Judge In Oracle-Google Case Given Crash Course in Java

itwbennett writes "Lawyers for Oracle and Google gave Judge William Alsup of the U.S. District Court in San Francisco an overview of Java and why it was invented, and an explanation of terms such as bytecode, compiler, class library and machine-readable code. The tutorial was to prepare him for a claim construction conference in two weeks, where he'll have to sort out disputes between the two sides about how language in Oracle's Java patents should be interpreted. At one point an attorney for Google, Scott Weingaertner, described how a typical computer is made up of applications, an OS and the hardware underneath. 'I understand that much,' Alsup said, asking him to move on. But he had to ask several questions to grasp some aspects of Java, including the concept of Java class libraries. 'Coming into today's hearing, I couldn't understand what was meant by a class,' he admitted."

Comments

Why not ?

[zero.kalvin]

I won't go to a Baseball match and be the whatever is the equivalent of a referee is! Judges need to be informed about the subject they are going to judge(?) on.

Re:Why not ?

[Mouldy]

You would need to be some sort of mechanical or engineering background information, however, to determine if party B's mechanical system is different enough from party A's mechanical system to rule on some sort of patent issue between the 2 parties. To one person, an engine is an engine. They don't care if it's petrol, diesel or rocket fuel. It's the loud bit that makes vehicles move.

Someone determining whether or not the latest engine by VW (for example) is too similar to an engine previously made by Alfa Romeo is going to need to know more than "an engine is the loud bit that makes vehicles move" - somebody is going to have to explain the details behind the mechanics. Based on this information, they might decide that the engine as a whole isn't a rip-off, but maybe the odd way the valves are set up is suspiciously similar. It would be unfair to make a judgement without the extra technical knowledge.

Re:Why not ?

[WrongSizeGlass]

I won't go to a Baseball match and be the whatever is the equivalent of a referee is! Judges need to be informed about the subject they are going to judge(?) on.

Exactly. Shouldn't most of us - those of us who want to see some sort of software patent reform, limitations, or even outright abolishment - be lauding the public display of a judge who wants and needs to learn about technology in order to address the software patent debacle? Just because he's already experienced in patent law doesn't mean he knows enough about the subject matter before his court.

Educating someone about the subject they are deciding, administering, governing, or even adjudicating over can't be a bad thing.

Re:Why not ?

[WrongSizeGlass]

Don't you also have special judges for traffic related incidents?

Yes, they have a special title: Punished

Re:Why not ?

[zevans]

You don't have to be a mechanic or even a licensed driver to detect when someone is speeding.

You don't. Until it's a sufficiently complicated case, and then, you do.

We had a case in the UK a couple of years ago where an econobox type small car was "detected" speeding at 90-some mph. The guy got an expert witness involved to prove that the car was simply not capable of the claimed speed in the prevailing conditions. The expert largely spent his time explaining advanced driving and mechanics to the magistrates.

Similarly with intellectual property... if you find a cut-and-paste into a document, complete with Mountweazles, then it's pretty black and white.

If on the other hand you are arguing about what's generic software and what is an interpretation and what is an algorithm, then, well, good luck with that; but you'll certainly need to understand the argot, and that is not standard International English by any means.

Re:Why not ?

[GooberToo]

Please flip a coin. Many people who professionally deal with technology fail to grasp many important aspects.

Re:Why not ?

[DarkOx]

Judges are not supposed to be specialists they should be generalists, because they need to understand the effects of the decisions they make in a broad societal context. What they need to be is REALLY SMART. We as a society need to do what we can to put some of the best and brightest on the bench.

This judge is a great example of how to do it right! He is not technical expert he is a legal expert and he smart enough to:

A) Know what he does not know
B) Recognize that he needs to be educated about what he does not know to make a good ruling
C) Learn and understand new possibly foreign concepts quickly enough to keep the legal system at least sort of efficient.

I applaud this fully!

Re:Why not ?

[ObsessiveMathsFreak]

It depends on who's doing the educating. The patent industry has done a fine job in "educating" Judges in the utility and indeed necessity of modern patent law as it is practised. Do you really want the software industry to start indoctrinating judges as well.

If the judge is unable to rule on the case, they should simply say so. If the government cannot find a judge to hear the case, then it is an overwhelmingly technical matter, not a legal one, and should be thrown out of the courts. If the government still needs to hear these cases, then they must either establish specialised courts or find some other method of dealing with them.

Far too much faith is placed in the court system by far too many. It is not the effective, impartial font of justice that many fantasise it to be. It's a creaking, rusty relic of a medieval property arbitration for the wealthy and powerful, with a criminal punishment system tacked on to protect that same property. It is failing miserably to deal with modern society, and sooner or later we're going to end up paying for it. The rise of private arbitration and legal fraud by major corporations is a symptom of how discredited rule of law by the court has become.

Re:Why not ?

[OzPeter]

Don't you also have special judges for traffic related incidents?

I got a speeding ticket a few years ago. I read up on all the relevant laws and knew that while I couldn't get off, I could get the fine reduced and also not get points if the judge allowed me to do a driver ed course (online even!). I also knew that what I was asking for was quite reasonable and done all the time. So I went to court and discovered that the normal traffic judge was out for the day and there was another judge in his place. As I was early for my case, I sat there and listened to all the people ahead of me asking for the same thing that I had planned for. Yet each time the stand-in judge kept saying "I can't do that". I felt like jumping up and yelling at the judge "damn well you can - you're the judge".

The state I am in also says that there is no justification for speeding of any sort, yet this judge let off a guy who said he was speeding to get away from a truck - even though the cop who booked him said that he did not see any such truck

So the answer to your question is that (from my 1 point datum) that not only don't you have special judges, judges don't have to know the law of what they are judging

Humility is great...

[bjourne]

But when you have no clue about what software development is, then how on earth would you be able to judge trials about it fairly? Are the lawmakers that arrogant that they think they can understand the basics of software development in about one sitting, when an engineer have to study the subject for several years? Why not turn the tables around. Let's take a prominent developer with a nice career. Have this judge and some of his fellows give him a a one day crasch course on business law, what to do when a lawyer says "objection" and so on. I'll bet my ass we would get much more reasonable and logical judgements that way.

Re:Humility is great...

[Anonymous Coward]

...Let's take a prominent developer with a nice career. Have this judge and some of his fellows give him a a one day crasch course on business law, what to do when a lawyer says "objection" and so on. I'll bet my ass we would get much more reasonable and logical judgements that way.

this happens everyday, its called a jury.

Re:Humility is great...

[hey!]

Oh, c'mon. The issues to be decided here aren't any more abstruse than in any other kind of engineering case. Nor are the *vast* majority software developers any more competent to make technical judgments than the judge would be after a few days prep time. Sure, developers more apt to have some idea of what a virtual machine is beforehand, but that idea is to vague to be of much use. How many could explain the difference between a stack machine and a register machine? Or give any kind of explanation of how a compiler optimizer does what it is supposed to? Not many. If you narrowed the pool of "judges" down to people who understood the technology involved to make *technical* judgments in this case, the pool would be tiny and probably consist of people who had a stake in the outcome of the case.

Fortunately, that's not what we need the judge to do. He doesn't have to make decisions about the applications of technology; he has to make decisions about the application of *law*. And it's way better this way. You get the software experts up to make their arguments, and see if they can convince somebody who doesn't have a pre-formed opinion. If one side doesn't have a leg to stand on, it'll show. If both sides have arguments that would sound reasonable to another expert, you want the judge to apply the law without ruling on which side of the dubious question is right. An expert in virtual machine technology couldn't do that impartially. If we followed *your* suggestion, we'd have courts making engineering decisions. What the court should do is rely upon engineering consensus where it exists, and not interfere with the development of that consensus where it doesn't yet.

The drawback of this system is that it depends on having a judge who is good at grasping the essentials of what is at stake. But software developers are ordinary mortals, and a little humility would do us a lot of good. True, not everyone can *do* what we do, but unlike particle physicists or poets we make our living doing things that don't take uncommon intellectual talent to grasp.

Of course, judges *do* often get things horribly wrong, but not more often than software developers get things horribly wrong. It's just the nature of the law that its failures have a wider impact. When you fail as a software developer, somebody who decided to rely upon you is disappointed. When you fail as a judge, people who had no involvement or choice at all are swept into that failure. I think where we've seen horribly misguided legal rulings on technical matters, its because the judge latched onto the kind of specious analogy that politicians often deal with when they talk technology. Even when a politician's intentions are good and the results of using the analogy positive ("information superhighway"), that's not a precise enough understanding to make legal rulings. This guy is doing the right thing and going to school, not exaggerating his prior understanding.

Think yourselves lucky

[Anonymous Coward]

We brits have judges who had to ask what a website is (in 2007!)
http://www.metro.co.uk/news/49376-judge-asks-what-is-a-website

kinda reminds me of the classic 'Not The Nine O'Clock News Sketch'

Counsel: This receipt is for the digital watch...
Judge: ...a digital watch? What on earth is a "digital watch"?
Counsel: Sorry m'lud. A digital watch is a watch worked by microelectronics.
Judge: Oh! How fascinating. Proceed.
Counsel: The next receipt is for an automatic video recorder...
Judge: ..."automatic video recorder"?
Counsel: Yes, I'm sorry m'lud. It's a machine that records television programmes on special tape.
Judge: Oh, how fascinating. What will they think of next? Proceed.
Counsel: Thank you m'lud. And finally, a receipt for a "deluxe model inflatable woman", whatever that is.
Judge: The Deluxe is the one with the real hair...

http://www.youtube.com/results?search_query=not+the+nine+o%27clock+news&aq=0

been before alsup before

[Anonymous Coward]

Judge Alsup is a very interesting Judge. He is a complete hardass who doesn't take any crappy and wants things done exactly his way. Often he does this beyond what is probably fair, but he will hold the parties feet to the fire on what they do. He has had software cases before him before, and he definitely has the intellect to be able to handle this one. People who make off the cuff commentator about how dumb lawyers and/or judges are have never work with or against them in the context of high stakes litigation. Tutorials such as this one are more the norm than the exception in most patent litigation.

Good lawyers are fast learners

[AlecC]

Some years ago I read the final judgement in a high-tech lawsuit involving detailed understanding of the internal function of disk drives (at that time - technology had moved on). I was impressed with the level of understanding achieved by the judge in the case. The explanation of the facts of the case in the judgment amounted to a fairly good tutorial in the internals of disk drives.

In another case I knew about, the two parties jointly hired a third-part consultant to write a tutorial for the judge on the underlying technology - the parts that both parties agreed on.

I cannot obviously speak for the case in question, but my experience is that judges and trial lawyers (barristers in the UK) are pretty savvy people. The are basically trained to go from 0-60 on a new technology within a few weeks. They may not be able to be creative in the technology, but they know enough to know when they do not know enough and ask further questions.

And that's the problem with IP law...

[itsdapead]

But when you have no clue about what software development is, then how on earth would you be able to judge trials about it fairly?

That is the problem with IP laws on technology: there are very, very few people with the combination of legal knowledge and technical knowledge needed to enforce them sensibly (and, in the case of patents, a lack of Renaissance Men with the required level of omniscience to make the hair-thin but crucial judgements about obviousness etc.*). Solution: don't pass laws that are impossible to enforce fairly (don't hold your breath).

(Does anybody know if Einstein was any good as a patent officer? "Sorry sir, what was that about claim 137b? I was daydreaming about riding on a beam of light...")

Great analogy, but ...

[brokeninside]

That's got the beginnings of a great analogy, but the implentation fails.

Traditional compilation: the grocery list is sent to a translater, who translates the whole list at once. This list is then given to a shopper who can read it.

JIT: An interpreter goes with the shopper and translates one item at a time as the shopper reads the list, but the translation for each item is written on a new list so that once it's translated, the shopper doesn't have to ask for a translation for that item.

Traditional interpreter: An interpreter goes with the shopper and every time the shopper looks at the list, the interpreter has to translate regardless of how many times a given item occurs on the list.

While many here pretend to be lawyers....

[gatkinso]

...he doesn't pretend to be a developer.

Awesome Judge!

Re:I'm glad, honestly.

[malkavian]

How is a software engineer, in a few weeks, going to learn enough about a particular field to go and be able to obtain a spec document from a client? Answer, they don't, they just learn enough to be able to communicate with the experts, and translate their knowledge into relevance for the task at hand.

Really, the Judge in question seems to be entirely on the ball. He doesn't need to know how to apply the relevant parts in a program of his own; he just needs to understand generally what the basic principle is. You know, the kind of thing that got explained in a couple of hours of a lecture at the beginning of a degree. A bit above layman, but nowhere near enough to be a full on practitioner. When you know enough of what the base principles are, you get the ability to make sensible questions on the deeper detail.
All the judge needs to know is how this language detail translates into the field of law,

Re:Oracle made a big mistake

[hairyfeet]

You know, despite all the "Oracle boo, Google yay" fanboyism we see here at /. I still haven't heard anybody give a reasonable answer as to why this isn't the exact same as the MS Java mess which everyone was cheering when Sun shut it down?

The point to having licensing with regards to Java at Sun was to keep third parties from fracturing Java and it sounds like this is what Davilk or however you spell it is doing. If MS renamed MS Java to MS Coffee would that have made it okay? After all it isn't Java, it just runs Java code! To me the whole thing smells like fanoyism and bullshit. But to me hypocrisy is hypocrisy, and if it was bad for MS to fracture Java then so too is it bad for Google.

And call me weird, but I still don't get why FOSS fans run to the defense of Google. They aren't letting you have the Android 3.0 source, they've historically kept the best bits like their file system (built on FOSS) to themselves to give them an advantage, and they've purposely gone out of their way to disallow any GPL V3 which means Android is most likely gonna end up TiVo'd, yet the FOSS fans run to their rescue? I just don't get it.

Re:I'm glad, honestly.

[WrongSizeGlass]

How is the judge going to learn enough about Java, in a way that is unbiased by the lawyers for each side, in a few weeks, to make a sensible decision? Doesn't he need to know a fair bit about how Java code is used, who uses it, what for, etc?

He needs to understand the concepts of Java, not all the implementation details and not how to code/debug/code/scream/debug/sigh/debug/relief java. And he doesn't need to be able to read Java or its syntax.

A quick overview of Java features that are not unique to Java will also help eliminate a lot of the clutter. He needs to understand using or extending classes (ie, the Java class libraries) but not have to worry about all the possibilities of public/private properties & methods, inheritance, polymorphism or other common OOP mechanisms.

Re:Oracle made a big mistake

[TheSunborn]

The difference is that Microsoft used and modified Suns software and source-code under a specific contract from Sun. Google don't use any code from Sun.

And I did really not cheer when Sun shut it down. What Microsoft did was to identify a real world problem where Java really missed some features(And java still miss this. Writing gui code in java is still painfull due to missing delegates/function pointers).

What sun should have done was to realize that Microsoft did find a giant feature hole in java. So sun should have changed java to add the features which Microsoft needed (And the rest of the gui developing world) missed.

Re:Oracle made a big mistake

[DrXym]

You know, despite all the "Oracle boo, Google yay" fanboyism we see here at /. I still haven't heard anybody give a reasonable answer as to why this isn't the exact same as the MS Java mess which everyone was cheering when Sun shut it down?

Because Google never claimed Android or Dalvik was compatible with Java, and hasn't used the term Java except in context of the language both VMs share. Indeed they take great pains in technical documentation to explain how Dalvik is not the same as Java and incompatible.

Microsoft basically tried to co-opt the Java brand, the trademark & logo, extended the system in some ways (delegates, CAB files) and omitting other parts (JNI, Jar files etc.) and palmed it off as Java / J++ even though it was not compatible.

Re:Oracle made a big mistake

[pem]

why this isn't the exact same as the MS Java mess which everyone was cheering when Sun shut it down?

From a legal perspective, Java wasn't open-sourced back in the day. Microsoft had a copy of the source and a contract, and violated the contract, by changing the source to make it incompatible and still calling it java. Google has written a VM from the ground up which is different than what MS did. Sort of like Tivoization -- RMS didn't like it, but his license didn't allow it.

From a political perspective, what google's doing is about making the software work well on the platform, not about introducing deliberate incompatibilities.

And call me weird, but I still don't get why FOSS fans run to the defense of Google. They aren't letting you have the Android 3.0 source, they've historically kept the best bits like their file system (built on FOSS) to themselves to give them an advantage, and they've purposely gone out of their way to disallow any GPL V3 which means Android is most likely gonna end up TiVo'd, yet the FOSS fans run to their rescue? I just don't get it.

Because they have let us have the source to the bestselling smartphone platform ever, and we expect they will hand over Honeycomb source when they're happy with it (just as they did with the previous versions), and because the "O" is FOSS encompasses a huge community, some members of which think that if you want to share your source, that's awesome, but if you want to keep something locked up, that's OK too, and that RMS is a dickhead who seeks to divide the community while pretending to unite it.

Swallowing the claim that the open source world is worse off because of google's actions requires a significant helping of RMS's kool-aide. Making the claim can be done either with RMS's kool-aide, or with astroturf money from Microsoft.

Re:Oracle made a big mistake

[walshy007]

If MS renamed MS Java to MS Coffee would that have made it okay?

Of course it would be okay, OSS advocates love forks so long as a spade keeps getting called a spade and so long as when something claims to be standards compliant to something it is. and how is that situation different from c#? java and c# share many similarities and most of the changes seem to be for the sake of change, because it made no illusion to being java at all nobody kicked up a stink.

People defend google because in comparison to most other companies they are pretty good when it comes to being mindful of the way they act with people.

Not everything has to be open source, what a lot of people here like is the ability to tinker and create your own things and being able to do what you wish with the things you make.

If I wrote from scratch something like android I wouldn't mind the ability to do with it as I please, just as google is. What people don't like are those that sue with patents and the like to stop people having the freedom to think up their own devices and applications.

The OSS advocates that think all source code should be open are a minority, the majority are simply pragmatic and realize the benefits of an open source development model and of using software that uses that model.

The slashdot crowd only have a few common things, but I would think one of them would be wanting freedom to do things as they wish, if they want to write closed software, they are entitled to.

Re:Oracle made a big mistake

[GooberToo]

why this isn't the exact same as the MS Java mess which everyone was cheering when Sun shut it down?

I'm sure you're get many an answer but here's the short answer.

Microsoft purposely failed to maintain currency as well as created incompatibilities. This had the effect of giving people a very bad taste in their mouth with Java since they were providing the defacto implementation on Windows; since it was bundled. This had the effect of damaging the Java brand; which was entirely Microsoft's intention, so as to push their own technology. It was part of their classic embrace and extend strategy and was extremely anti-competitive.

Google, on the other hand, is using the official java compiler and tools. Oracle's Java compiler then creates JVM bytecode. Google translates that JVM bytecode into Dalvik bytecode and packages it up into an application (apk). As such, its technically correct to say, Android programs are created using the Java language syntax and a subset of the Java framework. Their framework comes from freely available sources and is compiled using official, freely available, Java technology.

The differences are profound.

Basically it boils down to Oracle being pissed Google did an end-run around Oracle's Java ME implementation and licensing requirements by creating their own Java ME VM-like environment (Dalvik VM). Oracle simply doesn't have a case. At least not that I've seen so far.

Educated judges is a good thing

[realsilly]

With technology moving so fast, even developers have to read up on new technology to keep up with it; so it makes sense to me that we have judges who are facing these technological questions to be at least a little be knowledgeable about the subject matter before their court.

I would much prefer a judge to be knowledgeable about a subject.

This makes me wonder, should we now have judges who must have a background in the subject matter to be the only judge who would ever hear cases that pertain to that subject matter? How bad would this be?

Re:Lawyers?

[cerberusss]

Lawyers for Oracle and Google gave Judge William Alsup of the U.S. District Court in San Francisco an overview of Java [...].

Thinking about the tech knowledge of the lawyers I know, I am not sure that the Judge is a lot wiser now!

They could do worse! My wife knows everything better!

The judge has to ask the question for a reason

[Kupfernigk]

He is not allowed to use his personal knowledge as a matter of fact, and so when the trial may create case law he must ask these questions, so that the case record will explain something that may not be understood in future. Since "website" is not part of standard English, the question has to be asked. However, these apparently stupid questions only have to be asked when some facet of the case actually depends on the answer. Essentially, the judge has to elicit a citation.

It's also very good (and it happens) when a judge asks an apparently stupid question and it turns out that nobody can answer it.