Specifically, if we look at the PDFs from the original source here:
(BTW, the "org.apache.harmony" package prefix doesn't necessarily mean that the ASF ever distributed the particular file in question; that PolicyNodeImpl.java file was distributed by Google with an Apache Harmony package declaration, but had never been distributed by the ASF itself.)
That would have been better for two reasons - Oracle is screwing up bad with the JCP and it's going to hurt Java - May be Google could have been a better steward. Secondly Google relies so much on Java not only for Android but also for many other things - it is almost idiotic not to buy the IP.
I guess they had their reasons to not pursue - I can't imagine what those were.
Oracle bought Sun on a business model that saw them recouping the price of the acquisition not in proceeds from lawsuits but in shoe-leather sales to Oracle and Sun customers (which overlap intensely). Oracle already had the infrastructure to capitalize on that channel with Sun products.
Google had none of that. Had Google bought Sun, the first thing they could have expected would have been a shareholder revolt. The money isn't theirs to spend; it belongs, in large part, to Google's shareholders.
This is bad, but I still don't yet believe buying Sun would have been the better option.
Security from people suing you is probably the most compelling reason you gave here - I see no reason why Google would want to control Java IP and future direction when they can write all the software they need to in house to built on the Java platform to meet their needs.
So is 5-7 billion worth security for people suing you? I guess we'll see, but this is likely the biggest lawsuit related to Java they will ever face, and you can bet they won't make this same mistake again. Unless then pay out several billion dollars, which is unlikely given the tendency of these large technology firms to cross sue and settle with cross licensing agreements, I'd say Google's decision not be buy Sun is still sound at this time.
Just my two cents, I am watching just like the next person from the sidelines, this should be interesting.
But even if you just consider the suing part - this is potentially not just limited to Google but also their OHA members. So if Larry sues everyone in OHA and handset makers couldn't afford to pay what Larry asks - that could end up hurting Android.
i was far from all this politics, but made a few google searches - is this about ASF leaving JCP? then, as far as i understood, these issues are all related, to Harmony, suing etc.
so maybe Java wouldn't be hurt? just specs would be changed more tyrannically...
(java my favorite language, don't make me sad :)
It is hard to see it if you are not used to the page design. I do find the pages heavy on engadget. (which always cause problems when demoed on mobiles tee hee)
edit: yup, the version arn is referring to was the most annoying and hard to find IMO.
I'm not even sure how core those files are to running Android - it looks like they could easily be replaced by Google with new code.
However, to me, copyright infringement matters a lot more... If someone is getting free code then at least they could respect the LICENSE.
Unfortunately, I'm sure that now a few floors of lawyers are salivating at all the hours they're going to bill. In the end, the laywers will get a pile of money, oracle will gain some pocket change at google's expense, and the java community will be hurt the most.
Doesn't Oracle have better things to do? What do they gain from this?
Given the breadth of the infringement (making it open source w/ OEMs shipping tens of millions of devices based on it), the damages could be substantial.
What does Oracle have to gain from this? Money of course.
Unless Google has a good response to this, one thing I guarantee will happen is the engineer who checked in these files is going to be looking for a new job.
This is assuming it's a part of the code that actually makes it onto devices. Some of the other comments in this thread suggest they may be from a testing framework.
Sure, whatever opinion you have about Oracle might be valid, particularly with the way they're handling the JCP and OpenSolaris. Whatever. However, there is an underlying issue that is a bit bigger here, and the blog author is pointing that out in a reasonable fashion. It doesn't matter who is suing and who is sued; copyright infringement is copyright infringement.
You could make the argument that this isn't any of Oracle's business, but they absorbed everything about Sun: their IP, their liabilities, and so on. Sun was perhaps considering this action before Oracle bought them, and we'll never know.
What if the companies were reversed? Would you make the same remarks?
One of the few positive things about our legal system is that it at least attempts to take "intent" and "proportionality" into account when making judgments.
Hopefully who ever makes decisions about this case will think about what was actually done. There's a world of difference between a hard-working team inadvertently cribbing a little code, and a large company gaining a massive strategic advantage by violating copyright on mission-critical source code.
Given google's reputation for being open, they're not likely to pursue a law suit like oracle just because there's money in it. The media outrage will hurt their reputation more than the monetary gain.
It's silly to imply that Oracle brought this suit against Google just to score a few bucks, but easy to say when you dislike Oracle. The alternative, and very likely real reason for the suit, is that Oracle actually cares about its intellectual property and would prefer that Google not do what they will with it.
I think it's far more likely that Oracle has seen a way to get a slice of the Android pie, and the rapidly growing long-term revenue stream that it likely represents.
Again, they don't need a long-term revenue stream. I'd wager that their revenue stream from Oracle Database and other products, as well as the long-term support contracts for those products, shadows Android in comparison. Bold, I know, but Oracle is widely, widely used in the enterprise sector. The salary for an Oracle DBA alone speaks volumes regarding this.
In your world, Oracle executives sit around and think: boy, we could score a small percentage of that Android market. It pales in comparison to what we can make off of enterprise customers, but that cash flow is definitely worth the negative publicity and community recoil!
Your comment is merely another angle of repetitive arguments attempting to make Oracle out to be the bad party for suing Google. While I dislike Oracle just as much as the next geek for their treatment of open-source products, they might actually have a point with this lawsuit. That's all I'm saying, and watching my karma swing around wildly for taking the middle ground instead of jumping on the anti-Oracle bandwagon is amusing.
The people who buy their software understand things like duty to shareholders, and Oracle has historically determined that exercises like this are part of that duty.
About suing for money: The accumulation of patent in large companies is necessary because if they don't patent everything, someone else will and they'll be in trouble. But suing another big corporation "just to protect the patent" is mutually assured destruction because the company being sued has probably patented a lot of petty little things that the company suing is using and can counter sue. Unless there's a lot to gain, they don't because the only people that will get anything out of it is their patent lawyers.
Oracle does have a case against google, but saying that they're doing it just to protect their intellectual property is a bit naive.
I don't get one thing. I thought Java is supposedly open-source? How come there's proprietary source code? Or are these files part of a particular Java library developed by Sun that are not part of the Java that they open-sourced?
Basically you can't do shit if it isn't explicitly allowed.
As for could the just pay Oracle to license the code - they could have done at the time had Oracle, or Sun as it would have been at the time, agreed but they didn't. It would seem highly unlikely given that Oracle are currently suing them that they'd be willing to enter into such a license with Google now.
Not entirely; it might make a difference in how hard it is for Oracle to prove willful infringement (which increases the damages.
And even if it's irrelevant to the case, it's still an interesting question.
Google inherit everything the previous developer did including the intent or otherwise. It doesn't get to say "but we didn't know when we bought it", that's what due diligence and contacts are for.
(b) The longer ago it happened, the more likely it is that any evidence of how it happened has been lost. If so, even if it was willful, Oracle might have trouble proving it.
My guess is that the only out they'd have with this is if the purchase contract specified that any legal liability might remain with the previous owners but it's unusual to find anyone who will agree to that with a sale.
The possibilities are that they did not perform proper due diligence on the code that they bought, which should have brought this to light (and a giant the size of google really should know how to do a code audit) (bad), the second possibility is that they did do proper due diligence, were aware of the provenance of this chunk of code and chose to ignore it (bad), the third is that the code got incorporated after the sale (bad).
There are no 'goods' here, the central question is if this is a part that is critical.
That said, I'm fairly sure that any large enough software project that has had contributions from large numbers of sources is subject to this kind of pollution, unless very strict measures are taken to avoid it.
"In Florian's paper, he points these out as Sun PROPRIETARY / CONFIDENTIAL. However, it looks like several of the sources come from Sun's mmademo, linked here [java2s.com]. In this rendition of the document, each source file's license is a permissive one by Sun (i.e., not proprietary / confidential)."
IE, the header in the distribution may read incorrectly that the files are SUN proprietary.
There was confusion about this when the first bits of code that were obviously Sun's implementations were found in Android code... However, according to Apache, Google slapped the Apache Harmony license into the headers themselves and the files were never really part of Harmony.
BTW, I have no idea whether the author is or is not working for Microsoft.
Most of Engadget is accused of being an Apple fanboi site. Engadget's claim is that Apple just makes good products. I'm inclined to believe them. For the most part they seem really fair.
I believe Nilay (and Josh and Paul) all use Mac Books as their primary computer. And Nilay uses a Droid of some sort as his primary phone.
If they are on the MS payroll... MS needs their money back as they routinely slam MS. They slam MS so often, I think you'd even enjoy the site. :-)
Sadly, I have very little free time these days. I even gave up on Engadget's headline podcast...
Yea, it is the new name.
>IIRC that guy calls everyone who disagrees with him "Microsoft booster" so I'd take it at face value.
Yea, I do disagree with some of their claims. I just wanted to mention a source that I can think of off the top of my head to help readers.
Plus, the aspect of when Java was GPL'd does that pertain to everything in the distribution or does at eh time Sun's disclaimers of some being not GPL'd hold legal weight pertaining to whether its a legal enforceable GPL license.
"If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works."
Correct me if I'm wrong, but it doesn't mean that you can copy small portions of GPLed code as you wish and use them in your projects; it means that the original copyright holder for those small portions can apply any license to them, which means that by default all rights are reserved by the original author of those portions.
Neither section implies that including only a little GPL v2 code within your project is OK.