SoylentNews is people
SoylentNews
from the chickens-coming-home-to-roost dept.
Google has appealed its case with Oracle to the Supreme Court of the United States over a dispute about whether a java API may be copyrighted.
The ARS Technica Story: Google asks Supreme Court to overrule disastrous ruling on API copyrights.
The consensus among the comments on ARS seem to be that this will result in a substantial amount of litigation. I'm forced to ask whether this future litigation should have happened already, and that this paradigm shift is just a matter of catching up to the effects of a previous bad ruling Lotus v. Borland from the early 1990s.
I get that it is going to cause a lot of code refactoring. But won't that also create a lot of new products? The other thing to consider is that communications protocols are essentially API specifications. Historically intellectual property protection for protocols is very weak. Some (myself at least) would attribute the effectiveness of the EEE business tactic to the inability to defend protocol compatibility within the legal framework provided by the USPTO and Copyright Office.
Which is to say that an Oracle victory may expand the scope of FOSS licensing, giving FOSS developers more say over how their products are used. This would reduce barrier to entry in new communications product markets that are based on FOSS, and give more power to startups.
Yes it is going to be expensive for established players if Oracle wins. Most people seem to agree with that. That is the price of operating on a bad premise. Does it matter whether responsibility for the premise resides with a judge or a CEO? Aren't there also some upsides if Oracle wins? What are the trade offs?
Of course the whole thing could be a put on.
(Score: 4, Insightful) by aristarchus on Wednesday January 30 2019, @08:41AM (14 children)
Stupid questions beget really stupid answers.
The correct answer to this one is "no", followed by "damn no!"
Do not people know what a API is? Is it an Interface, a Public interface, that is Authorized. To copyright this would be to attempt to deny any one of these three things. I am shocked, (well, not so much after the mass rejection of insightful aristarchus submissions) that a Fallacious Article like this could eve make it to the front page. Proud Boy, but Not Too Bright Boys, strike again?
(Score: 0, Funny) by Anonymous Coward on Wednesday January 30 2019, @09:08AM (1 child)
Well then, why didn't YOU write the article so it would never see the light of day?
(Score: 4, Funny) by aristarchus on Wednesday January 30 2019, @09:12AM
I did. You didn't see it? In the light of day? Oh, dear!
(Score: 2) by MichaelDavidCrawford on Wednesday January 30 2019, @11:18AM
I don't authorize you to implement "void foo( void );"
Note that there is quite a lot more to an API than just a header file prototype; there is also - hopefully - between the caller and callee: for example, my above API holds that calling it will panic the kernel, take your daughter's virginity then feed your hamster. None of these are implied by its prototype.
Yes I Have No Bananas. [gofundme.com]
(Score: 2) by The Mighty Buzzard on Wednesday January 30 2019, @11:29AM (6 children)
The issue here is they're trying to copyright the names of the API calls so that someone can't clean-room up their own version of $whatever that otherwise violates neither copyright nor patents.
Now, am I looking at this wrong or could this be a very unfortunate thing for wine if the ruling stands?
My rights don't end where your fear begins.
(Score: 3, Interesting) by c0lo on Wednesday January 30 2019, @11:56AM (4 children)
[Citation needed]
The way I read [forbes.com], names are not copyrightable. One can write a book having Harry Potter as a character name, as long as the character and setup doesn't bear any substantial resemblance to JKRowlings novels.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 4, Insightful) by isostatic on Wednesday January 30 2019, @12:23PM (2 children)
Probably, you'd have to be careful over trademarks to avoid confusion. And that's a good thing - it protects someone who's looking to buy your book. If you released "Harry Potter the boy wizzard", or even "The tale of middle aged Harry Potter", it would be reasonable for a purchaser to think this was an authorised continuation of the Harry Potter world.
If you want to make your own version of Harry Potter, do so, and let it stand on it's feet. Seth McFarlane did it with Orvile (and trek), he didn't need to call the ship "Enterprise", or the captain "Kirk"
(Score: 2) by The Mighty Buzzard on Wednesday January 30 2019, @03:12PM (1 child)
That's trademark law not copyright law.
My rights don't end where your fear begins.
(Score: 2) by isostatic on Thursday January 31 2019, @01:36PM
Yes, hence me saying "you'd have to be careful over trademarks"
(Score: 3, Informative) by The Mighty Buzzard on Wednesday January 30 2019, @03:11PM
The initial judge taught himself java and agreed with you though not for that reason. The subsequent appeals courts did not.
My rights don't end where your fear begins.
(Score: 2, Funny) by Anonymous Coward on Wednesday January 30 2019, @03:50PM
Well, I'm not a oenophile, but I don't think this will affect wine in any way. ;-)
(Score: 5, Touché) by c0lo on Wednesday January 30 2019, @11:43AM (2 children)
All good and dandy, except:
(still grinning) So, you see, you are acronymically wrong. But I must admit, nobody surpassed you in the elegance in being wrong on purpose, magister, your style is the supremerest!
---
** All cited works belong to their authors. Their partial reproduction here was fully attributed and was made under the "fair use" doctrine.
So, esteemed author, not a cent from me.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by aristarchus on Wednesday January 30 2019, @11:56PM
Guilty as charged. Knew I would be caught, just didn't think it would be c0lo that would do the catching. Seems better this way.
(Score: 2) by All Your Lawn Are Belong To Us on Thursday January 31 2019, @05:38PM
A Punishable Interface?
This sig for rent.
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @05:17PM
The answer should be yes*.
*With conditions.
Because corporations fucked up what was supposed to be friendly use among people.
corporations are not people, but they will try to fuck people. Their use of APIs in truly novel profit seeking ways needs to be curbed, and they will not do so if it erodes their revenue--especially if publically owned, and then they will blame it on needing shareholder value or something like that (despite being a dick not also being a legal requirement).
I am totally fine with taking away perks and conveniences from the little person in order to majorly shaft the guy that is beating the little person with it.
Then, with the *conditions, let the little person go back to legally using the APIs as nature intended.
Nobel made a peace prize because of what happened with his good intention inventions. There will never be an API prize from any of those companies, thanking people for releasing stuff open source so that they can exploit it.
(Score: 3, Interesting) by pkrasimirov on Wednesday January 30 2019, @08:57AM (2 children)
Means I can haz (c) on
int my_precious(int)
then ask for compensation from everyone who dares to steal my property.
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @03:56PM
Just because you have a copyright on a novel you wrote does not mean that you also have a copyright on a random three-word sentence you wrote. Applying this principle analogously to your example API is left as exercise to the reader.
(Score: 0) by Anonymous Coward on Thursday January 31 2019, @01:28PM
The author of foo() and bar() is going to have a field day!
(Score: 5, Insightful) by stormwyrm on Wednesday January 30 2019, @09:09AM (6 children)
They'd be products that didn't need to exist, are a waste of programmer-hours that could have better been spent not reinventing the wheel. It will make interoperability expensive if not impossible. In many ways this misguided ruling is worse than software patents, since copyright basically lasts forever thanks to Disney. Imagine if Nokia (the current owner of Bell Labs) decided to exert its putative copyright over the foundations of the Standard C library API. Since copyright lasts essentially forever, they could still conceivably exert that right over this enduring legacy of their late, great, and much lamented employee Dennis Ritchie. They could sue any implementer of this clearly still-under-copyright API. If this were a thing in the early nineties Linux would never have existed either, as AT&T would then have had the legal basis to sue even Linus Torvalds back then for infringing on "their" API, just as they sued the BSD folks. I'll take the risk of Embrace, Extend, Extinguish instead, thank you very much. That's a pretty damn paltry upside to all the major downs.
Numquam ponenda est pluralitas sine necessitate.
(Score: 2) by krishnoid on Wednesday January 30 2019, @09:34AM
If copyright lasts forever, so would copyleft. Who knows -- requiring a break with old APIs could mean new copylefted systems targeting modern hardware written from the ground up, and lock-in or lock-out to a growing copyleft ecosystem. It would at least be fun to watch.
(Score: 2) by c0lo on Wednesday January 30 2019, @11:04AM (1 child)
Stop the fallacious breaking of the window of ©. Start breaking Windows™ instead, it's only right to do so.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 1, Funny) by Anonymous Coward on Wednesday January 30 2019, @06:54PM
Someones beat us to it.
(Score: 2, Disagree) by c0lo on Wednesday January 30 2019, @12:25PM (2 children)
While I somehow agree with the "don't re-invent the wheel if you don't want to" the "utilitarian" argument is flawed if pushed too hard.
In extreme, the "don't reinvent the wheel because it's a waste of programmer-hours" translates into:
* don't reinvent Windows, Linux has no need to exist and is a waste of programming hours
* don't sculpt yet another nude, thousands already exist and any new one is a waste of artist-hours
* "Westend story" is a total waste of playwright-hours, we already have "Romeo & Juliet"
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 0, Troll) by Anonymous Coward on Wednesday January 30 2019, @04:06PM
Linux is based on the Unix API, which is older than Windows. If anything, Windows is the reinvented wheel. And I'm sure a lot of people here do think it was a waste of programmer hours. ;-)
If it is just-another-nude, then most probably it is a waste of artist-hours.
If it simply retold "Romeo & Juliet" in different words, it probably would be.
(Score: 0) by Anonymous Coward on Thursday January 31 2019, @09:39AM
(Score: 3, Insightful) by pTamok on Wednesday January 30 2019, @09:52AM (8 children)
While the USA is doing it's darnedest to export the idea of extra-territorial applicability of its laws, the ruling, at present, applies to the USA. For at least a short period, use of APIs in software published outside the jurisdiction of the USA would remain non-copyrighted, until trade treaties and/or the Berne convention and related instruments catch up. That makes for an 'interesting' competitive environment.
Much like the USA's 'noble experiment [wikipedia.org]', it will be interesting to watch, but not to participate in.
(Score: 2) by MichaelDavidCrawford on Wednesday January 30 2019, @11:12AM (7 children)
"The Noble Experiment" was Prohibition.
While Article 1 of the 21st Amendment repeals the 18th Amendment, Article 2 permits "Any State, Possession or Territory" to forbid the "transportation or importation into" it, thereby enabling the Dry Counties that are still found here and there in the Deep South. I regard Article 2 as taking back the actual repeal, because it enables States, Possessions and Territories to Prohibit alcohol.
I'll Drink To That! hic!
Yes I Have No Bananas. [gofundme.com]
(Score: 2) by Thexalon on Wednesday January 30 2019, @12:59PM (6 children)
In Utah, bars and most other places are only allowed to sell 3.2% ABV beer, and if you want anything stronger you have to go to a government-run liquor store. This is largely due to the influence of the thoroughly anti-alcohol Latter-Day Saints, and thus shows the dangers of doing too much LDS.
I'm just as glad I don't live there for this reason.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.
(Score: 1, Informative) by Anonymous Coward on Wednesday January 30 2019, @01:30PM (3 children)
Huh?
Bars here sell all kinds of stuff way stronger than that.
You're confusing bar with grocery store. You can't buy anything at the grocery store including cooking sherry if it contains more than 3.2% alcohol. For that you need to go to the liquor store and yes those are state run.
There has been quite a bit of push back though on this and change is in the air as the state begins to realize what an embarrassment it is to force a professional chef to go to a state owned liquor store just to get a cooking wine. Or to have the zion curtain where the bartender in a restaurant cannot be visible to seated patrons lest they see him pour liqour into a glass as part of someone's meal.
A large part of this is the fault of the LDS church, but keep in mind there are entire counties outside of Utah that are completely dry and they have nothing to do with being Mormon. Puritans gotta be holier than thou and Teatotallers gotta teatotal.
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @04:26PM
According to wiki, beer only bars are restricted to 3.2% beer too.
(Score: 2) by Thexalon on Wednesday January 30 2019, @06:01PM
When I was in Utah visiting my sister a few years ago, all we could get in local bars was 3.2 beer. From what I understand there have been efforts to loosen the rules, but it's not like this isn't a real issue.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.
(Score: 2) by MichaelDavidCrawford on Wednesday January 30 2019, @10:38PM
Nova Scotia outlawed opening large stores on Sundays until I think 2005 when some worth-his-every-loonie attorney conceived of subdividing each large store among a whole bunch of really small stores.
The province still didn’t want to legalize Sunday shopping but relented due to the absurdity of everyone Sunday shopping anyway.
Yes I Have No Bananas. [gofundme.com]
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @06:01PM (1 child)
There is an important error here: the Utah limit to which you refer is 3.2% alcohol by weight, which is ~4% ABV (alcohol by volume).
(Score: 2) by Thexalon on Wednesday January 30 2019, @06:11PM
Either way, the stuff didn't taste half as good as the regular beer I could get in my home state. I don't enjoy being drunk, but I do enjoy decently flavored beer, and that wasn't it.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.
(Score: 2) by shortscreen on Wednesday January 30 2019, @10:03AM (1 child)
Push your left hand on the stack. Push your right hand on the stack. Jump to this location.
(Score: 3, Touché) by choose another one on Wednesday January 30 2019, @12:20PM
Sounds more like a derivative work of the hokey cokey...
(Score: 2) by MichaelDavidCrawford on Wednesday January 30 2019, @11:05AM (3 children)
Note that I do _not_ support improperly issued patents - those that are obvious, not novel, whose description of the invention is obfuscated or that is insufficient to enable one who is skilled in the art to reproduce the invention.
_None_ of these art particular to software patents; obvious inventions get patented all the time, which I feel is Just Wrong.
What I _do_ support about patents in general and software patents in particular is that to patent an invention that is simply Good Enough often results in Better inventions; for example Ogg Vorbis is better than MP3.
In 1995 I invented a lossless bitmap graphics compressor that I conceived of during my lengthy commute over the course of a month. I was only given three weeks to implement it, including the file format code. It's about as good as GIF's LZW - about 50% compression of typical photos - but is dramatically faster to decompress, which was important on the far-slower CPUs of the day.
I had never heard of the particular method I had for compression. I've not heard of it being used since.
I asked my employer to patent it. That would have named me as the inventor but Medior as the assignee - they were a multimedia CD-ROM consultant. I came up with the idea when, after specifically advising our product manager to create a space budget in a spreadsheet he said "Sixty hundred fifty megabytes ought to be enough for everyone".
And it wasn't.
The Classic Mac OS PICT compressor was blazingly fast, but using GIF on windows was quite slow, so that the user could not readily page through a home shopping catalog on CD. With the images compressed offline with my compressor, then decompressed on the fly as each new page was entered, because of the smaller size of the filesystem reads, to use my decompressor was faster than using uncompressed images.
They never filed for a patent as the company got bought by AOL during the Dot-Com Boom, who really only wanted our graphic artists.
At the time - maybe now I don't don't know - one had only one year to file an application after one "brought the invention to practice" - ie. made one. I've since come up with numerous improvements so that In My Infinite Free Time I _will_ file for a patent on my own.
(One can patent improvements to any invention at all.)
Yes I Have No Bananas. [gofundme.com]
(Score: 2) by isostatic on Wednesday January 30 2019, @12:16PM (2 children)
The problem with patents is that while you may have independently come up with that great method, and spent weeks/months/years implementing it, it's entirely possible someone else has too, and has either patented it already in some impossible-to-find patent, or will find that they have to stop doing what their nifty library does because you come along and claim ownership of their ideas.
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @07:30PM (1 child)
The patent rules are supposed to prevent that. This scenario would mean the patent fails to be non-obvious to experts in the field.
(Score: 2) by MichaelDavidCrawford on Wednesday January 30 2019, @08:00PM
The description of the invention is specifically required to be "clear", as well as to enable - I forget the exact words but something like - "one practiced in the art" can reproduce the invention.
A while back I went to the Sunnyvale Patent Clearinghouse where lots of hardcopy patents were kept; if I remember correctly, the photocopier was free.
The LZW patent was seventy pages. I found it completely unintelligible.
Despite this, I've never head of a patent being quashed for being unclear, and have been meaning - in my infinite free time - to point out to Richard that to challenge unclear invention descriptions could be a useful avenue of attack.
So why would I both support patents and Richard Stallman?
Because I oppose _bad_ patents, not so much software patents as bad patents are found in many fields.
Yes I Have No Bananas. [gofundme.com]
(Score: 3, Funny) by Debvgger on Wednesday January 30 2019, @12:11PM
Short answer: NO.
Long anser: NOOOOOOOOOOOOOOOOOOOOOOO!
(Score: 2) by bradley13 on Wednesday January 30 2019, @12:45PM (3 children)
An API is "a set of clearly defined methods of communication". In many, perhaps in most cases, the backend of the API is implemented. The implementation may well be a commercial offering. For example: if you use our web service, you have to pay us. The key question in this case is: If someone else implements a new backend that follows the API, does the originator have any legal claim?
For copyright, the key question becomes: Can you claim copyright on an API? Is an API a creative work, or just a mechanical listing?
Unfortunately, I can see arguments on both sides. On the side of the entire IT world, the argument is simply that no one has ever asserted copyright on an API. Hence, if precedent carries weight (which it does, under US law), APIs are not copyrighted. On the legalistic side: When you first create an API, you are not just mechanically listing methods and parameters - because these do not yet exist, and are yet to be defined. The API is, in fact, something that you create along with your initial implementation. If you software is copyrightable, the API is just an aspect of this. I don't like that legalistic conclusion, but I can see how nontechnical judges may follow it.
The real solution, of course, is to fix copyright law.
Everyone is somebody else's weirdo.
(Score: 3, Informative) by ilsa on Wednesday January 30 2019, @06:04PM (1 child)
While there are potential arguments on both sides, Oracle's side opens a ginormous can of worms that is liable to disrupt ALL software development.
Software developers code from the shoulders of giants. There is virtually nothing today that hasn't been in some way derived from past work. That includes virtually all APIs.
Everyone will have to completely and needlessly reinvent the wheel for every single god damn thing, just to avoid potential lawsuits.
It's the equivalent of saying that Ford has copywrited the gas pedal, so now no other car manufacturer is allowed to use one and every car has a completely different way to drive it.
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @10:30PM
The fun part is that if Oracle's argument prevails, they're going to immediately be sued for their use of Structured Query Language, as that would just as obviously be a copyright infringement under their own arguments here.
(Score: 2) by darkfeline on Wednesday January 30 2019, @07:46PM
It's important to remember that copyright is a hack. It is a limited license for a monopoly that is provided in the hopes that:
1. This encourages creators to create more stuff because they can exploit the monopoly to make money.
2. Creating more stuff is good for society.
If we apply this to APIs, it is clear that we do not want them to be copyrightable. The last thing we need is more APIs/standards, so providing an incentive for people to create more and more APIs (and license them for profit) is horribly bad for society.
Join the SDF Public Access UNIX System today!
(Score: 1) by shrewdsheep on Wednesday January 30 2019, @01:00PM (2 children)
When NeXT got into trouble they partnered with Sun. Their collaboration was to create OpenStep. My understanding at the time was (and still is) that this meant that the API could be freely re-implemented. It kicked of the GnuStep project (which is still somewhat alive BTW) as one of the re-implementers. Historically, the situations seems less clear-cut than Google claims. Certainly, nowadays the expectation is that a published API can be re-implemented. I do believe that the API is more relevant than the implementation and I would argue that it should be possible to protect APIs. Of course, that must be made explicit and documenting APIs and tolerating re-implementations (like is the case with java) implies a public license (I side with Google but for different reasons). Also a (near)monopoly should not be able to hamper re-implementations (Windows/Wine) on grounds of competition.
(Score: 2, Insightful) by Anonymous Coward on Wednesday January 30 2019, @02:26PM
If your implementation of $whatever is the only one, how can you NOT be having a monopoly on it?
If you then trap your users into your implementation through patents (for 20 years) or copyrights (for bloody ever) on interfaces, how are you NOT abusing your monopoly AND your "intellectual property"?
Such "inventiveness" got slapped down once already in the Lexmark case:
https://en.wikipedia.org/wiki/Lexmark_International,_Inc._v._Static_Control_Components,_Inc. [wikipedia.org]
(Score: 3, Informative) by stormwyrm on Thursday January 31 2019, @09:26AM
Numquam ponenda est pluralitas sine necessitate.
(Score: 2, Insightful) by Anonymous Coward on Wednesday January 30 2019, @03:28PM
Should pedals, stick gear shift and steering wheel be patented?. Imagine that, every car maker should make their own driving system. Some with buttons, some with sticks, some with levers. That would be a total chaos and don't mention getting your driver's license, because you'll need to take the car's maker driving course and get certified for it. And good luck trying to change car maker brand, It would be a really crazy world.
NOTE: In The begining it may have been patented, but in some pint in time the patent expired or some government office of normalization made it the standard that every maker should abide.
(Score: 2) by meustrus on Wednesday January 30 2019, @05:48PM (2 children)
What is the EEE and what are its business tactics? I searched for what that acronym stands for and got horse disease and psychedelics; the only computer-related thing was "Energy-Efficient Ethernet".
I read an implication that whatever this "EEE business tactic" is, it's bad news. I can't imagine the author's problem with the developers of "Energy-Efficient Ethernet", the IEEE. For that matter, EEE could just be a misspelling of IEEE. But it still makes no sense. Aren't we so much better off with portable standards than we were in the days where every computer had its own incompatible concepts for storing data, processing instructions, reading from disk, loading and running programs, and communicating over the network?
If there isn't at least one reference or primary source, it's not +1 Informative. Maybe the underused +1 Interesting?
(Score: 3, Informative) by http on Wednesday January 30 2019, @07:11PM (1 child)
Embrace, Extend, Extinguish.
Microsoft's original mantra - their perennial and consistent long game.
I browse at -1 when I have mod points. It's unsettling.
(Score: 2) by meustrus on Monday February 04 2019, @06:25PM
That makes so much more sense!
...can we not use unexplained acronyms in summaries, especially when Google doesn't know the acronym in question?
If there isn't at least one reference or primary source, it's not +1 Informative. Maybe the underused +1 Interesting?
(Score: 2) by ilsa on Wednesday January 30 2019, @05:52PM (1 child)
I hereby copyright the main() function, which is the standardized API call that an operating system uses to determine the starting point of an executable.
All you developers out there now owe me a lot of money.
(Score: 0) by Anonymous Coward on Friday February 01 2019, @07:58PM
Fair enough, I'll take _start().
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @06:35PM
Languages have typically not been copywrite-able or patent-able. An API is a language, or at least a sub-language. If languages are not copywritable but API's are, and the boundary between languages and API's is super fuzzy, then we'll have a messy unpredictable mass of lawyers involved.
An API is typically a sub-language that uses parts of a given language. If the sub-language is copyrightable but the "main" language is not, then people will end up making specialized languages that do what the sub-language would have done instead, just to avoid IP lawsuits. Everything would start to look like a custom version of COBOL.
(Score: 2) by shortscreen on Wednesday January 30 2019, @08:55PM
Based on the proliferation of "IBM-compatible" PCs and software that runs on them, if an API was copyrightable, 90% of everything since 1981 would owe royalties to IBM.
(Score: 0) by Anonymous Coward on Wednesday January 30 2019, @10:22PM
No. [wikipedia.org]
Next question.
(Score: 2) by All Your Lawn Are Belong To Us on Thursday January 31 2019, @05:56PM
APIs are a set of facts or ideas. Facts and ideas are not copyrightable, only a fixed expression of them would be (like a book describing an API).
Why isn't it that simple?
This sig for rent.