An API is like a menu

May 26, 2016 2:17 PM

Google prevails over Oracle: APIs are copyrightable but Android is fair use, jury says
posted by Monochrome (72 comments total)

DJ Khaled to do the next major Google company party.
posted by GuyZero at 2:21 PM on May 26, 2016

public class Reaction {
  public static void main(String[] args) {
    System.out.println("w00t! Suck it, Larry Ellison!");
  }
}
posted by w0mbat at 2:22 PM on May 26, 2016 [34 favorites]

Okay, everybody get on the BART and head over to 7th Street for the appeal!
posted by Holy Zarquon's Singing Fish at 2:23 PM on May 26, 2016 [4 favorites]

The APIs are copyrightable thing is fucking terrible news, mind, though I think that was decided ages ago.
posted by Artw at 2:27 PM on May 26, 2016 [25 favorites]

While I'm not thrilled with the "APIs are copyrightable" part of this (though we have no choice here thanks to the 9th Circuit, that wasn't the question put to the jury), this is great. If Android reimplementing the better part of the Java Class Library is fair use, it's hard to imagine what sort of use of an API wouldn't be fair use.

The scary thing is that people will still be able to use the uncertainty of a fair use determination to demand outrageous licensing fees in future disputes.
posted by zachlipton at 2:28 PM on May 26, 2016 [10 favorites]

Wow! I think the jury were just scared of google revealing their search requests, and acceded.
posted by marienbad at 2:28 PM on May 26, 2016 [4 favorites]

The APIs are copyrightable thing is fucking terrible news, mind, though I think that was decided ages ago.

Yeah, to be clear, the history of this case goes:

1) District court suit, judge holds APIs not subject to copyright.

2) Oracle appeals, 9th Circuit says APIs are in fact subject to copyright.

3) Google appeals to Supreme Court, which declines to hear the case. 9th Circuit decision stands.

4) Case returns to district court, which now has to decide whether Google actually violated Oracle's copyright.

5) Today's jury verdict says they did not -- it was fair use.
posted by Holy Zarquon's Singing Fish at 2:29 PM on May 26, 2016 [18 favorites]

Wow! I think the jury were just scared of google revealing their search requests, and acceded.

The jury-tampering potential of the worlds biggest search engine is intriguing.
posted by dis_integration at 2:30 PM on May 26, 2016 [7 favorites]

Well, that's a relief. At least here, sanity prevailed.

But I think in any case involving code, jurors and judges should follow a mandatory coding class for at least two weeks.
posted by MacD at 2:38 PM on May 26, 2016 [1 favorite]

This is from two weeks ago, but it's a really clear explanation of all this stuff.
posted by naju at 2:43 PM on May 26, 2016 [4 favorites]

But I think in any case involving code, jurors and judges should follow a mandatory coding class for at least two weeks.

Yeah, not sure if you're referencing this but the judge in this case actually took some time to learn Java and knows how to write code.

Sadly it probably wouldn't scale to make such things a requirement (and not just for software!).
posted by thefoxgod at 2:50 PM on May 26, 2016 [15 favorites]

The jurors, quite reasonably, ducked out the back rather than answer press questions after the verdict was read. But that means this is going to haunt me for a while.

(This is good news though, second only to Gucci Mane getting out of jail today.)
posted by sparkletone at 2:51 PM on May 26, 2016 [5 favorites]

While I'm not thrilled with the "APIs are copyrightable" part of this

My take on this is that the combined impact of the two rulings is that there would be a compounding factor if you copied more than just the API (or, say, republished an API that wasn't public).

This ruling doesn't seem to firmly establish that re-implementing public APIs is always fair-use, but it seems to strongly imply it.

This was likely a narrow ruling, because Oracle had very little working in their favor -- their lawsuit was a frivolous cash-grab. Sun and Oracle publicly supported Android right up until their Java ME revenues dried up (largely because, unlike Java ME, Android was usable).
posted by schmod at 3:05 PM on May 26, 2016 [6 favorites]

2) Oracle appeals, 9th Circuit says APIs are in fact subject to copyright.

3) Google appeals to Supreme Court, which declines to hear the case. 9th Circuit decision stands.

Isn't that only applicable to the 9th Circuit, though? So APIs could still be not subject to copyright elsewhere?
posted by cosmic.osmo at 3:05 PM on May 26, 2016

I'm still kind of shocked that the Supreme Court refused to hear the case considering the massive implications for software as a whole (and thus, by extension, a lot of the foundations of modern society).

Granted I am not a lawyer, so there may well be an excellent reason I know nothing about for why the Supreme Court decided what it did.
posted by Itaxpica at 3:09 PM on May 26, 2016 [2 favorites]

I don't understand any of this stuff at all but have been totally enjoying reading Sarah Jeong's twitter throughout this.
posted by latkes at 3:11 PM on May 26, 2016 [10 favorites]

While I'm not thrilled with the "APIs are copyrightable" part of this (though we have no choice here thanks to the 9th Circuit, that wasn't the question put to the jury)

2) Oracle appeals, 9th Circuit says APIs are in fact subject to copyright.

Isn't that only applicable to the 9th Circuit, though?

Actually, because there were initially patent infringement claims in the case, it got appealed to the Federal Circuit, which issued the ruling about APIs being copyrightable.

So that ruling only technically has to apply to cases that involve both issues of API-based copyright infringement and also patent infringement claims... cases with only copyright claims from Northern District of California would normally still be appealed to the Ninth Circuit, which could decide differently.
posted by cobra_high_tigers at 3:11 PM on May 26, 2016 [4 favorites]

More about the appeal to the Federal Circuit about halfway down this article from a few weeks ago. It suggests that other appeals courts are likely less sympathetic to the view that APIs are copyrightable than the Federal Circuit.
posted by DevilsAdvocate at 3:14 PM on May 26, 2016 [3 favorites]

The Federal Circuit appeared to fundamentally misunderstand the concept of APIs entirely as well as the principles of copyright. APIs are specifications that allow programs to communicate with each other, and are distinguishable from code that implements a program, which is copyrightable.

Copyright is fundamentally about protecting expression, not functionality. If you want to protect functionality, you go with patent law. Code is copyrightable because it's expression, albeit expression that has a functional purpose. APIs, on the other hand, are entirely about functionality and interoperability - there is no "expression" in and of itself. There are just a set of commands or instructions that facilitate interoperability when used in your code. It's a system or method of operation. And those things are uncopyrightable based on Section 102(b) of the Copyright Act.

So the Federal Circuit just, straight-up, fucked up entirely, and experts pretty much agree on this. Fortunately their opinion isn't binding on other appeals courts, which can decline to follow their ruling and instead declare APIs to not be copyrightable, as they should.
posted by naju at 3:31 PM on May 26, 2016 [28 favorites]

I'd forgotten this was originally a patent case, my bad. Thankfully they won't be looking at many copyright cases in the future.
posted by Holy Zarquon's Singing Fish at 3:32 PM on May 26, 2016

According to Mike Masnick, Oracle has a decent likelihood of winning an appeal (the Court of Appeals of the Federal Court leans copyright-maximalist), upon which it will probably end up in the Supreme Court.
posted by acb at 3:34 PM on May 26, 2016

Can't endorse reading Sarah Jeong's writing on this enough. Her description of how the Google lawyers brought out a SavaJe phone to talk about Oracle's success with Java phones was hilarious.
posted by fifteen schnitzengruben is my limit at 3:36 PM on May 26, 2016 [5 favorites]

It would have been interesting to see how this would have shaken out had the decision been made back in the 90's, when Microsoft was the overwhelmingly dominant force, and "embrace and extend" was their M.O. (and of which Java was one of the casualties).
posted by fragmede at 3:40 PM on May 26, 2016

This is good news, but it will be even cooler when it's moot after Android is rewritten in Kotlin.
posted by Apocryphon at 3:43 PM on May 26, 2016

With all the fuss about Peter Thiel, we forget Ellison was the original asshole SV billionaire.
posted by benzenedream at 3:52 PM on May 26, 2016 [3 favorites]

Wasn't Ellison partly doing this as a tribute to his friend Steve Jobs?
posted by Apocryphon at 3:55 PM on May 26, 2016 [2 favorites]

Thiel, Ellison, Jobs... the point is there are plenty of asshole billionaires to go around in SV.
posted by thefoxgod at 3:58 PM on May 26, 2016 [10 favorites]

it will be even cooler when it's moot after Android is rewritten in Kotlin.

how will that change anything? isn't the point of kotlin that it plays well with java? and doesn't that mean that the apis are going to stay the same for backwards compatibility? so even if both the android system level and the applications above were written in kotlin, they would still be talking via java apis?

(oe maybe i've misunderstood something and there's some ground-up rewrite with new apis? a quick google didn't enlighten me)
posted by andrewcooke at 4:07 PM on May 26, 2016 [1 favorite]

I haven't looked deeply into Kotlin, but my understanding is that it is indeed a JVM language and therefore the implementation of the core libraries would still be using Java APIs. So, switching languages isn't a way out for Google, unless they move to one that doesn't rely on the JVM.
posted by tobascodagama at 4:14 PM on May 26, 2016 [1 favorite]

Looks like Kotlin is reliant about the JCL, so I guess it would have similar dependencies with Java APIs as Android does. Never mind, I thought it was a completely different language that happens to run on the JVM and can interoperate with Java, not that it was still reliant. My joke was that Kotlin seems like a better language than Java so it'd be nice if that was the future of Android.
posted by Apocryphon at 4:16 PM on May 26, 2016 [1 favorite]

Well they could always migrate it from Java to JavaScript. That can't be any more inconvenient than migrating from VB6 to VB.NET after all.
posted by Bringer Tom at 4:19 PM on May 26, 2016 [3 favorites]

This was likely a narrow ruling, because Oracle had very little working in their favor -- their lawsuit was a frivolous cash-grab.

I mean it was, but I'm honestly surprised by this verdict. Oracle had a really simple case: here's thousands of lines of our very valuable code; Google stole it. Google's case is much more tricky. And they have to convince a jury that has no idea what's actually at stake here.

The Federal Circuit's decision was a real disappointment. Judge Alsup went to pretty extraordinary lengths (teaching himself some Java) and laid out a really strong case in his opinion why APIs should not be subject to copyright, based on his actual knowledge of the facts. The Federal Circuit pretty much ignored all that and spat something out that showed no real understanding of the issues, but would fulfill their job of being uncritical IP maximalists.

And that was all engineered from the start. Oracle threw in some pretty stupid patent claims so any appeals would have to go to the Federal Circuit.
posted by zachlipton at 4:24 PM on May 26, 2016 [8 favorites]

I'm still kind of shocked that the Supreme Court refused to hear the case considering the massive implications for software as a whole (and thus, by extension, a lot of the foundations of modern society).

It's possible (and honestly, pretty likely in my opinion) that the Supremes refused to hear the case because they have no idea about said implications. Like most people, they probably don't really even understand what's happening in this case, and as such they probably assumed that it is a much less important matter than it is.
posted by IAmUnaware at 4:41 PM on May 26, 2016 [2 favorites]

I think the Kotlin thing was a joke.
posted by schmod at 4:57 PM on May 26, 2016

The Supreme Court might have learned from their 2014 patent case Alice v CLS Bank that it is dangerous to meddle with highly technical areas where the implications and consequences are almost impossible to predict or enumerate unless you're a serious dedicated scholar. Those of us in the patent world are dealing every day with the fallout of that case, which pretty much threw everything into disarray.
posted by naju at 5:01 PM on May 26, 2016 [1 favorite]

With talk of rewriting Android in X or Y language - recall that their support of Java is crap. Their variant is ridiculously behind modern Java. This is a bit like the Gawker/Hogan/Thiel debacle - Google acted like complete asses as did Oracle; working together we wouldn't have quite the shit-show that is Android (it would still be a mess, but it would be a different kind of mess).
posted by combinatorial explosion at 5:01 PM on May 26, 2016 [1 favorite]

(at least they picked Java to copy and didn't try to shove Objective-C down developer's throats)
posted by combinatorial explosion at 5:03 PM on May 26, 2016 [5 favorites]

I have only heard my father speak ill of one person in my 33 years of existence.

"Larry Ellison is a real jerk."

I'm serious. I have a single instance of my dad expressing dislike for a person in MY ENTIRE LIFE, and it's somehow Larry Ellison.

If he was one to fist-pump, I'm sure he would privately do so after reading this article.
posted by slagheap at 5:13 PM on May 26, 2016 [11 favorites]

This is a bit like the Gawker/Hogan/Thiel debacle

Yeah, um, no.
posted by Artw at 5:20 PM on May 26, 2016 [12 favorites]

Not only are Sarah Jeong's tweets great, but so are the articles at Motherboard. I particularly recommend In Oracle v. Google, a Nerd Subculture Is on Trial.
posted by dame at 5:58 PM on May 26, 2016 [6 favorites]

1) I wonder if this has been tested elsewhere in the world.
2) Given how Disney has influenced US copyright law, could Google and it's buddies pay to have Congress rewrite the law to make API's not copywritable?
posted by Canageek at 6:37 PM on May 26, 2016 [1 favorite]

This is a bit like the Gawker/Hogan/Thiel debacle

Yeah, um, no.

Both parties were complete and utter jerks to each - yes, Gawker is much more in the right given some rather serious first amendment shenanigans, but in terms of behavior, neither Google nor Oracle were particularly in the right (Android was a massive fuck-you to Oracle's & Sun's ridiculous licensing schemes re:JavaME, Java SE, and Jave EE), which was perhaps the rather inelegant analogy I was attempting to draw.
posted by combinatorial explosion at 7:15 PM on May 26, 2016

...sigh. 'shenanigans that Theil is engaging in'...
posted by combinatorial explosion at 7:16 PM on May 26, 2016

Both parties were complete and utter jerks to each

Ok, fine, on the absolute most superficial level possible, you can maybe make some kind of half-baked equivalence. But in all the ways that actually matter, it's nothing like that other case at all.
posted by tobascodagama at 7:45 PM on May 26, 2016 [7 favorites]

2) Given how Disney has influenced US copyright law, could Google and it's buddies pay to have Congress rewrite the law to make API's not copywritable?

Explicit quid-pro-quo for legislation is of course illegal. But yes, I suppose. However, it always seems like its easier to corrupt Congress into passing stricter legislation, not looser. I can imagine the looks on their faces when they realize they're being lobbied for weaker, not stronger, IP laws.
posted by dis_integration at 9:21 PM on May 26, 2016 [2 favorites]

Is it just me or is that menu analogy not very good? You would think that for a case where understanding what an API is is so important, a lot of effort would be spent in clearly explaining to a lay audience what an API actually is. If there's no good analogy, it seems better to go without one than make it more confusing.

If we go with the menu analogy, let's say that I open a small food truck. I offer 3 choices of soup (tomato soup, pumpkin soup, and vegetable soup) and 3 choices of smoothie (strawberry, banana, pineapple). I also have two cakes (carrot and chocolate). Then, next to me, a new food stand opens with exactly the same choices, the menu looks identical to mine (though they also offer coffee and tea, which I don't). I don't think most people would be sympathetic to that stand, they would say that it's a copy cat, that the new food stand should just make their own menu's.
posted by blub at 1:37 AM on May 27, 2016

I don't agree with the ruling that APIs are copyrightable, but assuming they are, what's the actual argument that Google's reproduction of the Java APIs was fair use in this case? Seems like the article left that part out. My understanding of fair use is that you can reproduce a brief snippet of copyrighted material for the purpose of criticism, satire, reporting etc, but in this case Google have reproduced the API wholesale. Did they argue that it was covered by some other fair use exception? If so what is it, anyone know?
posted by L.P. Hatecraft at 2:23 AM on May 27, 2016

what's the actual argument that Google's reproduction of the Java APIs was fair use in this case?

Since it was a jury trial, and not something determined by a judge, we may never know what line of thinking brought them to their verdict. Maybe they looked at the API docs, saw all the methods declared public? Maybe they just liked Google's attorneys better? Maybe they asked a magic 8 ball?
posted by dis_integration at 4:55 AM on May 27, 2016 [2 favorites]

I particularly recommend In Oracle v. Google, a Nerd Subculture Is on Trial.

This is almost unreadable with her derogatory use of 'normals' to describe people that are involved with the case. There's some good data in there, but it's lost under her sneer.
posted by synthetik at 5:15 AM on May 27, 2016 [1 favorite]

I think that the key was that it was a transformative use-

Stanford has a brief guide to fair use that seems to be on point for this?

The Transformative Factor: The Purpose and Character of Your Use

In a 1994 case, the Supreme Court emphasized this first factor as being a primary indicator of fair use. At issue is whether the material has been used to help create something new or merely copied verbatim into another work. When taking portions of copyrighted work, ask yourself the following questions:

• Has the material you have taken from the original work been transformed by adding new expression or meaning?
• Was value added to the original by creating new information, new aesthetics, new insights, and understandings?

So copying the API was key to creating the larger transformative work. Fair Use seems reasonable to me. That said, it certainly looked like Oracle was doing a shitty shitty thing in grabbing a license to Google's original work, so the jury may well have been simply ruling against the nasty pack of sleazeballs out of not wanting Larry Ellison to seek yet more rent.
posted by jenkinsEar at 6:45 AM on May 27, 2016 [2 favorites]

One of the problems - the many, many problems - that arose out of the Federal Court decision that APIs are copyrightable is that until that point, the assumption based on practice and precedent was that they were not. Thus, there is no body of contract or license law that specifically covers API right granting or transfer. When you wrote a license for your software, you didn't cover what could or couldn't be done with the APIs because they just weren't something that fell into that part of legal physics. (Well, you could cover them under NDAs and confidentiality, because you can cover anything that way, but you couldn't control a published API, nor one that had become public knowledge). And clearly, you can't grandfather the pre-copyright usages of APIs, otherwise Oracle's case would not be able to generate the damages it claims, so is it possible for absolutely everyone who ever wrote an API to hit absolutely everyone who uses it with the troll gun?

And god only knows what an API actually will become, under this new interpretation of IP law. You can't write code that uses an API without replicating some or all of that API in mirror-image in your software - it's a rule set you have to follow, so your program will have to implement those rules to work.

I hope that the 'fair use' principle short-circuits a lot of this, but fair use is a minefield of its very own, even within the trench warfare of IP law.

The best outcome would be an Oracle appeal that does finally push this thing into the Supremes, which would then overcome their fear of scary tech iP law and shut API copyright down for good. Although I don't know how far outside the remit of a particular case a SC judgement can reach, whether it could go beyond saying 'fair use is correct in these circumstances' to saying 'because APIs are not copyrightable', or whether it could say 'Fair use always applies to APIs where no code copyright issues exist'. or some such.
posted by Devonian at 7:58 AM on May 27, 2016 [4 favorites]

This is almost unreadable with her derogatory use of 'normals' to describe people that are involved with the case. There's some good data in there, but it's lost under her sneer.

I think what you are reading as a sneer is a pretty clear understanding of how people on the "knows about computers" side of this argument feel about things.
posted by brennen at 8:30 AM on May 27, 2016 [2 favorites]

(Which is to say that yes, there is sneering to be found, but it is intrinsic to the subject matter of the article.)
posted by brennen at 8:32 AM on May 27, 2016

I read the "normals" bit as more of a joking reminder that, actually, the world doesn't revolve around Silicon Valley software engineering culture after all.
posted by tobascodagama at 9:26 AM on May 27, 2016 [1 favorite]

I have nothing but sympathy for the jurors. It's like asking people to rule on what model-view-controller actually means. (If you ever get bored, do an image search for that and see how many diagrams you see...)

Plus they were asked to look through source code with Notepad, which should be prohibited as it's cruel and unusual punishment.
posted by fifteen schnitzengruben is my limit at 9:55 AM on May 27, 2016 [5 favorites]

As I recall there was talk of just printing out Android for them to read.
posted by Holy Zarquon's Singing Fish at 10:41 AM on May 27, 2016 [1 favorite]

I see this as a pretty normal and somewhat beneficial consequence of a regulated society. You have a line between regulation that is created by experts (aka agency capture) and regulation/implementation that is created by normals (aka less corruptible but less informed individuals). These jury trials are an important check against the fact that Silicon Valley has blinders on its impact on broader society.

I understand the pain. I work in sales tax, and the line between taxable product and non-taxable service is razor thin in IT. Which means I have to explain that DNS management is not web hosting. Or Windows OS is not BIOS. Or the difference between Software and SAAS to government workers who are close to retirement age and not particularly invested in understanding technology.

Regulation of complicated industries is always going to to straddle between lax and dumb. That doesn't mean it's a bad thing, since the alternative is just letting Silicon Valley write their own rules and govern themselves.
posted by politikitty at 10:55 AM on May 27, 2016

...asking people to rule on what model-view-controller actually means

I really agree with this...hell, a lot of devs I talk to are completely unable to describe an original MVC versus the Model2 paradigm everyone from Rails to ASP.Net is calling "MVC". "Any software architecture that features three components or layers will eventually be described as an instance of MVC."-- MarcThibault

those ^^^ pedantic details aside, what's the best physical world analogy for the Interface vs Implementation differentiation in software development? Not 'menu', for sure.
posted by j_curiouser at 12:03 PM on May 27, 2016 [1 favorite]

those ^^^ pedantic details aside, what's the best physical world analogy for the Interface vs Implementation differentiation in software development? Not 'menu', for sure.

Cars make a good example. Every car has a steering wheel, a gas pedal and a brake (at least). This is the interface. What happens when you turn the wheel, or press the pedals, is the implementation. Some cars have better implementations than others, but they all have more or less the same interface.
posted by dis_integration at 12:06 PM on May 27, 2016 [2 favorites]

Op-ed: Oracle attorney says Google’s court victory might kill the GPL

This is an op-ed from one of Oracle's lawyers. I do not remotely understand it. She claims:

It is hard to see how GPL can survive such a result. In fact, it is hard to see how ownership of a copy of any software protected by copyright can survive this result.

But the jury here only found that the copying of the APIs was fair use. That was a central part of Google's argument: there are a gazillion lines of code that make up Java, and this much smaller percentage are the method declarations and field names. Google copied the APIs, not the implementing code.

The idea that this verdict means that all software will suddenly not be subject to copyright is ludicrous.
posted by zachlipton at 2:46 PM on May 27, 2016 [3 favorites]

[W]hat's the best physical world analogy for the Interface vs Implementation differentiation in software development? Not 'menu', for sure.

Outlets (electrical, network, audio, video, etc.) seem like a good analogy for APIs to me. Their matching plugs are like consumers. A physical device can't make use of a service (power, data, sound, etc.) provided by an outlet unless it has a matching plug. (You can, of course, buy adapters, but those adapters provide the matching outlet.) In this analogy, Oracle is claiming that they own the design of the Java outlet, and anyone else trying to build something that works with Java appliances (programs) has to secure Oracle's permission to manufacture Java-compatible outlets.
posted by reventlov at 2:53 PM on May 27, 2016 [5 favorites]

I work in sales tax, and the line between taxable product and non-taxable service is razor thin in IT.

This is simply an example of regulatory capture. Some lobbyist was able to convince regulators or lawmakers that certain forms of income should not be taxable. As soon as that occurs, the race is on to find ways to make taxable income look like non-taxable income. Without that discriminatory distinction, if all income were treated the same, there would be no conflict.
posted by JackFlash at 10:50 AM on May 28, 2016 [1 favorite]

Sarah Jeong captured all of her tweets and laid them out in Storify. Really great read.

Really not sure where Oracle goes from here. As long as there's still that theoretical payout of $9 billion, they can probably re-litigate appeals until the heat death of the universe or the dissolution of one of the parties involved.
posted by fifteen schnitzengruben is my limit at 7:51 AM on May 30, 2016 [3 favorites]

Basically they degenerate into SCO, if they are not there already.
posted by Artw at 7:54 AM on May 30, 2016 [2 favorites]

Oracle still has a huge business in high end databases, while SCO was basically dead except as a lawsuit machine. A little different.
posted by tavella at 11:33 AM on May 30, 2016

I am aware of their product line. It all has a very "IBM before its fall" vibe, if the fall hasn't indeed already started.
posted by Artw at 2:31 PM on May 30, 2016 [1 favorite]

Oracle's DB cachet does seem to be on the decline. I don't know if that's because they've declined in quality or because cheaper competitors are catching up to their performance and features or because of some other reason, but yeah. "IBM before its fall" does sound about right.
posted by tobascodagama at 3:24 PM on May 30, 2016

Basically they degenerate into SCO, if they are not there already.

Well, this is certainly the fate they deserve.
posted by brennen at 1:41 PM on May 31, 2016 [1 favorite]

if you're going to do a massive RDBMS with clustering because you still need ACID behavior and you haven't hit your scaling bound, you could hardly go wrong with Oracle. If you can afford the licenses.
posted by j_curiouser at 3:23 PM on May 31, 2016 [1 favorite]

Oracle sells a lot of other stuff that sits on top of databases these days. HR suites, marketing suites, ERP, supply chain management, etc. Raw databases are the least of what they do.
posted by GuyZero at 3:25 PM on May 31, 2016

And everyone is just overjoyed whenever they get stuck using any of that junk.
posted by Artw at 3:42 PM on May 31, 2016 [2 favorites]

Yeah, all that stuff Oracle makes to put on top of their databases is complete garbage. DBs are the only thing they could actually be said to do well. (And I think a lot of that ERP/HR/supply chain stuff is likely to be outsourced to cloud services these days anyway.)
posted by tobascodagama at 4:12 PM on May 31, 2016