tyrion

Thank God Oracle lost its case:

https://www.techdirt.com/articles/20...fair-use.shtml

Joe Bikerider

Yea!

ahsposo 

Of course, this is like Cthulhu losing to Satan.

tyrion

This goes way beyond Google and Oracle. If Oracle would have won it would have really screwed up the software world.

Juan Foote

I would not go so far as to say "screw up", but it damned sure would have made Oracle the Gatekeeper, big and rich as hell after everyone had to pay them royalties/licensing.

blacknbluebikes 

Poor Larry. Not.

J.Higgins 

I skimmed through that article long enough to realize how far it is beyond my capacity to understand. I envy you guys who live and breathe this stuff, which probably comes as easy to you as me tying my own laces. I'm glad to have people like you around to keep us straight!

Seattle Forrest

It really would have been bad in a lot of ways for most tires of computers including phones. We would lose a lot of flexibility and the ability of many things to work together. The rest would be more expensive. A lot of ISVs would be sued out of existence sooner or later.

Seattle Forrest

For what it's worth, the article made it clear as mud what the tech stuff meant. It was quoting the courts a lot and having what an API is explained by people who don't understand what it is just isn't going to be understandable.

J.Higgins 

Hah! For what its worth, I still don't know. Clear as mud. I like that expression. It fits!

Dan Burkhart 

Good to know I'm not alone. Heck, we may be the majority.

J.Higgins 

I'm thinking that I may have become my father. What is funny, is that I'm no stranger to the world of computers. I've had some form of home computer since about 1984. My brother and I would swap programs and floppy discs, and have conversations about computers and electronic games, and my dad would sit there smoking his Luckys and taking it all in. He'd irritate me with his stupid questions, and even though I was an adult, I didnt have the ability to understand just how foreign all this was to him. He didnt have a clue then, and I don't have a clue now.

zandoval 

Ha Ha... You got it... JAVA is just another buggered language. Microsoft, Google, Oricle... and the list goes on including Anderoid.

Been using Debian Linux for decades and have seen the JAVA fight go on and on... Even to the Supreme Court!

Seattle Forrest

API = application programming interface. It's the last word that's important.

An API is the way programs allow programmers to ask the programs to do something.

If you want to display a picture on the screen, there's a lot that has to happen, and you don't have to write most of it yourself. The computer needs to get the image data from a file, from the internet, or wherever, decode the way it's stored into something that can be displayed, and show it.

An API is the way you ask the operating system to do that. The API doesn't say how to do it. How to do it is what programmers are paid to write code for. How to do it is subject to copyright. How to ask for it to be done is what's at issue here. It's a little bit like how your house is your private property and your address isn't. Your address is basically how to get to the important part.

One API might require you to carry out a task in 3 steps, and another might let you do everything with one command. If you want to display a picture, maybe you have to load the image into memory first because the API for drawing on the screen requires an image; a different API will accept a file name or url instead of the 1s and 0s that make up the image. It's just two ways of asking other software to do the same thing.

Being able to copy an API (but still having to do the work of telling the computer how to carry those tasks out) means programmers can write their programs once and they can be used in more than one context, ie their programs can work with more than just one other program. Not being able to do that means those programmers have to write code specific to each context. Which means either their software is more expensive and less innovative, it's of lower quality, or it just doesn't support as much and is less useful.

I don't know if that's any more clear or not, I hope it helps a little.

rwh 

I was disappointed they didn't rule that APIs aren't covered by copyright. Calling it fair use leaves it open for another round next time someone decides to use another companies API.

RubeRad

I need to read that article, I want to understand it more thoroughly, BUT, without having read it yet, the keyword is INTERFACE (the I in API).

Here's another analogy: In a car engine, the water pump or alternator, has a certain interface to the rest of the engine; bolt holes this size in these places, a mating surface this shape, hosepipes or loopbelts of such and such length that have to be at this position relative to the bolts/mating surfaces, etc.

If a non-OEM company wants to manufacture replacement water pumps or alternators for a certain engine, they are free to do so. The interface is not protected. The inner workings of the Ford or Honda or whatever specific water pump or alternator might well have patentable technology, but not the interface. And the non-OEM implementation of the device may look different in its outer shape, color, or (separately patentable!) inner workings -- all irrelevant details to whether the device conforms to the matching Interface.

Another example; you can nowadays replace fluorescent light tubes with LED tubes. They work in existing fixtures because the conform to the T4 interface; a 4-foot tube with two prongs of a certain shape/size/position on either end, which respond to a certain volate/amperage of alternating electric current. But the inside is a completely different implementation. So if somebody had a patent on fluorescent lights, or even standard T4 tubes, would these LED replacement makers be in violation of that patent just for reusing the interface? I don't think anybody would reasonably think so.

So when Oracle complains that Google 'stole' the Java API, what does that mean, that Google wrote their own implementation of Java, which can also compile and run the same Java source code that Oracle/Sun Java can? Or is this about how other languages (C++? Python? Javascript? Go?) might access Java-implemented capabilities through an inter-language interface?

RubeRad

Is this case more or less important than when Unisys lost (?) their claim to IP rights to gif?

Or that case a few years back where a patent troll was claiming IP rights to "podcasting", and threatening pricey lawsuits against all kinds of big podcasters?

tyrion

Here's an analogy: I buy the company that made the first modern American standard electrical plug. Then I say "anyone that makes these kinds of plugs now has to pay me a fee", even though hundreds of companies have been making that kind of plug for decades. That's what Oracle tried to do - just a corrupt money grab.

tyrion

Software developers have been using other people's API's for decades. It's a long standing industry practice that Oracle tried overthrow.

Oracle's whole existence is based on SQL, which was invented at IBM, so it's insanely hypocritical for Oracle to be pulling this stuff now.

tyrion

Debian Linux wouldn't exist if APIs were copyrightable.

RubeRad

That's kind of the point of an API, you expose the API to the world in the hopes of attracting integrators that will create capabilities that use the API, which means that stuff you wrote 'behind' the API gets more use (and depending on your business model, you get more $$)

RubeRad

OK I skimmed the article, not completely thoroughly, but found some good stuff. From Breyer's majority opinion:



This is a great point, and actually one in Oracle's column. Once a great interface is designed, the implementation written to fulfill the guts behind that interface has a much better chance of being great. If the interface sucks, the whole enterprise is doomed. So a quality, usable, robust, well-architected API is a thing of craft and value, even though it's not the implementation itself. Much as a Frank Lloyd Wright blueprint is a purer representation of his creative genius (intellectual property?) than the resulting constructed house itself.

Here, the 'declaring code' is the API, vs the ''innovative implementing code" is what happens inside/behind the API when users invoke API functions

More about 'attracting programmers':

Not sure that's the strongest point there, it basically reduces to 'Google stole it because it was too hard to make something new itself'.

Interesting though that Breyer seems to have understood the heart of the issue so well. "Not a bad description for an 82-year-old Supreme Court Justice (kudos to his clerks and the various parties and amici that briefed this)"

RubeRad

https://barrgroup.com/software-exper...pute-nears-end

I did a little googling (actually ducking-ducking-going) and couldn't quickly turn up a reference showing exactly those 170 lines in question.

Anyone else know how to put their finger on them? Or the 11500 lines Oracle alleged?

tyrion

I believe the "11,500 lines" was a standard set of Java interface declarations (that one would routinely submit to the compiler when compiling a Java module), but upon analysis Google was only using a small portion of that in Android, "170 lines" of it. I've never seen those 170 lines identified.

RubeRad

I'd love to see those 170 (presumably they were entered as Exhibit X at some point?) It would help me understand it, and perhaps explain to kids what's the root of what's going on.