue from J2ME.

[ Oracle tries to ask a hypothetical question of witness: If Sun had given a license…" -- Google: Objection, hypothetical. -- Judge: Oracle, you could ask how he felt about it.. Objection sustained.]

Oracle: In the time of you being Chairman of the Board at Sun, did you consider an incompatible [ Java ] platform to be a bad idea?

Scott McNealy: Yes. THe difficulty with platform incompatibility is [ big ]. Huge economic value [ is represented ]. Sun wanted to preserve compatibility. We wanted to avoid 99% compatibility problem. [Uses the example of nearly-compatibly PC platforms in the '80's, companies that went out of business from *nearly* being IBM PC compatible ]. "Allowing forking was a negative economic strategy."

Oracle: Regarding TX16.

Google: Objection, foundation.

Judge: Sustained.

[ Oracle lays foundation.]

Oracle: In the first page, this is an email from you to Eric Schmidt?

Scott McNealy: Yes.

Oracle: [ At the time ] you were CEO of Sun, he of Google?

Scott McNealy: Yes.

Oracle: We offer TX16 into evidence.

Judge: Accepted. Proceed.

Oracle: The email is from February 2006. "… crucial to Sun for financial success" What did you mean?

Scott McNealy: We had a very lucrative product [ in J2ME ], a strong revenue stream that was almost pure profit.

Oracle: TX63(?), an email from you to Eric Schmidt in response, dated March 2007. Why did you write it?

Scott McNealy: [ something lost ], and to comment on the Google situation [ with licensing negotiations? ].

Oracle: Did Jonathan Schwartz ever tell you that he disagreed with anything in this email?

Scott McNealy: Not that I recall.

Oracle: TX565-- in evidence. Page 2 near the bottom, an email from Vineet Gupta- see that?

Scott McNealy: Yes.

Google: Objection- not to or from the witness.

Oracle: I haven't gotten to the substantive question, yet.

Judge: Overruled.

Oracle: Do you agree with what Vineet Gupta said:

1) If Google is still using something with Java in it, then

1a) They will have to come to us with a license.

Google: Objection.

Judge: Sustained. You can ask an "independent" question.

Oracle: So set that document aside. Think…

Judge: [To witness] Think back, what is your recollection?

Scott McNealy: Yes [missed] [ We had conversations with Google about various things.]

Oracle: Why [ were you talking with Google? ]

Scott McNealy: Several ongoing conversations with Google about possible infringing patents, licenses, plus Google toolbar.

Oracle: The specification License?

Scott McNealy: Yes.

Oracle: Also a GPL license by which any company could obtain access to the Java specification?

Google: Objection.

Judge: [To witness ]: Do you personally know the answer?

Scott McNealy: Generally.

[Discussion of how the witness can answer in general terms for the jury.]

Judge: Objection overruled.

Scott McNealy: [What Google was doing was] re-sharing innovation.

Oracle: [What do you recall about] GPL restrictions?

Scott McNealy: I don't recall officially. [ Something about ] no commercial use.

Google: Move to strike.[PJ: There is no such restriction in the GPL.]

Judge: Stricken. [Tells jury to ignore the answer.]

Oracle: [Do you know of companies who licensed Sun's Java technologies and took licenses?

Scott McNealy: Yes.

Oracle: [Please name some of them.]

Scott McNealy: Motorola, Nokia.

Cross-examination of Scott McNealy by Mr. Van Nest
Google: You are a personal friend of Mr. Ellison?

Scott McNealy: Yes.

[ missed one Q/A ]

Google: You made a lot of money when Oracle bought Sun?

Scott McNealy: Actually, I cashed out before the acquisition closed.

Google: How much money did you make?

Scott McNealy: A couple hundred million [dollars].

Google: You've made references to Larry Ellison as being a national economic hero?

Scott McNealy: Yes, for all the taxes he's paid.

Google: You proposed renaming the Norman Minetta airport in San Jose be renamed to the Larry Ellison Airport?

Scott McNealy: Yes.

Google: Do you remember the JavaOne conference, the one where you were on-stage with Larry Ellison?

Scott McNealy: I don't recall.

Google: You are a big fan of Open Source, right?

Scott McNealy: Yes.

Google: You said, "Open is good."

Scott McNealy: Yes.

Google: [You are quoted as saying "open and shared"; this is about Java, right?

Scott McNealy: Yes.

Google: You said, "Interfaces should be public and open"?

Scott McNealy: Yes.

Google: That nobody should own the fact that the brake pedal in a car is to the left of the gas pedal", right?

Scott McNealy: Yes.

Google: That's an API, right [ position of gas and brake pedals ]?

Scott McNealy: [Disagrees over this characterization.]

Google: Playback of conference.

Oracle: Was this disclosed?

Google: No. It's impeachment evidence.

Judge: Go ahead.

[ playback]

Scott McNealy: "Written, spoken language should not be copyrightable."

[ missed; essentially Larry Ellison saying what Google has asks him about previously in their cross about what Larry Ellison has said. ]

Judge: Record will be allowed in. Arguably impeaching the witness.

[PJ: In case anyone is new, impeachment is the legal term for when you are presenting information as evidence that demonstrates the witness isn't being truthful in his testimony. If you think of APIs as analogous to all the possible types of objections a lawyer can use, i.e., lack of foundation, assumes facts not in evidence, argumentative, leading, hearsay, etc., impeachment is one type of objection on the list. To actually use these "APIs", or objections on the list, the lawyer has to decide during testimony which type of objection he wants to use when he hears the other side ask the witness a question, then actually stand up and say the name of the type of objection, and then he has to respond to any questions the judge may ask or to any reasons the other side presents why the objection is not right. That list is the APIs in my analogy. The list doesn't do anything. The items on the list don't do anything. They are ideas. You can think, Impeachment. All the lawyers and the judge know what that means if you say it out loud. Saying impeachment isn't the definition of that objection, although in books you can find the menaing and a description of it, but everyone knows and has the idea in their minds when you stand up and just say the name of the objection, Impeachment. The implementation of impeachment objections is to stand up and start the process going and follow through on the steps. If you choose the wrong one from the list, it doesn't work. There you go: An API definition for lawyers. You are welcome.]

Google: You were referring to APIs.

Scott McNealy: For products.

Google: The whole point of an API is for interoperability, right?

Scott McNealy: [Disagrees, citing private interfaces as a counter example.]

Google: You never owned the Java programming language?

Scott McNealy: No.

Google: Did Sun ever take a public position with regard to APIs and blueprints?

Judge: We're not going to get into that [I don't understand why not, but IANAL.]

Google: In 2006, Jonathan Schwartz became CEO. He had the authority to enter into licenses, [etc ], right?

Scott McNealy: [Made it clear that the CEO had to answer to the Board of Directors.]

Google: Mr. Schwartz had a blog as CEO, right?

Scott McNealy: I didn't read it.

Google: You never shut it down, did you?

Scott McNealy: No.

Re-direct of Scott McNealy, by Mr. Boies
Oracle: Regarding the video, whether interfaces are public, does that mean that you wouldn't want a license?

Scott McNealy: No. We didn't license our technology with rights equivalent to ownership.

The next witness called by Google was Professor Owen L. Astrachan of Duke University.
Testimony of Professor Owen L. Astrachan
Google: [Established Professor's Astrachan's very impressive bona fides, etc. which you can read here.]

Google: What are the 37 things that are contested here, technically?

Professor Astrachan: Packages.

Google: Would you know what an API package is?

Professor Astrachan: Wouldn't use that term.

Google: You've heard references to "the API"?

Professor Astrachan: Yes.

Google: And it means lots of things to lots of different people?

Professor Astrachan: Yes.

Google: In Java 5, all 166 packages could [ accurately ] be referred to as "The API"?

Professor Astrachan: Yes. References the book, "The Java API: Core Interfaces"

Google: In this context, the title is referring to the Java API?

Professor Astrachan: Yes. I would use this term with someone who was conversant with much of the Java API.

Google: There are separate APIs for different languages?

Professor Astrachan: Yes, each of C, C++, and Java have their own distinct API.

Google: So the phrase "java.lang API"?

Professor Astrachan: Would be well-understood.

Google: "java.lang.math API", would that be understandable?

Professor Astrachan: That wouldn't be perfectly accurate, but well-understood.

Google: There is confusion regarding different parts or aspects of an API?

Professor Astrachan: [ missed ]

Google: Is it common to refer to an API as an abstraction?

Professor Astrachan: Yes. It helps programmers to understand. These days, interfaces have become pretty standardized.

[Discussion of standards, standards enforcement, interoperability.]

Google: Wants to show the witness the schematic diagram of Android.

[Time was spent trying to find/get the diagram up for demonstrating… and the Judge started asking questions.]

Judge: What is a specification?

Professor Astrachan: It tells you how to use something, how something behaves.

Judge: Is it the same thing as a declaration?

Professor Astrachan: The specification might include more information than just the declaration, for instance, error condition behavior. Take the java.lang.math.sqrt function. The declaration doesn't tell you anything about how it behaves when given a negative number.

Judge: Could a specification be one line of symbols?

Professor Astrachan: That would be overly terse; error behavior would be unknown.

Judge: Signature… what does that mean?

Professor Astrachan: A declaration and a signature are synonymous.

Judge: [Suggests that the trial takes a break soon and start fresh tomorrow with the rest of Professor Astrachan's testimony. It's 12:45.]

Google: [Gets admitted into evidence Dr. Astrachan's reports that he made in connection with this trial: TX2523, 2529, 2532, 3523 ]

Google: This is a good time to break.

Judge: You'll be able to finish [ with this witness ] tomorrow?

Google: Yes.

[ Jury exits ]

Judge and Lawyers Discussion

Judge: Next item: The jury verdict form.

[ break ]

[ 1:00PM, re-convene ]

Judge: Starting conference on jury instructions and the verdict form.

Judge: Submit to me by late tonight the 3 biggest areas of heartburn.

Judge: I want a definition of the compilable code, declaration, name.

Judge: Here is why I bring it up: English is not a part of the SSO.

Judge: What [ do we call ] the part that gets compiled? Code? Compilable code? "Package" doesn't get us where we need to be.

Google: Timeframe for submissions?

Judge: 5-6 pages per side. [ missed time frame… not long from now! ]

Judge: Time used up in the trial so far: Oracle: 944; Google: 900.

Judge: The gap is closing.

Judge: In these instructions, there will be 3 areas to confer:

a) ownership

b) subjective/objective [ from Apple case ]

c) [ not mentioned or missed ]

Judge: Regarding ownership, Oracle has to make up its mind to put in more… one clear way is to drop question #2 and stay with a global [ approach ].

Judge: 1) No evidence to support the 37; 2) Not in the pretrial orders; 3) comparison on a one-to-one basis means ownership problems.

Judge: [ to Google ] Is that your understanding on Question 2?

Google: [ Baber] No, your Honor.

Google: Question 1 is on solid ownership grounds; it addresses the whole.

Judge: To Oracle: If you need to re-open, you've got one hour and 16 minutes.

Google: 1) Oracle hasn't proved ownership. 2) They cannot prove what "the work" is. There is no evidence in the record.

Oracle: We have the title on the document.

Google: That's not enough.

Judge: Is this point preserved in the pretrial orders?

Google: Yes, your honor, Matters of Fact.

Judge: Question from the jury: did he say that the specification says what code to write in the implementation, and if so [ missed, moot ].

Oracle: What reports are Dr. Astrachan referring to?

[ I believe that his 4 submissions are what this was talking about. ]

[Adjourned for the day.]
Bloomberg reports it this way:
“I cashed out a couple of hundred million, $150 million, of stock,” McNealy said.

Van Nest played a video of McNealy speaking at a 2010 conference in which he said “interfaces should all be published and open.”

“Open doesn’t mean throw it over the wall into the public domain,” McNealy said.

To which one can only ask, then what *does* open mean?

Update 8: Jonathan Schwartz tweeted on April 27:

Under oath, u said my blog was personal, not Sun communication?
Srsly? :)
To which Simon Phipps responsively tweeted:
@OpenJonathan I have a Sun Chairman's Award that suggests otherwise...
Mr. Phipps's tweet says, "Tweeted on Android." I noted too that Mr. Schwartz links to a plaque, in the Attorney's Lounge in the courthouse where photographs by Judge William Alsup are displayed, and it reads
These photographs are dedicated to trial lawyers and their important work. They have the hardest job in the courthouse.

William Alsup
March 2011

Indeed, as Schartz writes, Judge Alsup is "a class act".

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.

  View Printable Version

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 )