It was a Friday, and the tone of the day was low stress, lots of good humor, and more of the judge trying to make sure he understands the tech. He indeed expressed that while the patent claims seem clear, the copyright claim is still a bit murky to him, or as he reportedly described it, "it's strange". So more briefs requested.

He then asked the parties, once the jury came in, to speak for ten minutes each, explaining what each side thought had been established so far. And then after the jury was excused for the day and the weekend, the judge significantly said something to the effect that it is one kettle of fish to say you need a license to use their source code; it's another when you say you need a license regardless of whether you've done an independent implementation.

So he's definitely getting the point.

Jump To Comments

[Update 1, Update 2, Update 3,
Update 4, Update 5]

Meanwhile, I'm trying to understand also. I asked a programmer who helped me in the SCO saga by explaining the tech to me to please explain APIs. All the conflicting imagery had me confused, file cabinets, verbs, blueprints. What exactly would he describe them as being like? Here's what he told me:

Essentially, an API is the way to interact with a system, programmatically (that's what the 'P' is for).

Let's say I'm writing a program to analyze a lot of data. I realize I want to compute averages, standard deviations, etc. But I don't want to have to write lots of code. So I go and find a statistics library somewhere that someone has already written. A library is essentially an implementation plus an API. The implementation is the code to do the work, while the API is essentially the way a programmer can activate the library.

Oh, here's a good analogy maybe. A 747, or planes in general. The "API" to a big jet is the control panel. There's a steering wheel. Maybe there are dials to set the altitude. There's a button to turn on autopilot, and to turn it off. The "implementation" is everything that the control panel is connected to. i.e. the engine, the electronics that control the engine, etc.

Now let's say you go to another plane, a 767, or whatever. Maybe it has the exact same control panel. But the "implementation" will be different - there are certainly a different number of engines, and (whatever else differs in planes). But by sharing the control panel / API between the 747 and 767, any pilot that can fly one can fly the other....

Back to the statistics example, more concretely.

Let's talk about averages and mins and maxes.

You could imagine an API for a stats library like e.g.:

Average(number list_of_numbers[]) returns number;

That means: I'm declaring a function named 'Average', which takes 1 parameter (list_of_numbers), and returns a single number. Here "number" would be a special type in the language. (Maybe the language also has "text" as another type.) There should also be a comment like

// This function returns the average of all the numbers in list_of_numbers.

so a programmer knows what the function does.

You might also have:

Max(number list_of_numbers[]) returns number;
Min(number list_of_numbers[]) returns number;

And so on... The API is essentially the set of these declarations.

The reason that programmers are pretty upset about this lawsuit is that sharing APIs is Good and Proper. It's really nice if people can agree on an API, and then lots of people can use it. Back to the Java plane - wouldn't pilots like it if there was a standard control panel, that worked across all airplanes? Programmers are the same way.

I think historically if someone made a new computer language (which typically include as a key component their libraries (and the APIs to those libraries) ), they made money by selling the compiler and/or the operating system. Programmers presumably place most of the value on the implementation, though, not the API.

I believe if Oracle somehow bought C++ (I don't think anyone owns that computer language, so that's probably not plausible, but just pretend), they would claim that anyone that implemented a C++ compiler would be stealing their APIs. So in their world, you either buy their C++ compiler, or you can't program in C++. I have no idea about any legal issues - but this would be bad for the advancement of technology.

Well, airplanes are easier for me than file cabinets and blueprints. But it raises a question in my mind. You might read this article or at least the Apple v. Microsoft ruling that is linked to on that page.

Then, this is my question: why wouldn't a menu bar be kind of like an API? Analogies are never perfect, but each item you click on calls up something, some functionality you want. I mean, what is there about 37 APIs or a million of them that isn't just a list that gets your computer to doing something you want it to do?

In other words, what isn't functional about an API? The class libraries would be what, in my menu analogy?

Maybe nothing, if the analogy is wrong, but in Apple v. Microsoft, the judge ruled you can't protect things that need to be more or less the same, so everyone is familiar and so the first to invent it doesn't own the entire market.

And what does this interchange mean about methods, etc.?

Judge: How many methods are in the 37 APIs?

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

This is from the trial, Day 2.

I think the "37" APIs is essentially like 37 header files - if you remember stuff from the SCO trial. java.awt.font, java.beans, etc. Those each correspond to a "package" in java. The package can have many classes, and each class can have many methods (which are just like functions).

As far as your menu bar analogy - yes, I think that's a pretty reasonable analogy. The menu bar is an interface for humans, while an API is an interface for computer programs.

I would agree that if you protect the Java APIs, then no one else can use Java without Sun giving the OK. Which I think is bad. And I guess the judge in the Apple v. Microsoft case thought it was bad.

: )

OK. But here's the deal. If you send me an email like that, you have to offer a *better* analogy. Also, here's a summary of the trial issues, the tech, and the history of Java by Stephen Shankland at CNET. I think he's about the best tech writer in terms of making the complex understandable.

Here's mirror_slap's report, which he'll be adding to as he gets his notes (33 pages) transcribed:

Oracle v. Google, Day 5

[I arrived at the courtroom at 7:20 AM. On my way into San Francisco on the BART, there was a planet rising in the East, brilliantly lit by the sun. Today I decided to try taking notes on my iPad to hopefully avoid the transcription time lag. If it turned out to slow me down, I could just use the trusty notepad. Which is what I ended up doing.

It seems that the demeanor at the Google table is more upbeat. There are even some smiles. At the Oracle table, nobody is smiling, except Michael Jacobs, the lead attorney from Morrison & Foerster. One of the attorneys at the Oracle table appears to be reading the laptop screen of one of the Google attorneys whose back is to him. Or maybe he is staring intently at the water cooler at the side of the courtroom for two minutes or more. They are both seated, back to back, only the Oracle attorney has turned and is facing toward the wall on Google's side. The Oracle attorney is a tall, slim, bespectacled guy with very closely-cropped light brown hair that is thinning on top.]

Judge Alsup comes in.

Judge Alsup: The lawyers seems to be doing pretty well. Usually by now someone would have come down with the flu or something.

Judge Alsup: With regard to the papers I asked for?

Oracle: Regret to inform Your Honor that we did not follow the letter of your order, but I think that we followed the spirit, so as not to subject Your Honor to my handwriting.

Google: We followed your order to the letter Your Honor... [chuckles from everyone.]

Judge Alsup: When you say class libraries, what do you mean by class libraries? Are they in source code form or compiled?

Oracle: I'd like to check with our experts.

Google: We refer to both source code format, as well as the byte-code environment. It's the same code, just in two different forms. The body of code that performs the function that actually does the work when you call it.

Judge Alsup: Is this the implementation?

Google: Yes.

Judge Alsup: Let me give Mr. Jacobs another shot at it.

Oracle: Defers to his client.

Judge Alsup: Next question... to Google. Does Google admit to factually copying the SSO, and declarations of the Java API elements (from the Java documentation)? Explains that this is for the jury to avoid having to explain to the jury about "substantial similarity".

Google: [Mr. Baber] I will answer the premise... there still must be substantial similarity in the work as a whole.

Google: Google did not copy the SSO or declarations. In their cleanroom Google used the names of the elements in the class libraries and the declaration elements. These are elements we all use.

Judge Alsup: Now to the point you were making... What is the concept of the works as a whole?

Google: [Falls back on pleadings.]

Judge Alsup: This came up on summary judgement, the de minimis argument.

Google: The plaintiff has to identify its copyrighted work. Oracle has not placed into exhibit any separate registrations for copyright for their works. Judge Alsup reserved judgement on this.

Judge Alsup: Is it the same work for the purpose of fair use, de minimis?

Judge Alsup: Mr. Jacobs, do you agree with this?

Oracle: [Indicates that each file is copyrighted. There is a lot of law that says that you don't escape copyright law by virtue of the amount that you add.] Judge Alsup: Nobody is arguing that stealing someone's work and adding to it is okay.

Google: What is the accused work?

Judge Alsup: How many lines in the 37?

Google: Maybe 100,000?

Judge Alsup: What is the work as a whole? Is it the 37 APIs? Is this a question for the court or for the jury?

Google: It is probably for the court to decide the scope. Many considerations.

Judge Alsup: It seems a little hard on the judge that the complaint could say... that we would get all the way to the end of the trial and I have to instruct to the jury as to what the scope of the decision is supposed to be.

Google: The works are the two you pleaded and if you [Oracle] need to …

Oracle: We are mixing two issues. 1) Versions of Java covered in this case are 1.4 and 5.0. 2) You are abandoning rights to intervening versions [I don't recall any significant versions between 1.4 and 5. There was a release numbering scheme changed. Was this a red herring? ]

Comparing A:B. This needs to be briefed.

Judge Alsup: Do the brief. It's due Sunday at 3PM [Same time as a previously requested brief is due.]

Judge Alsup: Here are the questions in my mind: Should the plaintiff be held to what they pleaded in the complaint? I find it strange that on the patent side it is really clear what is being infringed. With the copyright side, it's strange.

Yes, Mr. Van Nest? I saw you jump up when I said Sunday by 3PM.

Google's Robert Van Nest: The brief is limited to 10 pages?

Judge Alsup: Use good judgement. If you can express it in fewer than 10 pages, then by all means please do so.

Judge Alsup: Any other issues?

Oracle: We have stipulations [of exhibits that both side agree to be entered into evidence]
Oracle: 748, 749, 751, 752, 753, 2800, 2801, 2802, 3341, 3342, 3343, 3344, 3346, 3348, 3349, 45.1, 45.2, 45.3, 46.20, 46.21, 46.22, 46.23, 46.24, 46.25, 46.26, 46.28, 741, 767, 770, 771, 773, 862, 1030, 1031, 1032, 1033, 1034, 1035, 1036, 1037, 1039, 1039, 1040

Judge Alsup: Any more?

Oracle: No, your honor.

Judge Alsup: Okay, these exhibits are entered into evidence.

Judge Alsup: Didn't I ask for a one-page list of names of the lawyers?

Google: Your Honor, you asked for a one-page "Who's Who".

Judge Alsup: It would just be very helpful for the jury to know who you are and how to spell your names. Mr. Boies isn't here today, but they are spelling his names like "Boys", as in "Boys and Girls". It would help the jury be respectful to the lawyers.

Google: Do you want photos, too?

[Fun exchange between Judge Alsup and Mr. Van Nest about producing baseball cards for the lawyers.]

Google: Do you want our averages? [Big grin… something tells me he wins a lot.]

[Back to business.]

Judge Alsup: Regarding downloading something from a web site, if the download is stipulated, it doesn't follow that the date of the download is stipulated. Only the four corners of the exhibit itself are stipulated. Additional proof would need to be given regarding the date of the download.

Jury is seated.

Judge Alsup: Everyone doing all right? You are all smiling… it must be because it's Friday. We are about 1/3 of the way through the case. We are pretty much on schedule. I am going to do the following now: each lawyer will have 10 minutes to summarize what has been proven or not, where they are going next with the case, etc. It's a useful exercise. Again I remind you that nothing the lawyers say now is to be taken as evidence.

Oracle: [Mr. Jacobs puts up slides.] You've seen CEOs of both companies. "Our commitment is to make this as clear to you as possible". Java is licensed by companies all over the US and the world. There are different types of licenses for different purposes. The different types of Java implementations (standard, enterprise, micro) speak to the scale of the Java implementation. You wouldn't try to run a database on a credit card; there are differences in CPU and memory sizes. Sun/Oracle has IP and that's protected by copyright notices and copyright registration notices. Ref TX475 and TX476. [Missed some points.] You have heard that Google is an outlier [in terms of not taking a license]. You've heard from Larry Ellison that Google didn't take a license, and from Larry Page saying that he couldn't name a single company that hadn't taken a license. There was a licensing scheme that was available to Google. Google knew that they needed a license. Their CEO was repeatedly told that Google needed to take a license. Ref TX 7. Google understood that they had 23 options.

Sun/Oracle wants Google to be a member of the Java community. Google will try to say that this is trivial, that Android is huge. You heard from Eric Serevan that API design is incredibly creative. You heard from Mark Reinhold about how creative API design is. Mr. Bloch at Google said that API design is an incredible craft, a rewarding and noble craft. It involves esthetics; a nice-looking API is easier to use. Mark Reinhold said that the APIs for java.nio took 2 years to design. Both the API and the documentation come from the Java source code files. When we talk about the API, we are talking about the design of the [missed… Java environment? ] Regarding copying, in the next few hours we will see the nature of the copying, and the problem that the Google copying caused [some items including fragmentation], creating a Tower of Babel rather than a consistent programming language.

[Shows slide of relative size of Java language (tiny blip) versus the APIs (huge), with a small overlap. I didn't see any indication as to whether or not it's to scale.]

Apache Harmony did not get a license. Ref TX 326.

Google [Van Nest]: [In a strong, clear, steady, persuasive voice] Thank you (to the jury). We (Google counsel) have been attentive regarding whether you are being attentive, and we really appreciate your being so attentive. [Thanks Judge Alsup, too.] Here is the key evidence that is *not* in dispute: Use of the Java programming language; using the names of class files and methods; the source code in the libraries, comprised of original Google works and Apache Harmony. So what we are down to? -- the Sequence, Structure and Organization, the SSO, which is in the public domain. Apache Harmony and GNU used them and Sun did nothing. Then Google did what they did. Larry Ellison himself said that "Nobody owns the Java language". The APIs have been used for years as a necessary part of the Java language. There has been lots of testimony on this:

1) Waste of time trying to program Java with no APIs --Bloch
2) Google helped with the GNU classpath. Bloch, at Sun, helped GNU with it.
3) Dr. Kurian: Apache Harmony was using the same APIs from 2005-2009. The Apache Harmony license allows using any or all parts of Apache Harmony. Sun did nothing to stop this so long as Apache Harmony was not called Java.
4) Tim Lindholm's testimony: You need a system of organization to be able to use a programming language.
5) Google built their APIs from Open Source software.
6) Larry Ellison at the Java One conference: "We like Android and we know it's using Java.
7) Google's use of the SSO was absolutely fair. Google transformed something that wouldn't work on smartphones to something that would.
8) Larry Ellison knew that Sun had tried and failed to build a Java-based smartphone, and then Oracle tried and failed to build a smartphone.
9) Oracle even tried to build their smartphone on top of Android.
10) Oracle talked with Google, offering a business proposal.
11) Oracle wants to participate in the smartphone market without doing any work.
12) Mark Reinhold said that Write Once, Run Anywhere doesn't mean that there is code fragmentation in the JME, yet Sun launched Project OneJava to solve this very problem. But Oracle killed the OneJava project when they bought Sun.
13) This is all about Oracle trying to get a claim on the smartphone market.

First witness of the day: Brian Swetland

[Mr. Swetland sports a pony tail.]

Oracle: [Fred Norton?] You are employed by Google?

Brian Swetland: Yes.

Oracle: Working on Android?

Brian Swetland: Since 2005.

Oracle: Previously at Danger, Inc?

Brian Swetland: from 2000-2004.

Oracle: You worked with Andy Rubin at Danger, Inc?

Brian Swetland: Yes. Andy was the founder of Danger, Inc.

Oracle: [What smartphone products did Danger Inc produce]?

Brian Swetland: The HipHop, a.k.a. SideKick, which ran Java apps.

Oracle: [Did the HipHop run on a JVM?]

Brian Swetland: At the time that HipHop was released, the VM wasn't a JVM, and it was not compatible with Java.

Oracle: You never had access to Sun's Java source code?

Brian Swetland: No.

Oracle: Was the VM in HipHop incompatible with the JVM specification?

Brian Swetland: Yes.

Oracle: So Danger, Inc. took a license from Sun?

Brian Swetland: Yes.

Oracle: Danger then made the JVM compatible?

Brian Swetland: Yes.

Oracle: You worked with Tim Lindholm, who worked at Sun?

Brian Swetland: Yes.

Update 3: -- Here's Part 2b:

Oracle v. Google, Day 5, Part 2

Continued testimony of Brian Swetland

Oracle: You worked with Tim Lindholm, who worked at Sun?

Brian Swetland: Yes.

Oracle: Tim Lindholm said that Danger, Inc. needed to take a license?

Brian Swetland: [ Yes ]

Oracle: [Do you recall anything about Sun's copyright claims?]

Brian Swetland: I recall that Sun said that the method signatures were copyrighted.

Oracle: (Shows TX14.9 to Brian Swetland, a May 31, 2006 email to Andy Rubin and Dan Borenstein.) Do you recognize this email?

Brian Swetland: Yes.

Oracle: You wrote it to Andy Rubin and Dan Borenstein about Sun?

Brian Swetland: [ Yes ]

Oracle, quoting Brian Swetland: "whatever happened to the bullshit Sun copyright issue?". Did you write that?

Brian Swetland: Yes.

Oracle: [ To the best of your knowledge, ] nothing had changed from what you knew [ in ~2000 to 2006 ] about Sun's claim about copyright of the method signatures?

Brian Swetland: No.

Oracle: Did you start any investigation of Sun's [ current as of 2006 ] copyright claims?

Brian Swetland: That was not my role [ at Google ].

Oracle: [ Asks about employees at Danger who are now at Google working on Android ]

[ missed ]

[ I think that this is going to Oracle's claims about the Google cleanroom implementation? But Andy Rubin apparently got a clause added to the Sun agreement that Danger, Inc signed that specifically said that Danger, Inc. did not get access to Sun's source. ]

Oracle shows TX13, an email from Brian Swetland to Mr. Rubin and Mathias [?] dated 2006 - You wrote this?

Brian Swetland: [ the form of the email seems to confuse Brian Swetland a bit. Not sure what his exact response was, but I think it was, "I think so." ]

Oracle, quoting: "reason to shift to a primarily Java API"

Brian Swetland: There appears to be some other email that I quoted.

Oracle , quoting: "I think that we are pretty set on it…" Did you write that?

Brian Swetland: Yes.

Oracle shows TX123 from Brian Swetland to 3 fellow Google employees, Andy Rubin, _____, and Chris ____ [ missed ].

Oracle shows TX314 from Robert Reasoner to Brian Swetland dated August 2005, asks Brian Swetland to read a specific passage.

Brian Swetland: … a JAVA VM for Android…

[ All the above-mentioned exhibits were entered into evidence without objection. Google has been very judicious (IANAL) regarding which exhibits they object to, perhaps 1/3 (by Judge Alsup: 's count). The reason is that Google effectively uses a lot of the admitted evidence in demonstrating fallacies in Oracle's arguments and making their own arguments that (to me) are more straightforward and sensible. ]

Cross-examination of Brian Swetland by Google counsel Christa Anderson

[ A juror coughed. Judge Alsup asks if the juror would like a cough drop, which the juror gratefully accepts. Judge Alsup is very focused on getting this trial to a successful and timely conclusion, and has stressed his concern for the jury's health and his enthusiastic willingness to eject members of the gallery if they are "hacking or coughing" or are otherwise distracting the jury. This judge is awesome. ]

Google: Before we begin, some introduction [ of the witness to the jury ].

Brian Swetland: [ Lives in Palo Alto, graduated from University of Chicago at Champaign/Urbana. After graduating, worked for 2 years at Be, then at Danger, Inc. for 4.5 years. In late 2004, joined Android, and became a Google employee in July 2005 when Google purchased Android. Brian Swetland is the Systems Software Lead for Android at Google. ]

Google: What was your primary focus [ in your job at Google? ]

Brian Swetland: The kernel and device drivers.

Google: When did you first learn Java?

Brian Swetland: In 1995 at the U of Chicago.

Google: How did you learn to program in Java?

Brian Swetland: By tinkering with it [ more response lost ]

Google: Did you read any books?

Oracle: Objection, leading.

Judge Alsup: Sustained. [ I think in this exchange that Judge Alsup reminded Google to "use the magic words, To What Extent and [missed], to avoid being called for leading. I believe that a response was finally elicited from Brian Swetland regarding the book, The Java Programming Language Specification. ]

Google: What about the Java APIs? Are they free for use?

Brian Swetland: They have to be, otherwise the language would be useless.

Google: Regarding Danger taking a license from Sun, what was your understanding?

Oracle: Objection, hearsay.

Judge Alsup: Sustained.

Google: [ tries again ]

Oracle: Objection.

Judge Alsup: You know that I don't like that question [ when the witness has no direct knowledge ]… [ warns Oracle that they ] will have to live with that rule, too.

[ Different line of questioning ]

Google: Has Tim Lindholm ever been a member of the Android team?

Brian Swetland: No.

Google: What code did Tim Lindholm contribute to Android?

Brian Swetland: None [ to the best of my knowledge ].

Google: What was Tim Lindholm's role for Android?

Brian Swetland: Business development.

Google: Regarding TX149, what is the link in the email?

Brian Swetland: it's a link to Google's webkit. It allows you to convert Java to Javascript, which is a different language.

Google: The last line of the email, "Whatever happened to the bullshit argument over Sun's method signature copyright?". What did you mean by that?

Brian Swetland: I was trying to get at whether this was still an issue. The GNU GCJ, GNU Classpath projects and lots of JVM implementations had been done [ without a license ].

Google: Regarding TX13 [ not sure of this TX number ], in the double-caret-indented section, a portrion of the email that you did not personally write, in paragraph 5, [ do you see that? ]

Brian Swetland: [ Yes ]

Google: It uses the word "bright", yes?

Brian Swetland: It does.

Google: [ And that refers to you? ]

Brian Swetland: [ It appears to. ]

Google: But you didn't write this portion? ]

Brian Swetland: [ No. ] It refers to me in the 3rd person. It would be totally weird for me to write about myself that way.

Google: The reference to "Brian being scarred by the Danger experience", what was that about?

Brian Swetland: My unhappiness with Danger taking a license from Sun for Danger's [ already-implemented ] cleanroom implementation, and that Danger didn't need it [ a license ].

Google: Regarding X314, is this the email exchange in August 2005?

Brian Swetland [ Yes. ]

Google: At the top of the email, it says, "I really cannot comment on the project". The email references a lot of people, Tim Lindholm, Frank Yellin [sp?]. Did any of these people work on the Android project?

Brian Swetland: No.

Google: TX200? [ missed the question ]

Brian Swetland: No

Google: TX409? [ again, missed ]

Brian Swetland: No.

TX33? [ sorry, folks ]

Brian Swetland: No.

Redirect of Brian Swetland, by Mr. Norton for Oracle

Oracle: Whether you were scarred by the Danger experience or not, you know that Danger took a license?

Google: Objection, hearsay.

[ back and forth among Judge Alsup, Oracle and Google regarding this ]

Oracle: Android uses the API method signatures?

Brian Swetland: Yes.

Oracle: Other companies' and the GNU implementations were complete?

[ There is an exchange about what Brian Swetland knows and doesn't know. It boils down to the fact that he has not done a complete code analysis of the systems about which he is being questioned, and so he cannot say either way whether they are complete. ]

Brian Swetland: I don't know. [ Oracle seems to me to be badgering the witness in this string of questions. Brian Swetland, like most techies I know, really doesn't like to say, "I don't know" when he has a pretty good grasp of the subject, but he has been boxed in by the requirement that he can only testify about things about which he has direct personal knowledge. ]

Oracle: Is Android a complete implementation?

Brian Swetland: Based on your assertions [ of requiring direct knowledge ], I don't see how I can answer your question.

ReCross of Brian Swetland

Google: Mr. Swetland, what is your understanding about why Danger took a license?

Brian Swetland: To avoid the expense of litigation [ with Sun ].

End of Brian Swetland testimony.

Judge Alsup to Jury

Judge Alsup: (to jury) As I've told you in the past, there is discovery and pleading (accusations). Sometimes in the trial, something that gets admitted [ is changed from what was plead. ] Example: a traffic accident happens, and in the pleadings it is stated that the light was red. Then at trial, the witness says that the light was almost red. The judge can tell the jury that, in the pleadings the party said that the light was red. [ does Oracle see flashing red lights, I wonder? ]

There are 2 items that Oracle wants admitted as evidence against Google. They are Google's pleadings. Their pleadings are evidence. Not conclusive evidence. You have to use your own judgement in weighing the evidence.

1) Google has said that Sun released the specification for the Java platform, including the JVM, under a free license that allows developers to implement in a cleanroom… Sun would provide a license [ the JLS ].

The only way to prove compatibility is to run the TCK for a particular version of Sun's Java. The TCK is only available from Sun. It is not Open Source software. It is at Sun's discretion whether to provide access to the TCK. Developers could not easily obtain a critical component for [ testing a JVM that they had independently implemented ].

2) While Sun did open source Java, Sun added Field-Of-Use restrictions that precluded running the TCK on mobile device [ JVM's].

Update 4: Here's the testimony of Bob Lee and Dan Morrill:

Witness Bob Lee

Oracle: [Does some introductions… hmmm. it seems that when the witness is considered by Oracle to be helpful to their case that they do the introductions themselves. When the witness is someone that they want to tear apart, they let Google use their time to do the introduction. ]

Bob Lee is the Core Library Lead for Android from 2007-2010, and started working at Google in 2004. He joined the Android team "early, when there were 12 people on it" in 2006.

Oracle: The core libraries included libraries in the Java namespace?

Bob Lee: Package names, yes.

Oracle: Android implements part of the JSE5 APIs but not fully?

Bob Lee: Yes [partially implements].

Oracle plays video deposition page 48 lines 7-16, "Android uses Java APIs? Bob Lee (on video): Yes.

Oracle: Do you like the Java APIs?

Bob Lee: Yes.

Oracle: You consulted on the Java API specifications?

Bob Lee: Yes.

Oracle: [You consulted on the Java API specifications] while you were doing work on Android?

Bob Lee: Yes.

Oracle: You saw the copyright notices?

Bob Lee: Yes.

Oracle: Are you familiar with the TCK?

Bob Lee: Yes.

Oracle: You wanted to run Android's JVM through the TCK?

Bob Lee: Yes, but it was different versions of Java, [somewhere between] JSE, JME.

Bob Lee: "Because I work a lot on what I thought were open Java standards".

Oracle: You worked with an external company, Noser, that was hired by Google to work on implementing the Java class libraries?

Bob Lee: Yes.

Oracle: Ref TX405, email between Bob Lee and Eric Schmidt, May 30, 2008: You wrote to Eric Schmidt about Apache Harmony?

Bob Lee: Yes.

Oracle: In the middle of the paragraph that starts with, "I hope", what do you say about Sun's field-of-use restrictions?

Bob Lee: "Sun puts field-of_use restrictions in the JSE license which inhibit from using the JSE [ on mobile devices ], and "not to mention Android (because that's water under the bridge at this point)."

Bob Lee: There was no TCK available for using on Apache Harmony on Android.

Oracle: "These restrictions prevent Apache Harmony from implementing the JSE [on mobile devices]?

Bob Lee: Yes.

Oracle: Ref TX281- email from you to Hiroshi Lockymer [ sp? ] [ entered into evidence with no objection ]

Judge Alsup: How much longer will you be questioning this witness?

Oracle: About 10 minutes.

Oracle: You worked with Hiroshi Lockymer from 2006?

Bob Lee: Yes.

Oracle: When he became director of Android?

Bob Lee: Yes.

Oracle, quoting from TX281: [ Do you see the part where ] Lockymer says, "I am a little nervous about signing up Noser for further work from a business perspective; those guys (the managers) are pretty shaky."

Bob Lee: Yes.

Oracle: The comments were created in a clean room?

Bob Lee: No.

Oracle: What instructions were given in writing the documentation?

Bob Lee: To paraphrase, in their own words.

Oracle: [Compares taking someone else's work and changing a few words here and there to actually writing in one's own words.]

Google: Objection… notice. The documents were not provided to us with adequate notice. They were provided late last night in two large submissions.

[The witness is dismissed.]

Judge Alsup: What's the issue?

Google: Last night at 10 PM, Oracle identified two very large files of identifying experts and exhibits to use with 3rd party exhibits.

Oracle: Mr. Lee, as the core library lead is...

Judge Alsup: We are going to postpone this witness.

[To Google]: I am not going to tell the plaintiff that they cannot ever bring this up.

[To Oracle]: Have you followed my rules in your handling of the 3rd party exhibits?

Oracle: No, your Honor.

[Oracle has violated Judge Alsup's rules again and has to agree to let the subpoenaed witness step down, to be recalled on Monday, after Google gets a chance to go over the exhibits that Oracle has provided. Oracle has to also provide to Google the exact pieces of the exhibits that they will be using with each witness so that Google can prepare their case adequately. The witness will be recalled for Monday at 7:30AM. This judge is not going to allow a trial-by-ambush.]

Google: Question [to Judge Alsup] About the 3rd party nature of this witness, regarding whether he will actually be available on Monday morning.

Judge Alsup: Do I need to order him to be here?

[Not clear what decision is reached, if any, as to how this will work, but in the end it appears that Mr. Lee will be back on Monday morning.]

9:35AM, Judge Alsup calls a break.

[During the break, the lawyer for Oracle who appeared to be looking at a computer display at the Google table before today's proceedings started is talking with Mr. Jacobs. Mr. Jacobs' smile is still there, but still tight.]

[More smiles at the Oracle table now. Less so for Google.]

10:00 AM.

Witness Dan Morrill

Judge Alsup: Welcome, sir. You okay? [nods]

Oracle (Mr. Norton): You have been with Google since 2006?

Dan Morrill: Yes.

Oracle: Joined the Android team in 2009?

Dan Morrill: Yes, formally. I was working with the Android team informally prior to 2009.

Oracle: And prior to 2007?

Dan Morrill: I was on the Developer Relations team.

Oracle: You became the Technical Program Manager for Android?

Dan Morrill: Yes.

Oracle: What were your primary responsibilities [as Technical Program Manager] ?

Oracle: The Compatibility Definition Document?

Dan Morrill: Yes.

Oracle: Making sure that compatible Android devices can run third party software?

Dan Morrill: I'd say that I defined the criteria under which 3rd party apps would run.

Oracle: Google released the Nexus smartphone, right?

Dan Morrill: That was a co-branding. Google didn't actually make the device. [ some q/a missed ]

Oracle: Familiar with Java class libraries?

Dan Morrill: Yes.

Oracle: You use them [ Java class libraries ], right?

Dan Morrill: Not currently. It has been a while since I have written Java code.

Oracle: Refers to TX 5 [ ? ] This is an API packages listing?

Dan Morrill: Yes, for Android v2.1.

Oracle: On page 4, do you see java.cwp.font, and more packages beginning with java and javax on pages, 5, 6?

Dan Morrill: Yes.

Oracle: All those packages must be implemented on Android devices?

Dan Morrill: Yes.

Oracle: Android ships with implementations of all of these classes using these same names, the same as the Java API?

Dan Morrill: Yes.

Oracle: Not only does the Android API use the same names, it also uses the same form and organization as Java?

Dan Morrill: Yes.

Oracle: (Plays video deposition of Dan Morrill, page 57, lines 1-9, July 12, 2011)

Google: Objection. Mr. Morrill is not an executive or [director ?] at Google.

Oracle: 801 exception.

Judge Alsup: Let me be clear [to the jury , when we say "party admission" under rule 32, there is also an exemption to the hearsay rule. When something comes in as a party admission, it isn't written in stone. It's just a part of the evidence. There is no extra or conclusive value to it. No extra weight is to be given to it [compared to] something that has come into evidence [ via another route ].

Oracle: What you do in your job is to make sure that Android devices are compatible with Android.

Dan Morrill: That seems to be insufficient criteria for what I do.

Oracle: In your own words, then?

Dan Morrill: : [ missed ]

Oracle: A Java app cannot run on Android?

Dan Morrill: Correct, they cannot.

Oracle: Google produces an Android compatibility document that includes binding requirements [ for manufacturers ]?

Dan Morrill: Yes.

Oracle: [Tried to elicit information on how Google keeps OEM's in-line and not changing Android and breaking compatibility.]

Oracle: "Device implementation must not change the APIs for device management, [ missed ], must satisfy requirements in section 3.1?

Dan Morrill: Yes.

Oracle: Page 8, section 3.6, "To ensure 3rd party apps compatibility, device OEM's MUST NOT [ caps emphasized as being in the document ] change package namespaces, including java and javax, and must not change exposed public APIs". Is this correct?

Dan Morrill: Correct.

Oracle: Google has these requirements and [in ?] a compatibility test suite (CTS)?

Dan Morrill: Not all these requirements are testable.

Oracle: Tests for presence of the listed APIs?

Dan Morrill: Yes.

Oracle: Google has something called the "Android Market". What is that?

Dan Morrill: Android Market is a [missed, but to the effect of a program under Google whereby special placement privileges are granted to OEM's whose devices are compatible, or some such.]

Oracle: So in order to request access to the Android Market, a phone must be compatible with Android?

Dan Morrill: We do not have a certification program, but the devices must be compatible.

Oracle: Google prevents incompatible Android phones from accessing the Android Market?

Dan Morrill: I don't know. I am not sure.

Oracle: Motorola makes compatible smartphones?

Dan Morrill: Yes.

Oracle: Samsung?

Dan Morrill: Yes.

Oracle: HTC?

Dan Morrill: Yes.

Oracle: All of these have the Androidd brand?

Dan Morrill: Yes.

Oracle: What is the current number of Android device activations?

Dan Morrill: About 750,000 per day, last time I checked, which was several months ago.

Oracle: Each one has Java APIs?

Dan Morrill: Implementations of those APIs, yes.

Cross-examination of Dan Morrill by Google's Michael Kwan

Google: [Gets introductory information in front of the jury.]

Dan Morrill: BA in physics and CS from Clarkson, MSCS from RPI. Learned to program Java in 1998, self-taught, using online tutorials.

Google: These tutorials included the use of the APIs?

Dan Morrill: Yes.

Google: How long were you a Java programmer?

Dan Morrill: For 12 years.

Google: In that time, did you ever write a Java program that did NOT use the Java APIs?

Dan Morrill: No.

[On to Compatibility Definition Document (CDD)]

Google: Anyone can use the Android code. Why make it comply [ to the CDD ]?

Dan Morrill: To ensure a thriving ecosystem [in which to write and sell apps].

Google: Are there any [major] examples of non-compatible devices?

Dan Morrill: Yes. Amazon Nook and Kindle Fire.

Google: You said that Google doesn't have a certification program. What did you mean?

Dan Morrill: OEM's can make and get phones certified.

Google: Is there any way to determine by running the CTS that OEM's haven't changed the Android code?

Dan Morrill: No.

Google: You said that *you* could tell if they've change the code?

Dan Morrill: If the display looks visibly different, I know the OEM changed the Android code. New hardware (CPU's) can require changes to Android.

Google: Are Android apps written in Java?

Dan Morrill: Yes.

Google: When you said that Android doesn't support Java apps, what did you mean?

Dan Morrill: The Android VM is different than the JVM.

Google: What kind of programming language is used to write Android apps?

Dan Morrill: Java.

Google: Ref TX2301- Do you recognize this?

Dan Morrill: Yes, it's a web page of source code from the Google web site.

Google: (Goes to page 4 and reads title, "Compatibility". What does 'Compatible' Mean?")

Google: Page 5, "Compatibility: Math [ package ] mandatory"

[An explication of what is implied by compatibility ensues.]

Google: How long have you worked on Android?

Dan Morrill: Informally since 2007, formally since 2009.

Google: When was Android announced?

Dan Morrill: [ answered ]

Google: Do you recall reading anything from Sun [ at the time of the Android release ]?

Dan Morrill: Yeah. In fact there was a blog post from their CEO that we thought was pretty remarkable.

Google: You read this in 200? [ year of Android release ]?

Dan Morrill: Yes.

Google: Ref TX2352, the blog post.

Oracle: Objection, 403 and 410(2) [ not sure of 410(2) number ]

Judge Alsup: Overruled.

Google: Who was the author of the blog post?

Dan Morrill: Sun CEO, Jonathan Schwartz.

Google: Please read the first paragraph.

Dan Morrill: [ reads aloud ]

Google: Please read the 4th paragraph.

Dan Morrill: [ reads aloud ]

Google: You said that it was remarkable. Why?

Dan Morrill: I knew of discussions with Sun. The NetBeans [ development environment ] was particularly interesting.

Google: [Tried to get in some things that Oracle objects to… 403.]

[ sidebar ]

[ side note: Mr. Boies is not here today ]

[ sidebar ends. It's 10:45 ]

Google: Referring to the blog post, did you discuss this with other members of the Android team?

Dan Morrill: Yes.

Redirect of Dan Morrill

Oracle: There are approximately 750,000 Android devices activated each day?

Dan Morrill: Yes.

Oracle: That figure doesn't include non-compliant devices like the Kindle Fire?

Dan Morrill: Correct.

Oracle: [Asks a question. Google objects. Judge Alsup asks the witness if he knows the answer to the question, and Dan Morrill says no.]

Judge Alsup: Objection sustained.

Oracle: ref TX2352, is this a license agreement?

Google: Objection, argumentative.

Judge Alsup to Oracle: Arguments are for closing. This is not an argument. Sustained.

Oracle: [Pounces on TX2352.]

Oracle: On May 23, 2008, you were a Google employee?

Dan Morrill: Yes.

Oracle: Working on Android?

Dan Morrill: Yes.

Oracle: ref TX245-- an email that you sent?

Dan Morrill: Yes, the block at the top of the email, in reply to the others.

Oracle: Turn to page 2- a news article that appeared on CNET, May 22, 2006.

Dan Morrill: Yes, that's what it says.

Oracle: On page 4, this is an article about Android?

Dan Morrill: Yes.

Oracle: [Regarding the section about licensing issues, where the article expresses an opinion about Sun expressing concern about Dalvek fragmenting Java.]

Oracle: You forwarded the entire article "to the entire Android PR Group"?

Dan Morrill: Yes.

Oracle: Ref TX2352, dated 5 November 2007, on that day, Google had written most of the Android platform?

Dan Morrill: Not sure.

Oracle: The release of the Android SDK was a week later, right?

Dan Morrill: Yes.

Oracle: The Schwartz blog post was written before the Android SDK was released?

Dan Morrill: Yes.

Oracle: Mr. Kwan showed you an email from 2007, TX 245.

Dan Morrill: "Somebody said that Sun had expressed concern" [ somebody told the the author of the CNET article that Sun was concerned ]. "Sun didn't tell me that they had concerns".

ReCross of Dan Morrill

Google: Ref TX254, what were you forused on?

Dan Morrill: The explicit lines of code [that were claimed to infringe] in the source tree.

Google: On May 23, 2008, what concerns, if any, did you have about Sun's concern about fragmenting Java?

Dan Morrill: None.

Re-redirect of Dan Morrill

Oracle: On page one of TX254, Mr. Rubin got this email?

Dan Morrill: Yes.

Oracle: Mr. Rubin says in the email thread that Dan Morrill was misquoted in the article?

Dan Morrill: Yes.

Oracle: You were quoted in the article on CNET?

Dan Morrill: Yes.

Oracle: You would have read the whole article [ before forwarding it ]?

Google: Objection.

Judge Alsup: Would it have been normal in the course of everyday business for you [Dan Morrill] to have read the whole article before forwarding it?

Dan Morrill: Yes.

Witness dismissed.

Video Deposition

Oracle then played a video deposition of witness Rafael ????. Motorola Mobility Senior Software Director. In the video deposition, the witness has a pronounced accent, he speaks relatively softly, and the microphone is picking up a lot of room echo. In short, it was hard to hear what he said.]

[He has been working on Android since 2009.]

Oracle: What in Android did you work on?

R: Systems software.

Oracle: What did you do to prepare for this deposition?

R: Talked with a Motorola attorney, and [went over materials].

Oracle: Are you familiar with the Android class libraries?

R: Yes.

Oracle: Are the Android class libraries installed on all Android devices ?

R: Yes.

Oracle: What's your understanding of Google's compatibility test (CTS)?

R: You have to meet certain rules so that [Android] apps don't break.

Oracle: Motorola phones pass the Android CTS?

R: Yes.

Oracle: Have there been any failures of Motorola smartphones to pass the Android CTS?

R: Yes.

Oracle: Have all Motorola smartphones provided documentation of APIs per the Google comptibility requirements?

R: Yes.

[ Android version 1.6 specified ]

Oracle: What areas are covered?

R: [ missed item ] Platform API Tests (tests the platform, core libraries, android platform framework). Dalvik VM tests.

Oracle: For Motorola Droid, what version of Android was installed?

R: [ some version specified ]

Oracle: Did you get the Android software from Google?

R: Yes.

Oracle: Did you install it on Motorola smartphones?

R; Yes.

Oracle: Did Motorola make any changes?

R; Yes. We contributed Eclair.

Oracle: Any other changes?

R: No.

Oracle: So the version was entirely the Open Source Software version of Eclair?

R: Yes.

Oracle: All of Motorola's smartphones include all required Android class libraries and APIs?

R: Yes.

Oracle: They include Java and javax APIs?

R: Yes.

End of video deposition playback.

[11:10 -- 15 minute break.]

[11:27 re-convening.]

Oracle: We need to get the video deposition TX number entered into the record.

Judge Alsup: So ordered. It doesn't matter for the jury, just for the court record for appeal.

[ next witness: Leo Cizek ]

Update: Here's more from the day, with witness Leo Cizek on the stand:

Witness Leo Cizek

[The witness is sworn in.]

Oracle: [Introduces witness.] How long have you worked at Oracle?

Leo Cizek: Since 2010, when Oracle bought Sun.

Oracle: How long were you at Sun?

Leo Cizek: Since September 2000.

Oracle: What was your role at Sun?

Leo Cizek: Java Licensing Account Manager.

Oracle: And what is your role at Oracle?

Leo Cizek: Essentially the same.

Oracle: What customers do you work with in your job?

Leo Cizek: Device manufacturers and value-added software companies.

Oracle: Have you ever met Andy Rubin?

Leo Cizek: Yes.

Oracle: When was that?

Leo Cizek: Late fall in 2001.

Oracle: Under what circumstances?

Leo Cizek: Mr. Rubin was the CEO of Danger, Inc. They had implemented Java and needed a license prior to shipping products.

Oracle: [References exhibit TX2016] Who created this document?

Leo Cizek: I did.

Oracle: Why?

Leo Cizek: To summarize my notes over the years.

Oracle: How do you use this document?

Leo Cizek: To remind myself what happened [with various clients].

Google: Objection, relevance and 403b.

Judge Alsup: Overruled. Proceed.

Oracle: When you spoke with Andy Rubin, what was the topic?

Judge Alsup: Needs to be more limited.

Oracle: Why did you call Mr. Rubin in 2001?

Leo Cizek: I called him to tell him that he needed to validate compatibility and also take a Java Commercial License (JCL).

Oracle: Who else [was on the call]?

Leo Cizek: Tim Lindholm.

Oracle: What did Tim Lindholm do?

Leo Cizek: [some answer, missed]

Oracle: What did Danger call their device?

Leo Cizek: HipHop, a.k.a. Sidekick.

Oracle: When was it released?

Leo Cizek: October 2002.

Oracle: Ref TX 2016, bottom of the page…

Oracle: Did Danger have a JCL?

Leo Cizek: No.

Oracle: Had Danger done anything to ensure compatibility?

Leo Cizek: No. They hadn't signed any agreement HipHop?

Leo Cizek: I was disappointed.

Oracle: Did you have any meetings with Danger?

Leo Cizek: Yes.

Oracle: Who was there?

Leo Cizek: Me and Vineet Gupta.

Oracle: And for Danger?

Leo Cizek: Andy Rubin and their in-house counsel.

Oracle: Why did you have the meeting?

Leo Cizek: To let them know that they had to get the TCK and take a JCL. There happened to be a Java CLDC book on the conference table. Vineet Gupta opened it up and showed Andy Rubin and their counsel the copyright notices and the license.

Oracle: Did Danger take a license?

Leo Cizek: Yes, later.

Oracle: At the time that Danger took a license, was Andy Rubin still employed at Danger?

Leo Cizek: Yes.

Google: Objection, hearsay.

Judge Alsup: Overruled.

Oracle: Ref 1026, an internal document of executed Sun Community Source License between Sun and Danger.

Oracle: Move to submit into evidence.

Google: Objection, 403 and relevance.

Judge Alsup: May I see the exhibit? [Reads for a bit.]

Judge Alsup: Possibly counsel is correct. Do not show this exhibit to the jury. It will not yet be entered into evidence. I agree with counsel on 403 objection.

Oracle: To what extent, if any, did the agreement between Sun and Danger address Danger getting access to Sun source code, shown as "original source code" in the agreement?

Leo Cizek: Andy Rubin asked that a section be added to the agreement that stated that Danger had not seen Sun's original code [and did not want access to it]. Section 4, attachment F.

Oracle: To what extent did it address Danger's access to Sun's Java Specification?

Leo Cizek: That's all it granted, given that they didn't want access to it.

Oracle: Ref TX2001, page 2, halfway down, this is an email from Andy Rubin to you?

Leo Cizek: Yes. It was the first email that I had gotten from Andy Rubin since he transitioned to Google on August 4, 2005.

Oracle: Move to submit into evidence.

Judge Alsup: Submitted.

Oracle: Please tell me what it said?

Leo Cizek: It was to me, Tim Lindholm, and Vineet Gupta, and it referenced an internal Google license meeting about the Sun Java Technology for the Wireless Industry (JTWI).

Oracle: Did you meet with Mr. Rubin several times to discuss Java licenses?

Leo Cizek: We met 3 times.

Oracle: Who attended [the first meeting]?

Leo Cizek: Me and Vineet Gupta.

Oracle: And for Google?

Leo Cizek: Tim Lindholm and Andy Rubin.

Oracle: What was discussed regarding Android licensing Java?

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

Oracle: Where was this meeting held?

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

Oracle: Who attended meeting two?

Leo Cizek: Me and Vineet Gupta.

Oracle: And for Google?

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

Oracle: What was discussed?

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

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

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

Oracle: Why was Brenner there?

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

[Missed some, IIRC about royalties aspect of taking a JCL.]

Leo Cizek: The commercial use license specifies royalties.

Oracle: After 2005, were there any further discussions?

Leo Cizek: Yes, in 2009.

Oracle: Who was the Google employee [that initiated the discussion]?

Leo Cizek: Mark Bucholtz.

Oracle: Why were you talking with him?

Leo Cizek: He had contacted a colleague of mine about getting a J2SE license to gain access to some security technology. He was seeking a source code license.

Leo Cizek: [I called him and indicated that] a part would have to be fixed first: the Java license for Android.

Oracle: What did Mr. Bucholtz say?

Leo Cizek: That Google didn't need a license because they'd implemented [Java, or the class libraries] themselves, and that Android was using the Open Source Java specification.

Cross-examination of Leo Cizek, by Christa Anderson for Google

Google: During all your years at Sun and at Oracle, you have been an account manager for licensing?

Leo Cizek: Yes.

Google: For JME and JSE licenses?

Leo Cizek: Yes.

Google: And the JME is for smaller devices?

Leo Cizek: Yes there are 2 configurations of it [the CDC and the CDLC].

Google: One targets feature phones (CDC) and one targets smartphones?

Google: You never had a customer who manufactured smartphones?

[Missed some.]

Leo Cizek: Danger made smartphones.

Google: References video deposition playback [didn't get reference to page or line numbers]

[Playback starts.]

Google: What was the path for promoting the JavaFX platform?

Leo Cizek: Desktop and mobile devices. Virtually every phone that ships on Java is shipping the JME now. "This is a description for the herd…"

Google: TX2001-- email exchange with Vinoy Gupta: This describes Andy Rubin asking to set up a meeting?

Leo Cizek: Yes.

Google: It's an internal response-- Jonathan Schwartz and [ ? missed ] were in discussions about a broader license agreement with Google, regarding a toolbar?

Leo Cizek: Yes.

Google references TX2013, a presentation received from an email from Google. Move to place into evidence.

Judge Alsup: [No objection? None. Accepted.]

Google: This is a "Monetization Proposal" where "Google is seeking partnerships with major manufacturers… and to Open Source their software", right?

Leo Cizek: Yes.

Google: References TX2002 [not sure if this number is correct] -- an email exchange with Mr. Gupta: In this email, Mr. Gupta said that he wanted to change the plan that Andy Rubin had to open source their software. "I want to change this to *my* idea, to have Google ship their apps on J2ME." [Correct]?

Leo Cizek: Yes.

Google: Regarding the three meetings, you don't remember much [about the details], do you?

[No response.]

Google: Isn't it true thet] Andy Rubin said that Google had not yet made a decision regarding which technology to use?

Leo Cizek: Yes. Either CDC or CDLC.

Google: Plus other options?

Leo Cizek: Not mentioned to me.

Google: References TX2004, an internal Sun email that lays out action items after the second meeting.

Google: [Gets 3 items shown to jury]:

1) open source licensing model
2) CLDC-High applicability (vs. G internal implementation)
3) [ missed ]

Google: References TX2006 email including recipients Leo Cizek, Andy Rubin, Vineet Gupta, Ms. Garcia, Cole.

Google: Shows the following:

1st meeting is to talk about the OS
2nd meeting is to talk with Google about the benefits of CDLC

Google: ref. TX2008, a contact report prepared by Leo Cizek at Oracle.

Google: This report relates specifically to Google?

Leo Cizek: Yes.

Google: To Andy Rubin and Android?

Leo Cizek: Yes.

Is there a notation by date?

Leo Cizek: Yes.

Google: 5-25-2006 entry regarding Google communication about Android, would you read that?

Leo Cizek: "After many meetings, including Andy Brenner, we are unable to have a meeting of minds over CLDC-HI"

Google: ref TX2009, an email dated March 2007, Leo Cizek to Gupta.

Google: The email is several pages, based on an article about Google's intent to build a mobile phone, correct?

Leo Cizek: Yes.

Google: You had a discussion with Mr. Gupta about this?

Leo Cizek: Yes.

Google: You wanted to call Google and talk with them, yes?

Leo Cizek: Yes.

Google: And Mr. Gupta asked you to hold off?

Leo Cizek: Yes.

Google: He said that waiting will lead to a larger S&I interest?

Leo Cizek: Yes.

Google: And you followed Mr. Gupta's instructions?

Leo Cizek: Yes.

Google: ref TX2010, an email exchange amongst M. Purcey, Leo Cizek and Mr. Gupta, dated April 2, 2007, a continuation of TX2009.

Google: In this email you followed up with Mr. Gupta, asking if he could contact Google?

Leo Cizek: Yes.

Google: And Mr. Gupta told you to continue waiting?

Leo Cizek: Yes.

Google: You said that you'd hoped to discuss with Andy Rubin about the advantage of going with a Sun Commercial License (JCL), versus going the Open Source software route?

Leo Cizek: Yes.

Google: He was waiting for a "green light" from Rich Green?

Leo Cizek: Yes.

Google: You negotiated with Google the Sun StarOffice deal?

Leo Cizek: Yes.

Google: In fact the biggest deal that you'd ever negotiated with Google over the Google toolbar?

Leo Cizek: Yes.

Google: [What was the nature of that deal from the user's perspective?]

Leo Cizek: When customers downloaded a free version of Java, they'd also get the Google toolbar.

Google: Ref TX2019-- email to M. Lehrbaum [ sp? ], former employee of Sun/Oracle in the PSG Marketing group, from Leo Cizek, in 2009.

Google: At the same time that you were [IIRC, marketing JME to Google], you were marketing other products as well?

Leo Cizek: Yes.

Google: You sent this email because of the upcoming conference call with Google?

Leo Cizek: Yes.

Google: You were seeking to generate revenues for Sun that had nothing to do with JME?

Leo Cizek: Yes.

Google: Ref TX2021, email to Leo Cizek [incl. M. Sandman, Tom Harris ( a Sales Engineer for Sun in the Java Sales Engineering group) ], dated November 2007, regarding APIs.

Google: Page 2 "proposes some ideas that may work."

[ missed the specifics ]

Google: You were the account representative for Google?

Leo Cizek: Yes.

Google: But you represented Sun for other product licensing than JME?

Leo Cizek: Yes.

Google: ref TX2026, an email [ from? ] M. Lehrbaum, in May 2008.

Google: Mr. Lehrbaum says, "G's Android is a potentially dangerous competitor"?

Leo Cizek: Yes.

Google: Each of these conversations [from the TX's that have been entered into evidence] were amongst business people, not lawyers?

Leo Cizek: Correct.

Google: And there were no discussions of copyrights?

Leo Cizek: No.

Redirect of Witness Leo Cizek

Oracle: Back in 2005, in TX2004, paragraph 3, Vineet Gupta's email…

Leo Cizek: I dropped some of the action items.

Oracle ref TX2021- do you see the reference to "open social APIs"?

Leo Cizek: Yes.

Oracle: Do you understand that that had anything to do with Android?

Leo Cizek: It was totally unrelated.

Oracle: Regarding the "Ideas that may work", [ specifically ] "(once we understand the Android distribution (the SDK)…"

Google: Objection.

[Back and forth with Judge Alsup.]

Judge Alsup: Objection sustained.

Oracle: Did you ever tell Mr. Rubin that he could do an independent implementation and *not* get a JCL?

Leo Cizek: No.

Oracle: No further questions, your Honor.

The witness steps down.

Judge Alsup to the jury: I am going to give you a few minutes off.

Reminds jury not to do any research on the case, companies, or lawyers, plus no news, no talking with loved ones about the case.

Jury dismissed 12:43PM.

Judge Alsup: Summary of times used [in minutes]:
Oracle: 560 Google: 329

Judge Alsup: A total of 17 hours was allocated to each side. I don't have a fancy computer, all I have is this [ holds up a single piece of paper with columns visible and a yellow PostIt note sticking to it]. So if you think my numbers are off, by all means, please let me know.

Judge Alsup: [Admonishes Oracle to save some time for their cross-examinations.] Judge Alsup: "I question what that last witness added to this case."

Judge Alsup: "There will be no extensions."

Update 2:

And here's the final part of the day, after the jury has left for the day and the week-end, and the parties' lawyers are discussing issues with the judge:

Oracle v. Google, Day 5, part 6, (final)

[Jury has been dismissed.]

Judge Alsup: I have a question for both parties. I am confused about things… [it has been said that]

1) The source code was built in a cleanroom
2) It was built by Apache

Judge Alsup: Which one is it?

Google: Either. [Part of it was] was written from scratch and [part of it came] from Open Source (Apache).

Judge Alsup: Has anyone vetted that Apache (Harmony) was independently implemented?

Google: Yes, except for the 12 files.

Judge Alsup: Is this true in terms of direct copying, excluding the 37 packages and the SSO?

Oracle: What we have accused of copying and is in the list that you have. In addition, the documentation will be shown to have been substantially copied. Further, we will show that they (Google) really didn't implement a cleanroom in their choice of the people they chose to put in the cleanroom.

Judge Alsup: What about the cleanroom?

Judge Alsup: What if Apache did wrong?

Judge Alsup: What good does it do to have a cleanroom?

Google: Apache (Harmony) has been shown to be a completely independent implementation. The source code was copied by neither Google nor Apache.

Google: Google licensed the Apache code under the Apache License.

Google: Sun knew that Apache (Harmony) was (freely) distributed for 7 years [and did nothing to stop it].

Google: "The issues over APIs is just something that Oracle has made up after the fact."

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

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

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

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

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

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

Judge Alsup: What is the comparison made against?

Judge Alsup: What is the copyrighted work?

Oracle: There was a cloud over Apache (Harmony). Google did step into their (Apache's) shoes [when they chose to license Apache Harmony].

Judge Alsup: Was the cloud the 37 API's, or was it the trademarked name [Java]?

Oracle: Very much the former. "We recognize that the specifications are copyrighted."

Oracle: Mr. Rubin said that he knew the APIs are copyrighted.

Judge Alsup: Was the fight that you cannot use Sun's source or Sun's API definitions?

Oracle: The second. "Enter Oracle, exercising more energetic leadership."

Google: Oracle is trying to shut down the JCP. Now that they are the owner, they are trying to shut down the Open Source.

Google: Our main point about Apache is that it gives us [ missed ]

Google: Once it was clear to Sun that Google was going OSS, Sun moved on and tried to sell Google other things.

Judge Alsup: About next week, who is your [Oracle's] last witness?

Oracle: Dr. Mitchell.

Judge Alsup: How much time will you need for him?

Oracle: about 1.5 hours.

Judge Alsup to Google: [Similar inquiry regarding time usage.]

Judge Alsup: [Reminded by Google counsel] We will start with Bob Lee on Monday morning.

Judge Alsup: Some final thoughts: About 2/3 of the case could be stipulated with 1/3 in controversy on the copyright part.

Judge Alsup: Think of ways to help the jury understand specifically what is controverted."

Judge Alsup: "It is one kettle of fish to say you need a license to use their source code; it's another when you say you need a license regardless of whether you've done an independent implementation".

Judge Alsup: "Go home, or back to their hotel rooms and have a double." [Hilarity ensues… the double becomes a double of single malt scotch, and someone suggested that it would be nice if the Judge *ordered* them to do his... and as the good judge stepped through the door to his chambers, he said, "so ordered".]

[One of Google's main points today was that Sun/Oracle was trying to force Google to buy something that Google clearly did not want or believe that they needed. It seems clear to me that Oracle is going to spend a lot of effort trying to prove that Google extensively copied their documentation; that's about all they seem to have a chance at, at this point.]

Update 5: 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.