Oracle: Google Has "Destroyed" the Market For Java

itwbennett writes: Oracle made a request late last month to broaden its case against Android. Now, claiming that 'Android has now irreversibly destroyed Java's fundamental value proposition as a potential mobile device operating system,' Oracle on Wednesday filed a supplemental complaint in San Francisco district court that encompasses the six Android versions that have come out since Oracle originally filed its case back in 2010: Gingerbread, Honeycomb, Ice Cream Sandwich, Jelly Bean, Kit Kat and Lollipop.

Comments

Groklaw Needed More Than Ever

[linuxrocks123]

It's a shame Pamela Jones shuttered Groklaw ... her insight into this case would have been invaluable.

We need to stop the dangerous idea that interfaces can be copyrighted before it becomes as much a bane on software as software patents were before Alice vs. CLS Bank.

Re:Groklaw Needed More Than Ever

[Noah Haders]

I don't see what alice has to do with this.

Re:Groklaw Needed More Than Ever

[Luthair]

She was conspiring with Bob against Eve!

Re:Groklaw Needed More Than Ever

[phantomfive]

We need to stop the dangerous idea that interfaces can be copyrighted

I miss Groklaw too, but it's really too late for that. The Supreme Court upheld the earlier court's decision that interfaces can be copyrighted (or more specifically, declined to hear an appeal).

It's not the end of the world. Use of an interface for purposes of interoperability has been declared fair use. The Google vs Oracle case is still in court, trying to decide if Google's use of Java is fair use.

Of course, Java is under the GPL, so in most cases this is not even an issue.

Re:Groklaw Needed More Than Ever

[0123456]

If the API headers are the most valuable part of your software... you're doing something wrong.

Re:Groklaw Needed More Than Ever

[Zeroko]

Or doing something right but monetizing the wrong thing...it is not necessary that it be possible to make money off every valuable thing.

Re:Groklaw Needed More Than Ever

[Grishnakh]

Copyrighting interfaces is indeed a really bad idea, however I'm not so sure it's quite the bane you're thinking.

In Google's case, their problem was that they had already settled on using Java long before the troubles with Oracle came up, so it wasn't exactly feasible for them to switch to something else. They had already invested in Android, Dalvik, etc., so changing course midstream just wasn't worth it. However, suppose Google knew this was going to happen; instead of basing Dalvik on Java, interfaces and all, perhaps they would have simply come up with a Java work-alike but with different interfaces.

And now that Oracle's taken this path, what kind of moron would select Java as a business platform, knowing that Oracle will want to lock them in, and lock out any competing Java platforms using this perversion of copyright law?

If other companies pull this, it seems like it'd end up increasing adoption of open-source programming platforms and making customers avoid anything where the interfaces are copyrighted and proprietary. It's not like there's a shortage of programming languages out there to choose from.

Re:Groklaw Needed More Than Ever

[Grishnakh]

You've got to be kidding. The API headers are nothing more than an interface description. If you put more effort and money into the API headers than the actual product, then you have a really shitty product. Someone cloning it from the API still has to go through all the effort of actually implementing all the functionality.

Re:Groklaw Needed More Than Ever

[hairyfeet]

And the difference between this and the MS Java case is...what exactly? Because the only difference I can see is Google pulled a name out of their ass, which means all MSFT had to do was call it "MS Coffee" and it would have all been golden.

And what if somebody was to do this to Linux? After all they have access to the code, should be easy enough to just rip it off and take it proprietary by following the Google model, what would the difference be? None at all.

Re:Groklaw Needed More Than Ever

[flink]

And the difference between this and the MS Java case is...what exactly? Because the only difference I can see is Google pulled a name out of their ass, which means all MSFT had to do was call it "MS Coffee" and it would have all been golden.

The difference is trademarks. Microsoft called their unauthorized implementation Java(tm). You don't get to do that without passing Sun's certification process. MS never implemented the entire Java specification. They modified some parts and left others out (embrace and extend). So someone who wrote a Java program against the Sun JDK and brought it to the MS platform would potentially see it fail out of the box. Due to these issues Sun used it's trademark to sue for relief from having its brand damaged.

This is different from unauthorized implementations that did not claim to be official Java products. Indeed, prior to Sun open sourcing the HotSpot JVM, there were quite a few open source unofficial implementations: e.g. GNU Kaffe, Apache Harmony, GCJ, etc. Claiming ownership over interfaces/API is a new and treacherous behavior that came along with Oracle.

And what if somebody was to do this to Linux? After all they have access to the code, should be easy enough to just rip it off and take it proprietary by following the Google model, what would the difference be? None at all.

None. Linus owns the Linux trademark in many countries. Assuming someone didn't copy the source code and just re-implemented the APIs, it would be totally kosher as long as you didn't call it "Linux". How do you think Linux was allowed to exist in the first place? It's just an unauthorized implementation of a bunch of POSIX APIs, but because Linus didn't call his kernel a UNIX(tm) system or claim POSIX(tm) compliance, he didn't run afoul of trademark law.

Re:Groklaw Needed More Than Ever

[flink]

It's not the end of the world. Use of an interface for purposes of interoperability has been declared fair use. The Google vs Oracle case is still in court, trying to decide if Google's use of Java is fair use.

It's a serious blow to interoperability and to open source in general. Fair use is and affirmative defense, not an absolute right. It's very subjective. In order to even assert fair use, you have to be sued, refuse to settle, go to court, and convince a judge that the fair use defense applies... and then you have to actually litigate the case, with the risk you will lose, be out potentially millions in your own legal costs, plus damages, plus maybe paying the plaintiff's costs. This is a huge burden for anyone but a massive corporation to meet.

It is impossible to write a non-trivial Java application without extending or overriding some API "owned" by Sun/Oracle. This means that basically every Java application and by extension, every program that implements a public, non-open-source API or is written in a proprietary language exists at the sufference of the API/language creator. Maybe you could go to court and try to assert "fair use", but good luck doing that if you are not Google.

Re:Groklaw Needed More Than Ever

[davester666]

Ugh. Having to mount a fair use defense is basically "prove I'm innocent". Not really useful for anybody with say, less than 10 or 20 million /year in revenue, as you can't afford a defense.

Re:Groklaw Needed More Than Ever

[phantomfive]

To be fair, anyone with less than that much in revenue isn't really worth suing in a copyright suit anyway.

Re:Groklaw Needed More Than Ever

[davester666]

You are doing it to crush the idea. Getting a bunch of money as well is just a bonus.

Re:Groklaw Needed More Than Ever

[Wycliffe]

To be fair, anyone with less than that much in revenue isn't really worth suing in a copyright suit anyway.

Tell that to the RIAA. If Oracle claims copyright of Java, it would be fairly simple for them to start charging licensing fees. If they went after small shops, they could easily ask for $5k an executable and most shops would have to just pay it as it would cost more to fight it.

Re:Groklaw Needed More Than Ever

[phantomfive]

It is impossible to write a non-trivial Java application without extending or overriding some API "owned" by Sun/Oracle. This means that basically every Java application and by extension, every program that implements a public, non-open-source API or is written in a proprietary language exists at the sufference of the API/language creator

The Java compiler license specifically gives you all rights to programs developed in the Java language. So no, that is not a problem.

Re:Groklaw Needed More Than Ever

[phantomfive]

Oracle can't do that. The Java compiler license specifically relinquishes any right to programs written in Java.

Re:Groklaw Needed More Than Ever

[phantomfive]

You can't use copyright to crush an idea. Ideas are not copyrightable.

Re:Groklaw Needed More Than Ever

[davester666]

By crushing the implementation of the idea. You lose your current pile of cash due to IP problems, good luck finding investors to give you a new pile for the same idea "this time, we'll do it right!".

That is, of course, unless the winner of the lawsuit doesn't do the "well, we can settle this before you go out of business if you sign over 75% of the company".

Re:Groklaw Needed More Than Ever

[LordWabbit2]

Yeah, but there is a shortage of programmers who would know them. The cost to retrain an entire IT division to another language, rewrite what has already been written into the new language, AND then do a full testing cycle to make sure it all still does what it's supposed to would be HUGE! Unless of course your "team" is three people and all you have written is a porn website.

Re:Groklaw Needed More Than Ever

[phantomfive]

Is there any case in the history of software where an idea has been crushed in that way?

Re:Groklaw Needed More Than Ever

[davester666]

Oracle appears to be trying this with Google, and certainly the only reason it has gone this far is that Google is of a similar size to Oracle. If google were significantly smaller, Android would have to be radically altered or everyone would have to pay out megadollars for 'Java' licenses.

Re:Groklaw Needed More Than Ever

[a_n_d_e_r_s]

The java language is unimportant. This has to do with every programming language or anything similar to a programming language and their APIs.

This decision effects them all. And anyone who uses them.

Re:Groklaw Needed More Than Ever

[Kjella]

It's not the end of the world. Use of an interface for purposes of interoperability has been declared fair use.

But using an interface with the purpose of replacing the original software has not, otherwise Oracle vs Google would have trivially gone in Google's favor. And that is a very dangerous, because they go hand in hand. If LibreOffice can read and write MS Office documents, it's also looking to replace MS Office installations. If the scope of interoperability is limited to software that doesn't compete with the original, that would be a mess.

Re:Groklaw Needed More Than Ever

[Half-pint HAL]

We need to stop the dangerous idea that interfaces can be copyrighted

Various cases have established the principle that interface copyright cannot hamper interoperability -- this is good, right? Now, Google aped the JVM nit for interoperability, but just to reduce their own effort in creating a complete system, and this led to the creation of code that is not JVM compatible, and won't normally run JVM code. Embrace and extend...

Re:Groklaw Needed More Than Ever

[Half-pint HAL]

If LibreOffice can read and write MS Office documents, it's also looking to replace MS Office installations. If the scope of interoperability is limited to software that doesn't compete with the original, that would be a mess.

No, LibreOffice interoperates because you can transfer files both ways, and a user of one can collaborate with a user of the other. Dalvik does not interoperate with the JVM, as they will not run and load the same programmes. This is Oracle's point -- they are using JVM technology to their advantage without interoperating. Of course, this is irrelevant if Google has a license to use Java source code any way they like... isn't Java GPL...?

Re:Groklaw Needed More Than Ever

[Half-pint HAL]

Note -- Java platforms, not languages. Java can be compiled to various architectures other than JVM.

Re:Groklaw Needed More Than Ever

[TheRaven64]

Or you're doing something right. Good API design is good UI design. If your UI sucks in most pieces of software, your software is a lot less valuable. Most programmers are far too tolerant of crappy APIs and will happily use a poorly designed UI and blame other programmers who struggle.

Some bits are easy (for example, don't be the idiot LLVM contributor who thought that having isBundle() and isBundled() methods on the MachineInstr class that do completely different things was a good idea). Some bits are harder - making sure that you have consistent abstractions that you use across entire frameworks, making sure that you use consistent naming everywhere.

A good API is a considerable engineering effort. It's not excluded from copyright because it's not creative, it's excluded because interoperability is considered more important. The trend in copyright law has been removing exemptions that benefit society as a whole in favour of total control by the originator, so it's not a surprise that this is being weakened.

Re:Groklaw Needed More Than Ever

[Cederic]

The irony being that BEA bought JRockit because their JVM implementation was significantly better than Sun's on Intel, and Oracle bought BEA.

This is before Oracle bought Sun, so Oracle were themselves doing to Sun what they're claiming Google have done to them.

Fundamentally it all boils down to Larry Ellison and his company being cunts.

Re:Groklaw Needed More Than Ever

[Cederic]

I think Standards Committees get bogged down by politics rather than technical challenges, but otherwise I agree with you.

A good API is a work of art, an engineering masterpiece and a key success factor for adoption. Java grew popular not because it was faster than the alternatives, or because it had better IDEs, or because it was great for building UIs, it grew popular because it was bloody well designed and easy to use.

The solid reliable fast implementation came years later.

Re:Groklaw Needed More Than Ever

[dintech]

Good, it's about time someone got that malicious bitch back.

Re:Groklaw Needed More Than Ever

[gbjbaanb]

Yes, it means that is company x creates a language, no-one will use it unless they are explicitly told they have a free-to-use licence for it.

A bit like the existing Java licence.

What you will not get is a licence for other implementations to appear. So Google cannot take someone else's language and write their own versions of it., which may or may not be what you want - so again, if they do not release the language as an open standard, no-one will use it. Simple!

Other forms of API are exactly the same, you either get a free licence or buy a licence to access the system (exactly as it is today), you just cannot reimplement the product you're accessing without a licence.

The only difference is that it sucks to be Stallman, as this means all those GPL programs he wrote to replace the existing Unix commands would have been breaking copyright, but then, they changed things like the command arguments (eg instead of -h they use --help) which might be enough to say they are not exact replacements after all, possibly.

Re:Groklaw Needed More Than Ever

[cusco]

then you have a really shitty product

Well, it is Oracle . . .

Re:Groklaw Needed More Than Ever

[oh_my_080980980]

No, no we don't. She jumped the shark at the end.

Re:Groklaw Needed More Than Ever

[omnichad]

And what if somebody was to do this to Linux? After all they have access to the code, should be easy enough to just rip it off and take it proprietary by following the Google model, what would the difference be? None at all.

That's a bit of a roundabout way to recreate Unix.

Re:Groklaw Needed More Than Ever

[phantomfive]

The only difference is that it sucks to be Stallman, as this means all those GPL programs he wrote to replace the existing Unix commands would have been breaking copyright, but then, they changed things like the command arguments (eg instead of -h they use --help) which might be enough to say they are not exact replacements after all, possibly.

Remember that there is a difference between copyrightable and "fair use." Pretty near anything is copyrightable, even a line on a piece of paper. (Even a blank canvas with a slash in it!). But Stallman can argue his copying is fair use.

Re:Groklaw Needed More Than Ever

[Wrath0fb0b]

It is impossible to write a non-trivial Java application without extending or overriding some API "owned" by Sun/Oracle. This means that basically every Java application and by extension, every program that implements a public, non-open-source API or is written in a proprietary language exists at the sufference[sic] of the API/language creator

Can you explain this a bit more to an experienced dev with very little exposure to Java? Because I managed to write all manner of non-trivial applications in all manner of other languages without ever having to modify the language APIs. That's both low level (C/C++/FORTRAN) and high (C#/Python/Go).

This is really quite puzzling -- it seems like a real anti-pattern to modify language API, even in languages where you can do it. Not the least of which is that whatever dev inherits your code is going to curse you after they spend a day figuring out why std::foo (or whatnot) doesn't do the same thing it does everywhere else.

Re:Groklaw Needed More Than Ever

[phantomfive]

If LibreOffice can read and write MS Office documents, it's also looking to replace MS Office installations.

That's pretty clearly for the purpose of interoperability. ODF is the primary file format for open office.

Re:Groklaw Needed More Than Ever

[phantomfive]

What idea, exactly, is being crushed by the Oracle vs Google lawsuit?

If google were significantly smaller, Android would have to be radically altered or everyone would have to pay out megadollars for 'Java' licenses.

Or they could have actually used Java, the way the designers of Java intended. Or they could have used the GPL, the way the designers intended. Or they could have used Python, and avoided the problem entirely. Google had plenty of ways to easily handle the copyright problems here.

But they didn't. Instead, they waded into a grey area of copyright. They still might have a fair-use defense, which is why everyone is watching this case so closely.

Re:Groklaw Needed More Than Ever

[squiggleslash]

Which is presumably great if you're using Oracle's Java compiler. And not so great if you're not...

Re:Groklaw Needed More Than Ever

[phantomfive]

If you're not using Oracle's compiler, it's not a problem as long as you use a GPL compiler. Furthermore, there is a fair-use defense for interoperability, so it's not a problem on multiple levels.

Re:Groklaw Needed More Than Ever

[Coren22]

What happened to Charlie and Dave?

Re:Groklaw Needed More Than Ever

[squiggleslash]

I think you missed the comments earlier in the thread that I'm responding to. The TL;DR is that you're licensed to use the Java API if you use Oracle's compiler. That is, Oracle are giving you a license to use it.

Third party compilers, licensed under the GPL or anything else, don't have that licensing because it's not their's to license.

Re:Groklaw Needed More Than Ever

[mrchaotica]

Of course there isn't, but that means nothing because the Supreme Court [lack of] ruling that enabled it only recently happened.

Re:Groklaw Needed More Than Ever

[mrchaotica]

Dalvik does not interoperate with the JVM, as they will not run and load the same programmes. This is Oracle's point -- they are using JVM technology to their advantage without interoperating.

Bullshit. Google is using Java technology -- not "JVM technology" -- to its advantage, for the purpose of interoperating with Java.

Some other arbitrary code that also happens to interoperate with Java is irrelevant.

Re:Groklaw Needed More Than Ever

[mrchaotica]

a Java work-alike but with different interfaces

The trouble is, that's an oxymoron: a "work-alike" is what an interface is!

Re:Groklaw Needed More Than Ever

[phantomfive]

Your understanding of copyright law, and your knowledge of the history astounds me.

Re:Groklaw Needed More Than Ever

[phantomfive]

I think you missed the comments earlier in the thread that I'm responding to.

Probably not, since I wrote it.

Re:Groklaw Needed More Than Ever

[chooks]

Java uses a single-inheritance model wherein ultimately everything is implicitly a subclass of the Object base class. Therefore, you are unable to write code that does not ultimately extend or enhance the Object class.

To say this modifies the API is probably an overstatement, but technically (which is the best kind of right) it is extending it. Also, you do modify the intrinsic behavior of Object by overriding its methods (e.g. equals(...) and hash()) when needed.

Re:Groklaw Needed More Than Ever

[lgw]

The Java compiler license specifically gives you all rights to programs developed in the Java language. So no, that is not a problem.

Oracle has a long and established history of changing its licenses to screw over its customer base. I keep hoping Oracle finds a way to kill off Java, just so the industry can finally move on to something better (do we really need Java to live as many years as COBOL did?).

Re:Groklaw Needed More Than Ever

[phantomfive]

(do we really need Java to live as many years as COBOL did?).

'Did?' There are trillions of dollars worth of COBOL code out there still, with billions of lines of COBOL being written in COBOL every year. It is never going away.

Now, imagine Java went away. It would be a serious shame, because many shops would move on to .NET. Would you rather have that?

Re:Groklaw Needed More Than Ever

[lgw]

Now, imagine Java went away. It would be a serious shame, because many shops would move on to .NET. Would you rather have that?

Yes, that would be great, because C# is an amazingly better language than Java, and most shops wouldn't move to it unless MS and Xamarin really got the cross-platform thing sorted out properly.

But really, I'm sure there are better language ideas out there. There has been very little activity in the PL world for "curly brace languages", though Go was neat for what it was. C++, Java, and C# all have flaws for the modern world of backend development, though they're also all usable with various degrees of pain. I'd love to see a synthesis with a modern take on macros and generics and container classes and so on. The lack of macros in Java makes for a horrific amount of boilerplate -- heck, your IDE writes 80% of your code -- but macro hell is worse, and C++ template programming worse still - let's have something good instead. C# finally got properties right - whatever else, keep that. etc, etc.

Re:Groklaw Needed More Than Ever

[Grishnakh]

No, it's not. I'm talking about a work-alike to the thing on the other side of the interface. For an example, think about a USB thumb drive. They all have the same interface: a USB connector (the physical interface), plus the USB protocol (the software interface). You plug it into your computer, it communicates via USB and identifies itself, your computer mounts it, and you can now read and write data to it. You can get different USB drives from different vendors, and while they all look the same (except for USB vendor:device ID numbers, but these don't matter because they follow the Mass Storage spec so they don't need a special device driver, and of course how much capacity is has), how the drive is implemented internally can vary greatly. These days, they're all implemented with NAND flash chips, but they can be configured differently or have different grades of NAND (resulting in wildly different r/w speeds). But there's nothing stopping them from using other types of memory; USB hard drives do this, but if someone wanted to, they could make the drives out of NOR flash, PCM, or memristors. The implementation details don't matter because the *interface* is the same. And luckily, the USB protocol and the Mass Storage spec are all open, or else we wouldn't have all these nice, cheap thumb drives.

Re:Groklaw Needed More Than Ever

[phantomfive]

C# is an amazingly better language than Java,

Well ok, you just lost any credibility as a judge of languages. Code that looks like trash in Java typically looks about the same once it's translated to C#.

Re:Groklaw Needed More Than Ever

[mrchaotica]

Yes, and every USB connector and USB protocol is a "work-alike" to every other USB connector and USB protocol. Like I said, that's what an interface is.

Similarly, the Java API interface of the Dalvik virtual machine is a "work-alike" to the Java API interface of the Oracle JVM. That is separate and distinct from the functionality of the Dalvik virtual machine and the JVM themselves; those are only work-alike by coincidence (if at all).

If you swapped out Dalvik's Java API interface for, say, a .NET interface, it would no longer be "work-alike" in the same way that if you swapped out your device's USB interface for a Firewire one it would no longer be "work-alike" to USB.

Re:Groklaw Needed More Than Ever

[segedunum]

The Java compiler license specifically gives you all rights to programs developed in the Java language. So no, that is not a problem.

This has nothing to do with the Java compiler license and you're being disingenuous in your attempts to frame it in that manner.

Re:Groklaw Needed More Than Ever

[segedunum]

If you're not using Oracle's compiler, it's not a problem as long as you use a GPL compiler.

You have no basis whatsoever for saying this and I'm afraid you're plucking this out of your backside, or more specifically, the way Oracle's legal department wishes things to be.

Re:Groklaw Needed More Than Ever

[segedunum]

Your lack of knowledge as to the precedents on this that have gone long before astounds me, or maybe it doesn't.

Re:Groklaw Needed More Than Ever

[lgw]

Properties. LINQ. Proper generics. It's just a better language in the same space, with far less boilerplate.

Re:Groklaw Needed More Than Ever

[phantomfive]

Java was designed as a language to prevent incompetent programmers from messing things up too much. Generally I don't like the language, but I recognize that it succeeds at its goal, and appreciate it for that.

C# was such a close clone of Java that more than once, I've been reading one while thinking it was the other. LINQ and Properties are kind of cool, but they're syntactic sugar. At best, as you say, they reduce typing.

I absolutely despise the C# community (not you specifically, though) for its celebration of ignorance. The general attitude that, "geniuses at Microsoft did things so we don't have to learn them" which is captured by this speech. A good system gently encourages you to understand more deeply.

Re:Groklaw Needed More Than Ever

[rtb61]

Oracle does not need to kill off Java, that is exactly what this court case is all about, Oracle can see the writing on the wall for Java. Java had it's moment in the sun and now it is on the wain and other languages are starting to dominate. Oracle can see this and is attempting to make a two way bet, either a big grab for cash to make up for the lost investment in Java or disrupt Android and try to force Java back on top. Problem for Oracle even when weird crap can happen in corrupt US courts it will all be blocked in the rest of the world (the parts that count), leaving the corruption in the US visibly apparent and blocking it. So basically a pretty lame desperation move, likely pumped up by Oracle lawyers as a self serving grab for cash, for themselves from Oracles own cash hoard (don't trust lawyers, the best person to keep an eye on a lawyer is another lawyer, catch with that, endless loop). The US simply needs loser pays laws, quite simply this puts enormous pressure on lawyers to perform properly, else they can find themselves on the receiving end of a law suit from the ex-client for bad advice.

Re:Groklaw Needed More Than Ever

[lgw]

Every language is nothing but syntactic sugar over assembler. Syntactic sugar is, in fact, the only useful measure of a language: since all Turning complete languages are equivalent in capacity, easy syntax is what's left.

Re:Groklaw Needed More Than Ever

[phantomfive]

Technically, that's true.

Re:Groklaw Needed More Than Ever

[lgw]

"technically correct - the best kind of correct" - Bureaucrat #1

Re:Groklaw Needed More Than Ever

[lgw]

some of which can't be described in terms of other languages

You fail to understand Turing completeness. Anything that can be written in any real language can be written in any other. It's just a matter of how awkward that would be.

Re:Groklaw Needed More Than Ever

[jimbob6]

MS didn't call it java they called it J++. Got sued any way but not for trademark infringement. They got sued for violating the java license (which they paid for in '96) for incomplete implementation of java virtual machine built into IE, Java applets would crash in IE leading people to conclude that java was a big stinking pile. You had to install the sun java add-on to find out the truth about java. Which was that if you installed the sun java VM you could see that java was launched as a big stinking pile out of the gate.

Re:Oracle's monopoly?

[MightyMartian]

Because J2ME was such a brilliant mobile platform.

Profiting on the Backs of Others

[Great Big Bird]

Oracle (then Sun) could have created an operating system for mobile phones based around Java. But since Google did, they want to profit off of it? They should go to hell.

Re:Profiting on the Backs of Others

[Impy the Impiuos Imp]

I don't get it -- isn't Android programming primarily Java, with things like C relegated to support library status?

Sounds stronger than ever.

Re:Profiting on the Backs of Others

[binarylarry]

The problem with J2ME is that it's awful.

It's always been awful, I've always dreaded using apps on pre iOS/Android phones.

Re:Profiting on the Backs of Others

[swb]

Maybe that's the unintended consequence of a write once, run anywhere language -- they were supposed to transcend the operating system. They never made a Java operating system because of that concept, and Sun really wanted to sell Solaris, too.

Re: Profiting on the Backs of Others

[binarylarry]

Microsoft licensed the Java platform and trademark and then intentionally built an incompatible implementation, put the Java logo on it and claimed compatibility.

Android has never claimed to be a compliant Java platform, it merely lets you use the Java programming language and GNU's standard java library (gcj & libjava) to build apps for Android.

Re: Profiting on the Backs of Others

[ArmoredDragon]

Microsoft created libraries that were compatible with Sun's Java, and then added their own proprietary (and incompatible) extensions to pull developers away from the real Java. This was a deliberate move to make sure that developers had to target Windows and couldn't target ordinary Java (which could run on any other platform) By the way, this was the same motivation Microsoft had for creating Internet Explorer (that is, they didn't want developers being able to target a web browser instead of an operating system.)

The whole idea was to force end users to stay with Windows instead of anything else, as Microsoft wanted to maintain their monopoly status.

Android on the other hand wasn't attempting to do that. That is, it never made any effort to pull any developers away from the Java platform, nor was it ever intended to do that in the future, nor did they make any attempt at being compatible with existing Java applications. Sure, it would be easier to port Java applications over, but it's intended to be the same at all, whereas Microsoft's implementation was intended to be a drop-in replacement.

Furthermore, Sun won their case against Microsoft because it was proven that Microsoft did what they did for anti-competitive reasons; copyright infringement was never claimed at any point. And likewise, Oracle isn't making any kind of anti-compete claims towards Google.

Oracle is just saying "Hey, you created an interface with similar naming to something created by a company we purchased. Even though other companies have done the same thing numerous times and have never been sued before, we're going to shake you down because we happened to have noticed just how successful you are and we'd like to get on your gravy train without having contributed anything to it."

Which by the way, what I just said above is typical Oracle behavior. When somebody comes along that does something similar to what they do, then they first try to buy it out, and if they can't buy it out, then they sue it out. Having said that, Oracle is every bit as much of an asshole company as Microsoft has ever been, if not more so.

Re: Profiting on the Backs of Others

[ArmoredDragon]

Unless you're Nissan and you're hell bent on extracting a domain name that somebody owned before you even existed. Then you'll spend a shitload on lawyers fees only to lose and have your name dragged through the mud.

That said, never buy a Nissan car. Ever.

Re: Profiting on the Backs of Others

[pkinetics]

Nissan Hardbody pickup just works...

Re: Profiting on the Backs of Others

[fahrbot-bot]

That said, never buy a Nissan car. Ever.

Cars? I thought they made Cup Noodles

.

Re:Profiting on the Backs of Others

[Dutch Gun]

By "irreversibly destroyed Java's fundamental value", Oracle means "we should be getting payments from Google because they're using a version of Java that they didn't license from us to make money." Everything else is fluff.

Re: Profiting on the Backs of Others

[jhol13]

I disagree. I think it is obvious Google picked Java because it was popular. They could have made it compatible, but on purpose did not. So in effect they did pull developers from Java platform.
If Google did not want to do that, they would not have used Java, but made a better language. A better language closer to modern scripting languages (e.g. no "generics") and more sane arithmetic (integers which cannot overflow, not require sin(x) return exactly[1] same result in every platform with every argument, etc.).

[1] I do not remember what the Java requirements are, but they are silly for this kind of a language. IMHO sin(10^50) is pretty much "implementation defined" as sin(10^50 + epsilon) is in effect random. Sure there are cases where you really have to do "infinitely precise" arithmetic to calculate sin(huge_value), but for generic computation? No.

Re: Profiting on the Backs of Others

[fredgiblet]

If we refused to buy from any company that did anything assholeish then there'd be nothing to buy.

Re: Profiting on the Backs of Others

[SvnLyrBrto]

According to wikipedia, Nissan has existed since 1914 and began operating under that name in 1934. At that time "computers" were not even a piece of technology, but persons whose job was to sit around all day doing math by hand. So, I'm fairly sure that Nissan Motors has existed before the entire internet and, therefore, any domain registration.

Re: Profiting on the Backs of Others

[sg_oneill]

The reason they used their own bytecode was because of the Sun vs MS thing. Google wanted to add apis that where needed for modern android, but feared doing so would put them at odds with Sun, so they created a whole new bytecode system to avoid copyright entanglements.

Regardless, Androids about the only reason Java is still relevant. Sure theres the enterprise java thing, but even thats getting eaten away by web apps in more agile languages. Last job I had was at a government department where we where rewriting clunky old java apps to django and ruby on rails.

If it was about "the future of Java" Oracle should be thanking Google. But its not, its about getting a slice of that android pie.

Re: Profiting on the Backs of Others

[staalmannen]

I thought it used Apache Harmony with a strict "no GPL in userspace" rule for Android. Funnily Apache, IBM, Oracle and others were allies against Sun's rule over Jaba and wanted more equal footing of alternatibe implementations. Oracle quickly changed tune after buying Sun.

Re: Profiting on the Backs of Others

[ArmoredDragon]

Actually in that lawsuit, Sun brought up the emails uncovered in the DoJ investigation where a Microsoft exec (can't recall the name) specifically called out Java as a threat to Windows (the same email mentioned Netscape Navigator as a threat to Windows, and for the same reason.) This is why they won the lawsuit.

Sun did indeed try for copyright infringement, but that portion of the lawsuit was shot down, and the only one that stood was the violation of anti-compete laws.

Sun has argued in court that Microsoft viewed Java's "Write Once, Run Anywhere" capability as a threat to Windows, because Java reduced the incentive for software developers to write programs for the Microsoft operating system.

According to Sun, the version of Java distributed by Microsoft worked better with its Windows software. Such a move threatened Java's ability to provide a cross-platform development environment, Sun's lawyers said.

Microsoft has vehemently denied any wrongdoing and has maintained that it stuck to the letter of its licensing agreement with Sun. Any changes Microsoft made to Java merely allowed developers to take advantage of features specific to Windows, the company has argued.

The case has been watched closely, and Microsoft's dealings with Java were cited by U.S. District Court Judge Thomas Penfield Jackson in the U.S. government's antitrust case against Microsoft as evidence of the software giant's anticompetitive behavior.

Tom Burt, Microsoft's deputy general counsel for litigation, portrayed Tuesday's settlement as a positive outcome for Microsoft.

"Microsoft is very pleased with the successful conclusion of this litigation," Burt said in a statement. The agreement confirms Microsoft's ability to independently develop technology to compete with Sun's products, the company said.

Sun scored a victory in the case in November 1998, when Judge Ronald Whyte of the U.S. District Court in San Jose, California, ruled that Sun was likely to win its case based on the merits and issued a preliminary injunction in Sun's favor. The injunction forced Microsoft to modify the Java technology it had distributed in its operating system, Web browser, and development tools so that it passed Sun's tests.

A U.S. Appeals Court overturned the injunction the following year, questioning the grounds on which Whyte had based his decision. Whyte reinstated the injunction, but based his order on California's unfair competition statutes rather than on copyright law, as Sun had requested. The ruling was seen as a partial victory for Sun.

http://www.javaworld.com/artic...

Re: Profiting on the Backs of Others

[Anonymous Coward]

(gcj & libjava)

I need a source on that claim.

First GCJ has been deader then a doornail for over half a decade. The project is officially in deep maintenance, the main project page itself hasn't had any updates since 2009.

Second GCJ died after the statement that they added some 1.5 features, I see several sources mention Android with Java 1.6 language features.

Third the Android standard library was based on Apache Harmony not GNU GCJ. Which actually gave Oracle footing on its case against Google. If they had used a GNU project it would have been GPL licensed and with that compatible with the OpenJDK - since that is distributed by Oracle itself claiming copyright violation would have been rather questionable.

Fourth the Android runtime is the specially developed Dalvik VM using Dalvik bytecode or JIT compilation. GCJ acts either as a pure interpreter or a Java to native compiler.

Lastly if you had ever used anything build with GCJ you would be aware that it is painfully slow. Interactive software as needed for smart phones is not something it should be used for. Nor do I expect its effects on battery live to be anywhere near acceptable. There is a reason the project died the moment Sun published the OpenJDK source as GPL.

Re: Profiting on the Backs of Others

[Kartu]

Well, J++ doesn't sound like "Java" to me either.
Why did they force Microsoft to drop it altogether, instead of just "do not call it Java"?

Re: Profiting on the Backs of Others

[Kartu]

Java continues to rule on the server side, and Java was never used much on Desktop/Mobile.

Re: Profiting on the Backs of Others

[Half-pint HAL]

Compiler error: expression encountered; expected statement.

Re: Profiting on the Backs of Others

[TheRaven64]

The main issue for Microsoft was putting Windows-specific APIs in the java.* namespace. This broke the 'if you only use java.* stuff then your app will run anywhere' guarantee for Java developers and was in direct violation of the Java license (which included patent and copyright grants). Dalvik's register-based VM, which doesn't use the .class files at all, is partly a way of getting around the JVM patents and meaning that copyright is the only thing that they can infringe. As they're not using any code licensed from Oracle, the only argument left is that they're using the same interfaces.

Re: Profiting on the Backs of Others

[Cederic]

There was me thinking we have five thousand linux servers running Java in our UK data centre alone.

Bugger, wonder what we're using instead then?

Re:Profiting on the Backs of Others

[someone247356]

Oracle open sourced the desktop version of Java (Java SE) after the Google mess, but requires people to pay licensing fees to use the mobile version of Java (Java ME).

There were several alternate Java run times besides the official Sun one; Kaffe, SableVM, gcj, Harmony, etc.

Even for the relatively open/free Java SE, in order to pass the compatibility suite, needed to call your implementation 'Java' you had to agree to 'field of use restrictions'. Including agreeing NOT to use your version of JavaSE on mobile (so as not to threaten Oracle's license stream). It was one of the major sticking point with Apache's Harmony implementation. The 'field of use restriction' was incompatible with Harmony's license.

The Harmony implementation is what Google initially based Android's Java on. Eventually Oracle and Apache came to an 'understanding' when Oracle released the OpenSE JDK and IBM pulled support for Harmony. If you look carefully, you'll notice that there _still_ isn't an open Java ME.

So what Google did was to re-purpose Harmony JavaSE as an unencumbered JavaSE implementation. They didn't have any license agreement with Sun/Oracle, and didn't reuse the name, byte code or run time (as both were poorly suited to running under mobile). Sun's management boasted of their support for Google's actions. Oracle tried to memory-hole that support when they took over.

Re:Profiting on the Backs of Others

[oh_my_080980980]

Really Potsy? It's called Java. That's what Java was originally created for - mobile devices. Do some research before posting.

Re:Profiting on the Backs of Others

[oh_my_080980980]

Yup - you need the Java JDK to develop for Android. Seems to me Google could ditch Java and use something else - especially since Google repeatedly bashes Java. Google would rather use Java developers. Says alot about Google's motives and their inability to create their own programming language. Apple did it, why can't Google.

Re: Profiting on the Backs of Others

[someone247356]

Actually, the reason they used their own byte code was because the 'official' one is poorly suited to mobile use. One of the major changes between Dalvik and the original Java run time is related to stacks vs heaps. There are others as well.

Google redesigned the run time to work better on mobile. JavaSE wasn't designed for that, JavaME isn't much better.

You shouldn't confuse the byte code/run time with the API's. Google stripped what they felt were unneeded API's and added ones they thought were missing. What Google did was to leverage the existing JavaSE developer knowledge base to create a straight forward path for people to develop for Android.

The alternative, and what Google may end up doing, is to replace Java with Go, Dart, or some other home grown language for primary Android development. While this would have been unthinkable when Android was first released, it's now gotten enough market share that developers are more willing to switch development languages to continue to develop Android applications. Even Apple's now doing it (see Swift).
 

Re:Profiting on the Backs of Others

[ZeroWaiteState]

Well....you could say Apple had more of a head start. Work on Objective C started back in the 80's, and was licensed to NeXT in the late 80's early 90's. They've had around 20 years to work the kinks out in a production environment. And BTW, Google _is_ actually creating a language, called Go, but it's still immature at this point. It has the potential, however, to replace Dalvik/ART if they can get some kind of JIT compilation going, via LLVM probably. The key is the language needs to be able to elegantly handle multicore stuff in a way that is portable across devices with different hardware. Dalvik has to go through unbelievable contortions to get the Java programming model to work on a small memory-limited device: stuff like zygotes, bytecode-stripping, changes to GC, etc. It's very much a square-peg round-hole kind of thing. The real reason for use of Java was programmer adoption. Sun poured a huge amount of resources into lobbying public universities to churn out Java programmers, so that's what Google had to work with. Now that we have a lot of people in the boat, we can focus on getting where we want to go. Long-term, dropping Java would probably be a net positive.

Re: Profiting on the Backs of Others

[freezin fat guy]

I would mod "insightful"

Fuck Oracle

[binarylarry]

Java was never useful on phones until Google built something decent.

Sun/Oracle could never build a decent phone with Java, no matter how much money they pumped into it.

If you work somewhere that uses Oracle products or is considering an Oracle product, fight to the bone to get their shitware tossed out.

We need to end this company, it's a tumor in the software ecosystem.

Re:Fuck Oracle

[Bing Tsher E]

Don't you mean until Google bought something decent? Android does not originate with Google.

Re: Fuck Oracle

[binarylarry]

Well honestly the first few versions of Android weren't great.

I was an early adopter with the G1, etc.

Re:Fuck Oracle

[tuxrulz]

1. Google bought Android.
2. Java is used for the Apps, the core OS is not in Java.
3. You as a developer can use the NDK and native compilers like gcc (and soon clang) to compile native apps...
but that means losing the portability to any Android device independent of CPU in the process.

Re:Fuck Oracle

[Viol8]

Sybase? No, don't laugh. Its actually a good system especially from a dev point of view.

Re:Fuck Oracle

[myowntrueself]

Java was never useful on phones until Google built something decent.

Sun/Oracle could never build a decent phone with Java, no matter how much money they pumped into it.

If you work somewhere that uses Oracle products or is considering an Oracle product, fight to the bone to get their shitware tossed out.

We need to end this company, it's a tumor in the software ecosystem.

Imagine an Oracle smartphone...

Re: Fuck Oracle

[Hognoxious]

They might be better, in the sense that a slap in the face is better than a kick in the gonies, but are the newer ones that good? Perhaps it's because I was expecting it to be like Linux, and it's sort of, well, not.

Re: Fuck Oracle

[Bing Tsher E]

The first few versions of Linux weren't that great. I booted Yggdrasil's "Plug and Play Linux" in early 1994. If somebody like Google had come along and coopted it entirely, they would have made Linux into something completely other than what it has turned into.

Innovate, not litigate

[GigaplexNZ]

irreversibly destroyed Java's fundamental value proposition as a potential mobile device operating system

Well there's the problem. Oracle thinks the language and runtime are a complete operating system. There's nothing stopping Oracle making a different OS that uses Java. In fact, the vast amount of libraries for Android out there should be easy to port. Next we'll hear how Microsoft destroyed the value of C as used to build an OS.

Re:Innovate, not litigate

[binarylarry]

That's an interesting idea.

Could Bell Labs and Bjarne Stroupstroup take Oracle to court for using C/C++ without a copyright license in their products? Perhaps they use an implemention from a vendor that isn't the original one and thus don't have a license from the original author.

Think of the Billion$ they could make at Oracle's expense!

Re:Innovate, not litigate

[DoofusOfDeath]

Well there's the problem. Oracle thinks the language and runtime are a complete operating system.

Yeah! I demand they call it "GNU/Java"!

Re:Innovate, not litigate

[Grishnakh]

That doesn't help them with C++. Unix is intertwined with C, but definitely not C++.

Re:Innovate, not litigate

[TheRaven64]

Java was not designed for low battery power devices with hard real-time constraints

You might want to look at the origin of Java. It grew from the Green Project, which aimed to build a system for battery-powered devices with about 1MB of RAM that needed interactive UIs and to control other devices.

Re:Innovate, not litigate

[oh_my_080980980]

"In fact, the vast amount of libraries for Android out there should be easy to port. "

Because they're Java libraries. Nice own goal.

Re:Innovate, not litigate

[oh_my_080980980]

Yes Java was - trying researching sometime moron.

Re:Innovate, not litigate

[rkhalloran]

"Who owns the old AT&T Unix stuff? I have no clue, and I'm not sure even an army of lawyers could track it down at this point."

AT&T had partnered with Novell in the late 80s to promote System V as Unix System Labs (I remember visiting the Summit NJ offices while I was at BTL myself). Early 90s they sold off their share completely to Novell (prompting Dennis Ritchie's quoting Genesis and Esau selling his birthright for pottage). Novell took over the various OEM licenses to IBM, DEC, etc, but eventually outsourced the royalty collection business to Santa Cruz Operation as part of their licensing to produce UNIX on Intel CPUs. They very clearly kept the copyrights (such as they were - see Groklaw getting the AT&T v. BSD settlement unsealed), which was the basis for the Novell lawsuit after Caldera bought out the old SCO and claimed All The Rights while suing IBM for beeelyuns of dollars.

Novell has since been bought by Attachmate, who now nominally owns the source rights. The UNIX trademark was donated to The Open Group by Novell early 90s.

Re:Innovate, not litigate

[bluefoxlucid]

I'm more confused about the informal fallacy of begging the question.

Android has irreversibly destroyed Java's fundamental value position.... how?

Android has made Java the core, fundamentally-valuable language to know for the global mobile phone and tablet market. With eleven times the global market share of iOS, Android devices have literally decimated Apple's envisioned control of the market. It seems Android's use of a Java-compatible runtime has created a mobile market which bolsters Java's usefulness in other markets, as programmers in this substantial market will necessarily have such programming skills in Java so as to make the use of Java a good investment for those wishing to create enterprise applications with the most cohesive and readily-available team of programmers. Each programmer is now more likely to have competent Java programming experience.

Imagine if Google just used CIL. Everyone would be C#.NET and Java would be well and truly dead. Would that not have irreversibly destroyed Java's fundamental value position within and outside the scope of mobile device operating systems?

Re:Innovate, not litigate

[petermgreen]

Read "value" as "something that will make money for oracle" and it makes sense.

AIUI the java model was to give it away free on the desktop/server and then once they had an army of java programmers out there make money out of it in other product lines like mobile and embedded.

Mobile phones have moved from a variety of javame supporting system (which paid licensing fees to oracle) to andriod (which doesn't). Oracle is unhappy about this.

Re:Innovate, not litigate

[bluefoxlucid]

Read "value" as "something that will make money for oracle" and it makes sense.

My point was Android being non-Java would irreparably diminish Oracle's value proposition from Java in its entire, while rendering any Java-based mobile phone OS obsolete before its existence (look at what happened with Blackberry and Windows Phone offerings since Android/iOS, and try to talk about introducing a new phone OS with a straight face).

Mobile phones have moved from a variety of javame supporting system (which paid licensing fees to oracle) to andriod (which doesn't). Oracle is unhappy about this.

I realize it's trite, but Java is GPL now. Oracle doesn't have a claim here. Whether they're unhappy about it isn't an issue; their new monetization model is either A) sell Java platform consulting, Java contract programming, and Java training; or B) piss off.

JAVA FTW

[Anonymous Coward]

Java...[garbage collection] is the .[garbage collection] best .[garbage collection] programming .[garbage collection].[garbage collection] language for .[garbage collection] mobile devices .[garbage collection] .[garbage collection] because it is .[garbage collection] faster .[garbage collection] then C++ and .[garbage collection] more .[garbage collection].[garbage collection].[garbage collection] deterministic and .[garbage collection] .[garbage collection] .[garbage collection] nev .[garbage collection] er .[garbage collection] drops .[garbage collection].[garbage collection].[garbage collection] user .[garbage collection].[garbage collection] input.

I like Java .[garbage collection] becuase .[garbage collection].[garbage collection] it's write .[garbage collection].[garbage collection] once, .[garbage collection].[garbage collection].[garbage collection] and it runs .[garbage collection] .[garbage collection].[garbage collection] provided you have all .[garbage collection].[garbage collection].[garbage collection] the libraries, the .[garbage collection].[garbage collection].[garbage collection] correct java interpreter .[garbage collection].[garbage collection].[garbage collection] and enough .[garbage collection].[garbage collection]

javax.servlet.ServletException: Something bad happened
at com.example.myproject.OpenSessionInViewFilter.doFilter(OpenSessionInViewFilter.java:60)
at org.mortbay.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1157)
at com.example.myproject.ExceptionHandlerFilter.doFilter(ExceptionHandlerFilter.java:28)
at org.mortbay.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1157)
at com.example.myproject.OutputBufferFilter.doFilter(OutputBufferFilter.java:33)
at org.mortbay.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1157)
at org.mortbay.jetty.servlet.ServletHandler.handle(ServletHandler.java:388)
at org.mortbay.jetty.security.SecurityHandler.handle(SecurityHandler.java:216)
at org.mortbay.jetty.servlet.SessionHandler.handle(SessionHandler.java:182)
at org.mortbay.jetty.handler.ContextHandler.handle(ContextHandler.java:765)
at org.mortbay.jetty.webapp.WebAppContext.handle(WebAppContext.java:418)
at org.mortbay.jetty.handler.HandlerWrapper.handle(HandlerWrapper.java:152)
at org.mortbay.jetty.Server.handle(Server.java:326)
at org.mortbay.jetty.HttpConnection.handleRequest(HttpConnection.java:542)
at org.mortbay.jetty.HttpConnection$RequestHandler.content(HttpConnection.java:943)
at org.mortbay.jetty.HttpParser.parseNext(HttpParser.java:756)
at org.mortbay.jetty.HttpParser.parseAvailable(HttpParser.java:218)
at org.mortbay.jetty.HttpConnection.handle(HttpConnection.java:404)
at org.mortbay.jetty.bio.SocketConnector$Connection.run(SocketConnector.java:228)
at org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582)
Caused by: com.example.myproject.MyProjectServletException

Re:JAVA FTW

[rubycodez]

that's NOTHING. I can get even more spew at COMPILE TIME with C++, bee-yatch!!

Re:JAVA FTW

[The Evil Atheist]

Which is where you'd want most error messages to happen.

Re:JAVA FTW

[rubycodez]

or one could use a non-obtuse language that doesn't have Turing complete preprocessor macro system, and make less errors from the start

Re:JAVA FTW

[Xyrus]

1997 called. They want their overused Java meme back.

Re:JAVA FTW

[Luthair]

Have 4 bytes, aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa [buffer overrun]

Re:JAVA FTW

[0123456]

1997 called. They want their overused Java meme back.

I take it you've never used Eclipse?

Some days I'm lucky to be able to type three characters before it goes off and spends 30 seconds garbage collecting again.

Re:JAVA FTW

[sanosuke001]

It looks like most of your stack trace is from a third party library and not actually java. gg

Re:JAVA FTW

[phantomfive]

Java still has memory problems.
On a mobile device, garbage collection is fast, unless you are using up more than ~60% of RAM. Then it starts to get slow. So you have two options, either have unused RAM, or wait around for slow garbage collection.

On a larger server (like, ~30GB of RAM, not uncommon these days), garbage collection can take up to 10 minutes. A lot of people with clusters of Java servers have automated processes to detect and remove boxes from clusters while they are doing garbage collection.

Re:JAVA FTW

[geekpowa]

A total load of bollocks.

Re:JAVA FTW

[Jeremi]

or one could use a non-obtuse language that doesn't have Turing complete preprocessor macro system, and make less errors from the start

Or more errors, since your less capable preprocessor macro system can't do as much of the scut work on your behalf, which means you'll have to write (and debug) more code by hand.

Re:JAVA FTW

[The MAZZTer]

Preview of Java for Windows 10:

javax.servlet.ServletException: Something happened

Re:JAVA FTW

[edtice1559]

I'm not sure this is a true statement. There are lots of good alternatives to preprocessors that achieve the same goals without many of the drawbacks. The first one I think of is lambda expressions. Granted Java didn't get them until very recently. Templates/Generics also take away a lot of the need for the preprocessor. Most preprocessor work is around portability and the allowing compilation with different compilers. There was a time when you needed a wide swath of compilers to get code running on different hardware. For certain embedded device development, headers with conditional defines still make a lot of sense. But if you're running on phone or laptop hardware (or better), you're better off moving to a two-stage compilation like Java or .Net.

Re:JAVA FTW

[edtice1559]

The issue isn't that the messages happen at compile-time it's that they are very long and hard to understand. We want the errors to be found at compile time (good), but the reason for that is that we want to *fix* the errors. And that's much harder to do if you can't understand the output. Granted it's much better than producing code with undefined behavior. When I started working with C++11, I found it almost impossible to understand the compiler error messages if they involved anything related to templates. The messages have gotten better and, of course, I'm used to them now. But coming from the Java world, it really felt like an exercise in futility sometimes. More effort spent on understanding a minor syntax error than on actually writing code.

Re:JAVA FTW

[Anonymous Coward]

I use Eclipse every fucking day, no problems like this.

I even leave Eclipse running for several weeks at a time; only restarting it when I restart my computer, no problems like this.

Currently my system is fairly small, I'm only running with 10 projects at once. Normally my environment consists of between 30 and 50

I also have Java applications that run unattended for several months straight; nothing like this.

Sorry biach, the problem the problem isn't with Java, but the incompetent bozo behind the keyboard.

Re:JAVA FTW

[viperidaenz]

I use it all day, it's not too bad. When it's not overloaded with useless plugins, it runs fine with 512MB of heap space.

Re:JAVA FTW

[viperidaenz]

The "core event loop" doesn't need to do any garbage collection, ever.
Most phones sold now are dual or quad core and garbage collectors can run on their own thread mostly in parallel with application code.

fyi: the "core event loop" is a thread. Threads don't do garbage collections, the garbage collector does. It runs in it's own thread(s) and can run on any CPU core.

Re:JAVA FTW

[viperidaenz]

The only relevant part of the stack trace is the top line, which shows the coding error in is the doFilter method of com.example.myproject.OpenSessionInViewFilter

Not caused by Java OR the third party libraries.

Re:JAVA FTW

[rubycodez]

no, with superior langauges we make subroutines/functions/methods/closures instead of needing any preprocessor at all

Re:JAVA FTW

[phantomfive]

My worries were put to rest when the ops team told me how they're using JVM's with 6GB heaps with millisecond response times to serve websites with 1000's of concurrent users.

6GB and 1000s of concurrent users is small in today's server world.

Re:JAVA FTW

[The Evil Atheist]

I'm currently writing something using Javascript and WebGL. I spent a lot of time actually writing code. But "actually writing code" doesn't make me more productive. I think there's a tendency to conflate productivity with writing code. Certainly, the relationship between writing code and testing/debugging code is also forgotten. I find using templates (and their errors) eliminates the need for a lot of tests, especially those surrounding polymorphic aspects of design. I don't have to run the program in debug and step through if I can get the compiler to do that for me.

Re:JAVA FTW

[The Evil Atheist]

I'm pretty sure the "Turing complete preprocess macro system" was aimed specifically at templates. I don't think C macros are Turing complete.

Re:JAVA FTW

[GigaplexNZ]

There are lots of good alternatives to preprocessors that achieve the same goals without many of the drawbacks. The first one I think of is lambda expressions.

Ironically lambda expressions are essentially little more than just syntactic sugar (ie, preprocessor) implementations of functor objects in C++.

Re:JAVA FTW

[ckatko]

Actually, it's pretty pathetic that C++ doesn't give you a stack trace for exceptions.

Though, as an aside, that just reminded me of the equally-as-pathetic amount of Stockholm Syndrome exhibited by C++ programmers on Stack Overflow:

http://stackoverflow.com/quest...

You don't need it! They're useless! If you use it you're not a good programmer! Why would you want C++ to be like other languages?!

Re:JAVA FTW

[Anonymous Coward]

you make it sound as if turing complete-ness is something "big"... give me a break... *brainfuck* is turing complete.

Re:JAVA FTW

[TheRaven64]

I can't remember if I committed it in the end, but I had code in libcxxrt that did printed a stack trace for uncaught exceptions. It's not very hard. The Itanium EH ABI (i.e. the one that everyone except MS uses) is a two-phase unwinder. First it walks the stack to find where the handler is, then it walks it again actually doing the unwinding. If you don't find a landing pad, then the throw function fails and you can then walk the stack again to get a call stack. You could equally keep track of the function addresses during the initial unwind and wrap them up in the exception object, though you'd probably have to do this in the generic unwinder, as the C++ personality function is only called for C++ stack frames (you can throw C++ exceptions through, C, Objective-C, Ada, Fortran, and so on).

Objective-C's NSException class does a stack walk when it's instantiated (before it's thrown) and records the stack trace for debugging. It's fairly trivial (half a dozen lines of code) to make an exception class in C++ that does the same. The main reason that it's not done by default is that the stack walking is expensive. It's also expensive in Java, but a JITing JVM can optimise based on where the exception lands whether it will ever have its stack trace used (if nothing either rethrows the exception or reads the trace, don't bother computing it).

Re:JAVA FTW

[Assmasher]

Maybe you should use C++ before posting stupidity like this. If you mean to say C++ doesn't have built-in stack tracing you'd sound less stupid. Getting a stack trace, like many things in languages built before JAVA is a matter of including some other code and turning on a feature of your compiler. I get stack traces just fine in my exceptions.

C++ has mountains of capabilities in libs/code outside the standard, JAVA has it all packed into the standard - what's the big deal?

conclusion: languages are tools in a toolbox, stop getting pissy about any one of them. Hammer? Good. Screwdriver? Good. Right tool for the right job.

Re:JAVA FTW

[serviscope_minor]

Actually, it's pretty pathetic that C++ doesn't give you a stack trace for exceptions.

Given compilers are allowed, but not required to do things like tail call optimization, what precisely would you like the stack trace to be? Once you throw in the requirement for accurate (repeatable on all platforms?) stack traces, you hobble the optimizer. There's a reason C and C++ are consistently faster than just about anything else except FORTRAN.

Given the philosophy of C++ is not paying for what you don't use, this would be a pretty big penalty for those of us who don't need accurate stack traces. If you don't mind prefectly accurate backtraces, then there are a few different options which will give you the desired backtrace.

Tell you what, why don't you prove it's pathetic. Write a workable backtrace proporal for inclusion in the standard. If it's so pathetic, then fixing it ought to be really easy.

Re:JAVA FTW

[Beezlebub33]

They are not, because they lack recursion. However, you can use a script to iteratively run C macros so they are Turing complete. See stackoverflow discussion here.

And you can see a program that does it as part of the IOCCC here with hint file here

Re:JAVA FTW

[Kyogreex]

I much prefer C++ because it'Segmentation fault (core dumped)

Re:JAVA FTW

[The Evil Atheist]

If I remember correctly, there was a recent analysis done on popular open source C projects that showed most people don't abuse language features, so such examples simply do not warrant the hysteria surrounding the use of C features like macros.

Re:JAVA FTW

[Wrath0fb0b]

Lambdas are absolutely great (we have them in C++ even, I use them and evangelize 'em). But they don't quite solve the same problem as CPP. There are a few things that really need to happen at compile time ...

Re:JAVA FTW

[B4D BE4T]

You either:
1. Don't have enough memory installed on your system
2. Installed way too many plugins
3. Installed one plugin with reeeeally shitty memory management

Re:JAVA FTW

[ADRA]

It's a bad comparison since:

1. The error they spewed was a runtime error that could never be caught in compilation (duplicate primary key in the DB table)
2. The static analysis built into Javac is way more than I've personally seen in stock C++ compilers (though my knowledge is limited). Hell, most of the time people leave compiler warnings off even when it keeps bad smells out of code (through some safety flags are too aggressive on error setting)
3. There are a boat load of off the shelf OSS static analysis tools for Java that find significantly more complicated coding errors if the built-in javac variants aren't sufficient

The good thing is a knowledgeable developer could look at that and find the source of that specific problem in 5 seconds.

Re:JAVA FTW

[swillden]

On a decent OS, you get something better than a stack trace... you get a core dump. It takes more knowledge to use and work with, but it contains a great deal more information.

Re:JAVA FTW

[The Evil Atheist]

Yes, runtime errors that are environmental cannot be caught at compile time. But many other runtime errors are the result of design errors - designs that lead to more possible mistakes. When given a bit of thought, you can make those errors get flagged at compile time.

For example, the classic criticism of C++'s resource management. C++ doesn't have those problems if people used the RAII that's provided by default. You don't need to analyze for leaks or null pointers if those cost-free features (which are also semantic signals) are used. And by leaks, even stuff like managing database resources can be taken care of by RAII, leading to reduced environmental errors that may lead to things like duplicate primary keys being created.

Static analysis tools are okay if it comes to snippets of code changes, but can't tell you much about design flaws. I too often see people making local changes to "fix" the error when they should have given more thought about reorganizing even just the code of that function a bit. The other thing that is popular these days in C++ are generic algorithms. The properties of generic algorithms are known, including the elimination of most range errors. Static analysis doesn't tell you that you should replace your hand-crafted for loop (or thread synchronization) to use a more suitable higher level construct. In Java, that may as well be a good thing, since high level implies efficiency cost in a way C++ generic high level constructs don't.

Re:JAVA FTW

[edtice1559]

I can't really say anything about lambdas in C++ as I don't think I've ever even encountered something. In Java and C#, they are "syntactic sugar" in that they get expanded into more fundamental constructs. But they have stronger type-checking than an equivalent implemented with a preprocessor and are first-class entities. Passing a lambda to a function is not the same as passing the result of something from the preprocessor. The lambda gets executed in the future.

Re:JAVA FTW

[allo]

today it's more true than ever. java scaled with your ram. today a single java program eats 1 GB RAM, where it ate 20 when 64 MB were the norm for your computer.

Re:JAVA FTW

[allo]

4. made the mistake to start eclipse

Oracle/SCO

[Runaway1956]

People are reverse engineering their shit, and Google is destroying value. This is a pretty sure sign that Oracle is going down.

Ooops, misread the headline

[gnunick]

At first, I read that as "Oracle Has 'Destroyed' the Market For Java"... which, of course, seemed quite plausible.

RIP SUN

Re:Ooops, misread the headline

[garyisabusyguy]

Yes, I had a couple of Sparc Stations in the 90's and admined SunOS and Solaris on those and some enterprise server systems, but

Fuck Sun, they favored proprietary server systems that lined their sales-reps' pockets with cash while the world changed around them and then sold all of their knowledge lock stock and barrel to Oracle, simply because Oracle users were their largest remaining customer base

I feel the same way about DEC, who flushed thirty years of Alpha architectural superiority down the drain because they couldn't sell their way out of a wet paper sack

We get what we deserve because we let the free market reign supreme where the most cut throated business-people win and the rest go down the drain

Re:Ooops, misread the headline

[hey!]

At first, I read that as "Oracle Has 'Destroyed' the Market For Java"... which, of course, seemed quite plausible.

Too late. Sun already did that -- at least if we're talking about Java as a mobile platform. I spent years tracking J2ME as a potential target for our apps. Java may not have been all things to all people, but back in the day (late 90s early 00s) personal basis profile would have been ideal for what we were doing. Even MIDP would have been a good match.

The problem is that there never was *a* standard J2ME implementation; J2ME was only a set of specifications. Implementations came from third parties and they were either incompatible in various tricky ways or they were impractical to distribute to customers (e.g. IBM's Palm implementation of J2ME, which was almost impossible to buy unless you knew the right people at IBM, none of whom were interested in app developers). In practice that meant that if you wanted to develop for J2ME you were tied to phones from a particular carrier, and couldn't target PDAs or early tablet computers at all.

When I heard about Dalvik I was overjoyed. Not because I thought it would be any fundamentally better than J2ME, but because Google was going to use it to create a cross-vendor, cross-carrier market for apps. I have no idea what the whole point of J2ME was for Sun. So far as I can see just about all the effort Sun put into it accomplished nothing.

Re:Ooops, misread the headline

[dfghjk]

"I feel the same way about DEC, who flushed thirty years of Alpha architectural superiority down the drain..."

What 30 years is that? Alpha lasted 12 years, not all of it (if any) with "architectural superiority". Loved their integer divide...or total lack of one. ;) It's not so hard to design a processor to be fast at doing very little. Power consumption was absurd as well. The Alpha failed for well deserved reasons, not the least of which was lack of a value proposition.

Re:Ooops, misread the headline

[myowntrueself]

At first, I read that as "Oracle Has 'Destroyed' the Market For Java"... which, of course, seemed quite plausible.

RIP SUN

Who the FUCK buys or pays for Java??? Any server I maintain thats got Java (Elasticsearch anyone?) uses OpenJDK.

Heres the news, Larry; Java is FREE and people use it free, gratis and for NOTHING. There is and never was a market for Java.

Seriously, people pay for it?? Cry me a river.

Re:Ooops, misread the headline

[garyisabusyguy]

When Alpha was introduced they laid out thirty years of growth for the platform, primarily because they designed for the long run. When HP was replacing Alpha with Itanium, they had to suppress Alpha benchmarks because it made their 'enterprise' chip look like the garbage that it was

FWIW all RISC chips sucked at integer division, to quote:
"When integer multiplication is cheaper than integer division, it is beneficial to substitute a multiplication for a division."
https://gmplib.org/~tege/divcn...

Re:Ooops, misread the headline

[Trailer Trash]

We get what we deserve because we let the free market reign supreme where the most cut throated business-people win and the rest go down the drain

None of that stuff would have existed in the first place absent the free market :-/

Re:Ooops, misread the headline

[Just Some Guy]

We get what we deserve because we let the free market reign supreme where the most cut throated business-people win and the rest go down the drain

Oh yes, I'd much rather have a centrally planned technology base, because the best ideas always come out of committees and government procurement processes.

Re:Ooops, misread the headline

[gnunick]

Sun was no saintly operation, to be sure. I worked there briefly and hated every moment of the mega-corporate lifestyle (not to mention lack of a sense of direction), and went back to working for small companies. I was never a fan of Scott McNealy, but of course it was (and is) difficult to despise anyone more than Larry Ellison.

There were a lot of good people working at Sun, and they accomplished some great stuff, but I have to admit the only reason I wax nostalgic for the company is because I do lot of work with Java (though I thankfully there is OpenJDK) and I feel a sense of shame to be even *that* loosely associated with Oracle.

Re:Ooops, misread the headline

[Christian Smith]

Yes, I had a couple of Sparc Stations in the 90's and admined SunOS and Solaris on those and some enterprise server systems, but

Fuck Sun, they favored proprietary server systems that lined their sales-reps' pockets with cash while the world changed around them and then sold all of their knowledge lock stock and barrel to Oracle, simply because Oracle users were their largest remaining customer base

Eh? While SUN's machine were proprietary, in that they designed and sold their own machines, they were far from locked in platforms.

SPARC was covered by independent standards, and available from multiple sources. IO buses were all standard (VME, then SBUS, then PCI.)

They even produced open specs for their firmware. What more could a proprietary vendor do to be open?

Really???

[Anonymous Coward]

Does anybody feel sorry for Oracle???

Re:Really???

[Anonymous Coward]

Saddam Hussein . He loved shitty companies like Oracle.

Re: Really???

[Anonymous Coward]

Gartner probably feels sorry for Oracle.

Give me a break

[Anonymous Coward]

There was no "value proposition" for Java, Sun eventually made it open source with new APIs and language features created through a community development process, in order to compete against Microsoft and others. That's why IBM, the biggest corporate cheerleader for Java, refused to increase its $8 billion acquisition offer for all of Sun Microsystems including their server and storage businesses, as well as Solaris, Java, and MySQL.

But Larry was/is intent on pulling a Darl Mcbride and figuring out how he can take Java closed so Oracle can charge users serious dough. He can't, that genii flew.

Re:Give me a break

[garyisabusyguy]

Darl is the equivalent of Larry's inbred idiot second cousin
Do not predict Larry's failure because Darl failed

Oracle, please look in the mirror

[JPyObjC Dude]

The contenders for biggest enemies of Java are:

1) Micro$oft - Effectively killed the JavaBean web plugin market with their own lackluster JVM via EEE (Embrace, Extend, Extinguish).

JavaBeans is the technology that has the biggest negative view on the net and rightly so. If Microsoft had not done such a good job killing it, Java would likely be in a different light today as more energy would have been spent making JavaBean libraries better while the real engineers at Sun still had control of the source.

2) Oracle - They just do not get open source or anything that came from Sun.

Google has popularized Java way more than Oracle could ever imagine.

"You don't anthropomorphize your lawnmower, the lawnmower just mows the lawn, you stick your hand in there and it'll chop it off, the end" - B. Cantrill

Re:Oracle, please look in the mirror

[phantomfive]

More specifically, Oracle doesn't understand you can't monetize everything.

They understand it. They just don't care about stuff that doesn't bring in money.

Re:Oracle, please look in the mirror

[Anonymous Coward]

1) If you are ever trying to make a point or prove your words hold some validation, don't call them Micro$oft. Unless you are working for some hipster company, then nobody even remotely feels that Microsoft is an evil empire anymore. Bill Gates has proven to do more for computing and charity then 100 Steve Jobs' could ever do.

Bill Gates isn't at Microsoft anymore. Besides, just because he runs a charity doesn't wash away the evil he's done to the industry nor the gains he's made. You do not get a membership to sainthood just because you spend some of your blood money here and there.

Re:Oracle, please look in the mirror

[GigaplexNZ]

Microsoft is run by a different set of people now. Your stance practically states that future generations of Germany as a whole can't be trusted because of Godwin's Law.

Re:Oracle, please look in the mirror

[JPyObjC Dude]

I knew I would get some flaming for that. However, considering that I was pretty hot on Java back in the day and I was super pissed with what Microsoft did by single handedly killing JavaBeans, I can't help but slip in a $ in once in a while for old time sake ;)

Today, I feel bad for all those system admins out there today who need to administer IPMI's. Talk about a massive pain in the butt to get to a damn IPMI KVM window to open let alone work properly.

Ask.com has destroyed the market for java

[De_Boswachter]

FTFY

Oh brother

[Tablizer]

Oracle.self.project.sins(google);

Palm

[ChadSmith4920]

Google's fall back plan switching to PalmOS in C++.

Sorry... to what market are they entitled again?

[Falconnan]

Quick message to Oracle: Between the security officer coming off as, at best, a self-entitled, over-inflated executive believing her services are better than her customers deserve, and now the company claiming they have a "right" to market share, I think the psychology of the company is becoming quite clear. It isn't that their products are poor (which they are not), it is that they seem to believe that they deserve their piece of the market by divine providence.

Please note, this is opinion, and only my own. Note to Oracle: One of the only things anyone is actually entitled to. Meaning: You are welcome to the opinion that you are above competitive market forces, but reality may or may not disagree. Be humble, quietly make better products than your competitors, and demonstrate effectiveness in the market rather than in the courts.

Java could be great. Make it so, or let those who can do better do better. That is the nature of the free market large corporations are supposed to embrace.

Re:Sorry... to what market are they entitled again

[binarylarry]

Oracle's products are extremely poor.

Cry me a river

[chromaexcursion]

Sun destroyed the market for Java.
Sun wanted to sell hardware, and they designed Java to run well with their hardware. Sun's ideal was the network is the computer. Java is/was a client language that could run on a lot of platforms, with in Sun's mind a Sun server at the other end. Didn't quite work out that way. Sun was going belly up, Oracle bought the carcass. Sun gave Java away. You can't put the jinni back in the bottle.
Java was worthless when Oracle bought Sun. They're engaged in revisionist history trying to milk a dead cow.

Re:Cry me a river

[binarylarry]

Uh most of the world's business and web software ran on Java when Oracle bought Sun.

It was a huge reason why Oracle bought Sun in the first place, gave them some leverage over IBM, Google and other companies.

Re:Cry me a river

[MouseR]

Oracle had (and still have) much reliance on Java and Sun hardware for their server & middleware tiers. They simply could NOT let it die along with Sun or, worse yet, let it pass to a competitor.

Disclaimer: that's as far as I will comment on that issue, as I am an Oracle employee. Though I have nothing to do with the Java or Sun group, the native mobile apps I develop eventually touches Java code, server side. It's everywhere.

Re:Cry me a river

[edtice1559]

Prior to that, most software ran on C/C++ on Unix. Then people moved to Linux. Oracle's argument here would be that implementing libc on Linux shouldn't be allowed as it destroyed the value of Solaris. Maybe they will litigate this next. They can buy up SCO and make the same argument.

Re:Cry me a river

[ADRA]

Java has immense value, may the single greatest value outside of Windows/Office/Linux/C&C++ compilers, but it has practically no revenue. Almost all very large enterprise systems that sell for obsence amounts of money are built on or inter-operate with Java in fundamental ways. They don't sell PC language/runtime, so instead, Sun monitized:

1. Mobile platforms, because the 'market' for the runtime was only a few carriers they could get away with charging them big bucks for the dev. This market is now essentially dead with the end of 'dumb' phones and carrier/manufacturer locked platforms (IOS started the ball rolling, but Android gave it the knockout punch)
2. Very niche JVM improvements (like RTJava) which are technically still kicking around but very low on revenue
3. Add-on's to core functionality (Middle-ware servers). Despite there being significant OSS competition, they still make boat loads of money here

Oracle bought and kept Java for no other reason than it was a cornerstone to so many of their profitable businesses that they couldn't have it ruined by outsiders. The sad fact is that the biggest detriment to Java has been Oracle themselves, but that's a completely different matter.

Keep up, or fall behind

[Nyder]

The tech industry, just like every industry, improves as people discover new and better ways to do things. If you can't keep up Oracle, you fall behind. And since you've chosen to litigate instead of innovate, you have fallen behind.

No one is guaranteed profit.

Re:Keep up, or fall behind

[brxndxn]

Is there anything that Oracle does that people (tech people and developers) really like? I still don't understand their killer platform.. All I see is that they buy old proprietary software and support it for companies that are already locked into it. Of course they should end up in court.. It's where the last remnants of SCO ended up. If Oracle doesn't have some kind of new technology that is 'awesome' or even relevant, then court and more court..

Re: Oracle's monopoly?

[Anonymous Coward]

Are we supposed to believe that business models in general must be protected from disruptive technologies?

Re: Oracle's monopoly?

[garyisabusyguy]

Yes, and what exactly did Sun do with that?

Oh yeah, there is JavaFX, which still requires an OS to run on

That is where the disconnect is

Re: Sorry... to what market are they entitled agai

[binarylarry]

I have bought them and Oracle products suck. For developers and end users.

Oracle works its miracle

[Anonymous Coward]

You must give Oracle credit.
Oracle has taken the remnants of Sun and has now, through hard work and determination, turned it into the World's Greatest Patent Troll.

Re:Oracle works its miracle

[gnupun]

Java is no longer the carrot/bait to get users to buy Sun computers. Java is far more important than commodity hardware and Sun took too long to realize that or monetize Java's value like Oracle is trying to.

Frankly, IP laws are quite lax if Google can simply take major parts of Java, reimplement the remaining parts, and pay Oracle exactly $0.

Re:Oracle works its miracle

[invictusvoyd]

IBM is the worlds greatest patent troll . Oracle and the jolly fellas are not even close.

Re:Oracle works its miracle

[erice]

IBM is the worlds greatest patent troll . Oracle and the jolly fellas are not even close.

IBM arguably has the world's largest arsenal of dubious patents. However, they don't seem to use them for anything but defense. If they've done any trolling at all, it has been on a small enough scale that few noticed.

Re:Oracle works its miracle

[tsotha]

In addition to the largest arsenal of dubious patents, IBM has a huge arsenal of non-dubious patents, some of them quite revolutionary (like the 1997 patent for copper interconnects). I don't begrudge them patents on the wheel as long as they don't start to bully other companies.

Re:Oracle works its miracle

[TrueSpeed]

Java is no longer the carrot/bait to get users to buy Sun computers. Java is far more important than commodity hardware and Sun took too long to realize that or monetize Java's value like Oracle is trying to.

Frankly, IP laws are quite lax if Google can simply take major parts of Java, reimplement the remaining parts, and pay Oracle exactly $0.

Your recollection of history is quite lax. The VM patents Oracle asserted against Google were all defeated. All Oracle had left was the SSO of 37 Java API's. If the SSO of a subset of an API collection can be copyrighted then the US software industry is in for a world of hurt.

Personally, if Oracle wins on copyright, I hope IBM goes after Oracle.

Let me rephrase that.

[fahrbot-bot]

Oracle ... Now, claiming that 'Android has now irreversibly destroyed Java's fundamental value proposition as a potential mobile device operating system, ...

... for profit.

Oracle Bundles ASK toolbar with Java

[Anonymous Coward]

To put this in context, Oracle thinks so little of Java it slaps a malware (^h^h^h^h) adware bundle with the installer that installs ASK toolbar to take over searches.

http://www.zdnet.com/article/oracle-extends-its-adware-bundling-to-include-java-for-macs/

One of the reasons I've stopped installing their 'critical updates' and instead turned it off.

Re:Sun

[garyisabusyguy]

Larry doesn't want to sell, he wants a percentage of the action

Haul both to court

[VikingNation]

Oracle's lack of quality control has brought untold numbers of zero days which has eroded confidence and trust in the 'compile once run everywhere promise.' Likewise Google has fractured the mobile handset and tablet eco-system with too many variants of a 'standard' framework . Both companies should be taken to court by a class action lawsuit and sued for a gazillion dollars.

Re:Haul both to court

[TrueSpeed]

Oracle's lack of quality control has brought untold numbers of zero days which has eroded confidence and trust in the 'compile once run everywhere promise.' Likewise Google has fractured the mobile handset and tablet eco-system with too many variants of a 'standard' framework . Both companies should be taken to court by a class action lawsuit and sued for a gazillion dollars.

By that logic, Microsoft should be exterminated for the havoc they've inflicted on the computing industry for releasing an OS as porous as a cheese cloth and a house of cards for a foundation.

Re:Haul both to court

[VikingNation]

How many zero days were there in Java before Hor-acle took it over? Enough said.

Re: Oracle's monopoly?

[binarylarry]

Lol javafx!

I'll check for the latest news...

[JBMcB]

I wonder if I can use the Oracle for Android app to keep up to date on the lawsuit against Android...

http://www.appszoom.com/androi...

Re: Google/Android should replace Java with C#

[binarylarry]

Yeah because why only have a legal battle with one evil software compnay, when you could battle two at the same time! Just to show off how much of a silicon badass you are.

Cmon Oracle and Microsoft, Google will fuck your shit up 2 on 1!

What losers

[Anonymous Coward]

“Given the widespread dominance Android has achieved with its continued unauthorized use of the 37 Java API packages over the past few years, Android has now irreversibly destroyed Java’s fundamental value proposition as a potential mobile device operating system,” Oracle wrote.

Judge: So I see here that Google has made Android a very popular and successful mobile OS over the past five years. What has Oracle done over the past five years to promote the JAVA platform as a potential mobile device operating system?

Oracle: Well Your Honor we have studied new and innovative ways to litigate our position.

Judge: No, you misunderstand. What have you done to improve and promote JAVA as you say Google has done?

Oracle: As we said Your Honor, our lawyers have devised a strategy ...

Judge: So you really haven't done a single thing other than plan legal ways to squeeze money out of people actually developing useable mobile systems?

Oracle: Ummm, we will take the fifth on that Your Honor.

Re: Oracle's monopoly?

[goarilla]

Reminds me of Project Looking Glass back in the day.

Oracle Rage

[Anonymous Coward]

Someone should make one of those raging Hitler movies (from Der Untergang) with quotes from the Oracle tirade, especially "Oracle: Google Has "Destroyed" the Market For Java"

Mobile OS?

[Rassleholic]

I thought Java was only good for CS college courses and Minecraft.

Re:Java was GPL'd

[gnupun]

a lot of vague claims about the interface being copyrightable (based on the names of the classes! i.e. trying to claim the interface as copyright material).

Fine. Next time I have a large project I will ask you to analyze the problem and design the interfaces/class declarations. I will then take/copy those interfaces from you without paying you a penny. Any half-competent programmer will implement those interfaces you donated for free.

Oracle is...

[subk]

Oracle is the new SCO

Jython

[tepples]

Can you imagine a interpreter of perl or python written in Java where no native option possible?

Yes, and it's called Jython.

Oracle confuses language and operating system

[Morgaine]

'Android has now irreversibly destroyed Java 's fundamental value proposition as a potential mobile device operating system ,'

Java is a programming language, not an operating system. Examples of operating systems are Linux and Unix.

Nothing could have "destroyed Java's fundamental value proposition as a potential mobile device operating system" because the value proposition of Java as an operating system is zero, and always has been. It's like the value proposition of an orange to be an apple.

Oracle's nonsensical claim might be merely a case of lawyers or managers showing their ignorance of the computing subject domain or just being sloppy with their terminology, which is not uncommon. However, it gets worse.

A proprietary software package may have a calculated expectation of market share and profit if there is no competition, but this is not the case with programming languages because they always have competition from countless other languages. It is especially not the case with open source programming languages because they typically enjoy multiple implementations, and these make captive markets almost impossible to maintain.

It seems therefore that Oracle's market expectations were based on a flawed analysis.

That mistake would have made any market expectations unsafe, but any expectations were dealt a further blow by Oracle's highly abusive attempt to copyright SSO in their litigation against Google. This must have alienated practically everybody who knows anything about programming, and the likelihood is high that many Java programmers who had other languages available must have abandoned Java like the plague to avoid potential SSO copyright liability.

In other words, if anyone killed off interest in Java, it was probably Oracle themselves.

Re:Oracle confuses language and operating system

[timrod]

"Examples of operating systems are Unix and Linux.

I'd like to interject for a moment. What you're referring to as Linux is in fact GNU/Linux or GNU plus Linux. Linux is not an operating system unto itself, but rather another free component of a fully functioning GNU system made useful by the GNU corelibs, shell utilities, and vital system components comprising a full OS as defined by POSIX... etc etc.

Re:Oracle confuses language and operating system

[bluefoxlucid]

The core services of the system required to actually function include, mainly, kernel event systems not supported by GNU kernels; message passing buses using systems not supported by GNU kernels; graphics display systems developed well before GNU; init systems that don't even work on GNU kernels (systemd; upstart was also made by Canonical, though), and which use non-GNU shells to bring up the system (/bin/sh is the alquist shell, ash; Debian uses their own version, dash); and so forth.

The entire support system to bring up the OS, get you logged into a desktop environment, and run Chrome or Firefox to get on Slashdot is non-GNU, save for glibc--which is a highly-replaceable core C library (some Linux distributions even use a different libc). Further, modern glibc and other core GNU libraries contain specialized code to access Linux-specific kernel functions, thus making them, at best, GNU for Linux, rather than a primary and defining component of GNU/Linux. The GNU components are the smallest and most insubstantial parts of the operating system as a whole.

Re: Google/Android should replace Java with C#

[edtice1559]

Google doesn't have to do any such thing. They can just get rid of the legacy Java APIs and, instead, provide modern ones using lambda expressions. Then they can provide a tool chain that goes directly from source to .dx files and skips the whole Java .class file intermediate step. I'm surprised actually that we still even have to compile to Java classes. Then, in order to have write-once, run-anywhere we can get an ART VM for non-Android operating systems.

IBM will provide OS/2

[jfdavis668]

IBM will finally find a use for OS/2, as a mobile OS to replace Android!

Re:What losers

[viperidaenz]

Oracle: Ummm, we will take the fifth on that Your Honor, by the way, did you get the invite to party on Larry's superyacht?

Judge: Oh yes, the court rules in favour of Oracle.

It always sucked

[jakeelee]

Java sucked from day one. Anything that can be "destroyed" by MS, Oracle, IBM or Sun was never worth a crap to begin with. I've programmed with MS .Net and Java and every Java solution took longer to code, was less reliable, slower and consumed more system resources. It got hyped because it was an alternative to MS back in the days when Ellison and McNealy went crying to the government over MS's supposed "monopoly".

Re:It always sucked

[TrueSpeed]

Is that why Java rules the enterprise and Microsoft is relegated as a distant outsider? Java pretty much dominates enterprise development. C#? Not so much. Just ask the London stock exchange for experience with Microsoft and C#.

Re:It always sucked

[Anonymous Coward]

Is that why Java rules the enterprise and Microsoft is relegated as a distant outsider? Java pretty much dominates enterprise development. C#? Not so much. Just ask the London stock exchange for experience with Microsoft and C#.

Since the London Stock Exchange switched to Linux/C++, your comment is irrelevant to this discussion.

http://www.computerworlduk.com...

Convert Android code to Go

[Anonymous Coward]

Maybe Google should convert Android code to their own Go programming language.

Re:Google/Android should replace Java with C#

[TrueSpeed]

You should have a look at the C# license. I believe it stipulates that you must maintain compatibility with core classes or face the consequences.

Re: Oracle's monopoly?

[MacTO]

In general, no. Unfortunately, Oracle owns the rights to Java. That means that Google must comply with Oracle's terms within the limits defined by law. Doubly unfortunately, that means that everyone with a vested interest in the situation is going to have to watch a drawn out soap opera as the case winds its way through the courts.

Re:Sorry... to what market are they entitled again

[Falconnan]

I don't know about that. I do perceive them to have the unfortunate combination, however, of being priced to make their customers poor(er) while not being anything special. In the end, high-priced mediocrity is merely high-priced.

Re:Oracle's monopoly?

[Z00L00K]

Oracle are the ones that have destroyed Java since nobody trusts Oracle and their licensing.

Re:Java was GPL'd

[GigaplexNZ]

And unless you pay me some fraction of the costs up front, I wouldn't even bother doing the work to design the interfaces. That's a completely different situation.

Re: Google/Android should replace Java with C#

[GigaplexNZ]

What legal battle with Microsoft? As the OP already stated, Microsoft has put C# out there for 3rd party implementations with a legal guarantee not to sue.

Re:Google/Android should replace Java with C#

[GigaplexNZ]

So, a few issues here and there, but really, why not?

The amount of effort required. Not only does Google have to rewrite the relevant portions of Android to use C#, all the apps in the store need to be ported too.

Re: Oracle's monopoly?

[nikkipolya]

I have used iPhone and iPad. They gave me a shitty experience. I felt like my devices were still owned by Apple and whenever I was using them I was just doing time. I moved to Android and it feels much better. I have a lot of freedom.

Had Google just straight up used C or C++ we wouldn't be in this predicament.

I sort of agree with you on that point. But C++ lacks core language support for so many fundamental things. Then there are a myriad 3rd party libraries with their own quirks.

Re: Oracle's monopoly?

[gl4ss]

sony-e had a working prototype all java(with I suppose their own os underneath). basically android was a clone of that.

but what sunoracle fucked up in the mid 00's was being too slow in developing j2me extensions(and the 'all java' phone os that they kept in different projects for years and years) and just badly managing how they could be used(four security dialogs for creating a file in a folder on the sd card each with two clicks from the user, for example - NO MATTER WHAT SIGNING YOU PAID FOR), thus the market for android was there when android emerged.

as for j2me, the process a new API went through to be an approved API was just stupid. the end result was api's that had always some flaw on them or were just unusable from the day 1, like the j2me 3d _scene_ descriptor shit, which was just a wrong, wrong way to go about it on the hw and use it was launching for(like, the api might have been ok for making some animation suite or whatever, but shitty for making games).

there was a market for a java development based smartphone os all right.. they just dragged their feet on it for way too long, so that market hole is now filled with android.

they just didn't care about it enough to make sure that the shit they were certifying and dictating how it should be was usable at all.

Java is not an operating system

[viperidaenz]

I don't see how this argument is any different than "Apple's used of an operating system has destroyed the market..."

BTW Oracle, the Android OS is not built in Java. At it's core it is a Linux kernel. There just happens to be a Java-like API and VM for running applications.

You are wrong

[Anonymous Coward]

To avoid data corruption, race conditions, and deadlock, the Java garbage collector must obtain exclusive access to memory. While the garbage collector is running, no thread on any core can access the same memory as the garbage collector. Java's garbage collector is called a "stop the world" garbage collector for a reason.

From Oracle's own site: "Major garbage collection are also Stop the World events. Often a major collection is much slower because it involves all live objects. So for Responsive applications, major garbage collections should be minimized. Also note, that the length of the Stop the World event for a major garbage collection is affected by the kind of garbage collector that is used for the old generation space." http://www.oracle.com/webfolder/technetwork/tutorials/obe/java/gc01/index.html

The approach used in Objective-C, Swift, some Python, some Ruby, and possibly in new C# provides the benefits of garbage collection along with deterministic behavior and no "stop the world" situations.

they're just dickheads

[markhahn]

Java was never that great, and Sun/Oracle tried their best to monetize it, maximize the leverage they had over their network-effect monopoly. So boo-hoo that reality didn't go their way.

The courts are not a valid way to fix your competitive failures.

Re: Oracle's monopoly?

[Wycliffe]

But I digress, Android is an even worse clusterfuck by itself because nobody wants to develop for it, and only develop content for it because some bigwig in their company says so.

People look at iPhones and iPads and then get bait&switched by AT&T/Verizon/Rogers/Bell/etc being told that these Android devices are exactly the same device and experience, but cheaper. Then they get a substandard experience and believe that all mobile devices are a shitty experience.

I've have iphones, ipads, and android devices. I much prefer the android devices. Despite apple's reputation of "just working", I have found
that for me, the androids "just work". They android devices have a better user experience, have more configuration options, have less "gotchas"
and just are all around more pleasant and easier to work with not to mention that google is considerably better than siri as are google maps.

Ditto for developing. I much prefer developing android apps vs iphone apps. The only downsize with android is having to pay attention to different
resolutions for different devices. This is a minor inconvenience for a developer but gives the end-user a lot more choices on what type of phone
they want so I think it's worth it. Have you actually owned and used both an iphone and multiple android devices as your primary device for any
length of time? Yes, there are some crappy android phones out there but the top of the line android phones are every bit as good as the iphones
if not better. I'm currently using the iphone 6 as my primary phone but will probably switch back to the android when it's time for a new phone
as I much prefer the android. I have a friend who used iphones 4-6 and recently switched to android as well and doesn't plan on going back.

Still don't understand

[bayankaran]

Why Google did not buy Sun?

Re:Still don't understand

[allo]

why? Seems to work good for them without buying it.

Java is slow

[Karmashock]

I've never used a Java program had equivalent performance to a similar compiled program.

Just my experience. And in that context, I have a hard time respecting the idea of lots of java nonsense on the android. Maybe I'm being unfair. But I've found Java to be more of a problem over the years than a solution. It starts out as a nice thing. Its easy to develop something in Java and easy to tweak and fix stuff. But it has limitations.

Re: Oracle's monopoly?

[wierd_w]

Adnroid lives on top of linux kernel. Dalvik, the Google fork of Java that Oracle is crying to mommy over, still has this problem. Thankfully, the OS underneath android is very powerful and useful, and it can be tamed for use by the user if they get root access to it.

Re:Oracle's monopoly?

[Znork]

No shit. If not for Android, maybe Nokias or Microsofts efforts would have been more successful, but Oracle has managed to create enough antipathy that people will go out of their way to avoid them by now.

Oracle has destroyed the market for Java

[Anonymous Coward]

With these ridiculous claims that interfaces can be copyrighted, Oracle themselves killed java.
No company is going to use a language where someone claims to assert copyright over the APIs.
At my company, when management discovered this, all Java development was killed right away.
It's a real shame, but when a company's management goes deranged, you don't want to be anywhere near their path.

Re: Oracle's monopoly?

[Threni]

Why didn't Google leave Java as soon as this lawsuit kicked off? Just move to some other language, or provide some wrapper around c++ to keep the noobs happy. We'd be 3 years further along now. Must be galling for anyone at Google to see the "3 billion devices run java" splash screen when over a third of those are android devices they're getting sued over.

Re: Oracle's monopoly?

[gl4ss]

qt was owned by Nokia at the time, being developed in house in Nokia to be basis of programming for their mobiles.

gtk was and still is a piece of shit, abandoned for use on mobiles already because coding for it is so shit.

anyhow, google went with java because a lot of mobile devs were already familiar with java syntax and if they just offered them something better than j2me the devs would just jump straight over.

"only because some bigwig in the company says so" in regards of java is _BULLSHIT_. perhaps you're thinking about windows phone for which it is true - not because wp development would be a pain, but because wp has no market share - but there's plenty of willing android developers out there, perhaps because it's not such bullshit as objective c.

with android you get choice from 70 dollars to 700 dollars, that's the whole point. but then again some devs could only do windows applications for hardcoded 1024*768 size too so whatever..

Oracle has destroyed all achievements of SUN

[Anonymous Coward]

And now they even want to destroy the only working free and open-source mobile operating-system .. this is like destroying the ocean for profit, destroying the rain forest for profit or destroying free wisdom by patents .. so in the end Oracle yes indeed is as evil as any other terrorist / capitalist.

Whoracle is Full of It

[BrendaEM]

If Java was profitable, wouldn't Sun been profitable in the first place?

Re:Whoracle is Full of It

[e70838]

When Oracle has bought sun, all their prices have increased by 100%. If sun had increased its prices by 100% a couple of years before, it may still be there.

Wh Oracle is Full of it.

[BrendaEM]

If there was a good market for Java, wouldn't Sun have been been profitable in the first place?
(Let's see if that title is acceptable for the filters.)

Re:Oracle's monopoly?

[Kartu]

I don't see that happening server side.
And desktop/mobile, cough, Java never really got serious traction there.

Then you don't quite get a number of things

[Kartu]

1) Nothing stops JIT compiler to produce code that is faster than static compiler, on the contrary, JIT compiler can do faster, knowing runtime data (i.e. which branch is more likely to execute) OK?
2) I've written "parse file and feed it into DB" Java code that beat C code written by a colleague. (used JavaCC/JDBC)
3) In general, Java loses vs C on the following fronts:
a) operations like sin/cos, since it must produce exactly the same result on all platform
b) memory footprint
c) uh, no stack, just very fast (compared to C) heap

All that is not because of JIT compilation, but rather because of automatic mem-management. That's the price you pay for having easier time diagnosing / fixing errors.

In real life, mem footprint is the major difference you'd notice. It's in no way "times slower".

Re:Then you don't quite get a number of things

[Karmashock]

1. I've never seen it. Every Java program I've handled has been slower.
2. I've never seen it. I've used a lot of java programs and they've always been slower than C/C+ etc programs of similar nature and complexity.

3b. I've noticed the memory issue. I've also noticed a lot of java programs seem to have a hard time going beyond 1GB of ram. I'm sure there is a way to make them do that but... I've had to screw around with work arounds more than a few times to deal with that issue.

As to your claim that it isn't slower if it has enough memory... That's not my experience. I'm sure I could get you testimonials and links to people talking about Java being slow. But I rather suspect you won't listen to it or will say it is invalid for some reason.

I've got no problem with java on simple things but when java programs start getting really complicated I frequently find them to frustrating. I think part of the problem is the java libaries. A java program coded for one doesn't seem to behave the same way if given a newer version sometimes. And that's one of the more irritating things about Java that I have to fix on a regular basis. Making sure all the java applets have the EXACT version of java they're supposed to have. And when that becomes problematic the java applets have to be recoded.

We have less bullshit to deal with the compiled programs. They just "work" more reliably.

Re:Then you don't quite get a number of things

[IamTheRealMike]

3b. I've noticed the memory issue. I've also noticed a lot of java programs seem to have a hard time going beyond 1GB of ram. I'm sure there is a way to make them do that but... I've had to screw around with work arounds more than a few times to deal with that issue.

There are people happily using Java with 300 gigabyte heaps. Look at the Azul Zing JVM for examples of this. Also: they're using it in ultra-low latency financial trading apps. Just because you haven't seen this sort of thing personally doesn't mean it never happens.

As to your claim that it isn't slower if it has enough memory... That's not my experience. I'm sure I could get you testimonials and links to people talking about Java being slow. But I rather suspect you won't listen to it or will say it is invalid for some reason.

Performance is complicated. There are lots of cases where a Java program is just as fast as a C++ program or even faster. PIC-optimisable virtual method dispatch in a tight loop is one example of where Java/JVM stuff has stomped C++ for many years, with devirtualisation optimisations only appearing very recently in GCC stuff it seems. HotSpot is an excellent compiler and can do a lot of interesting things.

Moreover, it's not like for any program there's a choice of Java or C. Many developers use languages like Ruby or Python. It turns out that there's an advanced research JVM which allows you to co-compile Ruby and the C source code of Ruby/MRI extensions together with performance that's radically faster than the original code.

But mostly, people use Java because the performance is good enough, and the benefits over the C/C++ ecosystem are big. For instance, you get reliable debuggers, stack traces that are never corrupted, no manual memory management, ultra-fast compiles, a huge and standarised package repositories/dependency management system, high quality profiling tools, lots of libraries etc.

We have less bullshit to deal with the compiled programs. They just "work" more reliably.

I don't doubt your experience but it has nothing to do with AOT compiled vs JIT compiled. Applets that stop working on newer JVMs are probably relying on bugs in earlier versions. This can happen any time there is dynamic linking. Every time I upgrade MacOS X some apps I use stop working properly, even though they're all compiled. Apple just isn't very good at backwards compatibility.

Re:Then you don't quite get a number of things

[Karmashock]

Yes... I've used Java based ultra low latency trading apps... I believe the last one I used was "ThinkOrSwim"... and as I remember it basically crashes or gets very slow if you try to make it use more than 1 GB of ram.

My work around in that case was to launch two version of the program with each one doing different things so that neither execution went over 1 GB.

I told you that I'm sure someone is able to get beyond the 1 GB limit... I'm just telling you that most Java programs I've ever used can't.

I'd be happier with Java if you could compile it into independent executables that didn't depend on external java libraries. I think that would give us something of a best of both worlds here.

If you want to keep the java in an uncompiled format that's fine. YOU can do that. I don't like it. Part of my issue is security. Securing java or any kind of scripted language program is a lot harder than controlling finite executables. I'd like the entirety of the code to be in as tight and discrete a package as possible so I can track it and control it. I don't feel comfortable giving permissions to java itself which is something I have to do when java programs are being run. Especially when I only wish to permit SPECIFIC java programs to be able to run. Now if those java programs were finite and self contained then I wouldn't have to worry about it.

I've found ways to make this work and restrict java programs so that only specific java programs can possibly run. However... it was a giant pain in the ass and I have to go through some trouble every time a new program is added... and sometimes even when they're just changed.

The security situation is honestly a big part of my problem with scripted languages. The bias most people have is having the core of the system just hanging out there for anyone to send scripts to... including malware, hackers, whatever. And then people will restrict the java programs on the theory that if people can't trigger a given script that secures the system. But it doesn't because if a script can be passed to the generally unlocked core... it will run them. And I can't have that.

I don't like having Java generally installed or Ruby or anything. I lock down power shell and shell script as well.... I'm okay with those because there's no reason for a user to need permissions for it. So they just don't get it. I've had a few exceptions. But what I do is give them a program that sends a command to a different account on the terminal server and THAT account has permissions... and that account executes the command.

Sorry if I sound crazy.... It is my job to make inherently insecure systems secure. And frankly Java fucking annoys me.

Re: Oracle's monopoly?

[Half-pint HAL]

It's not about language -- Dalvik is a full-blown virtual machine, based on the JVM architecture. Using a different language wouldn't get them away from the lawsuit, as they'd still be compiling to the Dalvik VM as target architecture.

Re: Oracle's monopoly?

[Trongy]

Android doesn't use Java. Android used to use the Dalvik runtime environment. In Android 5 they introduced ART (Android Run Time) to replace Dalvik.

Dalvik and ART work like Java in that they execute compiled bytecode, but they are separate implementations and are not compatible with Java bytecode.

You can write your own source code in the Java language and compile it to Dalvik/ART bytecode.

Oracle is mad that Google made something that worked similarly to Java, but is not Java and isn't bytecode compatible.

Oracle didn't need any help destroying Java

[DrXym]

They managed to do it all by themselves. I used to program set top boxes with J2ME and dear god was it awful. J2ME was so stripped down that it simply wasn't fit for purpose by the end. It didn't even contain fundamental classes that had been in Java since 1.2 like ArrayList! And it was very expensive to licence too.

So we ended up using another VM called Skelmir which was a clean room Java, roughly analogous to Java 1.5 SE albeit missing some stuff mostly in the javax & sun namespaces. Performance was better, it was cheaper and it was possible to develop normal Java code with a reasonable expectation it would work on the STB. I'm sure the same sentiment was felt everywhere. Companies resented being charging an arm and a leg for a piece of shit runtime which was barely fit for purpose.

As for why Google succeeded where Oracle failed... It's because they offered more or less a full Java SE API and a rich mobile API that allowed developers to write apps without making compromises. It didn't really matter that the byte code was compiled into something else because they also provided excellent tools that integrated with Eclipse to take care of all that.

I don't believe for a second that if Google hadn't used Java as their API that Oracle would have triumphed. Not in the slightest. If anything Google did Oracle a favour by using their language and therefore keeping it relevant for portable devices.

Re: Google/Android should replace Java with C#

[Anonymous Coward]

Still, it would be a pretty good environment, though. It would attract and cross-pollinate some apps and games with the Windows ecosystem as well. Solid language, solid framework, solid tools, and fucking Oracle = a pretty good move.

One little problem...

[jcr]

Java was never any good as a mobile platform.

-jcr

Re:Oracle's monopoly?

[Anne Thwacks]

nobody trusts Oracle

Hardly surprising really - after being repeatedly shafted by Oracle, anyone who has had dealings with them would probably rather not have a phone at all than be dependent on an Oracle product.

Witness the way that their take over of Sun was seen as a step backwards - its not like Sun was revelling in a great reputation for its ways with customers.

I suspect their motto is:

"Let's be evil".

Non sense

[e70838]

There are two things that have destroyed Java market:

• The bad handling of security issues by Oracle
• The threat of litigation as demonstrated in the trial against Google.

I read: Oracle Has "Destroyed" the Market For Java

[Carrot007]

I agreed with that sentiment.

Re: Oracle's monopoly?

[gbjbaanb]

if it looks like a duck, walks like a duck, quacks like a duck then its a duck, even if you artificially grew it from duck embryos in a lab.

They didn't make something that worked similarly to Java - that would have been OK, C# is similar to Java after all. They made something that was *identical* to Java. If they didn't want to be sued they should have made their own API and their own language, Oracle couldn't have done anything about that then, but Google wouldn't have been able to take Java programs and run them without change either.

Re: Oracle's monopoly?

[AC-x]

That means that Google must comply with Oracle's terms within the limits defined by law.

But Google doesn't use Java, they use Dalvik/ART, which aren't written by Oracle and therefore don't have Oracle's ToCs attached to them.

They do happen to be compatible with Java, but if you are allowed to copyright APIs (which is what Oracle are pushing for) then that would be absolutely insane for the IT industry, as you wouldn't be able to implement an API (or possibly even access an API) without the permission of whoever wrote that API.

Java fanboy here

[SpaghettiPattern]

Java fanboy here.

Java is the new COBOL. that's a status very few general purpose languages have reached. It runs everywhere, can do crazy stuff and banks have embraced it. I'll not jump ship for a long time.

But calling Java "a potential mobile device operating system" is bat shit crazy.

Re: Oracle's monopoly?

[AC-x]

They didn't make something that worked similarly to Java - that would have been OK, C# is similar to Java after all. They made something that was *identical* to Java. If they didn't want to be sued they should have made their own API and their own language

What it comes down to is should APIs be copyrightable. Google created their own implementation of the Java API, if companies are allowed to copyright APIs then you can kiss WINE goodbye immediately, anyone wanting to implement an existing API would also be in trouble, and you might not even be able to create a program that even accesses an API without explicit permission.

To come back to your metaphor just because something implements the IDuck interface doesn't mean it's the same kind of duck.

FTFY: Oracle has destroyed the market for database

[Anonymous Coward]

LE - was an asshole

Re: Oracle's monopoly?

[l3v1]

" Dalvik, the Google fork of Java"

Dalvik was not a Java fork. It's VM, and it was written from scratch, not forked from a Sun/Oracle VM.

So, before smartphones there was no market for it?

[Assmasher]

Sensationalism at its worst...

Re: Google/Android should replace Java with C#

[Anonymous Coward]

Because with a single, seemingly innocuous loophole in that "legal guarantee" MS could tie Google up for years in court. The anonymous OP was trolling plain and simple. Google would lose half their market cap overnight if they ever bought into that ridiculous line of reasoning.

Seriously folks, the leopard's spots may have changed color slightly but there all still in the same spot. Microsoft is never to be trusted.

Re: Oracle's monopoly?

[Antique Geekmeister]

And Linux is not UNIX. But it was designed for compatibility, using most of the same system architecture and API's, and software is designed to be compatible between them.

Or at least, it was increasingly compatible until systemd came along.

Re:You are talking about 2001-2004 technology!

[gl4ss]

there was another update in 2008, with more ui building blocks. for touch phones. s60 5th gen had it and symbian^1.

but they still didn't make enough of the api's sane enough to use(practical. many api was unpractical due to asking the user 10 times yes/no to create a file.. )

Oracle complaints.

[kurt555gs]

I'm calling the Waaaaaambulance.

Re: Oracle's monopoly?

[mariox19]

What Oracle fucked up on in the mid-Aughts was not copying Apple's iPhone innovation.

Re:Oracle's monopoly?

[oh_my_080980980]

Oracle owns Java - how is that a monopoly. That's like saying Nike owns a monopoly of Nike Air sneakers....

Re: Oracle's monopoly?

[oh_my_080980980]

What freakin freedom? Why the fuck do you have to register an account at Google Play (and a credit card too boot) in order to download a fucking app! Cell phone carries lock down your phone - restricting access to certain features of that phone. And some phones may or may not be able to upgrade to newer versions of Android; the restriction is not based on hardware specs.

Spare us your notion of freedom.

Re: Oracle's monopoly?

[D.McG.]

NDK development and debugging of C++ code on Android is no where near as polished as it is in Xcode for iOS. Debugging Java on Android is trivial, but that's not where the big money's made. Writing JNI wrappers for everything is getting tedious. I've also had to rewrite Java nio channel code that was cloned poorly in Android, causing deadlocks; whereas the same code worked perfectly in Sun's/Oracle's implementation that Google couldn't touch; only the interfaces were carried over.

Java? Seriously?

[JustAnotherOldGuy]

I haven't used Java on anything device in years. I haven't installed it on the last 5 computers I've owned (about 10 years worth of computing hardware) and have no intention to do so, ever. Java has no place on any of my hardware and never will.

Re: Oracle's monopoly?

[omnichad]

You can install an APK without that. Blame the developers for not releasing free apps directly as an APK.

In that case, it's no different than Windows 8 or 10 or even OS X. There's an app store, but you don't have to use it.

Re:Oracle's monopoly?

[omnichad]

Write once, run nowhere.

The future Python monopoly...

[__aaclcg7560]

I was under the impression that Python was replacing Java as a teaching language in the community colleges, ensuring a monopoly of Python programmers.

Re:Sun

[mark-t]

Google didn't see the point.... previously, Sun offered to license Java to them for 100M, but Google decided they would make a their own clean-room implementation that wouldn't be called Java in the first place, and so would not have to pay them anything. Later, when Sun was floundering and about to be bought, why would Google want to buy the company for access to a technology when they had already decided that they would go in a direction where they wouldn't need to pay licensing fees for that technology in the first place? The only thing Google can be faulted for with not buying Sun themselves is not anticipating that the company that *DID* buy Sun would turn around and say that even a clean-room reimplementation of Java wasn't enough to be free of obligation to them, when Sun had already apparently let the matter drop when Google announced their intention to not try and actually use what Sun called Java in the first place. This is just IMO, but I think that Google probably would have decided to buy Sun before Oracle did if they had realized the headache it may have saved them, but I doubt anyone could have reasonably foreseen at the time that is how things would ultimately go down.

The biggest problem that I see, however, is that If Oracle is allowed to win this, then absolutely every single clean-room reimplementation of anything can become a target for a copyright infringement claim. And if something like that had been a precedent over a decade ago, the whole SCO vs. IBM thing over Linux could have gone *VERY* differently than it did.

Re:You are talking about 2001-2004 technology!

[IamTheRealMike]

It was not "pretty amazing". I have written J2ME apps. It was a disaster zone, mostly for policy not technical reasons.

Problem one: its conformance testing was crap and the licensing for the upstream implementation was expensive. So, guess what, phone OEMs made their own. And did it badly. EVERY J2ME phone was full of bugs, often incredibly basic and obvious bugs like camera APIs that leaked every image taken (take three photos in a row->OutOfMemoryError), or drawing APIs that crashed the device if you tried to draw a bitmap to negative coordinates (correct behaviour is to clip).

This meant that in practice you had to test every version of the app on every device, because bugs were so common.

Problem two: it was tiny. Almost every API was optional, and Java has no good support for on the fly adaptation to missing APIs. So apps ended up needing a C style macro pre-processor to customise the app for every combination of bugs and missing features. You think Android is fragmented? I rofl in the face of Android fragmentation, because I've seen J2ME's equivalent.

Problem three: the CLDC VM was unbelievably sluggish, even compared to the early Dalviks.

Problem four: many APIs were protected by a code signing requirement that was painful to meet and often very expensive for no good reason. Forget about writing free hobby apps.

Problem five: no app store. Every carrier ran its own, and if you wanted distribution ....... yup. They wanted money. Often, meetings and contracts too. Just forget it.

Re: Oracle's monopoly?

[cant_get_a_good_nick]

if APIs were copyrightable, Linux would be in jeopardy. Linus pretty much created Linux by looking at BSD man pages and implementing that. (At least initially) I remember - I had a bug in some code because Linux wasn't bug compatible with BSD. Linus actually implemented "send whats left in the time" for select(2), just as the man page said, just as BSD didn't do.

Re: Oracle's monopoly?

[phantomfive]

but if you are allowed to copyright APIs (which is what Oracle are pushing for)

You are allowed to copyright APIs. This case went through the appellate court, which ruled that APIs are copyrightable, and to the supreme court, which let that ruling stand.

The case has moved on, and is now trying to determine if Google has a fair use defense. But there is no doubt that APIs are copyrightable under current law.

Re: Oracle's monopoly?

[phantomfive]

if companies are allowed to copyright APIs then you can kiss WINE goodbye immediately

No. There are two levels of defense. This is something you need to understand if you want to understand copyright.

1) Is a thing copyrightable? Does the author own the copyright? The answer is clearly (according to the Supreme Court) that APIs are copyrightable.
2) Is there a fair use defense? This must be answered after question 1 is answered. WINE has a clear fair use defense, because it is built for the purpose of interoperability. Interoperability is well established as fair use.

Re: Oracle's monopoly?

[shadowknot]

I remember when some video of them presenting Looking Glass at some conference and the cheers that went up when they demonstrated the rotating windows. I also remember spending a lot of time getting the LG3D WM working on some very old hardware. Fun times!

Re: Oracle's monopoly?

[Skuld-Chan]

APIs are copyrightable.

I think if this is true - you have to wonder what the statute of limitations is on this concept. One could easily argue that Java, C#, AS were all inspired by C/C++ - which was developed by AT&T Bell Labs. I'm sure there's some lawyer who could craft a case that they need to pay royalties now.

Mod Parent +Insightful

[handy_vandal]

A good API is a work of art, an engineering masterpiece and a key success factor for adoption.

True, insightful, and comprehensive. Well said.

Re: Oracle's monopoly?

[Skuld-Chan]

It's still not a fork - if you're talking about code repositories. No-one went to Sys 5's repository and made a fork to develop Linux.

There are licensed variants where this did happen - Linux isn't one of them ;).

Re: Oracle's monopoly?

[phantomfive]

I think if this is true - you have to wonder what the statute of limitations is on this concept.

70+ years after the death of the author. Yes, it can get messy, which is why we have court cases.

One could easily argue that Java, C#, AS were all inspired by C/C++ - which was developed by AT&T Bell Labs.

These could be called "derivative works," and the courts use several tests to determine if something is a derivative work. "Inspired by" is probably not enough for it to be a derivative work.

Also, even if they are derivative works, they may have a "fair use" defense.

Re: Oracle's monopoly?

[AC-x]

But it's not resolved because if it's fair use to re-implement an API then everything's fine. The problem is if APIs are copyrightable with no fair-use exemption to use/re-implement then that's an issue, because anyone who writes a compatibility layer or service that adheres to a 3rd party's standard is just one copyright claim away from ruin despite the actual implementation being an original work.

Re: Oracle's monopoly?

[phantomfive]

But it's not resolved because if it's fair use to re-implement an API then everything's fine.

In some cases it's fair use to re-implement an API, in some cases it's not.

Seriously, go read something about copyright and fair use, then come back. This conversation will be a lot more interesting.

Re: Oracle's monopoly?

[AC-x]

Alright, the court case is still on-going to decide if reimplementing an API is fair-use. If it is fair-use it would put us back to where we were before, if it's not then that's set a precedent where there is no fair-use protection for someone to re-implement an API, which again is what WINE, Mono, GNUstep and many other projects do.

Re: Oracle's monopoly?

[AC-x]

Ok then, list a few examples of where it's fair-use to re-implement an API, and where it isn't.

Re: Oracle's monopoly?

[lgw]

"Install an APK"? Wait, I didn't think you could edit your hosts file without rooting the phone.

Re: Oracle's monopoly?

[phantomfive]

You don't understand copyright law, it's clear. Before continuing the conversation, please:

1) Re-read my earlier comment, which explains why WINE has a fair-use defense.
2) Read the Wikipedia entry on fair use.

Educate yourself.

Oracle shot itself in the foot with a gun

[dafelcardozo]

... and then it is suing Google for not picking the same gun to put a bullet on it's head. Oracle/Sun never went too far way from the 'plugin' mentality: yes, it succeeded in launching a server platform and became the industry standard, but it also failed on he desktop and web applets. And that's so sad, because at some point Sun could have owned the whole stack, they even launched a 'Java OS', which really was just a mimic... Sun lacked the money to create Android, and Oracle lacked it's vision. So sad. Now, give all you money, or else...

Re: Oracle's monopoly?

[multi io]

WINE has a clear fair use defense, because it is built for the purpose of interoperability. Interoperability is well established as fair use.

Any API reimplementation is built for the purpose of interoperability. So by your logic, they should all have a fair use defense anyway.

Re: Oracle's monopoly?

[Antique Geekmeister]

I agree with you. I was trying to point out that most people won't care about the subtle distinctions, and will insist that they are they same even when there are clear historical, architectural, and API differences between them.

Re: Oracle's monopoly?

[phantomfive]

Any API reimplementation is built for the purpose of interoperability.

No. To take the present example, Google's Java API is clearly not interoperable with the original Java code.

Re: Oracle's monopoly?

[swillden]

Ok then, list a few examples of where it's fair-use to re-implement an API, and where it isn't.

The thing about Fair Use is that there are a set of guidelines, but no hard and fast rules. Everything is a case by case, situational analysis. Further, since it just recently became the law that APIs are copyrightable, there isn't yet any body of legal opinions on what does and doesn't constitute Fair Use.

Those facts mean that this is green field new law territory and, even worse, that it will never become thoroughly-settled law. The nature of Fair Use ensures that every single implementor of an API will end up in court unless the owner of that API gives them explicit permission to use it.

My position is that this won't be as disruptive to the industry as people expect, because once it's known that implementing an API without explicit permission can land you in court, no one will implement APIs without explicit permission. If that had been known in the early days of Android, Java wouldn't have been used... except actually it would have because Sun was quite happy with Android's use of Java, and encouraged it. So Sun would have given Google a blanket license and we'd have no issue.

So, that's what the copyrightability of APIs is going to mean to the industry: Everyone who has an API they want others to use and implement will give blanket licenses. Everyone who doesn't liberally license their APIs will find that no one uses them.

Re: Oracle's monopoly?

[leonardluen]

you forget that Sun (before oracle owned them) sued Microsoft over J++. Microsoft lost and then C# was born

Re: Oracle's monopoly?

[segedunum]

In some cases it's fair use to re-implement an API, in some cases it's not.

Well that really cleared things up.

Re: Oracle's monopoly?

[segedunum]

No. I think you genuinely believe you have something to say here, but you simply don't.

1) Is a thing copyrightable? Does the author own the copyright? The answer is clearly (according to the Supreme Court) that APIs are copyrightable.

We've been round this set of houses before in the 80s and 90s with DR DOS and IBM and Phoenix BIOSes. Every time someone has tried to claim that APIs are copyrightable, and courts have agreed, the inevitable problems crop up. Hence:

2) Is there a fair use defense? This must be answered after question 1 is answered.

This makes absolutely no sense whatsoever. The very reason someone else uses the same API is, by definition, interoperability. At the very least It is done to allow the same code to be recompiled and for developer familiarity with the API. 1 simply becomes a nonsense when you understand this, but unfortunately that decision was made by people with no knowledge of the topic they made the ruling over nor the precedents set here before. If APIs were practically copyrightable then the fair use clause shouldn't even be necessary.

Unfortunately, yet again, legal people don't understand the difference between interfaces and software. It really isn't as if we haven't been here before and unfortunately we will go through the same thing all over again in this drawn out legal battle.

Re: Oracle's monopoly?

[segedunum]

1) Re-read my earlier comment [slashdot.org], which explains why WINE has a fair-use defense.

It doesn't explain why WINE has a fair use defence at all. Define interoperability.

However you're looking at this the wrong way specifically because you can't provide an example as to why WINE doesn't have a fair use argument. By definition, any software that uses the same API does so for the purposes of interoperability. There is no other reason. Ditto Google. Developers could take existing Java code they were familiar with, recompile it and have it run in an environment that Oracle had absolutely nothing to do with and didn't build. All you end up doing is spending time in court arguing something that is already crystal clear - APIs can't be copyrighted - which I suspect might be the point.

Educate yourself.

Funny considering we've been over this very same ground in the 80s and 90s. These precedents have been set and we're just covering old ground.

Re: Oracle's monopoly?

[phantomfive]

This makes absolutely no sense whatsoever. The very reason someone else uses the same API is, by definition, interoperability

You would look smarter if you said, "I can't think of any other reason for someone else to use the same API."

At the very least It is done to allow the same code to be recompiled

Java code can't just be recompiled under Android. There are a number of incompatible changes.

Re: Oracle's monopoly?

[segedunum]

No. To take the present example, Google's Java API is clearly not interoperable with the original Java code.

Total and utter nonsense and this gets to the heart of the matter. I've no idea what you mean by the original Java code by the way and this is a pretty big clue that you don't have an argument here because you haven't been able to say why. There is nothing clearly about it and I'm afraid it really doesn't matter what adverbs you use for embellishment. I'm sure Oracle's lawyers will try using the same language in their arguments and why probably why they've had to come up with an additional claim.

Google's API means you could take existing Java code, recompile it to run on a Dalvik VM and have it run on an Android system Oracle didn't build. You could reverse the process. Using this nonsensical logic it would be illegal to take code written for Java and recompile it to run in a .Net VM or vice versa or in any of the countless compilers out there for C, C++ and other languages. This happens in countless development tools between various programming environments every day of the week. The parts of Google's API that aren't interoperable because Oracle doesn't have an implementation of them don't apply because they aren't part of the Java API as it is. If you're arguing that this is a problem then you've just argued that a programmer cannot implement their own classes, interfaces and API in general on top of an existing one or that look likes one. This is even more nonsensical than anything else and would destroy programming completely.

Either way, you have nothing to stand on here because you can't argue where interoperability doesn't apply. This is what happens when you get legal people making judgements on things they clearly haven't taken any effort to understand. We have been through this all before.

Re: Oracle's monopoly?

[segedunum]

The legalities over J++ had nothing to do with the Java API. It had to do with Microsoft creating their own, subtly incompatible, version of the JVM they had actually licensed. It then became a trademark argument as Microsoft wasn't using a compatible JVM that passed compliance tests. Microsoft and Sun then settled, so this was not a court ruling. J# still existed afterwards.

It's not like we haven't been here before.

Re: Oracle's monopoly?

[phantomfive]

Google's API means you could take existing Java code, recompile it to run on a Dalvik VM and have it run on an Android system Oracle didn't build.

No, that's not true lol. Try using a Java Point class, for example.

Anyway, the case is currently back in court, waiting for Google to come up with a fair use defense. Why don't you offer your professional services to Google? I'm sure they'll be happy to hear that the case should be thrown out.

Re: Oracle's monopoly?

[segedunum]

You would look smarter if you said, "I can't think of any other reason for someone else to use the same API."

You don't have an answer for when something isn't fair use, as I knew you wouldn't ;-). It then becomes trivial for Google to argue.

Java code can't just be recompiled under Android. There are a number of incompatible changes.

You have no idea what you're talking about. You take existing code and put it through a different compiler. Happens every other day of the week and arguing incompatible changes has no relevance. These hairs have been split many, many times before

Unfortunately Oracle have a bunch of lawyers who think they know how things work and Google have a bunch of technical people, with lawyers, who actually understand how this works. The arguments then fall away.

Re: Oracle's monopoly?

[segedunum]

No, that's not true lol. Try using a Java Point class, for example.

Just because Google's API doesn't implement it, or indeed implement anything in the same way, has no relevance. There are countless C compilers with a long list of exceptions they don't implement. You're trying to split technical hairs from a legal perspective I'm afraid and this is where Oracle is coming unstuck very badly. These arguments have been made before about APIs that 'look' like others many times in the past. This isn't a trademark issue in the case of J++ where Microsoft licensed to be Java compatible. That's where Oracle falls foul here because they think they can frame it in those terms.

Anyway, the case is currently back in court, waiting for Google to come up with a fair use defense.

I don't know where you get that idea. Google already have a nailed on fair use argument, and as I've explained you can't miss, and Oracle are now desperately trying to get around it by claiming damage by the defendant, Google, on the value of their work (Java). To call that flimsy doesn't really do it justice and frankly, their claims are outright laughable on that front as that's their position on why this isn't fair. However, as I've explained, arguing contrary to interoperability can't even be done. This isn't a Visual J++ trademark case. Oracle wants it to be black or white and that just isn't possible.

Why don't you offer your professional services to Google? I'm sure they'll be happy to hear that the case should be thrown out.

That is incredibly nice of you but they really don't need my help. In fact, Oracle and the technical knowledge of their lawyers are doing most of the work for them. It's been amusing to see them celebrate over the concept of copyrighted APIs getting reversed, slip in the fair use doctrine to try and get around the very, very obvious problems that causes for software in general and then see them fail to argue a contrary position or argue how much of something isn't fair use. Just where is that line?

Litigation is all Oracle has as a business, and they are dying.

Re: Oracle's monopoly?

[MemeRot]

Why is everyone surprised about this? Microsoft was successfully sued for their jvm ages ago.

Re: Oracle's monopoly?

[MemeRot]

Not sure why that's related. There's just a security setting to allow our disallow installing apks. You need to load the amazon prime video app that route, since they have shitty app store and don't want to be on Google's. Or install their app store as an apk, and then download prime video from there. My old HTC had amazon store preinstalled but none of my other phones did

Re: Oracle's monopoly?

[lgw]

That whooshing sound you hear is an AC post in a mix of bold and caps extolling the virtues of host files as the solution to every problem.

Re: Oracle's monopoly?

[gbjbaanb]

The very reason someone else uses the same API is, by definition, interoperability.

Except that Google's reimplementation was designed to be un-interoperable. They wanted the Java programmers to use their system instead of any officially-licenced Java system. They went to all this trouble deliberately, not to make something that worked with other Javas but to break them.

I don't think "interoperability" means allow people to write the same programs for. Microsoft went down that route with J++, it was interoperable with Java programs too. If they didn't want interoperability with Java, they would have made their own very-similar-but-not-the-same language and VM. This is what Google should have done, rather than poach the entire API and call it "not-Java " in the hope they could have all the Java they wanted without any of the disadvantages such as licencing.

Re: Oracle's monopoly?

[phantomfive]

I don't know where you get that idea. Google already have a nailed on fair use argument, and as I've explained you can't miss,

Then Google has nothing to worry about in court.

Re: Oracle's monopoly?

[multi io]

Any API reimplementation is built for the purpose of interoperability.

No. To take the present example, Google's Java API is clearly not interoperable with the original Java code.

Yes it is. You can compile and run source code that uses the common subset of Google's and Sun's APIs for both Google's and Sun's VM. That's not accidental, it's intentional, because the thing was built for the purpose of interoperability.

Re: Oracle's monopoly?

[phantomfive]

The Google has nothing to worry about, they have a fair-use defense.

Re: Oracle's monopoly?

[LinuxLuver]

Well done, Google, if that's the case.

Why?

[allo]

It seems more, like google android was the breakthrough for mobile java.

Re: Oracle's monopoly?

[Kirth]

Kiss most programming languages goodbye, especially Java. Because they would violate the API of C.