Oracle V. Google Being Decided By Clueless Judge and Jury (vice.com)

theodp writes: The problem with Oracle v. Google," explains Motherboard's Sarah Jeong, "is that everyone actually affected by the case knows what an API is, but the whole affair is being decided by people who don't, from the normals in the jury box to the normals at the Supreme Court." Which has Google's witnesses "really, really worried that the jury does not understand nerd shit." Jeong writes, "Eric Schmidt sought to describe APIs and languages using power plugs as an analogy. Jonathan Schwartz tried his hand at explaining with 'breakfast menus,' only to have Judge William Alsup respond witheringly, 'I don't know what the witness just said. The thing about the breakfast menu makes no sense.'

"Schwartz's second attempt at the breakfast menu analogy went much better, as he explained that although two different restaurants could have hamburgers on the menu, the actual hamburgers themselves were different -- the terms on the menu were an API, and the hamburgers were implementations." And Schwarz's explanation that the acronym GNU stands for 'GNU is Not Unix' drew the following exchange: "The G part stands for GNU?" Alsup asked in disbelief. "Yes," said Schwartz on the stand. "That doesn't make any sense," said the 71-year-old Clinton appointee.

Comments

"The G part stands for GNU?"

[Anonymous Coward]

Please let them bring up PHP for some reason...

Re:"The G part stands for GNU?"

[Plus1Entropy]

Honestly, I think we made our beds, and now we're being forced to sleep in them.

Re:"The G part stands for GNU?"

[hcs_$reboot]

I suspect a bit of snobbishness in telling what the acronym GNU stands for ( "GNU is Not Unix" ).

Re:"The G part stands for GNU?"

[ShanghaiBill]

I suspect a bit of snobbishness in telling what the acronym GNU stands for ( "GNU is Not Unix" ).

Many non-nerds don't understand that nerds are often whimsical about things that normal people would take more seriously. Like naming a project that encompasses the life work of many people. Non-nerds not only fail to understand our technology, but they also fail to grok our culture.

Re:"The G part stands for GNU?"

[Beeftopia]

Recursion is not a simple concept to someone who has never heard "recursive case" and "base case" and "function". It's not a simple concept even to those who have. So, no snobbishness I think.

They're like engineering jokes: "What do you get when you cross an elephant and a grape? Elephant grape sin theta." No one who doesn't know what a cross product is, is going to get that.

This speaks to the weakness of the jury system. It worked well enough in agrarian times for simple concepts. But this is not a jury of (programming) peers. So they're not being judged by jury of their peers.

Re:"The G part stands for GNU?"

[GPS Pilot]

Recursive acronyms can be fun and clever. Not everyone immediately groks the concept, but that's no reason to beat ourselves up.

Re:"The G part stands for GNU?"

[Austerity Empowers]

Honestly, I think we made our beds, and now we're being forced to sleep in them.

Amen. Although I question what the value of elaborating G.N.U. adds to the conversation, particularly given that it is idiotic and essentially meaningless. The reason there is so much whimsy in the computer geek world is that many of us have a heightened distaste for social hang-ups that serve no useful function. If a noun is required to identify something, any one will do, provided it doesn't violate trademark . In this case, that noun is GNU, and if asked to elaborate the correct answer "It is just a name, what does Alsup mean?"

Re:"The G part stands for GNU?"

[ItsJustAPseudonym]

"...that's no reason to beat ourselves up."

We can tell ourselves that, after the judge has become irritated and secretly prejudicial, due to a sarcastically-recursive acronym. It's a sad state of the courts, for sure, but it's also a case of nerd isolation.

Re:"The G part stands for GNU?"

[Plus1Entropy]

Also, TFA talks about this sound-byte quote, but what bearing is it really going to have on the case? What was said afterward? Did the judge simply shrug it off as a stupid joke and move on? My guess is probably, but of course that doesn't fit the narrative TFA is trying to get across.

Re: "The G part stands for GNU?"

[Anonymous Coward]

I bet they would prefer Gnuy McGnuface.

Re:"The G part stands for GNU?"

[ShanghaiBill]

They're like engineering jokes: "What do you get when you cross an elephant and a grape?

What do you get when you cross a tsetse fly with a mountain climber?

Nothing. You can't cross a vector with a scaler.

Re: "The G part stands for GNU?"

[Anonymous Coward]

Explaining this to the judge and jury would surely win you extra points in court. :-D

Re:"The G part stands for GNU?"

[jellomizer]

To be fair, this isn't isolated to just technology. Nearly every industry has its details that "normals" outside of that field just have no idea, and are considered idiots for not knowing what that is. That is the thing about specializing in a field, you know stuff that the non-specialist in that area don't know.

But I do believe that in the tech industry we had reveled a but too much in that idea of our superiority and sometimes try hard to make things more complex to outsiders than they have to be. And now it can hurt us.

Normally the idea that non-specials in the area for a court hearing is a good thing, because a specialist would be fixed on their point of view, so they would already be pro-Google or pro-Oracle, without the rest of the facts and not with the idea of justice, but with pushing their agenda.

Re:"The G part stands for GNU?"

[yithar7153]

Yeah, we should have added a base case somewhere. I'm pretty sure Alsup is looking at from the stance of Mathematical proof of Induction, and that's why he said "That doesn't make any sense."

Re:"The G part stands for GNU?"

[K. S. Kyosuke]

Nearly every industry has its details that "normals" outside of that field just have no idea, and are considered idiots for not knowing what that is.

The problem is, not many fields are based on specialized mathematics, and programming is specialized mathematics, and most people are probably worse at abstraction than at memorizing random tidbits (including many programmers (!)). So the idea of a normal-order-evaluated self-referential structure is probably more alien to the average person than most, if not all things from your randomly picked industry.

Re:"The G part stands for GNU?"

[serviscope_minor]

Because the judge didn't say it's a stupid joke he said he didn't understand it. Big difference there. In the first case, the judge was ignorant of programming so taught himself before the case started. This case follows the proud British legal tradition (which you inherited) of proudly ignorant old dinosaurs ruling on stuff they consider it to be beneath them to understand.

Re:"The G part stands for GNU?"

[rtb61]

All they need is a working class geek to do the interpretations for the court. Using an intellectual from more affluent circumstance will produce problems. So the proper definition for an application programming interface for oldies, especially Luddite is, "it's morse code". The interface being the commonly accepted pattern of dots and dashes to communicate information, but does not include the information to be transmitted, which is protected and copyright, where as the agreed pattern of dots and dashes is just an agreed pattern and the nature of the pattern is completely arbitrary as long (which patterns of dots and dashes represent what information), as long as there is agreement between sender and receiver with regards to the arrangements of dots and dashes are required to represent components of that communication (they have no inherent value, no worth, beyond being an agreed pattern of dots and dashes to facilitate communications). It is the total arbitrary nature of the pattern that eliminates it's value, no pattern is worth more or less than another, they just represent an agreed pattern between the morse operators, so they can send and receiver information and understand what those messages mean, that message itself having the worth, the equipment involved being patentable but the arbitrary pattern has no value, it is arbitrary.

Re:"The G part stands for GNU?"

[Gravis Zero]

Honestly, I think we made our beds, and now we're being forced to sleep in them.

I think you mean YABWBFSI (Yet Another Bed We're Begin Forced to Sleep In). ;)

Re:"The G part stands for GNU?"

[JaredOfEuropa]

There's a big push in the industry to make things easier for non-experts. For entirely selfish reasons, to be sure...

try hard to make things more complex to outsiders than they have to be

If that accusation ever came from a lawyer or a judge, I'd say that's the pot calling the kettle black.

Re:"The G part stands for GNU?"

[JaredOfEuropa]

Maybe it speaks more of the weakness of the witnesses. I don't buy the idea that the concept or the importance of an API cannot be explained to a layperson of average intelligence. I'm not sure I'm on board with that breakfast menu analogy either.

Re:"The G part stands for GNU?"

[fph il quozientatore]

Typical sloppy engineer lacking rigor. Actually, that's norm of elephant norm of grape sin theta.

Re:"The G part stands for GNU?"

[thegarbz]

It is. Fun and clever can also be interpreted as childish and confusing. To understand GNU is to first understand Unix, understand where GNU comes from and understand how it is different. To those not willing to give it that level of thought (the 16.97bn people not on Slashdot) it just draws a large "WTF?"

Re:"The G part stands for GNU?"

[ragahast]

In the first case, the judge was ignorant of programming so taught himself before the case started.

It's the same judge (William Alsup), and he still knows how to program (in Java). The menu analogy is poor, and recursive acronyms aren't funny to everyone, I guess. It doesn't make him clueless, and anyone who has followed these cases could tell you that Alsup is considers Google to be in the right (thus his ruling in the first case).

Re:"The G part stands for GNU?"

[serviscope_minor]

Huh OK. I stand corrected.

Re:"The G part stands for GNU?"

[cr0nj0b]

you do realize the judge in this article is the SAME Judge William Alsup? So, I'm having a hard problem with this current situation. The Judge learns programing a few years ago. Then has problems with FOSS and GNU now...

Re:"The G part stands for GNU?"

[bytesex]

So, if I understand correctly, to make all the diagrams and illustrations in your case documentation, you have used a GIMP? I see... Say, what else does this GIMP of yours do?

Re:"The G part stands for GNU?"

[loufoque]

I think that's a good analogy, but I wouldn't say that the communication protocol is worthless and not subject to copyright.

As a professional software developer, I've found that the most difficult thing to do is to come up with the right interface; designing the right software architecture is much more challenging than implementing algorithms.
Copyright applies to interfaces as well, regardless of what Google would like.

Re:"The G part stands for GNU?"

[Plus1Entropy]

Because the judge didn't say it's a stupid joke he said he didn't understand it.

Well, we don't know everything that he said, because the article stops that particular branch of dialog with that statement. I mean, did they just sit in silence after that? The statement, "That doesn't make any sense." can be interpreted in a few different ways, it doesn't necessarily mean "I don't understand the subject matter.", it could just as easily mean "I don't understand why they would do that."

Re:"The G part stands for GNU?"

[Zero__Kelvin]

Yes. It is a lot like being born intelligent. We deserve what we get for having the audacity to be unlike the vast majority.

Re: "The G part stands for GNU?"

[Zero__Kelvin]

GNU and PHP are not a recursive acronyms; they are self-referential acronyms. Self-reference is as important to programming as recursion as anyone versed in OOP knows, but the two are not the same, and people who refer to GNU as a recursive acronym rather than a self-referential one are as confused as the layperson.

Re:"The G part stands for GNU?"

[Zero__Kelvin]

What is wrong with words that are well defined and in the Oxford dictionary? Personally, I'd rather see Slashdot dweebs who purport to be English experts, but are actually clueless, neutered (We all know you can't spay a Slashdotter.)

Re:"The G part stands for GNU?"

[Avarist]

Recursive acronyms can be fun and clever. Not everyone immediately groks the concept, but that's no reason to beat ourselves up.

Let us drink water to that.

Re:"The G part stands for GNU?"

[Archtech]

And yet "a jury of your peers" must surely demand some understanding of the domain of expertise and the issues being tried. I don't think the phrase originally meant "a bunch of perfectly average 'normals' too dumb or lacking in initiative to get out of jury service".

Re:"The G part stands for GNU?"

[Archtech]

If that accusation ever came from a lawyer or a judge, I'd say that's the pot calling the kettle black.

Actually, the legal profession is even more guilty than IT. Because lawyers and judges stand to earn vastly more money when the public at large doesn't understand their guild secrets. Whereas there is nothing in particular to stop the ordinary citizen from learning as much as he or she wants to about computers.

Re: "The G part stands for GNU?"

[ChrisMaple]

Sadly and more importantly, the attitude of young people toward politics is the same. They vote for the people who offer shiny stuff and make them feel virtuous, with no thought to whose efforts are needed to make the shiny stuff.

Re:"The G part stands for GNU?"

[ChrisMaple]

For an explanation that reaches the most people, call the API a language. Signs reading "ALTO" and signs reading "STOP" are instructions from 2 APIs commanding a temporary end of motion.

Re:"The G part stands for GNU?"

[ChrisMaple]

Much humor comes from surprise. The surprise here is changing the context from the expected (genetics) to the unexpected (math) while still providing a sensible answer.

Re: "The G part stands for GNU?"

[ChrisMaple]

I also like the idea that lawyers must be permanently prohibited from being legislators, and vice-versa. There is an inherent conflict of interest involving legislators making the law complex and incomprehensible, so that only lawyers can decode it.

Re: "The G part stands for GNU?"

[russotto]

Sadly and more importantly, the attitude of young people toward politics is the same. They vote for the people who offer shiny stuff and make them feel virtuous, with no thought to whose efforts are needed to make the shiny stuff.

So it has ever been since at least Roman times.

Re:"The G part stands for GNU?"

[currently_awake]

This might be a good time to require Judges to be knowledgeable about the trial subject. Obviously you'd need judges to specialize, you can't expect 1 man to be an expert with everything. I would think a "jury of your peers" would require sufficient background knowledge of the subject at trial as well. Yes this would be a burden, but it would make trials more fair.

Re: "The G part stands for GNU?"

[currently_awake]

You want our laws written by people who don't know what they are doing? That's like prohibiting engineers from designing cars. We already have special interest groups writing our laws, at least we can say the senator knew what he was voting for.

Re:"The G part stands for GNU?"

[currently_awake]

A better analogy is a vending machine. You push this button you get a bag of chips, that one gives a chocolate bar. My program wants to save a file it pushes this (API) button, it wants to show a movie it pushes that (API) button.

Re:"The G part stands for GNU?"

[pla]

"Childish and confusing" does not equal "illegal".

The real problem here comes from how our judicial system interprets the word "peers". Any sane interpretation of that word would have the present jury packed with programmers. Instead, we usually take it to mean "complete idiots", to the point that if one juror actually understands anything technical brought up during the trial and tries to explain it to their fellow jurors, that juror has potentially committed a crime.

Re:"The G part stands for GNU?"

[UnknownSoldier]

So you're saying:

* Clever puns are not allowed? Gee, thousands of years of writings shows people love puns.
* The intelligent must pander to the whims of the stupid?

Get off your high horse already.

Re:"The G part stands for GNU?"

[flargleblarg]

Yes, but are there really 12 programmers who don't all hate Oracle?

Re: "The G part stands for GNU?"

[N3wsByt3]

Panem et circenses.

It is as true in current modern times as it was in ancient times: the only thing the masses and the hoi palloi need to remain relatively docile is food and entertainment.

Re:"The G part stands for GNU?"

[ItsJustAPseudonym]

But I do believe that in the tech industry we had reveled a but too much in that idea of our superiority and sometimes try hard to make things more complex to outsiders than they have to be. And now it can hurt us.

Yes. You have said it well.

Re:"The G part stands for GNU?"

[ItsJustAPseudonym]

I think you are correct, that the judge is out of his element, with the programming aspect. However, we are concerned about another aspect. One that ought to be trivial and have no bearing on the case. The 'GNU' acronym. And that's where things get weird and sad.

The judge certainly has prior experience with acronyms, and he plainly expects acronyms to be created using a simple set of rules. The 'GNU' acronym has clearly violated his expected set of rules. Now we are faced with the prospect that the oh-so-clever GNU acronym is helping to irritate the judge. It's a stupid situation, but there it is. That's not a situation any lawyer wants to be in. (irritated judge)

Re:"The G part stands for GNU?"

[e432776]

I disagree that "This speaks to the weakness of the jury system"- the opposite is true. The whole point of the article, which you have encapsulated perfectly in the quoted fragment, is that "nerds" and their pursuits are/should be apart from the rest of society. The opposite is true, as we exist in society and so are judged by the standards of the same society. If we want the public at large to understand us and what we do better it is on us to disseminate what we do and bring more folks in. System working as advertised.

Re:"The G part stands for GNU?"

[ItsJustAPseudonym]

We deserve what we get for having the audacity to ACT unlike the vast majority.

Re: "The G part stands for GNU?"

[Zero__Kelvin]

So it takes an incredible amount of audacity to act like oneself? I'm going to go with no on that one, and hope you aren't serious.

Re:"The G part stands for GNU?"

[jedidiah]

No. It's just math and Americans have been conditioned to be afraid of math. At most, something like GNU is just silly and nothing really to get terribly excited about.

Re:"The G part stands for GNU?"

[jedidiah]

No, we just mate within our own species and all remains well.

Re:"The G part stands for GNU?"

[jedidiah]

Then the facts don't add up. If he really "knows Java" then the GNU acronym really shouldn't be a problem. If GNU hurts his head then he probably doesn't really know as much as the media likes to claim he does.

Re:"The G part stands for GNU?"

[NicBenjamin]

Playing around with Java means you played around with Java, it does not mean you magically gain a strong opinion on the eternal Starship Enterprise vs. Star Destroyer debate. So it's hard to know how much background he got on the cultural stuff, particularly the FOSS community.

Moreover it's entirely possible he knows what's going on better then he lets on, and he's trying to get Google's guys to explain so that the dumbest possible Juror knows what's going on.

Re: "The G part stands for GNU?"

[GPS Pilot]

Then please go make a correction to the second paragraph of this Wikipedia article: https://en.wikipedia.org/wiki/...

It says "GNU is a recursive acronym".

Re:"The G part stands for GNU?"

[UnknownSoldier]

I really shouldn't feed stupid Arrogant Cunts ...

pun:

* noun, a joke exploiting the different possible meanings of a word or the fact that there are words that sound alike but have different meanings.
* verb, make a joke exploiting the different possible meanings of a word.

Why do you think the GNU logo is a gnu animal.

The recursive acronym, GNU, means 3 things. So, YES, GNU is a pun. But I guess some self-righteous people feel that computer scientists / programmers can't have a sense of humor.

.

Re: "The G part stands for GNU?"

[Cederic]

You think that the fact that this supposed loophole is only questionably legal and not blindingly obviously legal and encouraged doesn't automatically mean the US legal system is fucked?

Thanks for pointing out Canada but you have serious issues to sort out a little nearer to home first.

Re:"The G part stands for GNU?"

[Plus1Entropy]

So you're saying:

* Clever puns are not allowed?

Nope, I'm not.

The intelligent must pander to the whims of the stupid?

Funny, that's what I feel like I'm doing now... No, I didn't say that either.

Get off your high horse already.

Oh, the irony... Tell you what, let's joust; you on your horse, and me on my gnu. En garde!

Re: "The G part stands for GNU?"

[Zero__Kelvin]

You can certainly feel free to do so. I didn't get the level of skill and knowledge I have by believing Wikipedia is a definitive source, or by spending my time updating it so that people who don't get that will be able to continue falsely believing it is. Here it is pal ... Your chance to contribute rather than complaining and expecting someone else to do it. Have at it ...

Re:"The G part stands for GNU?"

[Carewolf]

No. It's just math and Americans have been conditioned to be afraid of math. At most, something like GNU is just silly and nothing really to get terribly excited about.

No at worst it is stupid (and it basically is, if you are naturally inclined to lazy evaluation). Image if the underpants businesplan in South Park had gone:

1. Collect underpants
2. Step two
3. Profit

And the joke being that step two is step two...

  That is not very funny. Or even that clever unless you already know what a recursion is and recognize it, for everyone else, it is just stupid.

Re: "The G part stands for GNU?"

[ItsJustAPseudonym]

I would not say it's necessarily audacious to "be oneself". However, you wrote "unlike the vast majority", which carries a different implication, specifically an implication of isolation or separation. Given such isolation or separation, there's a choice: do you act in a manner that increases that separation, or not?

It is audacious to act needlessly clever, which increases that isolation, possibly to the detriment of oneself. That is the situation now, with the "GNU's Not Unix" pun and this judge. Oh boy, how clever! They couldn't just have a common acronym? No? And this judge is now irritated at it.

Personally, I find the judge's opinion preposterous and superficial, but he's in a position of power, and his opinion is about to influence something real.

Re:"The G part stands for GNU?"

[ItsJustAPseudonym]

It is silly, until someone like a judge gets prejudiced by it. Then it is both foolish and problematic.

Re:"The G part stands for GNU?"

[stealth_finger]

Saying grok makes you sound like you want to be in a clockwork orange.

Re: "The G part stands for GNU?"

[Zero__Kelvin]

That would make sense if the term wasn't coined by Robert Heinlen in a book called "Stranger in a Strange Land", which is exactly what it feels like to be highly intelligent in a world surrounded by those of average intelligence by the way.

Re:"The G part stands for GNU?"

[WallyL]

I don't know, how many employees does Oracle have?

Re:"The G part stands for GNU?"

[Austerity Empowers]

This case follows the proud British legal tradition (which you inherited) of proudly ignorant old dinosaurs ruling on stuff they consider it to be beneath them to understand.

This is clearly true, but one sided. A proud tradition of technical people, from ivory tower scientists to lowly plumbers is often to talk over the heads of the audience, or else to mix in some really terrible and confusing metaphors. Or both, because when you order pancakes, you want them entangled with bacon, at 350 Kelvin to ensure a flavorful protein denaturation.

The reasonable solution is to get more technical people through law school, but technical people frequently have a strong disdain for law or else don't want to let their technical skills go to waste while they learn about hundreds of years of human stupidity, greed, sloth, avarice, malice and precedents set for curbing such behavior therein. As a result, there's a strong shortage of capable people in the field.

Re:"The G part stands for GNU?"

[UnknownSoldier]

> Tell you what, let's joust; you on your horse, and me on my gnu. En garde!

I've spoked my peace.

See what I did there. :-)

Re:"The G part stands for GNU?"

[Coren22]

So, ah, we noticed your blinker fluid was getting low, so we ah filled it up. That will be an additional $20 for your oil change, you really should check that blinker fluid more often.

Re:"The G part stands for GNU?"

[Coren22]

17 Billion? What time period do you call home?

Re: "The G part stands for GNU?"

[Coren22]

You do realize that that comment indicates that you yourself feel you are a smug doofus right?

Re:"The G part stands for GNU?"

[Coren22]

Say, what else does this GIMP of yours do?

He runs around in a skin tight vinyl suit.

Re:"The G part stands for GNU?"

[thegarbz]

"Childish and confusing" does not equal "illegal".

No one said it did. In fact none of what is going on is illegal as it all amounts to a trademark and copyright dispute. In those cases it's up to you to convince a braindead version of a jury of your peers of why you're right. Childish and confusing may not be illegal but it sure as hell doesn't help you in court.

Re: "The G part stands for GNU?"

[GPS Pilot]

Everyone is free to do so. I've added a lot of information to Wikipedia, on topics I care about. I don't care about the topic of whether GNU is a recursive acronym... but I thought you did, since you took time to express disagreement.

Re:"The G part stands for GNU?"

[PCM2]

See, I thought we were over all of this. GNU doesn't stand for anything, end of story. It's just a name given to a group of associated free software projects.

Re: "The G part stands for GNU?"

[GPS Pilot]

AC, if you are really Zero__Kelvin,

I don't ask people to contribute to Wikipedia if I think they're wrong.

But if you removed "recursive acronym" from the GNU article, it would be very interesting to find out whether Wikipedia's editors would let that stand.

I'm not going to throw around unsubstantiated accusations like you have done, but it seems like you don't want to make that edit because you're afraid it won't stand.

Re:"The G part stands for GNU?"

[PRMan]

But it lacks the comedic timing to actually BE a joke.

Oh my god

[NotSoHeavyD3]

He couldn't just bring up steering wheel, accelerator, brake, and gear shift as an example of an interface?

Re: Oh my god

[meerling]

I know what APIs are, and I couldn't figure out that menu 'explanation'.

Re: Oh my god

[the_Bionic_lemming]

The analogy wasn't simple enough.

Here's what I would of said.

"Ok, so I want to say hello to someone that doesn't speak my language, so I get an interpreter.

I say "Hi" and the interpreter turns to the person I want to speak to and says "Labas" (they are Lithuanian) the person who hears "Labas" replies with "Laba Diena" and the interpreter turns to the guy who said Hi and says "good Day to you"

The interpreter is the API - and then go on to explain the case.

Re: Oh my god

[ShanghaiBill]

I know what APIs are, and I couldn't figure out that menu 'explanation'.

The "menu analogy" didn't make sense to me either. The power plug analogy was better, since (unlike a menu) that really is an interface. It lets you use power from any source (solar, wind, coal, nuke) to power any device (computer, TV, microwave oven). You can swap any source or device in-or-out as long as it adheres to the spec (analogous to the API).

The power plug analogy also demonstrates why copyrights/patents on interfaces are a really bad idea. If everyone need a separate plug for every power source / device combination, then our walls would be covered with outlets, and you would need to hire an electrician every time you bought a new lamp.

Re: Oh my god

[Darinbob]

I'm not sure I understand APIs. I saw someone wearing a tee shirt that said :"I APIs". That made no sense. Then I heard people say that they worked on APIs. The APIs I know about you don't fall in love with, much less treat them as the focus of a product.

Re: Oh my god

[roman_mir]

Except that is not true. API does not modify or interpret anything, an API is an agreement, a contract, a promise to accept inputs and to produce outputs in the way that can be incorporated into another system.

API only exists as an ephemeral description. The description is used in an applicatiin that expects the form described by the API to provide connection to the functionality behind it. once the compiler produces machine code the API form is gone, it ceases to exist in the same form, the implementation takes over.

API is the name and a promise to accept inputs of certain form and to provide outputs of a certain form. API does not have functionality beyond a promise (like translation you are using as an example). API is more like a silhouette of an object, the object itself may be guessed by the contours, but the object is not its contour.

API is a shape, a form, not any implantation of a shape or a form. I don't think it is difficult to explain but bad analogies are very dangerous when trying to explain in court.

Re:Oh my god

[roman_mir]

Or he could have thought about it for 10 minutes and said something like this:

API is a silhouette, a contour, an outline of an object, but it is not an object itaelf, it is a promise that the object will provide functionality that the contour is hinting about.

To copyright a contour while maybe possible should not penalize those, who want to provide their version of an object that is projecting the same contour. A contour of a woman's body is clearly recognizable but it does not say anything more than 'it is a woman'. A contour of a car promises that the object behind it is a car but the car itself with all of its parts cannot be seen.

Applications depend on such contours to request the functionality of the objects behind the contours. To allow a company to put a lock on a contour would destroy ability of applications to use each other's functionality and would significantly and negatively impact the economy.

To prevent others from projecting a promise of functionality by using an existing and well recognized description of that functionality through the means of these API object contours is to stop all development of alternative systems unless sanctioned by the current legally recognized owner of the specific object providing such functionality. But an outline of a system is not a system itself. An outline of a door is not a door yet it makes it clear that there is a door and it can be used.

Should a particular door maker be able to prevent others from making doors that people can walk through because we recognize a rectangle on a wall as a passage, as a door regardless of the company that made the door?

Do we want a single company to control all doors?

Re: Oh my god

[Hognoxious]

I saw someone wearing a tee shirt that said :"I APIs".

You should get one with "I an emoticon".

Re: Oh my god

[Hognoxious]

I think the restaurant menu sucks too. I would have gone for paper forms. They have different items depending on their purpose, each is laid out in a specific way so that the recipient can easily understand them rather than being free format etc.

At least some of the jurors would remember filling out a purchase requisition or leave request, or a deposit slip at the bank.

Failing that, something about plumbing. But definitely not dump trucks.

Re: Oh my god

[Mondragon]

The "menu analogy" didn't make sense to me either. The power plug analogy was better, since (unlike a menu) that really is an interface. It lets you use power from any source (solar, wind, coal, nuke) to power any device (computer, TV, microwave oven). You can swap any source or device in-or-out as long as it adheres to the spec (analogous to the API).

The power plug analogy also demonstrates why copyrights/patents on interfaces are a really bad idea. If everyone need a separate plug for every power source / device combination, then our walls would be covered with outlets, and you would need to hire an electrician every time you bought a new lamp.

This falls flat in reality. Many devices have their own power receptacles on device-end (micro-USB, firewire, etc.) and those are all licensed. The in-wall receptacle in various countries (and in this case we consider the US) were also once licensed, but most of those have expired at this point.

If Oracle wants to demand licensing fees on the Java API, it's hard to argue that they shouldn't be allowed to do so. We might argue this is really stupid, economically, but the legal equivalencies are not hard to find. Even more importantly at issue in this case is that Google has taken the Java APIs, but not actually implemented Java - they've used these APIs to do something else entirely. This is somewhat akin to taking an L5-15 and only delivering 10 amps. Sure it works most of the time, but it's not the same.

Re:Oh my god

[myid]

Just because the judge and jury don't know what an API is now, doesn't mean they can't understand it, given a good explanation. An API has to do with code. Show them some code.

If I were explaining an API to them, I would say that it's mainly a set of commands for certain tasks. If a Java programmer wants to instruct a Java program to do one of those tasks, then he/she must type in the corresponding command. (A "set of commands to type" is easier to understand than "set of rules", "interface" or "contract".) There is more to the API than commands - there are data values (ex: Math.PI), and the way that the API is divided into classes, interfaces, and packages. But mainly programmers care about the commands.

I would go to https://docs.oracle.com/javase.... I would show the judge and jury the Math.random() method description, and briefly go over the description.

I would keep the description of Math.random() up on one screen. On a second screen, I would show them the source code of a very simple Java program that calls Math.random(), and then prints the random number. I would point out the line in the program that contained the call to Math.random(), and say, "See, that's an example of using the Java API method that you see here on the first screen".

I would say that we don't know how the Java program determined the random number. We don't have to know. We just have to know which command in the API to use.

They'd be able to understand that.

Re:Oh my god

[r0kk3rz]

...and this is why so many people are mystified by computing, the people that understand it are really bad at analogies.

Re: Oh my god

[Darinbob]

Hmm, supposed to be I love APIs...

Re:Oh my god

[smallfries]

Or he could have said something that simplified it instead of romanticising it:

The English language is an API: the alphabet and the dictionary. People use that API to write books, emails, webpages and a million other things. The things that they create are copyrighted as creative expressions using combinations of letters and words.

Oh whoops, it is a bit awkward that the law no longer fits that explanation after the previous ruling...

Re: Oh my god

[Anonymous Coward]

HDMI is an excellent idea to merge audio and video leads into one cable and improve bandwidth that was turned into pure shit by tech companies adding DRM to the specification negating the benefits

That is how I do it. YMMV

Re:Oh my god

[roman_mir]

Do you want to be happy or do you want to be right? It is a court where people have to decide a number of things based on the current and very unfortunate copyright and patent law. A precise architectural drawing can most certainly be copyrighted and maybe patents apply as well. A precise architectural drawing describes the internal workings of an object, API does not. A contour or an outline without the description of the internal structure is what API provides.

Re:Oh my god

[roman_mir]

Yes, but did you see my nick? You cannot say 'romanticise' without 'roman', can you? Besides, language is not precise either, you can say the same thing in a dozen different ways. A contour of an object has a promise of that object but nothing more than a promise. As long as your side of the system fits to the contour of an object your system has a belief that it will have the underlying object available to it. This works for static but also for dynamic linking.

Language is used to describe ideas and thoughts and feelings and facts. How is language an API when it is actually the structure?

Alphabet is an API to language (maybe) but not really, language came to exist without any alphabet, alphabet was built on top of language. Actually alphabet serves as superstructure, language is not derived from it.

In case of an API the syntax allows the API to exist, not the other way around. But it is getting too muddy here, too meta.

I like your approach better than what was said in the courtroom though.

Re:Oh my god

[roman_mir]

Come up with a better one if you think this one is flawed.

Re:Oh my god

[roman_mir]

Well, we could start by selecting people for jury who would know the difference between their ear canal and their asshole... Both are orifices and yet...

Re:Oh my god

[tomhath]

Do you know the Morse Code for "Help"? Every kid learns it: SOS - three short, three long, three short. That's the interface, the API.

How do you send the signal? Light flashes? Horn blasts? Radio or telegraph clicks? Palm fronds on the beach? Doesn't matter, those are all implementations that adhere to the API for SOS

Re:Oh my god

[Chrisq]

...and this is why so many people are mystified by computing, the people that understand it are really bad at analogies.

Yes, it's a bit like a dolphin trying to talk to an elephant .... oh wait!

Re:Oh my god

[roman_mir]

Nonsense. SOS is a message itself, it is data, not an interface.

Re:Oh my god

[roman_mir]

Maybe they should hire a cow as an interpreter.

Re:Oh my god

[perpenso]

API is a silhouette, a contour, an outline of an object, but it is not an object itaelf, it is a promise that the object will provide functionality that the contour is hinting about.

To copyright a contour while maybe possible should not penalize those, who want to provide their version of an object that is projecting the same contour. A contour of a woman's body is clearly recognizable but it does not say anything more than 'it is a woman'.

That's not true. A contour can be far more specific than a general "woman". With a sufficient description it can be trademarked. While not a woman, what if your female contour resembled Minnie Mouse? Expect a call from Disney attorneys.

I'd argue that an API is not a general contour but a very specific recognizable one. So this analogy is not a good one. I hate to say this, but the car user interface analogy is a good one. Accelerator on the right, brake to its left, etc.

Re:Oh my god

[roman_mir]

Actually the original ruling didn't say that API cannot be copyrighted. That API in fact *can* be copyrighted is still true. The issue is fair use of an API, not inability to copyright it.

Re:Oh my god

[perpenso]

A trademark is a very different type of intellectual property than a copyright. Its fair use may be limited to reviews, not something functional/operation like an API. To me copyright fair use seems to make this analogy weaker not stronger.

Re: Oh my god

[the_Bionic_lemming]

An API transfers a variable to a function or sub. Remote app says Hello, it's dumped into the function waiting for labas. Function takes the expected input and returns the result to the calling code.

So I disagree with you.

Re: Oh my god

[mrchaotica]

A cookbook analogy would be better than a menu analogy: code is like the recipes, and the API is like the table of contents.

Re:Oh my god

[TranquilVoid]

You've got +4 Funny, but I think you're serious and I agree with you, so maybe I'm a clown..

Analogies tend to make more sense to the person who actually understands the system in the first place. Rather than explaining the unknown system, an analogy creates a pattern link between two systems the alleged explainer already understands. Sometimes the best way to understand is to get in there. Showing them basic APIs is the way forward, then they can form their own analogies based on other systems they personally understand. Most of the presented analogies have an in-built bias anyway.

Re:Oh my god

[Tablizer]

Come up with a better one if you think this one is flawed.

I have to agree the silhouette analogy will go over like a lead balloon, based on my experience trying to explain technical things to "regular folks".

But I also agree agree that good analogies are hard to come by.

Maybe addressing an envelope: a convention of communication is established whereby the destination address goes in the middle, the return address in the upper left, and the stamp(s) in the upper right.

Re: Oh my god

[Coren22]

This falls flat in reality. Many devices have their own power receptacles on device-end (micro-USB, firewire, etc.) and those are all licensed. The in-wall receptacle in various countries (and in this case we consider the US) were also once licensed, but most of those have expired at this point.

Which is kind of exactly the point being made. All of those different connectors is due to differences in the specifications or due to copyright. You have a bigger plug on a 30 amp power supply, or a 220 V source than a 110 V or 15 A. Micro, mini, lightning, these are all copyright or physical size decisions, mini stopped being used when cell phones became too small to accommodate the connector.

If Oracle wants to demand licensing fees on the Java API, it's hard to argue that they shouldn't be allowed to do so.

If Oracle wants to go proprietary, then maybe we need to invent a new Micro USB connector to get around the idiocy of the lightning connector being locked up in licensing fees. If Oracle goes with a fee structure, someone will just invent a new Java and no one will use that crap that Oracle makes you pay for.

Re:Oh my god

[Some_Llama]

but where would they hold the meaningless meetings?

Truly unprofessional headline and story

[Anonymous Coward]

It's not clear to me that the problem is the judge and the jury. The analogies seem like they're pretty bad analogies, especially the breakfast menu. It seems like the problem is just as likely that the lawyers and expert witnesses are doing a bad job of explaining things to the jury. Why do things have to be dumbed down, anyway? Why not directly explain what an API is instead of resorting to simplistic analogies? Instead, the judge and jury are accused of being clueless. Perhaps you should listen to the judge that the analogies are confusing instead of claiming the judge and jury are idiots. Maybe people don't like being talked down to. Somehow I have a feeling the judge and jury actually care about understanding what an API is, and resent that witnesses are talking down to them.

Re:Truly unprofessional headline and story

[Anonymous Coward]

I'm not sure the summary is accurate wrt to the article or if the articles writers are slanted, because there was praise for Alsup in Oracle v. Google (2012) for actually knowing how to program. Would argue that maybe the better interpretation of Alsup's questionings and responses were that the programmers and lawyers were dumbing down the explanations too much.

https://tech.slashdot.org/comments.pl?sid=2887227&cid=40175735

by gcnaddict ( 841664 ) on Thursday May 31, 2012 @10:17PM (#40175735)

via c|net [cnet.com]:

On many days, the San Francisco courtroom where he presided was more like a computer science classroom. Alsup acknowledged during the trial that he had learned about Java coding to better prepare for the case, and it showed. On a daily basis, he would deftly query the lawyers and expert witnesses on the structure, sequence, and organizations of APIs to assist the jury in understanding the key facets of the copyright phase of the trial.

This is why I have respect for Judge Alsup. In order to apply the law in a complex engineering-related case, he worked to learn the subject matter in order to properly apply the law to the material. That's how I expect every Judge should apply the law rather than just sit and "trust the experts" per-se.

Re:Truly unprofessional headline and story

[meerling]

Geek or not, the technical stuff actually is technical and norms usually don't even come close to having the necessary knowledge to understand even the basics.
Then there's the whole brain factor. Simply put, we think differently than they do. We tend to think using more reason and logic than they do, and unlike the common misconception, we have just as much emotion and humor which makes us unlike the Mr Spock they think of us as, at least when they aren't thinking of us as raging drunk/stoned hippies at a movie stereotype college frat party.

Re:Truly unprofessional headline and story

[Tailhook]

The breakfast menu is a fine analogy and has the immense benefit of being automatically familiar to everyone in the Western world. I can easily imagine someone with sufficient social skills successfully conveying the concept of an API using a restaurant menu analogy. I think the real problem is the choice of using these celebrity geeks, all of whom appear to be suffering various degrees of Aspergers. They can't explain things to `normals' whether they use analogies or not.

Re:Truly unprofessional headline and story

[DaHat]

Precisely, and this is an interesting quirk of our legal system where we are to be judged by ones 'peers'... but where both sides will attempt to reject as may potential jurors as they can who may tilt the scales too much in the opposite direction.

As a software developer well into his second decade, I know I would never be able to serve on a jury with regards to anything technical because, just as I would expect anyone with any semblance of affiliation with #BlackLivesMatter activist to be rejected from a jury regarding a cop killing someone.

Re: Truly unprofessional headline and story

[Anonymous Coward]

The breakfast menu is a good description, but the menu is really the documentation not the api. The api is the waitress, she understands certain requests defined in the menu, the kitchen is actually the implementation and the cheeseburger is a possible output

Re:Truly unprofessional headline and story

[Anonymous Coward]

Not surprising when the technical people they deal with act like douchebags.

Re:Truly unprofessional headline and story

[St.Creed]

The conceptual modeling ISO standard already declared that a conceptual model is as applicable to modeling the judicial domain as it is to modeling business information. So I would argue that judge Alsup behaved like lawyers should, which is as information analyst.

The problem being that most lawyers do not actually understand their own domain in a mathematical sense, from a conceptual modeling view. It would improve laws a lot if they were treated as conceptual models that need to be disambiguated before use.

Re:Truly unprofessional headline and story

[tomhath]

It's not clear to me that the problem is the judge and the jury.

It's clear to me that the problem is with the witnesses. If Schwartz had prefixed his explanation with "GNU is a little programmer's joke, it stands for..." the judge and jury would've understood perfectly.

And describing a hamburger menu as an API is a horrible analogy. I see where he was going with it - the API describes how you request a service, rather than being the service itself. But that's way too subtle for a non-programmer to understand the difference between the API and the implementation.

Re:Truly unprofessional headline and story

[drinkypoo]

A restaurant menu is a terrible example. That's a good example of a pick list of some kind, like... a menu. A better example is anything actually designed to interoperate. How about a plug and a socket?

Re:Truly unprofessional headline and story

[Livius]

The analogies seem like they're pretty bad analogies

I don't know why they couldn't explain a programming interface by analogy to a user interface, and I'm guessing everyone knows what a user interface is.

Not wrong

[Anonymous Coward]

The judge isn't wrong, having a recursive name like GNU is weird and something only nerdy programmer types really appreciate.

This trial is a prime example of a concern I often have with the legal system in particular and the government in general: people who do not understand something are being asked to decide an issue. Government officials, whether they are judges, lawmakers or the leader of the country are usually well versed in law, but not medical research, technology, engineering, education and rarely have first hand experience with poverty, womens issues, etc. I think it's an unfortunate side effect of our system.

Re: Not wrong

[Anonymous Coward]

That's what expert witnesses are for. The judge also has the power to say that something doesn't make sense and ask questions. The problem here is that the judge is providing feedback and, instead of addressing the feedback, the judge is being treated like an idiot.

Re:Not wrong

[Rick Zeman]

And PINE Is Not Elm.

Anyway, so much for a jury of their peers.

Re: Not wrong

[Anonymous Coward]

That's what *expert witnesses* are for. The judge also has the power to say that *something doesn't make sense* and ask questions. The problem here is that the judge is providing feedback and, instead of addressing the feedback, the judge is being treated like an idiot.

I am more concerned with tonight's realization that there's a cultural tendency in the US for fools to immediately say "... does not make ANY sense" without thinking why they came to you for help in the first place. They are OFFENDING you.
This judge didn't say "sounds odd to me" or "*I* do not understand" or "you lost me at this part..." and it drives me a little nuts. In my experience, half the time we just repeat ourselves more slowly and non-geeks will understand (or drop the subject, which the judge here CAN'T, hah). Nobody wants to look "dumb" when they need something explained, and I hate the implication that the explainer is incompetent. I know certain people who are not technical and here is why they succeed at my explanation: enough mental acuity to avoid being lazy during the explanation, and good follow-up questions that sometimes come in the form of a forgivable interruption.

Re:Not wrong

[Darinbob]

If they do understand an issue, then it becomes hard to be impartial. If they do not understand an issue, then they're likely to be swayed by whoever speaks the best. It's a lose-lose situation.

Re:Not wrong

[gweihir]

It is not. A recursive definition is a perfectly valid and well-understood concept from Computer Science or Mathematics. The only thing this shows is that the Judge is _not_ an expert in this field at all and that is what he should have concluded. Obviously he is not smart enough for that and that is a real problem. When a judge does not realize he is not an expert and needs to listen to the expert witnesses and believe them, things are screwed up to an extreme. Also, what business does a jury of non-expert have judging an expert issue?

Re: Not wrong

[Anonymuous Coward]

This judge didn't say "sounds odd to me" or "*I* do not understand" or "you lost me at this part..." and it drives me a little nuts.

But in this case, the judge is right. A recursive acronym does not make much sense. Or, to put it another way, it only makes sense as a joke.

Maybe the judge was trolling a little bit too -- just like asking a catholic why he has to eat his saviour's flesh

Re:Not wrong

[yithar7153]

But he has a B.S. in Mathematics. I would think he would know what a recursive definition is, or at least proof via induction. I think it just doesn't make sense because there's no base case.

Re:Not wrong

[phantomfive]

Not to mention the definition of GNU is completely irrelevant to the case at hand, so even a perfect judge would have tried to move on from that point.

Re: Not wrong

[tgv]

> I am more concerned with tonight's realization that there's a cultural tendency in the US for fools

Can you FUCK OFF on the high horse you came in on, please? Are you calling a Supreme Court judge a fool? They probably have more intelligence in their index finger than you in your entire body.

Re:Not wrong

[Anonymous Coward]

This is common in FOSS - "people dont understand what im saying therefore they are idiots. i need to only deal with an expert"

If a jury does not understand your argument, the issue is with you, not the jury, to make it make sense

Not act like a condescending prick

Re: Not wrong

[publiclurker]

you apparently have not looked at who is on the supreme court in a long, long, time, have you.

Re: Not wrong

[Agripa]

But in this case, the judge is right. A recursive acronym does not make much sense. Or, to put it another way, it only makes sense as a joke.

So, it makes perfect sense?

Re: Not wrong

[Agripa]

Nothing, but the judge has already decided what the result of the case will be and this was his chance to further it.

Re: Not wrong

[shutdown -p now]

Problem is, who determines whether an expert witness is indeed an expert?

I mean, we're currently seeing whole large areas of forensics being demonstrated to be built on outright lies. Yet, before that, plenty of "experts" testified that they were certain that such-and-such has committed a crime because their teeth marks matched, or their hair matched - and juries have delivered guilty verdicts based on such testimony, with no-one the wiser.

"That doesn't make any sense," said the 71yo judge

[Snufu]

Unless the gnu lives on Endor.

Now Nerddoom is biting back ...

[angel'o'sphere]

Why was SVN so slowly adopted?
Well, which CEO wants a software tool in house that is called "subversion" and the main IDE integration was called "subversive"?

Why do we have a program called "less"? Who is old enough to remember that the original program was called "more"?
A program that displayed its input page by page and had in the last line "hit space for MORE". Calling the improved version of "more" "less" might be funny ... once ... but actually it is not. From a today perspective calling a program "less" which's only purpose is to display "more" makes no sense at all.

There are more funny plays of word like this, e.g. the first improved "man" program, that was worth installing was called "woman". Yea, funny ... I was like 23 when I installed it (on request of my 'boss' a guy making his PhD).

Hm, to tired now, I had 3 or 4 more examples but nothing comes to mind right now.

I would not be surprised if there is an improved version of "cat" that is called "dog" ...

Re:Now Nerddoom is biting back ...

[gweihir]

Whetstones and Drystones. And some more. I do disagree on "less". If you do not see the beauty of that naming, then you are blind to aesthetics.

Re:Now Nerddoom is biting back ...

[golgotha007]

>> From a today perspective calling a program "less" which's only purpose is to display "more" makes no sense at all.

Less is more than more, sheesh.

Re:Now Nerddoom is biting back ...

[ChrisMaple]

There's a backwards version of cat called tac.

Re:Now Nerddoom is biting back ...

[djinn6]

"Less" is correct. Without less, all of the program output gets displayed on screen. With less, only one screen of text is displayed at a time. So it is displaying less (than the original).

Re:Now Nerddoom is biting back ...

[Shortguy881]

That's better than windows 8's:
hover on right side,
click on settings,
select power settings,
select off.

But, that still beats windows 10. I have yet to find a power button there. Its ok though; my computer shuts off nightly with automatic updates.

Re:Now Nerddoom is biting back ...

[angel'o'sphere]

Perhaps you should reread what I wrote.

Renaming an improved version of "more" as "less" makes no sense, regardless how you want to rationalize it.

According to you the program "more" was badly misnamed :D

Perhaps it would make sense if the first version of the two programs already was called "less".

Same thing as democracy

[Kjella]

What the f... does the average person really know about running a country? Nothing. Why should he have any say in that, really? Couldn't we just leave it to a bunch of experts? And what's to say the experts are really neutral at anything? For example if you wanted "experts" on copyright law you'd probably end up with a jury full of MPAA/RIAA/BPA members, oh and maybe a couple from the EFF for balance and surprisingly most their verdicts would go in favor of big business. Having ignorant people on the jury is the worst of all systems, except every other system we've tried. If you think you can design a system that won't have these problems you're either absolutely brilliant or extremely ignorant. And I know what my money is on.

Re:Same thing as democracy

[meerling]

We tried leaving the software issues to the software, but most of it had nothing to do and just stopped running. The few that took up the task were MCP and Skynet, and of course that went completely off the rails so we had to terminate that process and reboot the timeline.

Democracy is for people, software are tools, and though some people are tools, they are allowed to participate not because they are tools but only because they are people. Any tools that are not people don't care about it in the first place.

Re:Same thing as democracy

[DaHat]

Or... only in 14% of cases have the police screwed up in a major way either with regards to not collecting enough evidence, not doing so in the right way, or worse charing the wrong suspect.

The more time you spend around someone involved in criminal law (on either side) you realize that the police do not screw up as much as you would otherwise think.

Re:Same thing as democracy

[afgam28]

Very rarely do democracies require citizens to actually make detailed decisions about how a country is run(*). We usually get people to vote on something much more general - which politicians get into office - and then we let them handle things without micromanaging them. If they fuck up, we get pissed off and vote for someone else in the next cycle.

This jury on the other hand is being asked to decide the outcome of a single case. The case is fairly technical, or at least more technical than the general public is able to understand.

There are better ways to do it. We could be more selective in the jury selection process, and ensure that the jury consists of technical people. Or we could have expert witnesses instruct the jury on how to make findings of facts.

But we shouldn't just throw our hands up and say that it's "the worst of all systems except every other system". It's easy enough to identify problems with the system in general, and this case in particular, and then propose solutions.

* Yes, I'm aware that there are some places (e.g. California) that have people vote directly on little things, but this tends to lead to bad results (the state is almost broke and a lot of the cities have dysfunctional politics).

Re:Same thing as democracy

[swb]

Isn't there some philosophical argument (at least) to be made that says that laws that can't be understood by ordinary men shouldn't be enforced?

It seems like there's both a basic democratic element to it -- if the rule of law derives its legitimacy from the consent of the governed, how can they consent to what they can't understand? I have to live my life knowing that I may be subject to laws made in such a way I can't know if I am even obeying them.

It also seems to be kind of a streamlining effect -- if laws can't be easily understood, maybe they're unnecessary or not really enforceable, Or they lack the basic coherence that says they represent a concrete idea. Complex laws are more likely to represent the interests of narrow constituencies.

I think in the legal system also has a vested interest in the complexity of law. If laws were understandable and enforceable in plain language, the legal system would have less purpose and standing.

Re:Same thing as democracy

[afgam28]

Anyone who has the capability to break this would-be law (reimplementing an API) also has the technical capability to understand it, and therefore consent to it. Those who don't, don't need to worry about it. I don't think this is a problem, and there are many branches of the law where this is the case, especially in specialized areas.

For example there are laws around the minimum legal tread depth on car tires. Someone who does not have any experience of driving might not understand that there's a relationship between tread depth and water dispersion, but any driver who intentionally lets his or her tires go bald is being irresponsible. When someone takes on a technical activity, it's their responsibility to also educate themselves on the risks involved, and the branch of law that applies to that activity.

Re:Same thing as democracy

[ChrisMaple]

You have used an untrue initial assumption, that a country should be "run". Free countries are not run. The government in a free country may be run, but the government is not the country. In a free country people run their own lives, and the job of the government is to protect people so that they can continue to run their own lives -- no more, no less.

Re:Same thing as democracy

[Hognoxious]

Exactly. If you aren't running a restaurant, there's a lot of laws to do with food hygiene that you don't need to bother with. Likewise for pretty much any trade or profession.

This "there are too many laws for everybody to know" meme is talk-radio twaddle.

Re:Same thing as democracy

[AutodidactLabrat]

Not really, given the wholesale and flagrant death penalties given to innocent men discovered and documented by Project Innocence.
When cops screw up, they bury the evidence of the screwup, and often enough, the victim of their acts.

Re:Same thing as democracy

[AutodidactLabrat]

Oh, and don't forget the 28 other officers of various jurisdictions (sheriff, Highway Patrol) who were filmed watching the Rodney King beating, 100% of whom perjured their sworn reports.
Not one of them even got Mark Furhman's $200 fine for the felony.
What we know about criminal cops is what leaks through the blue wall.
LOVE the body cams and auto-uploaded cell phones. Expect lots more calls for special prosecutors for Blue Wall Thugs.

Re:Same thing as democracy

[shutdown -p now]

There are many areas where the laws are extremely convoluted, and apply the moment you step your foot there, even unknowingly.

For example, if you go and buy a GMRS radio on Amazon (where they're readily available and advertised as walkie talkies), and actually try to use it, you're violating FCC regulations, because the use of GMRS requires a license.

Or, say, gun laws. Buying a gun is easy (if you're not a felon already), but installing an innocuously looking gadget on it (for example, a vertical forward grip on a pistol) can instantly turn it into a highly restricted firearm that requires a license or a tax stamp to "manufacture", and violation of that law can land you in prison for 10 years.

Re:Same thing as democracy

[Hognoxious]

I don't see how either of those is different to learning what roadsigns mean before driving a car. Planning on doing an activity? Do your homework first. Ignorance of the law is no defence.

Re:Same thing as democracy

[shutdown -p now]

The problem is the same one that was stated originally - the sheer amount of laws in that area, and the fact that they're written in a way that is very hard to understand and parse (i.e. is not "common sense"), and at times very vague. For gun laws, for example, ATF itself has repeatedly made conflicting rulings on various aspects in the past, literally based on a slight change in the implied meaning of some word (like "possession" or "manufacture").

Also, with cars, we at least force people to learn the rules by licensing the activity and predicating the license on training. With all this other stuff, it is not the case. If you're a buying a bubble pack radio, it is not at all obvious that it requires a license to operate (especially since it's not the case for all such radios, and in some cases depends on which channels you transmit on), and that there's a large body of laws regulating such operation that you need to be well acquainted with.

Your approach would mean that literally every time you engage in some activity - any activity - you need to consult the laws to determine if any apply. Which brings us back to the original point - there are so many laws in effect that rigorously implementing this approach is unfeasible for an individual in a modern society - we'd spend all of our time reading laws. The end result is that many people do break laws unintentionally in the course of their routine activities, and the majority that does not, doesn't do so by accident rather than knowledge. It is not really an acceptable state of affairs.

HS diploma who failed geometry

[Overzeetop]

These are people (the jury) who failed geometry in high school. And that was when he subject material of proofs and theorms (and logical arguments like programming) was fresh in their minds. The average American struggles with 6th grade pre-algebra - and I'm talking college grads more than 10 years out! Most of my (non-tech, 40-something, BS or MS degreed, commercially successful) parent friends are basically tapped out of helping their own kids in math by 7th grade.

You could explain APIs from now until 2020 and half of them still wouldn't get it. An analogy involved food/restaurants (which they DO understand) may be your only hope, since sex and excretory functions are off the table in terms of polite conversation.

Re:HS diploma who failed geometry

[PopeRatzo]

These are people (the jury) who failed geometry in high school.

Let's all remember that this case never had to come before a jury. The sides could have chosen for it to be a bench trial, but I'm guessing both sides liked their chances better if they could find a group of people who really didn't understand what was going on.

Hell, they could have probably gone to arbitration and avoided a trial altogether.

A pox on both their houses.

Re:HS diploma who failed geometry

[meerling]

I don't know about that.
In my opinion, Larry Ellison is a rather unstable nut job, and that's from a long time before Java was ever written.

Re:HS diploma who failed geometry

[Ferocitus]

An analogy involved food/restaurants (which they DO understand) may be your only hope, since sex and excretory functions are off the table in terms of polite conversation.

How can you explain Microsoft's Win 10's strategy without referring to a shit sandwich?

Re:HS diploma who failed geometry

[msauve]

"You could explain APIs from now until 2020 and half of them still wouldn't get it."

APIs are like dog commands - sit, down, stay. Implementations are the individual dogs and how a dog was trained to follow that command.

Re:HS diploma who failed geometry

[AK Marc]

It's nearly impossible to get two corporations into binding arbitration. Your legal rights are greatly limited, and one bad day from one "impartial" person or panel, and you've lost everything.

Re: HS diploma who failed geometry

[Applehu Akbar]

I would say that is a failure of the lawyers during the process of voir dire.

If you want to get out of jury duty, standard procedure is to admit being an engineer at voir dire. The result in this example is a jury that is not competent to try the case.

Re:HS diploma who failed geometry

[DoofusOfDeath]

How can you explain Microsoft's Win 10's strategy without referring to a shit sandwich?

It's not difficult, here's an example:

Imagine it's Valentine's day, and you go to the store to get your bonnie lass a box of chocolates. You have them gift-wrap it, and you go to the fancy restaurant where you have dinner reservations. The evening goes beautifully. You take her home, and she gives you a very exciting come-hither loop. You sweep her off her feet, and carry her to the bedroom. After some passionate kissing, you gently start to lift her blouse, and there's a shit sandwich.

Sorry, my mistake. It can't be done.

Re: HS diploma who failed geometry

[yithar7153]

I think it's a failure of the system. Normally you want people who have no bias whatsoever about the case so they can produce an opinion based on the facts, or what the lawyers and witnesses and such say. But this backfires here, because here you have people who have no idea how this stuff works. IIRC there was one guy who actually worked with computers, a network manager, and Oracle didn't like him for some reason, so he was out.

Re:HS diploma who failed geometry

[smallfries]

I don't think that is very fair at all. Does a shit sandwich constantly bug you if you have decided to dine on something else tonight?

Re:HS diploma who failed geometry

[loufoque]

I don't understand your argument, geometry is one of the most difficult mathematical fields, much more difficult than calculus.

Re: HS diploma who failed geometry

[AthanasiusKircher]

If you want to get out of jury duty, standard procedure is to admit being an engineer at voir dire. The result in this example is a jury that is not competent to try the case.

Actually, I'd say it's broader than that -- lawyers generally want jurors who aren't capable of independent, rational thought.

There's a common perception that lawyers are looking for stupid people for juries. That's not generally true. Lawyers don't want people who are so stupid that they don't understand what's going on and thus will make random decisions in the end. They want people who are intelligent enough to follow along, but NOT ones who are capable of critiquing complex arguments.

Why? Because in most trials at least one side realizes that they have what might be perceived as a weaker set of facts. Thus, they will depend on convincing a jury to follow an argument through steps A, B, C, and D -- while there might be a bunch of things that make steps B and C rather murky from a strict logical standpoint. Lawyers want people who can overlook those conclusions and go along the chain from A to D if it "feels right" and "makes sense" to them.

Engineers, scientists, other lawyers, philosophers, etc. generally have training about how NOT to fall into misleading traps of argument. So they are more likely to follow the facts, no matter where that leads them -- and that's distressing to lawyers, who sometimes call such people "rogue jurors," since they are less susceptible to "herd mentality" and insist on thinking for themselves.

Also, I've heard (from lawyers) that there's a perception among many lawyers that "experts" are bad people to have on juries. The reason seems somewhat irrational: they often believe that experts are "too biased" by their own understanding of a case to make an impartial judgment. That makes little sense to me in most cases, though I can see that an expert who ALSO has strong preconceived views might be a greater problem than a layman with preconceived views (if nothing else, for the way the expert might sway the rest of the jury due to expertise).

But in most cases, I think it's a combination of what I said above (experts in an area may have more critical thinking skills in that area to pick apart arguments) and that lawyers recognize it's harder to present an "air-tight" argument to an expert, particularly given that many lawyers are not subject matter experts on cases they are trying.

Re: HS diploma who failed geometry

[ajakk]

As a patent lawyer, I can tell you that technology lawyers don't like engineers on the jury because they jury will give that juror's opinion inordinate weight. If that engineer completely misunderstands the technology, or doesn't like one of the parties, they can steer the entire jury one direction because they will use their outside knowledge to "teach" the jury. Rather than relying upon what evidence was given in trial, the jury will follow the lead of the one engineer, even if that one engineer is wrong.

Re: HS diploma who failed geometry

[Applehu Akbar]

The legal system perceives the ideal juror as being of fairly normal intelligence but easily led. What's scary is that both prosecution and defense have the same view.

Re:HS diploma who failed geometry

[ChrisMaple]

More difficult than symbolic multivariable integration by parts, where you have to keep guessing the form of the ultimate answer (assuming one exists) until you find one that works?

Re:HS diploma who failed geometry

[Koen Lefever]

I don't understand your argument, geometry is one of the most difficult mathematical fields, much more difficult than calculus.

Opinions and experiences on what is more difficult will be different for different people.

According to Henri Poincare (in La valeur de la science) there are two kind of mathematicians: analysts and geometers.

It is impossible to study the works of the great mathematicians, or even those of the lesser, without noticing and distinguishing two opposite tendencies, or rather two entirely different kinds of minds. The one sort are above all preoccupied with logic; to read their works, one is tempted to believe they have advanced only step by step, after the manner of a Vauban who pushes on his trenches against the place besieged, leaving nothing to chance. The other sort are guided by intuition and at the first stroke make quick but sometimes precarious conquests, like hold cavalrymen of the advance guard.

The method is not imposed by the matter treated. Though one often says of the first that they are analysts and calls the others geometers, that does not prevent the one sort from remaining analysts even when they work at geometry, while the others are still geometers even when they occupy themselves with pure analysis. It is the very nature of their mind which makes them logicians or intuitionalists, and they can not lay it aside when they approach a new subject. Nor is it education which has developed in them one of the two tendencies and stifled the other. The mathematician is born, not made, and it seems he is born a geometer or an analyst.

I should like to cite examples and there are surely plenty; but to accentuate the contrast I shall begin with an extreme example, taking the liberty of seeking it in two living mathematicians. M. Meray wants to prove that a binomial equation always has a root, or, in ordinary words, that an angle may always be subdivided. If there is any truth that we think we know by direct intuition, it is this. Who could doubt that an angle may always be divided into any number of equal parts? M. Meray does not look at it that way; in his eyes this proposition is not at all evident and to prove it he needs several pages.

On the other hand, look at Professor Klein: he is studying one of the most abstract questions of the theory of functions to determine whether on a given Riemann surface there always exists a function admitting of given singularities. What does the celebrated German geometer do? He replaces his Riemann surface by a metallic surface whose electric conductivity varies according to certain laws. He connects two of its points with the two poles of a battery. The current, says he, must pass, and the distribution of this current on the surface will define a function whose singularities will be precisely those called for by the enunciation.

Doubtless Professor Klein well knows he has given here only a sketch: nevertheless he has not hesitated to publish it; and he would probably believe he finds in it, if not a rigorous demonstration, at least a kind of moral certainty. A logician would have rejected with horror such a conception, or rather he would not have had to reject it, because in his mind it would never have originated. (PDF English translation)

Re: HS diploma who failed geometry

[Agripa]

It did not work for me either last time I was called. Next time I will say when asked that I will not follow the judge's instructions.

Re:HS diploma who failed geometry

[sysrammer]

"...come-hither loop."

nice

Re: HS diploma who failed geometry

[loufoque]

Definitely. Integration is methodical, a computer can do it.
Geometry is much more abstract.

Re:HS diploma who failed geometry

[dingleberrie]

Actually, I liked where this was going. Let me try my ending:

Imagine it's Valentine's day, and you go to the store to get your bonnie lass a box of chocolates. You have them gift-wrap it, and you go to the fancy restaurant where you have dinner reservations. The evening goes beautifully. You take her home, and she gives you a very exciting come-hither look. You offer her a chocolate, and she says she's not wanting that, but that doesn't dissuade you. You push it toward her mouth, pressing it against her teeth as she ever-increasingly tries to push your hand away. You switch hands and finally get it down her throat.

As for the rest of the chocolates? You drop them in her mouth while she's sleeping, like windows updates.

Re: HS diploma who failed geometry

[ItsJustAPseudonym]

"...admit being an engineer..."

Lawyer: Are you an engineer?
Potential Juror: Not necessarily.
Lawyer: You sound like you might be an engineer.
Potential Juror: That's not a sufficient condition for being an engineer.
Lawyer: Look, you are not fooling anybody.
Potential Juror: I don't grok your meaning.
Lawyer: Bzzzt! You can go.
Potential Juror: What gave me away?

Re:HS diploma who failed geometry

[DoofusOfDeath]

FTW. Well done, sir.

Re: HS diploma who failed geometry

[Overzeetop]

If you only know how to solve a problem one way, you probably don't understand the underlying concept of the mathematics. As the parent of a teen going, I'm amazed at how many people know the rote method they were taught and are utterly confused when presented with another method to achieve the same solution. Not just your run-of-the-mill parents, but parents with engineering degrees and technical jobs where math is used regularly. I get that you may not understand the specific method on first reading, but this is grade school math and you already know how to solve the problem - what's so hard about understanding an alternate method. (okay, lattice multiplication is on the weird side - I'll give you that one)

Re: HS diploma who failed geometry

[Overzeetop]

Yup - I can absolutely see that happening. I'm also imagining, with a bit of sadistic enjoyment, a jury comprised completely of engineers. I mean it's possible you get a super efficient deliberation....but its more likely that a fist fight* breaks out over some minutia.

*and by fist fight, I mean 3-4 engineers yelling at their shoes about how the other ones are wrong.

**disclaimer: I'm an engineer. I've been in those meetings.

Re:HS diploma who failed geometry

[Tablizer]

That's what you git for dating Billy Jean Gates

Re: HS diploma who failed geometry

[Some_Llama]

i told them that i hate cops because they all lie and prosecutors will say whatever they can to ensure a conviction because that is their job.

I was politely excused. I have another jury duty coming up soon and we'll see if i have to tell them the same.

How our justice system works

[linear a]

This actually sounds somewhat like what the American justice system is *supposed* to be like. The goal is justice, not understanding by the jury. Both sides present their best cases - hopefully both do so competently so that lay people can understand what the important facts are. The jury then decides the various disputed points based on that evidence and their understanding as lay people. Experts are supposed to explain things to them clearly. The judge's actions also appear correct. His job is to facilitate the understanding that the jury forms of the evidence. If something seems muddy to him then he can expect it to be muddy to the jury and the presenting side can, as in this case, present that part more clearly. The counsel of the side not presenting the muddy breakfast menu may be displeased when the judge causes that to be explained again more clearly. At the end of the day, jury needs to have enough understanding to know which side should prevail on which issues and also to decide the penalty.

Re:GNU is Not Unix

[PPH]

This could work to our advantage.

Judge: So, exactly what is recursion?

Definition: recursion; see recursion

Supreme Court is now stuck in an endless loop, never to bother us again.

Re: And Clinton screws us again

[AutodidactLabrat]

And if intent meant anything, Gore won Florida. By more than 48000 votes (butterfly ballot).
but it doesn't.
Hillary had nothing to do with it

Clueless Judge and Jury

[AutodidactLabrat]

Is there any other kind?
I give you Loreena Bobbit, who emasculated her drunk husband for sexing it up with another woman, then cried "rape attempt" and was LET OFF though there was zero evidence of the claimed rape.
As the biker's said years ago " Remember that at trial, your future will be decided by 12 people so stupid they could not think of an excuse to get out of jury duty.

Normals, huh?

[Anonymous Coward]

>normals in the jury box to the normals at the Supreme Court

Who talks like this? The whole tone of TFA is cringe-inducing. Couldn't the submitter find another article about this lawsuit that reads like it was written by adult?

Not ready for prime time.

[westlake]

The whole point of pre-trial proceedings is to frame the issues in a way that a judge and jury can understand them. If you are not focused, if you are not making yourself clear, this is the time and place to fix the problem.

Erm Guys... This Judge writes code..

[andydread]

This judge is not clueless
Judge Alsup writes code

Quote from this judge from this same case before the appeal I believe
I have done, and still do, a significant amount of programming in other languages. I've written blocks of code like rangeCheck a hundred times before. I could do it, you could do it. The idea that someone would copy that when they could do it themselves just as fast, it was an accident. There's no way you could say that was speeding them along to the marketplace. You're one of the best lawyers in America, how could you even make that kind of argument?

Re:Erm Guys... This Judge writes code..

[Anonymous Coward]

If he writes code and doesn't know what GNU means he's even *more* clueless than if he didn't write code.

Re:Erm Guys... This Judge writes code..

[gweihir]

I was wondering that too. Maybe he has some expertise but is vastly overestimates its worth? A recursive definition is not advanced stuff, we had the first in the first week of CS and then again in Mathematics, also first week. Maybe he has never heard the definition of "GNU", but if he understands recursion, he should have immediately recognized the pattern. And it is not even a tail-recursion.

Re:Erm Guys... This Judge writes code..

[Anonymous Coward]

Despite what Slashdot would like to think, not all people live and code in the open source world, nor come across these OSS-associated names daily.

Re:Erm Guys... This Judge writes code..

[radarskiy]

The judge did not say he didn't know what GNU stands for. It is counsel's job to present the evidence for what GNU stands for, not the judge.

The judge said the definition as stated doesn't make sense, which is correct. To explain the definition of GNU you must also have an explanation of recursion. In conversation between experts the definition of recursion is assumed as given, but for the trial it must also be entered into evidence.

One of the big problems of this trial is that different people are using different definitions of terms like API, declaration, etc. Getting counsel to clearly and completely back up their usage is part of the judges job.

Re:I'm on oracle's side on this

[Harlequin80]

Really? I see this as no different to a 3rd party manufacturer making coffee pods or oil filters. The API is the bit that connects / fits with something on the outside. And I can do it both ways. I can make a coffee machine that uses nespresso pods or I could make pods that fit in a nespresso machine.

What I want to know is...

[CMU_Ken]

Who puts hamburgers on a breakfast menu?

Re:What I want to know is...

[gweihir]

Americans...

Re:What I want to know is...

[MouseTheLuckyDog]

My thought exactly. No wonder the jury was confused.

API

[fyngyrz]

Good grief. Really? REALLY? From first principles, assuming only that someone has used a reasonably modern computer:

1) The purpose of a computer is to perform various series of small steps, called "instructions", assembled in an order such that each whole series of instructions can perform a more complex task than a single instruction can by itself.

2) These assemblies of instructions are called "programs."

3) Computers can have previously prepared programs built-in, or installed later, that are designed to provide certain services to other programs, so that these services do not have to be re-created for every new program that needs them, and so that these services are performed in a standardized way for all programs that need them.

4) An API is the part of the previously prepared program that provides the means to access these services. The term means "Application Programming Interface" or "Application Programmer Interface."

5) An example of this is the window that opens when you want to select a file from within a program such as a spreadsheet or a word processor. This is often called a "file dialog." In order that users of computers only have to deal with one set of tools to open files, this service is provided by modern computer operating systems, and is often preferentially used instead of creating one's own version of such a function. The API for this service allows for asking that the dialog be opened, and then, when the user chooses one or more files from the list in the dialog, the returning of which file(s) the user picked to the program that requested the file dialog service. There are other services provided in the API, including "cancel", when the user changes their mind about choosing a file; change the storage location where the chosen file will, or does already, reside, and so on.

---

It's not trivial to actually explain, but it isn't all that difficult, either. I'm sure there are others here who could do much better than I. Without ever mentioning a... menu, etc.

Re:API

[U2xhc2hkb3QgU3Vja3M]

But to open a file, don't you usually need to go into the menu?

Re:API

[Ferocitus]

Good grief. Really? REALLY? From first principles, assuming only that someone has used a reasonably modern computer:

1) The purpose of a computer is to perform various series of small steps, called "instructions", assembled in an order such that each whole series of instructions can perform a more complex task than a single instruction can by itself.

2) These assemblies of instructions are called "programs."

3) .

At this stage your best bet would be to line up some potatoes on the bench in front of the judge.

Re:API

[Bongo]

I think you hit the nail on the head, in that, the only sane way to explain this stuff, isn't with clever analogies, it is by just setting out the actual things themselves. And it isn't hard to do that. There's only three or four things which need to be understood and then people can see where API fits.

Re:API

[St.Creed]

I doubt Google would want to do this. I'm afraid that under current IP law they'd be very much screwed. The confusing analogies may be entirely on purpose.

Re:API

[tomhath]

4) An API is the part of the previously prepared program that provides the means to access these services.

Your definition is what the trial and all the confusion are about.

Is the API an implementation, i.e. "the means to access these services"?

Or is an API the documentation of an interface, the implementation of which is a black box?

Oracle claims it's the interface specification, Google claims it's the implementation

Re:API

[fyngyrz]

The application programming/er interface is the implementation. It's what the program uses. The documentation is the description of the application programming/er interface. The program will not run because I read the manual.

Just as the manual for a car's interface is not the interface itself, but a description of the car's interface. I'm not going to be able to steer the car by twisting the manual around in the air.

If your description of the issue at hand is correct, there's no issue here at all.

Re:API

[fyngyrz]

I should have said:

An application programming/er interface is a means to access implementation of a service. The API is not the implementation of the service(s) themselves, but an means of access to the service(s.)

Sorry. My coffee level is low as yet this morning. :)

Re: API

[Anonymous Coward]

But what if the judge can't be bribed?

Re:API

[tomhath]

Well, that's your personal definition of an API. I'm not disagreeing with you, but as I said previously - Oracle is essentially claiming that their JAVADOC is the API that they own, Google is claiming that they can write their own library of routines that happen to have the same JAVADOCs (so their implementation works fine with code that uses Oracle's library). Is the JAVADOC an API? That's what the jury will decide.

Meh...

[__aaclcg7560]

If you think "normals" are bad, try dealing with users as a system admin tech. "No, sir," I would explain. "The computer doesn't belong to you. It belongs to the company. And I have full authority to fix your computer." Some days I wish I could replace their computer with a box of crayons, which is all they need to do their job anyway.

Re:Meh...

[l0n3s0m3phr34k]

"so...you'll pay my bill if I fix it? Oh, so, then, it's NOT your computer..." Go read all of the Bastard Operator From Hell, then you'll find your capabilities as an effective sys admin increase a thousandfold. Just the excuse list is useful..."solar flares" is the best non-answer ever for all wifi/cellular issues you don't feel like actually explaining.

Re:Meh...

[__aaclcg7560]

Just the excuse list is useful..."solar flares" is the best non-answer ever for all wifi/cellular issues you don't feel like actually explaining.

I've been using "gamma radiation" for decades now. If it works for the Hulk, it should work fine for everyone else.

Re:Meh...

[Livius]

"The computer doesn't belong to you. It belongs to the company. And I have full authority to fix your computer."

Refresh my memory, which of those statements is factually incorrect?

Re:Meh...

[__aaclcg7560]

Refresh my memory, which of those statements is factually incorrect?

According to corporate policy, all these statements are factually correct. Problem?

Re:I'm on oracle's side on this

[Anonymous Coward]

If they're copyrightable then surely there aren't any *books* about Java not written by Oracle right? After all, a book couldn't mention them without violating copyright. Yes?

GNU is a recursive acronym

[TerraFrost]

GNU is a recursive acronym. The best non-tech example I can think of is VISA, which stands for Visa International Service Association. The judge probably has a VISA card himself.

Re:GNU is a recursive acronym

[gweihir]

Nice find! So this _is_ used outside of the IT world as well. As a recursive definition is an idea from Mathematics, this makes sense to me.

Re:GNU is a recursive acronym

[Aristos Mazer]

Judge Alsup is a very well informed judge for technical issues... take a look at the original 2012 rulings from Alsup in this case. He is trying to elicit from the lawyers an explanation for the jury. It's the judge's job to predict the questions the jury would have about what they've just heard. It is NOT his job to try to explain the information himself, even if he knows the answer ... he's there as judge, not to provide testimony.

quickest way off a jury

[dltaylor]

From what I've seen during jury selection, demonstrate either knowledge or the ability to think for yourself and you will be dismissed post-haste. The lawyers for both sides (criminal or civil) want more-or-less house plants that will follow their version of "logic".

Anyone who has followed the United States political scene since, essentially, forever (about 1796, or thereabouts) knows that our system is fully intended to maximize the power of the dimmest bulbs in the shed.

Re:quickest way off a jury

[DoofusOfDeath]

the dimmest bulbs in the shed

Please tell me that phrasing was intentional.

Re:quickest way off a jury

[Livius]

maximize the power of the dimmest bulbs

It would all depend on whether the power output of the bulbs was a limitation of the bulb wattage or an issue of amperage.

Not clueless

[roman_mir]

You know, people maybe uninformed on minutia of a specific field but they are still trying their best to understand it, the first court decision probed that point. I wouldn't be so quick to call anybody clueless. On the other hand this could be a case of corruption, after all we are talking about Oracle and Larry here.

AFAIC Larry and Oracle lawyers in this case should be held personally responsible for all of this. If I were running Google (Alphabet?) I would be taking other measures, planning steps to stop these proceedings by means that have nothing to do with any court... We are talking about billions ( or more) here and about every application and system out there. Beside planning a hit on Oracle and its key people, I would also be working on the government level, trying to dismantle the entire copyright/patent system. Government must not have the authority to protect any company or individual from competition.

Re:Not clueless

[ChrisMaple]

That's a nice book you just finished after 10 years of constant work. Too bad for you I'm going to provide copies to all at cost. You'll never make a penny.
Libraries would be empty without copyright law.

Re:Not clueless

[roman_mir]

Not a government problem.

Re:I'm on oracle's side on this

[tp_xyzzy]

> 3rd party manufacturer making coffee pods or oil filters.

This analog does not hold, for the reason that it's not covered by copyright. Copyright involves text content, basically sequence of letters. Those sequence of letters get copyright protection, if it's unique. Api definitions are text content like that.

The key in regognizing copyrightable content is that there exists large number of possible arrangements (of letters), but authors choose their unique combination. Simply duplicating it is illegal, because doing so is easier than choosing the combination from scratch.

Re:I'm on oracle's side on this

[tp_xyzzy]

> Why do I suspect that you don't really know what an API really is?

I'm a programmer. Even though api has smaller state space than ordinary text content, it still contains enough freedom for authors to choose their unique implementations. Thus it gets copyright protection. Assuming otherwise is dangerous.

Re:I'm on oracle's side on this

[afgam28]

The entire software industry would grind to a halt if a copyright on APIs were enforced. Do you think someone should be able to sue Oracle for using SQL as the interface to their DBMS? Or using the Unix API in Solaris? Solaris ships with a C/C++ compiler and I'm pretty sure Sun/Oracle "stole" the APIs from AT&T and UCB. I don't see how this is any different from Android.

Re:I'm on oracle's side on this

[tp_xyzzy]

> The entire software industry would grind to a halt if a copyright on APIs were enforced.

I think you underestimate the problem. It's much worse than that. Companies would need to build their own markets for their software, instead of always building their wares utilizing someone elses market. The trick is that this would be useful activity for a company to do. Now that they skip the hard work, consumers will suffer because the services are simply not available where they're needed.

Re:I'm on oracle's side on this

[gweihir]

If you make APIs copyrightable, you do extreme economic damage by enabling monopolies. Copyright is not a natural right. Its only purpose is to assure that people cannot just copy creative works by others and thereby stop the creation of such works. In the case of an API, this is not a risk, because an API without an implementation is worthless, and the implementation is covered. It is also highly doubtful that an API is a creative work at all.

Clueless, You think not.

[www.sorehands.com]

In the first trial, which Alsup ruled that an API is not subject to copyright. He learned Java for the case and written that even a high schooler could write it in 10 minutes.

https://developers.slashdot.or...

Re:I'm on oracle's side on this

[Harlequin80]

But my understanding is that they are duplicating an implementation via black box reverse engineering. Anything can be described in words, this doesn't extend copyright protection to the final object.

If your interpretation is correct there would be no need or desire for software patents of any kind because software is collections of letters and hence fall under copyright. This just simply isn't the case.

Re:I'm on oracle's side on this

[tp_xyzzy]

> You think the government should protect corps from entering each others' markets?

That's why companies apply for patent protections, so that other companies don't immediately conquer your market with significantly bigger resources than you have. Copyrights have similar purpose, authors don't want publishers conquering the market with _copies_ of his own books. It takes several years to write a book, and if other publishers don't need to spend that time to get the same result, it's unfair to the author who did need to spend that time.

Re:I'm on oracle's side on this

[Anonymous Coward]

The problem here is that oracle don't have a design patent, which is what you are talking about. If you have a design patent for the API, then this whole thing would have been a different story. For example the pod is patentable in its design. You are not allowed to make one that looks identical without violating the design patent. The invention itself of the idea of a pod is also patentable, and that patent would be violated by making any pod.

Instead of oracle arguing its a design infringement, they are instead only have an invention patent on the way java works as a virtual machine concept.

So google will win as they haven't taken the concept of the whole java machine, but they have taken the design in terms of form and fit the design takes.

Re:I'm on oracle's side on this

[Anonymous Coward]

By default, everything you scrible to a piece of paper, or a text file gets copyright protection

Right, but the scribbles aren't the recipe - they're a *representation* of a recipe. The recipe itself is an abstract idea. Should I have copyright on the number 5 just because I wrote it in text? Similarly, the code that describes an API is not the API.

Ceci n'est pas une pipe.

Re:I'm on oracle's side on this

[tp_xyzzy]

> But my understanding is that they are duplicating an implementation via black box reverse engineering.

from the paperwork, it didn't sound like black box reverse engineering... Instead they copy-pasted the api definitions and then implemented the missing functions. This requires no reverse engineering activity. It just requires a text editor + some programmers to write new code based on _existing_ api specification. But they had no reason to assume that they are allowed to use that specification.

but if you're reading some other paperwork, maybe i missed the information ;)

There's a lot of entertainment in the article

[phantomfive]

Hysteria aside, there's a lot of fun stuff in the article. The lawyers were brutal:

No one bothered to challenge Schwartz’s apparent belief that hamburgers are commonly featured on breakfast menus, as he had already moved on to confusing the jury on another front

“You don't remember this article about being one of the fifteen worst CEOs in American history?” [Oracle's lawyer] asked him.

Schwartz congratulates Google on developing an open source mobile platform to its face, and then calls it “Scroogle” in a private email

btw, this case isn't the end of the world, it won't affect your ability to use APIs, only to copy them wholesale. And even then you can have a defense if you do it for interoperability purposes, so in 90% of the cases where you'd want to copy an API, you'll be able to.

In the other cases, make up your own stupid language, it's not like Java is that great.

The author got it wrong

[BigU 03C0mpin]

This isn't the nerds versus the suits, it's one businessman-nerd against a bunch of non-businessmen nerds.

These are super nerds. They're so smart that they cast baffled-courtroom-of-nerdrage at the jury, unfortunately lining up to only win this dungeon on the next roll with a perfect 12.

Re:I'm on oracle's side on this

[radarskiy]

-1, irrelevant

The Appellate Court stipulated that APIs are copyrightable when they remanded the case back to the lower court. Not argument that they are not copyrightable is being entertained in this court. They have to get back to the Appellate Court to challenge that stipulation.

Naif, or disingenue?

[radarskiy]

Be careful of mistaking a judge who is unaware for a judge that is letting counsel have enough rope to hang themselves.

Re:Naif, or disingenue?

[ragahast]

Be careful of mistaking a judge who is unaware for a judge that is letting counsel have enough rope to hang themselves.

Especially when you'd have to be clueless about the two cases to think Alsup is himself clueless.

Re:Naif, or disingenue?

[ajakk]

Anyone who has ever practiced in front of Alsup knows that he is far from clueless.

Re:I'm on oracle's side on this

[Mondragon]

Well, yes, I actually do think that someone (IBM) should be able to sue Oracle for using SQL, in the abstract. In reality when IBM offered SQL to ANSI they agreed to make it zero license. Of course, all users of SQL add non-standard extensions to make it not-really-interoperable, in order to keep their own edge. Those extensions are often not licensed to third parties (although sometimes are, and those licenses end up making strange movements over time based on mergers and such, which actually often ends up with them being opened, or they find their way into future standards with zero-license clauses).

The AT&T SVR4 API was in fact licensed by Sun (and HP and IBM), at considerable expense. These licenses continued to be paid until Novell released the SVR4 license and Unix name to The Open Group. SCO has attempted to collect on later SYSV licenses, with little success, but that is only because most forks were legal variants of SYSV made before the revisions that SCO controlled.

It's not different from Android, and historically APIs have been licensed.

Re:impossibru

[grahammm]

Are you saying... the jury is not "of your peers"?

In this case yes. The definition of peer (From Oxford Dictionaries Online) - "A Person of the same age, status or ability as another specified person" As those involved in this trial are legal not natural persons, the 'age' part does not apply, for the 'status' to apply the jury would have to be legal, not natural persons and as to whether individuals people can have the same abilities as a corporation ...

Re:I'm on oracle's side on this

[tricorn]

The text that expresses the API is what's copyrighted, the idea that's expressed is not. 17 USC 102(b) denies protection to "any idea, procedure, process, system, method of operation, concept, principle, or discovery". This is deliberate, denying protection to functional aspects among other things.

This is even more explicit in "pictorial, graphic and sculptural works", where the very definition of them excludes utilitarian aspects.

While 102(b) does not deny protection to the expression of those functional ideas, when there are very few ways to express that idea, when the expression is inseparable from the idea, then the idea and expression are said to merge, and the denial of protection to the idea applies to the expression as well.

With source code implementing an API, the idea expressed is that API. When there is only one way to express that idea, then the expression itself is denied protection.

The parts of the Java API that define the names and types of the various elements are the declarations, and there is only one way in Java to express them without changing the meaning and functional specifications for the API being expressed. Thus, Google's declarations are going to be almost identical to Oracle's, and that should not be infringing.

The code that implements the methods can be written in many ways, usually, so protection remains intact for that portion, even though the idea (what that method is supposed to do) is still an unprotected idea. Google is allowed to re-implement the code for a method because the idea is not protected, but they can't copy the actual code from Oracle, since that is protected by copyright.

Clean room reverse engineering can be used to prove you didn't copy protected expression, that any similarities are purely by chance, but since the jury found that none of Google's implementation code was infringing, that isn't even an issue here, all of their code was sufficiently different, or was the same only because that was the only reasoanble or efficient way to do it (i.e. merger).

Clean room reverse engineering is also not the issue with the declaring code. It's the same because it has to be the same. No amount of "clean room" can change that, so it doesn't matter whether they disassembled the byte code from Oracle's compiled library, looked in the documentation, or looked in the source code. It's always going to be the same.

Alsup's original decision was not only technically accurate, it was logically sound and based on solid precedent. The CAFC decision overruling him contained logical fallacies, and apparently deliberate mistatement of fact and precedent at pretty much every turn.

He may not know what GNU is, or appreciate the joke, but he's nowhere near the clueless judge depicted in the article. And the menu analogy was lame.

Bad analogies can do a lot of damage.

[Anonymous Coward]

Part of the problem in this court case is the poor explanations offered by the technical experts. What they should have said to explain it:

A programming language is like a human language. It defines the rules and syntax for constructing sentences for those wishing to speak the language or those wishing to understand the language.

A programming API is like a lexicon, or a vocabulary. In order to speak a language, you have to know what words it has available and what they mean. Without that, you cannot actually communicate anything using the language.

An API implementation is like a printed-and-bound dictionary. It offers a full explanation for the words in the lexicon, so people can make use of them. It is not the vocabulary itself, that is an intangible concept. It is merely one person's expression of the semantics underlying that vocabulary.

A dictionary can be covered by copyright.

A lexicon cannot.

An implementation of an API can be covered by copyright.

The API itself cannot.

Re:Bad analogies can do a lot of damage.

[seyfarth]

That's a good analogy. I suggest that the people involved could understand an API vs the implementation by observing the prototype for a function and then the actual function. They don't need to understand the code. They need to know that the provider for the implementation allows people to call the function by observing the API and not the code. It is simple enough to discuss without analogies. The decision needs to be made on the real thing and not on analogies.

My favorite engineering jokes...

[shanen]

Are lightbulb jokes, but they're all obsolete. Does anyone have any good LED lightbulb jokes?

And I still regard google as EVIL ever since they changed their motto to "All your attention are belong to us."

Re:I'm on oracle's side on this

[NoOneInParticular]

The number 5 would be hard to copyright, but any long enough number will do. Essentially that's all a digital file is: a large number. And these can be copyrighted.

Re:I'm on oracle's side on this

[NoOneInParticular]

Mods, why is this marked troll? It seems to be an honest position all the way to the end. Nothing really outrageous about it.

Re:I'm on oracle's side on this

[tp_xyzzy]

> The parts of the Java API that define the names and types of the various elements are the declarations, and there is only one way in Java to express them without changing the meaning and functional specifications for the API being expressed. Thus, Google's declarations are going to be almost identical to Oracle's, and that should not be infringing.

No, this does not hold. Nothing in the law guarantees that google is allowed to express the _exactly_ the same idea than oracle is using. In fact, the opposite is true -- it's explicitly illegal to copy large number of elements from a plot of a book, even if they didn't use exactly the same words to express it. Manual translations of the original to different form creates a derived work, instead of losing their copyright protection. And the current situation seems to be that they used exactly the same api definitions.

It's a strange development that people tend to utilize every alternative available to go towards the edge of what is allowed by copyright law. The default still is that those actions are illegal, and instead of trying to find loopholes in the law, people should just accept that the actions are simply illegal. Examining these situations shouldn't require a lawyer's degree -- the real problem is that people want to stretch the permissions they're given, instead of actually fixing the problem. Fixing it means that you simply do not have access to other people's technology when you implement your own technology. Why should google have _exactly the same_ technology than what oracle is using? It's important that every company is doing slightly different stuff, instead of just copying what other companies are doing.

If they want to use a copy of oracle's tech, they can always license the technology from oracle. But guess the tech was more difficult to create than they expected, if they didn't get licensing deal done for faforable enough terms.

Disdain for jury duty.

[hrieke]

To each any every one of you who claimed that jury duty was a waste of their time, effort, and for all of the other idiotic variations of not wanting to do their civic duty, this is what happens.

Re:GNU is Not Unix

[alexhs]

Supreme Court is now stuck in an endless loop,

I heard that Linux does these in six seconds.

Re:I'm on oracle's side on this

[tricorn]

You're simply wrong. The statute explicitly says, and with the intent to exactly do that, allow copying the functional idea:

102(b) In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.

A plot is not functional. The elements in a plot are not essential parts of an "idea, procedure, process, system, method of operation, concept, principle, or discovery", so the copyright protection of the expression, the written words, extends to the plot.

This isn't some strange obscure loophole, it's a basic principle of copyright. Copyright is not supposed to protect functionality. APIs have been free to copy and adapt and enhance and extend since the IBM PC BIOS was first legally cloned.

If you want to protect function, get a patent. Otherwise, it's free to use and build upon.

The primary objective of copyright is not to reward the labor of authors, but "to promote the Progress of Science and useful Arts." ... To this end, copyright assures authors the right to their original expression, but encourages others to build freely upon the ideas and information conveyed by a work. ... This result is neither unfair nor unfortunate. It is the means by which copyright advances the progress of science and art.

Re:So quit with the analogies...

[Time_Ngler]

Surly given the time and cost involved ...

Yep, definitely, Joe Six Pack is going to be surly after you've tried to teach him all that...

Re:I'm on oracle's side on this

[tp_xyzzy]

> APIs have been free to copy and adapt and enhance and extend since the IBM PC BIOS was first legally cloned.

This sounds like very dangerous idea. Cloning of PC BIOS most likely involved very accurate legal manovering, which normal companies simply do not have access to. This idea is clearly at the edge of what is allowed, and popularizing it to be something that everyone should be doing just sounds like dangerous for the people involved. Basically every person must ensure that the operations they're doing are legal, and going to the very edge of what is allowed, is simply very stupid idea. People should assume this stuff is simply illegal, even though there exists people who have managed to do it without getting sued.

Re:I'm on oracle's side on this

[gnupun]

Wouldn't explaining how an API works fall under fair use? You're not copy-pasting the API for use in your software product, and therefore not unfairly competing or stealing from the API vendor.

But the courts should decide if books explaining how APIs work are violating the copyright of the API vendor.

Re:what is an API?

[gnupun]

A spot reserved to write the date is not a creative work, it is merely something that is required for any cheque to be usable.

A spot for dates on cheques can be creative work and can only be protected by a utility patent, not something as weak as copyrights. This is about copyright infringement, or exact copying of somebody's original work without payment/licensing. /. is full of pirates who hate creative people and just want to steal their work without paying a dime.

Judged by a pool of your peers....

[geggam]

... when that pool is obviously not your peer.

Re:I'm on oracle's side on this

[tp_xyzzy]

> This isn't some strange obscure loophole, it's a basic principle of copyright.

Well, the court paperwork explicitly rejected this idea, for the reason that api definitions are part of program source code, and thus covered by copyright. The exact quote is as follows:

"petitioner argued that theSSO constitutes aâoesystemâor âoemethod of operationâthat is ineligible for copyright protection under Section 102(b)of the Copyright Act."

"The court of appeals rejected that argument,ex-plaining that Section 102(b) âoerestate[s]*** the basic dichotomy between expression and idea.â"

"The court concluded that, for these purposes,computer code is âoeexpressionâ despite its functional character. "

Re:I honestly wish you were dead

[__aaclcg7560]

people like you are the reason most people would gleefully execute all user-facing IT workers,

I don't work as an user-facing IT worker. I'm the IT security guy who works behind the scenes to make sure computers are updating properly and all software conforms to the baseline standard. Users who reboot the computer while I'm on it are the ones who complain to management and are shocked to get a verbal warning not to do that again. Corporate policy states that computers don't belong to users and users can't prevent technicians from working on the computers.

I hope Trump wins and puts you into camps.

If Trump ever releases his corporate policies for IT, I wouldn't be surprised to see identical language.

Now get get back to work computer janitor.

How productive would you be at work if people stopped cleaning up your shit?

Mmmm Breakfast

[JBMcB]

I like the breakfast menu analogy better, as it's closer to what an API actually looks like and, usually, behaves. It's really just a list of things that trigger off actions. Those actions vary from restaurant to restaurant. The car analogy isn't quite as close, IMHO, as mechanically you are much closer to doing the work.

Re:I'm on oracle's side on this

[tp_xyzzy]

> Science and technology can't move forward if we don't allow ourselves to stand on the shoulders of giants.

You can just get a license to the work. People already are forgetting that it's possible to spend your money and buy a permission to use the underlying technology.

How else are the developers of the technology going to get their invested money back, if people just steal the technology, fill the market with copies of the tech, and then hide illegally obtained money to luxenburg?

Re:I'm on oracle's side on this

[mark-t]

Copyright is not a natural right

This is only partially true. At its core, copyright is simply the right of exclusivity to control who may copy a work. In fact, one does naturally have this right of exclusivity if they do not ever share their work in the first place. After all, if nobody else has it, nobody else can copy it. Copyright suggests to the person it is being offered to that legally speaking, they would still have this exclusivity even after publishing (even if they do not any longer "naturally" possess it). Theoretically, this might encourage people who would otherwise be so concerned that people may copy their stuff without authorization that they would resort to self-censorship to publish their works anyways, and society could be enriched by them. Of course, that incentive only works to the extent that people voluntarily respect that law. Theoretically, the general public's incentive to respect copyright is supposed to come from the enrichment that the society receives from creators that may have otherwise not published at all, offering a continually diverse set of published works. If these incentives are not strong enough for either one party or the other, then copyright starts breaking down. This is where we seem to be at today, but I don't want to get into that too far.

Anyways, while copyright may not truly be any kind of natural right (although if you think about it, very few things really are), it *is* a kind of logical extension of something that could otherwise be an entirely natural right, but has been augmented to try and function in the real world.

Re:I'm on oracle's side on this

[gnupun]

While 102(b) does not deny protection to the expression of those functional ideas, when there are very few ways to express that idea, when the expression is inseparable from the idea, then the idea and expression are said to merge, and the denial of protection to the idea applies to the expression as well.

With source code implementing an API, the idea expressed is that API. When there is only one way to express that idea, then the expression itself is denied protection.

You make it sound as if Google had no choice but to copy the APIs, coz there's no other way. But the reality is, there are multiple ways to express the same ideas. For example, Java itself has two different APIs just for file access: the regular one is in the java.io.* package and the other API for files is called NIO. Clearly, there are multiple ways to express file and other APIs and Google could easily create their own API for file access.

But the reason they chose to copy Sun/Oracle's API could be because they wanted something that programmers were already familiar with and to free-ride on the success of the Java language + API. Creating their own API would be risky, expensive, time-consuming, and prone to failure as it has not been tested.

"Fact" not presented as evidence is a no-no ...

[perpenso]

And yet "a jury of your peers" must surely demand some understanding of the domain of expertise and the issues being tried.

Absolutely not. The use of "facts" not presented as evidence is considered "bad" by the court system, there was no debate, no cross examination of the "expert" who provided the "fact" to the juror. Plus there is the technical problem that many "facts" are in fact "opinions", dearly loved and deeply held opinions that the holder can not imagine not being true.

I don't think the phrase originally meant "a bunch of perfectly average 'normals' too dumb or lacking in initiative to get out of jury service".

The original phrase had more to do with social class, a peasant having a jury of peasants not a jury of lords, and little to nothing to do with technical expertise, for example a blacksmith having a jury of blacksmiths. At best there was being of a sound mind and unbiased disposition. Perhaps unbiased implies not having a bunch of dearly held opinions in a relevant subject matter?

Re: I honestly wish you were dead

[__aaclcg7560]

Lol you're such a cunt.

No, I'm an asshole. I wouldn't be working in IT if I wasn't.

Well, then they should get the expert in

[Krishnoid]

And Schwarz's explanation that the acronym GNU stands for 'GNU is Not Unix' drew the following exchange: "The G part stands for GNU?" Alsup asked in disbelief. "Yes," said Schwartz on the stand. "That doesn't make any sense," said the 71-year-old Clinton appointee.

In follow-up news, Richard M. Stallman was served a subpoena to thoroughly explain the 'recursive' acronym, and additionally, to brief the jury on the facts of the case.

In an unusual exception, both the prosecution and defense agreed that popcorn would be provided free to the court, both parties, and spectators to the case.

HEIL

[tepples]

Heil Environmental Industries Ltd. is a company that makes garbage collection trucks.

Re:what is an API?

[AutodidactLabrat]

Exact copying? Not even close.
Substantial similarity so as to confuse the market
That is the legal standard. Thus Microsoft got away with stealing Windows from Mac, who got away with stealing from PARC, by making the "market" issue irrelevant

Re:I'm on oracle's side on this

[tricorn]

No, Google had no other way to express that particular API. Yes, there are plenty of different ways to do the same type of functions, but they would be different. Google is permitted to use that specific API.

Besides the quote from the Supreme Court above ("This result is neither unfair nor unfortunate. It is the means by which copyright advances the progress of science and art.") there's the CONTU report from which which the Ninth in Sega, and the Second in CAI, quote:

In some circumstances, even the exact set of commands used by the programmer is deemed functional rather than creative for purposes of copyright. "[W]hen specific instructions, even though previously copyrighted, are the only and essential means of accomplishing a given task, their later use by another will not amount to infringement."

Implementing a different API is not "accomplishing a given task". In particular, replacing the class files in the Oracle version of the library with class files compiled from Google's code would not accomplish the same task if the API was different.

This is how the courts and the Congress have shaped copyright law, and it's the way they intend it to work. The CAFC decision is an aberration, filled with mistatements, misapplication of precedent, and major logic failure.

Re:I'm on oracle's side on this

[gweihir]

You have no idea what an API is, do you?

Re:I'm on oracle's side on this

[tp_xyzzy]

> >They couldn't enter the market that oracle/sun built for themselves

> A market is not something you can build for yourself. You can only build artificial barriers for the others to enter, which is often illegal in a regulated market.

Why would hostile takeovers of companies markets be allowed? For normal large companies, it's extreamly important that they don't _accidentally_ enter markets which they don't have license paperwork in good condition. It's possible to do accidentally, and large companies know to avoid such problems. But in the current situation, it seems google is trying hostile takeover of sun/oracle market using oracle's own tech.... This kind of situations are exactly what copyright/patent laws are designed to solve.

Re:I'm on oracle's side on this

[tp_xyzzy]

> If Google could create an alternate API they would. Except they don't want to spend tens of millions and several years to come up with a new API and implementation.

I don't think this is the reason why google doesn't implement their custom api. They could create completely new api very easily. The real reason is that programmers will reject it because it doesn't provide the same functions than java standard library. The momentum is in java api's structure, the actual implementation issues are minor problems compared to building your own communities and popularity to the technologies in question.

Google's android is clearly freeriding on java's popularity. The reason is that everyone who has ever looked at android, will think that it's an implementation of java platform. They simply can't avoid this, because any programmer that looks into the technology will immediately regognize the structure of java platform in android technology. So the analysis is that they do need java license for that technology. It's amazing that they do not have a license, even though the tech is clearly available all over the world.

Re:I'm on oracle's side on this

[david_thornley]

And then it may be up to Congress to change the law. I'd rather not have judges changing the law just to get a result I see as reasonable.

Re:I'm on oracle's side on this

[david_thornley]

There's another complication here. How much are regular Java programs interoperable with Android programs? If I were implementing a version of Java, I could claim that I need to copy the API to create my own compiler. I'd be on much shakier ground if I were writing a compiler for a new language and decided to use the Java API just for convenience. AIUI, Oracle is arguing that Google produced something that won't run Java programs, and therefore used it to make things more familiar rather than for interoperability.

Re:I'm on oracle's side on this

[david_thornley]

However, why did Google choose to use that API? Was it to allow Java software to run under Android, or was it to present a familiar interface to Android programmers? Oracle is arguing that Android is sufficiently different that Google could have designed its own API without losing functionality.

Re:I'm on oracle's side on this

[david_thornley]

Buying licenses is not going to work for Free/Open Source software. For that to thrive, there has to be some legal mechanism to allow use of the APIs.

Re:I'm on oracle's side on this

[tricorn]

Both, but it doesn't matter. Copyright law doesn't protect functionality. That's what patent law is for.

Re:I'm on oracle's side on this

[bingoUV]

Do you understand the difference between API documentation (e.g. javadoc in this case) and the API itself ?

Re:I'm on oracle's side on this

[david_thornley]

Functionality is if I write a Java runtime to run Java programs on. I've got a really good case for doing that, which is really what the industry needs in copyright permissions. If I create a different language that's going to run on the JVM, that's functionality. If I create a different language that's not going to run on the JVM, the functionality question is much less clear.

The usual idea of functionality is whether I could write something different and still get the same functionality. In the case of a system that runs Java programs or runs on the JVM, I have to use the API as written. If I'm doing something different, I can get essentially the same functionality with a different API. Oracle is claiming that Google is not intending to run standard Java programs on Android, and is using Dalvik

Re:I'm on oracle's side on this

[tricorn]

The functionality, the "given task", is "run this Java source code".

If it can't do that, then it isn't the same functionality.

Re:I'm on oracle's side on this

[david_thornley]

Precisely.

Oracle is arguing that there's enough difference between current Java programming and Android programming that the functionality argument fails. They claim that Google could have used different APIs without losing significant functionality, since programs for the Android use sufficiently different facilities.

Re:I'm on oracle's side on this

[tricorn]

It doesn't have to reproduce all functionality. There's no such requirement. If I copy the entire API, make a drop-in replacement for Oracle's library, some parts of the source code must be the same for it to work, to have the same function. The expression is inseparable from the idea. Copying those elements is not infringement.

Once I (conceptually) do that, there's nothing that says I can't change my version, whether that's removing parts, changing parts, or adding parts. Copying those elements is not infringement, even if I don't copy all of them.

Re:I'm on oracle's side on this

[david_thornley]

APIs have been held to be copyrightable. That means that, to copy without a license, you need to have a valid defense against infringement. You don't get to start by saying that copying an API is not infringement and doing whatever you like with it; you present your use and say why it isn't really infringing, typically by invoking an argument about functionality. If you had to copy the API to provide functionality, that's cool. If you did it for some other reason, such as not wanting to make your own, that isn't.

APIs are creative. To offer some service, the API designer must select which functions to provide, the names of the functions, and the parameters. If you and I were to design APIs for the same thing, we'd wind up with similar overall designs and different sets of functions, function names, and parameters. The functionality argument only comes in when the API is in use in some way, and there's a reason to be compatible with it.

Oracle is arguing that, since regular Java programs and Android programs aren't at all interchangeable, the API copying is for convenience, not functionality, and hence is infringing. Whether this is true is a question of fact, as far as the courts go, and I'm not going to say one way or another.

Re:I'm on oracle's side on this

[tricorn]

APIs are copyright in one case. In other courts, it's the expression of an API, the source code, that's protected, not the actual API. Just as Google could reimplement the idea of a procedure to e.g. compare two character strings, or store a set of references to objects in a container, without infringing on that idea, the idea of how to invoke that procedure (call a method named java.lang.String.compareTo, passing two pointers to String, returning an integer) should also be able to be reimplemented without infringing.

If I said the way to invoke that was to call subroutine number 3246 instead of giving it a name, would you have a problem with that not being protected? Is it the names?

The only thing an API provides is functionality. An API is a functional idea. The source code is copyrighted. The functional idea is not supposed to be protected by copyright. It doesn't matter why you copy it, it's not protected, just as I can copy an algorithm (unless it's patented), for any reason I want.