Submitted by The ongoing legal saga known as the Oracle-Google copyright battle took a huge leap Wednesday when Oracle claimed the last six Android operating systems are “infringing Oracle’s copyrights in the Java platform.”
That’s according to the latest paperwork Oracle filed in the five-year-old closely watched case that so far has resulted in the determination that Application Programming Interfaces (APIs) are, indeed, copyrightable.
Oracle is the cesspit of the industry. What a horrible, horrible company.
James Gosling, http://nighthacks.com/roller/jag/entry/my_attitude_on_oracle_v
Edited 2015-08-14 17:33 UTC
Using APIs is not stealing. Even portraying it as copyright infringement is highly questionable.
Using APIs and therefore header files should be “fair use” I agree as a matter of compatibility especially since it would severely damage an already artificially strained software industry with all the questionable software and design patents issued by a rubber stamp USPTO. Under US copyright law it’s not at all clear if header files that form an API can be covered by a copyright. The Appeals Court has already ruled that the judge erred on point of law on that ruling and it requires further review. Artificial words and languages *are* covered by US copyright law as meriting copyright protection. And whatever we may consider computer languages as generally fair use, they are entirely artificial.
I certainly agree that Oracle is a company with no ethics and a horrid PR record, but they have brought up valid questions as far as US copyright law is concerned and the question on functional header files needs to be addressed rather than assumed it’s alright to borrow them. All players in the software industry would rather that can of worms be left closed, but since it is opened, this needs to play out and US copyright law needs to be brought to the reality of what copyright is going to have to become in an increasingly electronically connected global community. Will US copyright (and patent) laws reasonably accommodate interoperability, or will they continue to stifle it?
The whole issue of header files came up in SCO vs IBM
Google for “SCO errno.h” and you can see the whole story.
That case came down on the side of IBM and that files like ‘errno’h’ were not subject to copyright In that case
While (IANAL) header files do not an API make, they go a long way towards making the case that they aren’t copyrightable.
As Thom says, Oracle (in particular their CEO and legal dept) are scum. (I was a former Oracle DBA, much to my shame now)
A bit irrelevant here though, as Java does not have header files.
Temcat,
The court with the highest authority in the US refused to block API copyrights, so now we’re stuck with them. “Highly questionable” indeed. Actually no, that’s too kind, it’s f#@*!1ng moronic!
I think it’s bad for developers, but good for businesses. If I spend hundreds of thousands developing and implementing an API for my platform, at least I know my competitors can’t steal it for themselves and poach my customers without giving me something in return.
It’s bad for businesses too. If you think the patent trolls have gotten out of hand now, imagine what we’re going to get when everyone can claim copyright of a print function?
But is it? It may be a good marketing ploy for companies to say “buy our code because we won’t screw you over by enforcing API copyrights”. To paraphrase a well known adage: modern software development sees copyright as damage and codes around it.
Nobody’s going to claim copyright on a print function. If you rip off entire schemas, however …
WorknMan,
You suggest the motivation for using the same API is to rip off the original work. However I think that’s a very limited world view and you are missing the most common reason for copying an API to begin with: compatibility.
Asserting that an API is copyrightable implies that third parties cannot create their own compatible implementations of software. Obviously this was never the intention behind software copyrights.
Look at ReactOS or Wine, those explicitly copy the windows API so that they can run windows API programs (especially those which microsoft holds no copyrights to). These projects, in your words, “rip off entire schemas”. The fact that the implementation is completely independent is irrelevant, since API infringement by itself is now a thing.
Maybe you overlooked this, or maybe you really believe that third parties should not be allowed to build compatible alternatives? I do hope you can come to appreciate that there is public value in allowing the copying of APIs and it’s not all about ripping off the original creator.
Yes, I understand all of that. But sometimes, public value isn’t always worth those who spend a tremendous amount of time and money creating a thing having to take it up the ass when every ‘me too’ copycat artist comes along and piggybacks off their work, without giving any compensation in return.
Of course, this could cause some fragmentation in the industry, but I also think it forces companies to think outside of the box and come up with something more creative than just, “Hey, this is the same as the other guy’s thing, but just on a different platform …”
To be clear, I don’t entirely disagree with what you’re saying. I’m just trying to present another point of view here.
Edited 2015-08-16 01:35 UTC
WorknMan,
Fair enough.
Well, you can copy IDEAS and make a similar story, but you can’t reuse the exact characters and plot points, and then simply rewrite the story in your own words. Or, at least not without obtaining the rights to do so. For example, if you wrote a book about a girl named Dorthy who takes a trip down the yellow brick road and meets a scarecrow, tin man, etc, you’re probably going to get sued into oblivion.
Bad example, as the ‘Wizard of Oz’ is public domain, but I get your point.
The key to this case is going to be the fact that Oracle is trying to use copyright law to do the same thing patent law does. That is, control the implementation of an idea, not an expression of it. If APIs are copyright and cannot be used without permission, that means the rights holders can stop any clean room implementations (such as Dalvik) from being created.
Personally I feel that we need to stop using copyright to protect code. It was necessary at one time, but it’s a poor fit. The algorithms we use are pretty universal, often differing ONLY by names, and how we choose to express them has more to do with the language and platform than personal taste. But I also feel patent law is a poor fit too, and even harmful to the progress of the sciences and useful arts when applied to code. Sun used Trademark law to protect Java from incompatible forks simply by saying you couldn’t call such forks Java.
Code occupies that weird netherworld where Expression IS Implementation, and I don’t think things will get better until we (and the law) recognize that.
I am pretty sure that that “tremendous amount of time and money” went into the implementation of the API, not the design of it (you could clearly see this for some of the Java API’s anyway…). Hence your point is mostly moot. Not fully; it is conceivable that somebody spends a lot of time and money to come up with THE perfect API, but most of the time, take any API, and you could devise an alternative one in half an hour tops. So unless Google also uses Sun’s implementation, I really don’t see how this case could stand in court.
So, why don’t they do that then? Obviously because the value in an established platform for developers is its API.
Tell that to Righthaven and those like them. Won’t take much for them to go from patent to copyright trolls. Then again, from the way you’re acting, anyone who dares make a compatible API is a thief. There are so many things wrong with that way of thinking that I don’t think I can even try to enumerate all of them.
I’m afraid that working out where this line is drawn, and what comes under fair use as a result, is where Oracle will come a cropper.
You know that we are talking about the abstract API and not the actual code behind it, right?
Not necessarily, according to this article, Oracle are alleging “these six Android releases copy thousands of lines of source code from the Java platform, as well as the structure, sequence and organization of that platform” as well.
http://www.computerworld.com/article/2970944/android/oracle-google-…
If that is the case, they’re funked.
Oracle are having to allege that because their case is dropping to pieces. They’ve tried to go the copyrighted API route, and got the Supreme Court to back them, but they still have absolutely no case. Either that or their lawyers are just outright confused.
There has been no code theft at all. A lot of Google and Android’s API looks similar, and some code you can recompile from Java straight over, but the implementation behind it is completely different. They’re now trying to argue that they have been ‘harmed’ to get around arguments of fair use.
They just don’t know the difference between APIs and software.
Edited 2015-08-17 11:47 UTC
Copying lines of code is very different from implementing an API. The code is, and should be, copyrightable. The API, not so much.
Yeah, it’s worth to remember Gosling’s words, that give a non-Oracle perspective, but why Gosling ended up working for Google?!
OK, it was only for 5 months but still a questionable choice
Just because he created the (horrible) language doesn’t mean his right on this account.
That’s why Sun is bust and Oracle really doesn’t have a future. Sun should have made something spectacular out of Java long before Android ever came on the scene. However, they sat about on their backsides and then expected to sue when someone did something with Java’s API – and all Google really took was a limited API so Java apps and developers could switch over. Sun could have tried creating some decent development tools, taking Dalvik, getting involved….but no.
The court should go back and fix that. Next time they should get judges who can understand technology.
Yeah….I’m not sure you’re ever going to see that.
A lawyer who is fluent in IT, coding, or technology? Good luck with that.
EFF has some. What we need is a judge like that. And the one who ruled first that APIs can’t be copyrighted actually knew how to program. It’s the appeals court that got some dumb judges who had no clue what they were doing.
Edited 2015-08-14 22:17 UTC
In the US, judges don’t even have to understand law. They are elected.
For example, this case (http://q13fox.com/2015/08/14/group-files-complaint-against-judge-wh…) indicates a judge with no knowledge of law, or even decency.
In this country? You’ve got quite a powerful idealist hat on today.
It’s a pretty random thing. That’s why sometimes such rulings are sane, and in other cases they are insane. Kind of a roulette outcome.
It will be. We’ve been through this before in the 80s and 90s and been round in circles so many times.
What’s funny is that they obviously knew that copyrighting APIs was a problem so they had to drop in the ‘fair use’ clause. However, everything comes under fair use because that’s the reason you use an API from another language. Google can’t fail to argue it.
is hyping the wonders of its OpenUI environment for various Siebel enterprise sales and service database apps we are efforting to upgrade this year.
Oracle’s design defects and inadequate support for fixing same are creating a minor cluster f*ck in our ability to complete SIT and UAT testing within required time frames. A clown show with the big ‘O.’
Java blows