Slashdot Mirror
Oracle and Google To Finally Enter Courtroom
Fluffeh writes "After around 900 motions and filings, not to mention a timeline of two years, Google and Oracle are finally putting their case before a jury which will be selected on Monday. While Oracle originally sued for billions, the possible damages have come down to a more reasonable $30-something million (the details vary depending on if you ask Google or Oracle). However, the sides are still far apart. Oracle's proposal was a minimum, not a maximum, and Oracle has asked for a tripling of damages because of the 'willful and deliberate nature of Google's infringement.' For ongoing royalties from future sales, Google has proposed payment of just over one-half of one percent of revenue if patent infringement is proven, but Oracle wants more. Beyond financial damages, Oracle has asked for a permanent order preventing Google from continuing to infringe the patents and copyrights. The case is planned to start on Monday afternoon, after jury selection or Tuesday at the latest."
jing jing jing!
https://www.youtube.com/watch?v=TAX0gJt-aZg
I still have no god damn idea why Oracle is doing this other than amazing short sightedness.
Android is one of the few things left stopping coders fleeing to dot net , its literally a lifeline keeping java alive, and Oracle in their stupidity want to sever that.
*WHY* would they engage on a path so god damn harmful to the health of one of their most important intellectual properties. Its frigging bizare.
I mean ok, sure get a pound of flesh for licencing costs, whatever, billionaires suing billionaires is not my interest. But their "rememdy" seems to effectively involve killing davlik, which would be catastrophic to java coders who have had a huge new source of work from android.
Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
... the trend now is the lawyers, who invented nothing but hot air, gonna be the ones who rake in the $$$
Muchas Gracias, Señor Edward Snowden !
Actually no...Android is here to stay and won't move away from Java and Oracle knows that very well. So they're trying to have their cake (Java made more popular by way of Android dev) and eat it too (grab lots of monies from Google for using Java in that manner).
Seriously dude. Oracles remedy seems to involve killing davlik. That means no java on the android. Its a scorched earth aproach to IP litigation, and you better hope oracle fails on that.
Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
Actually no...Android is here to stay and won't move away from Java and Oracle knows that very well. So they're trying to have their cake (Java made more popular by way of Android dev) and eat it too (grab lots of monies from Google for using Java in that manner).
This is one of the worst "have your cake and eat it too" uses I've seen - it's equivalent to "make a popular product, and earn money on it too!".
will be well enough educated in technology to make a reasonable decision based on evidence. The last time I had jury duty on a first degree murder case, the person selected from my pool brought a herd of ants into the jury room with his lunch bag (plastic bag from store checkout) and kept going on about how special he was because he and his wife had the only set of twins in the world with identical fingerprints. I am a biologist and was strucken from further review by the defense because I answered the question "Do you believe that DNA technology is accurate?" with "Yes sir, I believe it is accurate." It must be great to be a lawyer.
Koalas. They're telepathic. Plus, they control the weather. -Margaret
Well, at the very least it furthers googles reputation as a company not to screw with.
Even if they lose, the fact that they were willing to go this far is a pretty big message to anyone they threaten in the future.
As to why they did it.. I think this goes way beyond Java. I think they just want to crush google plain and simple. Infact, I think the only reason they bought sun was for the IP to bludgeon google over the head with. I'm not quite clear why though.
*sigh* I remember a simpler time when companies would make a good product and sell it for a profit. Good times..
...Google are probably planning on stealing Oracle's lawyers.
You have to be kidding. Java is so firmly entrenched in the enterprise application space that Android is a blip on the radar. It could go away tomorrow and people who write real applications - booking engines, investment monitoring, vehicle tracking, stock management, supply line tracking - will never even blink.
In spite of the sunk cost of Davlik, I think at this point it would be better for Google to simply deprecate Java and tell developers that new development will happen in some other language (like Dart, python, whatever). They could continue to support the Java API indefinitely, but give new apps all the new features and optimisation. Android has had a lot of stick for being a slow, unpolished platform, and this is an opportunity to ditch some of that reputation, at the same time as ditching an unwilling partner (Oracle) who obviously doesn't appreciate what Google have done for Java with Android. The alternative doesn't bear thinking about - constant antagonism with the management of the language standard they are using. If Oracle loses this case they will not take it lying down - expect other moves against Google in the future, hell, even if they win I expect they'd come back for more at some later date. Oracle is obviously in a death spiral and determined to take the rest of the world with it - Apple has also ditched them recently, it seems because of friction with Oracle and new licensing terms, so it's not as if this is going to get better.
Java has caused Google serious issues with performance on a mobile platform anyway - they'd be better off with a language and platform that they control entirely. Unfortunately changing the platform like this would be a huge wrench and would have to be managed very carefully over a period of years, but it can be done.
* oracles reputation
Except of course a bunch of lawyers getting wealthy off these two companies. Case in point: The SCO debacle is still ongoing. There are lawyers still wringing cash out of that mess and it's been nearly a decade. And that is some dinky shell company versus IBM. Two giants with deep pockets slugging it out? It will never end. And why should it? Lawyers get paid by the hour.
Ten years from now when this mess is still ongoing it will be a shining example of why our patent system is broken. It protects nothing, can come to no resolutions, stifles innovation and is really nothing more than a toll booth manned by lawyers.
Weaselmancer
rediculous.
You asked "WHY"...Contrary to popular belief (sorry Starbucks) money makes the world go round. Not java :-)
Oh.. I must be in the wrong thread. Sorry, carry on.
Captcha: laughed
Yup.
Java is huge in the kind of stuff that doesn't make the news very often.
More importantly, a lot of these systems are so large that "switching to .NET" isn't really a practical option.
Even if all Java development ceased tommorow.. I suspect Java would still be around for a long, long time. Java could become the next COBOL!
They're probably playing hardball...ask for too much in order to receive a lot. Like you said, killing Dalvik would only hurt them, it wouldn't make any sense to actually enforce that even if they get the authority to do so.
Unfortunately, that's business....
Seriously dude. Oracles remedy seems to involve killing davlik. That means no java on the android. Its a scorched earth aproach to IP litigation, and you better hope oracle fails on that.
Yes, I noticed the scorched earth approach. However, there may be a ray of hope against that approach. Because Java is a standard [if only de facto], Oracle may be compelled to offer a license under FRAND [fair, reasonable, and non-discriminatory] terms. Google's offer of 1/2 of 1% is in the FRAND ballpark for a mass market item.
In the Apple/Moto fight in Germany, Moto got an injunction against Apple for infringing some Moto patents. They got the injunction because Apple had not negotiated in good faith [stalling for five years]. However, latest ruling appears that Apple might reverse this on the FRAND argument.
Like a good neighbor, fsck is there
Clearly, Google having to pay Oracle millions of dollars is no big deal - chump change to them.
But Oracle has asked for a permanent order preventing Google from continuing to infringe the patents and copyrights. Clearly Oracle is willing to go to the wall to get its way. It would certainly appear that the future of Java is of little import to Oracle compared to winning this battle and getting as much compensation from Google as they can.
If Oracle wins, what they demand will only be limited by the importance of the patents and copyright in question. Leaving the copyright issue aside for now (since it is less clear), how important are the patents? Can Google work around them?
It is my understanding that the principle concern is this patent:
Method and system for performing static initialization.
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=/netahtml/PTO/search-bool.html&r=18&f=G&l=50&co1=AND&d=PTXT&s1=6061520&OS=6061520&RS=6061520
How significantly would Davlik be affected if they had to work around this patent?
You say that like it is a long time. This is going at light speed compared to the SCO saga.
Quattuor res in hoc mundo sanctae sunt: libri, liberi, libertas et liberalitas.
To be fair, I'm not sure this sort of endless litigation is necessarily caused by patents, it's more a result of the legal system we have, and the perverse incentives for lawyers to keep themselves in work. Jarndyce v Jarndyce is a good place to start for an example of this which doesn't involve patents.
Go would actually be an excellent option. Its a really clever language that solves a whole ton of C related pain-points, and compiles surprisingly snapilly.
I mean google might be concerned that not many people know it, but Apple took the exact same punt with objective C, but ultimately objective C's strengths as a rapid development platform won over a lot of coders who might otherwise be spooked away from it.
Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
I'm just trying to imagine what both companies could have done, if the money for this had been spent on R&D projects. Probably both companies and their ecosystems would have been better off. Conflict between two titans rattles the earth, and shakes and frightens smaller beings.
Two years of hard core litigation? Which small companies can afford that? Even if a small company is clearly in the right, a giant can litigate them out of existence, before the truth comes to light.
'tis uneasy waters, in which we tread today, my fellows.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
Haven't really looked at Go, thanks for the tip.
There are a lot of options, and as you say Apple is a good example here - it is possible to take most of your developers with you, with the right combination of threats, cajoling, and incentives, even through multiple huge transitions, as Apple have managed over the last decade.
oh that's right, the company who squanders and kills anything decent they might have acquired (cause sure as fuck they haven't developed anything in the last decade +) for a quick buck
Not that I feel anything for google, but its fun watching Oracle piss on their 50$ shoe to win a 3$ bet
Maybe you've overlooked something
The broken patent system has made a lot of lawyers very very rich
And rich lawyers will see that the broken patent system stays broken
Muchas Gracias, Señor Edward Snowden !
So they're trying to have their cake (Java made more popular by way of Android dev) and eat it too (grab lots of monies from Google for using Java in that manner).
Are you sure you understand the meaning of that proverb? You usage of it suggests you don't because you most certainly can have a company make your product popular through their use of it and also reap the rewards of that popularity.
Go is suffering from significant memory leak problems in its garbage collector that makes it problematic to run on less than 64 bit systems.
They may be approaching from a scorched earth perspective from spectators viewpoint, but you can bet that is just an opening position from which they can negotiate a settlement. In the end Oracle want a deal that is best for them and a better negotiating position to be in is take what we offer or we will take our bat and ball and go home.
Why use an unproven and uncommon language? Qt is already here, and if C++ isn't quite your cup of tea, Qt-Quick gets you pretty far, plus Qt itself already have bindings for most languages.
All in all, a far more sensible approach, that would even bring some sort of compatibility with Symbian and Meego fwiw, besides lightening the load of the device.
Java constantly is listed among the top three most popular programming languages. It's not because of Android.
Android chose Java because Java was popular, not the other way around. You must be unaware of the other uses of Java in this world.
"First they came for the slanderers and i said nothing."
I noticed too that Java systems tend to be large. Just like Cobol.
Oh, if only it could be that way.... >:)
Two rich lawfirms leave richer...
Android is ... literally a lifeline keeping java alive
As an enterprise infrastructure technologist, I can tell you that Java is very much alive. With or without Android, it is not going anywhere anytime soon.
> have no god damn idea why Oracle is doing this
Because Google can't axe Java now, they in their infinite wisdom allowed it to proliferate. If only they have kept C and let developers to add Java, Python, Go, Haskell runtimes (all derived and compiled from C) they would have a great and truly free&open platform, the whole Java thing would get offloaded to third parties, something that smart companies do. Now Java is mandated, and of course you can't compile Go from Java nor Python from Java etc. as it all requires C to be the default underlaying SDK, which for some uniquely flawed executive reasoning is not. So Java is now the huge drag anchor of Android development, not only creating nightmares to developers, but also this patent/copyright Oracle stink.
I'm just trying to imagine what both companies could have done, if the money for this had been spent on R&D projects
Good question maybe.....
Google oceanSpray! a new product that twists the web, attracts millions of users, but never leaves beta and is closed after a few years to the protests of the small but devoted community that stuck around.
or
Oracle Cloudsense Exalogic. A new database that analyzes the cloud, provides a solid support package, but uses an obscure syntax that mostly annoys the programmers who need to work with it, and costs $3million a seat.
"First they came for the slanderers and i said nothing."
Steve jobs was great friends with Larry Ellison (CEO of oracle) Steve wanted nothing more than to kill/destroy/wipe android off the face of the earth. Maybe Larry is helping him fulfil a dying wish.
There ARE alternatives to Java on android - BUT if Oracle wins on copyrighting APIs their name will be more mud than it is now....
I have friends that work there now and they are already on the defensive when they admit they work there - it will only get worse....
People who have a choice will not work there - and over time - that WILL hurt oracle...
These "permanent injunctions" are rather stupid. They do no one any good. The only question in cases like these is: "how much does the infringer owe?" If someone figures out how to make more money with your patent than you do, then they should be allowed to do it, but they should have to pay for it. Presumably they'd pay less if they got a license first rather than going to court. Never, ever, should a court grant a permanent injunction, or stop the sale of anything. It harms the market, harms innovation, harms the free flow of ideas.
We need compulsory licensing of patents. (And copyrighted material too, for that matter, now that the marginal cost of distribution is zero)
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 1=0.
Android is here to stay and won't move away from Java and Oracle knows that very well
Yeah, it's not like Google has made their own language or uses popular, high level language internally that could replace java.
Since there's already C++ support for those needing the support, python could easily replace new development. Freeze the java API, only release the goodies in the new python API, and watch as java rides off into the sunset wrt new development.
Infact, I think the only reason they bought sun was for the IP to bludgeon google over the head with. I'm not quite clear why though.
Walter Isaacson's bio of Steve Jobs sheds some light on it. Jobs and Larry Ellison were BFFs. Together, they had a long history of conspiring to advance each others' agendas.
Ellison, for instance, was prepared to launch a "hostile" takeover of Apple if they didn't bring Jobs back on board. Even after Jobs's death, rolling boulders downhill at Google just for the lulz would be precisely Ellison's style. He has nothing to lose and potentially a lot to gain.
I'd say your original wording was correct. After this companies should hesitate in screwing Google. They will fight.
In the Apple/Moto fight in Germany, Moto got an injunction against Apple for infringing some Moto patents. They got the injunction because Apple had not negotiated in good faith [stalling for five years]. However, latest ruling appears that Apple might reverse this on the FRAND argument.
They can argue FRAND in that case because Moto has actually signed a bunch of disclaimers when they submit their patents to the standard org. I very much doubt you can argue FRAND on a random technology by claiming that it is a "de facto standard".
I don't know if that's still in the game, but previously Oracle was also claiming patents on JIT that Dalvik was violating (and that, apparently, pretty much any JIT-compiling VM would violate; so MS is paying royalties for .NET, for example). If that's still the case, switching languages won't help.
IANAL, but I'm not completely unfamiliar with concepts involved in litigation. "900 motions and filings" makes it sound, to me, like rather a complex case. Reading the first article linked, then, it looks like the case will boil down to some analyses of patent claims, on one hand, and secondly, the question of whether a programming language can be a copywritten work. I would wager that the court's decision in the second matter of those, that it may ultimately have the widest affect on the industry, overall, in any eventual repurcussions of the case. (At the least, it's the matter I find myself most intersted about, so maybe that's just my bias speaking.)
As far as trying to guess out Oracle's strategy, I don't know if we really can, this far along in the process. My own guess is that they just want some money out of Google, plain and simple - but certainly, there must be more details to their strategy.
Somewhere at Google is a Java fanboy, they do far to much mission critical stuff in it (while still developing their stuff) for it to be anything but a fanboy reasoning. They even generate their javascript by writing java instead. No people, javascript is NOT a bad language, it is just a different one that requires a different mindset. Writing a program to be able to generate code for one language in another SCRIPTING language... that is just insanity.
And as has been pointed out, turning Linux into a java only platform is just insane as well. One of the major strengths of Linux is it wide coding support, why on earth would you limit it to just Java a language owned by a rather dubious company before and after (Sun's handling was just as insane) over the countless truly free alternatives? By all means let developers develop in it if they want to but keep your options open.
No, a fanboy exec has declared Java to be the way and business logic be damned.
Meanwhile Apple has forced developers to learn their own language and has the more healthy market. Go figure. Where is the java advantage?
python could easily replace new development
you're a bit late for april 1st.
Does google even have any direct revenue for android?
Jason
2 corps enter! 1 corp leaves!
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Not going to happen. Android platform is BIG. And it works quite OK, actually.
Python is not going to cut it. It's interpreted (i.e. 'too slow') and has miserable multithreading. Dalvik VM is by now JIT-compiled with decent multi-threading. Besides, Python is a dynamic language and they are a pain for complex apps.
Go isn't going to cut it either because it's a purely compiled language.
Yet somehow with necessitas (http://necessitas.sf.net/) people manage to run Qt-applications written in C++ on Android without much problems.
Android chose Java because Java was popular, not the other way around. You must be unaware of the other uses of Java in this world.
Exactly. Java's got a huge number of server-side programs written for it, and it mostly gets on and supports those pretty well. Its somewhat chunky startup costs aren't a big problem in that situation (you don't need to start processes very often and you can usefully throw hardware at it) and its both fast and safe; fast because this is the case that JITting does best with, and safe because there's no loading of strange native code or user-supplied classes. The only real competitor in this space is .Net (yes, I know its not a language but a group of them targeting a single runtime, but then again what I say about Java really applies to a suite of languages too) and that only really has traction on Windows; Mono isn't very trusted yet, and most of the organizations with these sorts of code bases are very conservative (the only reason that Java and C# have had real traction into this space is because a lot of effort was made by Sun and Microsoft to enable relatively low-skilled programmers to work on connecting nasty legacy databases with relatively modern front-ends; that finally displaced a lot of companies away from COBOL and MUMPS).
The amazing thing is that people still write new desktop apps in Java...
"Little does he know, but there is no 'I' in 'Idiot'!"
Java - the COBOL of the 90's
Ellison or Brin? If I was a betting man, well, I am not, so Ellison it is...
I wonder how the jury will be selected: will owning an Android phone be considered a factor for disqualification? And what about owning an iPhone (apple fan => android hater => oracle sympathizer)?
Google's business model for Android is based on people using Google search from their phone.
In my opinion, this is essentially illegal dumping - Google is using it's search monopoly to undercut competitors in the mobile OS space.
But apparently people only care about these things when MS does it, so...
That's because it's using the bdw gc 6.8 in the golang 1.0 release. It's not some unfixable design problem - in fact, they have already written a new gc and it's pending inclusion after 1.0.
Yes a statically typed language with no generics. That means type checking at run time. So wont be much speed increase over java unless you wanna copy and paste every container for each of your types...
Qt is a zombie. Maybe it will survive the death or takeover by MS of Nokia, and the current churn in the codebase which is trying to reposition it for mobile, but I wouldn't like to place bets on it, or bet the future of a healthy platform like android on it. Symbian and meego are never going to go anywhere, as Nokia has made clear, so support for them is totally pointless at this stage. Far more important than a vast catalogue is having a clear and consistent ui on the core apps which people use every day.
If they managed a transition right though, they wouldn't even need to give up the android java apps, they could just let people gradually migrate.
Google would be far better to look at the progress made in mobile uis, the mistakes they've made with android, and try to learn from them to produce something truly new and exciting - if I were them I'd make it web based, but with a nicer language than JavaScript driving interaction, and of course hooks for using c or java libraries. Webos so nearly fot there, and then failed for other reasons, and ios tried to start that way and only changed tack because the tools were not mature enough and they were not eating their own dogfood (if all the native apps had been local webapps on ios, things might have been different). Google have enough momentum and talent to create their own platform and push a lot of developers onto it.
I summary, google needs qt far less than it needs a committed sponsor.
Yeah, how dare Google give away their OS and render Microshit a hasbeen in the mobile space.
Last I heard of it go while good was still incomplete in a lot of ways. Worst of all using a generic conservative garbage collector instead of one that actually frees all unused memory. Go even has a howto for 32 bit programs to "reduce"(not avoid, you can't be sure) the chance of random out of memory crashes, the current GC leaks memory like a sieve. Then there are things like missing basic functionality, from recent reddit posts it seems that it does not even support dynamic linking.
In other words: while nice to look at go still has a long way to go before it can replace something as big and mature as java.
they did that research already and came to conclusion that going with something like dalvik(a java ripoff) is the best way for them to take, they didn't want to go with native code for whatever reasons.
besides, mostly the lawsuit is not about the syntax of the written language, but of the bytecode / vm side.
and if they switched away from dalvik they might just as well call the new product cyborgzzz or whatever since it wouldn't be android..
thing is - there were platforms like android developed within and in association with sun in the mid 00's, basically j2me with much wider apis. that's one of the things why oracle would consider android a ripoff of their research, which it is - however in my opinion that research was obvious and shouldn't be under protection.
world was created 5 seconds before this post as it is.
Yup.
Java is huge in the kind of stuff that doesn't make the news very often.
More importantly, a lot of these systems are so large that "switching to .NET" isn't really a practical option.
Even if all Java development ceased tommorow.. I suspect Java would still be around for a long, long time. Java could become the next COBOL!
Java is also huge in the kind of stuff that does make the news. It's either the #1 or #2 most used programming language for applications, depending on what you try to measure.
The reason why switching to .NET isn't practical doesn't have anything to do with size. There is nothing preventing anyone from developing a Java-to-CLR compiler (google says http://www.janetdev.org/ but I haven't tried it), and writing any new parts of your application in some other CLR language. I think the biggest hurdle would be switching the IT infrastructure to windows and then being committed to sticking with that choice for ever.
By the way, if you think all COBOL development has ceased, you are wrong.
Instead of fighting it out in the courtroom, at this point the computer wars have gotten so personal we really need a steel cage match between Google and Oracle...
Now you might think Google would have the advantage, what with Larry & Sergey tag-teaming against Ellison.
But if you ever look deep into the eyes of Ellison, you will have strong reason to think even the Datariffic Duo will have trouble indeed bringing down the mighty L.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Unfortunately, that's business....
I've been in the business field for decades, and I will tell you that 99.9% of business people on this earth do not include extortion as a part of normal business practice.
What Oracle is doing is extortion, pure and simple, and unfortunately, Google isn't their only target.
Hundreds of million Android users will also be affected
Muchas Gracias, Señor Edward Snowden !
Say what now? The Oracle/Google lawsuit is about copyright infringement of the Java library APIs, as thoroughly documented at Groklaw. I don't know which lawsuit you're thinking of.
If you were blocking sigs, you wouldn't have to read this.
I hope Oracle wins and forces people off their product the same way Novell shot themselves in the foot.
People who haven't tried many languages check some benchmarks and say - well, Java is almost half the speed of C++ on many tasks so it's not so bad.
What people don't realize is most non-trivial implimentations bog down when they run out of RAM and page to disk. The difference in execution speed is dwarfed by the orders of magnitude slower swap to disk.
You need to see the benchmarks to believe them. Maybe you need to run them yourself to be sure.
For many benchmarks Java took over 50x as much RAM as C and over 2000x as much RAM as Pascal.
Yes, check that again. Your hardware could support 50x as many users in C and 2000x as many with Pascal before running out of RAM.
Java is slower than FORTRAN and uses more RAM than LISP.
Where's the progress?
Google thought they were in the clear using Java in the grounds that they re-implemented, similar to how many compatible products have been created in the past using publicly available specifications. I am somewhat surprised the even offered 0.5% to Oracle, but I suppose if it makes them go away it might be worth it.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
"Java is so firmly entrenched in the enterprise application space that Android is a blip on the radar."
Agreed, but Java's problem is that it's usage is not growing as fast as it once was. COBOL faced the exact same problem, it wasn't that it wasn't in use in many places, it's that less and less new dev was being done in it, until it reached the point the only dev being done on it was basically maintenance work.
In contrast, languages like PHP and C# are now growing much more rapidly for new development, they're taking far and away the lion's share of the pie.
If you're a Java dev you have plenty of work available for you for decades to come, as those systems wont just dissapear. There is still some new dev too even, so it's not like employment is an issue, but unless Oracle do something soon, Java is going to see continued decline.
I knew you'd make that mistake. Qt != Nokia. And it's not a zombie either. Sorry. You fail.
Feel free to educate yourself.
Sorry, but practically all the "trash" about google is nothing but a smear campaign. The smear campaign is just part of the coordinated, under-handed, attacks on Google, but shit companies who don't want fair competition, namely: oracle, microsoft, and apple.
I'm hoping we start to see some jury's just watch the show and return the nope verdict. Of course, if juries do get smarter and start voiding these suits, the lawyers will scramble to east texas.
'Qt is already here...'
"There is no Master but the Master, and QT-1 is his prophet."
It isn't Java and it isn't Oracle's. The usage is intended to proffer the idea that Oracle want their cake (Java is a portable product, not tied to Oracle) and eat it (tie Android to Oracle).
And in this case, it isn't appropriate to have this as Oracle making Android popular and reap the rewards of that popularity.
Google is making Java popular (according to GP), and Oracle want to reap the rewards of that popularity.
Do you think Google didn't know exactly what they were doing when the developed Dalvik? The company is awash with ex-Sun execs.
There are already companies licensing Java on mobile who are at a distinct competitive disadvantage because they have to pay a licensing fee to Oracle for the use of Java, whereas Google (and its partner manufacturers) do not license Java on Android. If Oracle wins my guess is Android will die slowly because whatever the PR may be, Android is successful because the cell carriers make more money from the devices than any other.
Sun knew exactly what is was doing when it opened up the non-mobile Java SDK but failed to do the same with Java ME.
Oracle is in an all or nothing move to either generate revenue from Java on mobile or kill it off. Whatever the damages they're claiming, the real cost will be in the massive license fees they negotiate if they win.
RIM's Playbook device is the first example of what is happening. The device has no Java unless you use Android apps. Despite the fact RIM's entire app catalog on App World for its earlier OS was Java based. Whatever their public statements on the subject, the truth is this is about licensing and competition (otherwise why didn't RIM release the Java VM already available on the new OS they bought).
oh, I don't know. If Oracle manage to kill Java entirely due to this lawsuit then we might all be winners. :-) :-)
If Dalvik is killed, there is always the option of porting the official Oracle produced Java stack to Android with little or no effort. Personally I think this is what Oracle is after -- they want royalties and for Android to pay for their Java branding, the same as IBM does with their stack and Websphere, or a host of other vendors that use Java.
Where this might be more of an issue is if Oracle continues to demand that only Java ME can be deployed to portable devices like Android. Now that would be short sighted of Oracle.
Yeah, yeah, yeah. Oracle Java isn't open source.
Nor are the Oracle database, IBM DB/2 UDB, Sybase ASE, and a host of other products that get deployed on Linux stacks. That doesn't stop people from using them; to most potential customers (i.e. Smartphone vendors) having to pay an Oracle fee would just be a cost of doing business, the same as paying for the various wireless patents they use. And the odds are the expense wouldn't be onerous.
Sure there will be much crying and gnashing of teeth, but the vendors will pay rather than give up Android entirely. They have too much invested already.
I do not fail; I succeed at finding out what does not work.
and came to conclusion that going with something like dalvik(a java ripoff) is the best way for them to take, they didn't want to go with native code for whatever reasons.
what reasons? I mean, someone at Google might have said "and it'll run our own version of Java" and that was it, or whoever did the initial prototyping wrote some code in Java and that was enough for it to be Java from then on. No-one really needed to go into in-depth analysis of the benefits of Java v Python or C++ or javascript.
At least, my explanation has as much validity as your assertion.
No the patent side of the lawsuit is pretty much gutted. The patents Oracle was suiing over were mostly tossed by the patent office. That is why the $$ went from billions to millions.
Where's the -5, retard moderation when you need it?
If for some crazy reason, the judge ends up giving Oracle everything it wants, what happens if a bunch of companies panic and try to replace their Java apps? What language and platform is ready to step in? I'm guessing C# is the most likely winner, but is there anything else that doesn't have all the possible baggage that Java now has?
Are you forced to use Google for searching on Android? I know I am not.... Am I forced to use IE after installing Windows? Do you see a SMALL difference there? Any user can type "www.bing.com" into a URL box, very few people can start an FTP session to download Firefox, or Opera, or Safari and/or can manually install the patches and updates(require IE to keep your machine safe ... cough, cough, BS requirement alert).
Illegal dumping is the wrong term, but assuming you just made a typo; how is (and since when) was giving something away for free illegal? MS never got in trouble for giving anything away for free... they got in trouble for making it nearly impossible to use anything except what they gave away. Again, a huge difference.
If you've used a true enterprise java app and it was slow, it was due to 1.) overloaded/underdspec hardware 2.) poor coding practice 3.) your workstation. All of these issues should be identifiable and correctable with proper testing (including perf testing) and planing from the beginning.
I manage java apps in an enterprise environment and only have 2 or 3 applications people complain about, out of 50 or 60 spread across 2 clusters. Oh, and generally, the apps that are slow are heavily dependant on external systems (database, web services hosted elsewhere, mainframe, etc...). I will admit, java WAS slow in the 90's and maybe even into the early 2000's, but it's just fine for 99% of desktops running today.
He has nothing to lose and potentially a lot to gain.
Nothing to lose... you mean other than having most of Oracle's Java patents invalidated and spending an obscene amount of legal fees against the prospect of not recovering anything?
Right. That's kinda like saying that internet just a small fad that'll go away soon.
While the last Java developer certainly won't be retiring in the near future, there's a chance we're actually past "peak Java" by now. Normally that would be bad news, but well, it's Oracle we're talking about so what's going to happen fits quite nicely in its business model (ie. sucking the lifeblood out of its customers). I just wonder what they gonna do after decade or three when they've finished with the carcass.
Oracle sucks, that's all there is to it.
Copyright violations over APIs? And Java is supposed to be F/OSS?
I think the biggest hurdle would be switching the IT infrastructure to windows and then being committed to sticking with that choice for ever.
See Mono
AccountKiller
Python is not going to cut it. It's interpreted (i.e. 'too slow').
Go isn't going to cut it either because it's a purely compiled language.
Wait.. what?? What's left?
If they don't use them to sue people, what would be the point of having them.
Not meant as a snarky reply but a legitimate question. Seems the only reason anyone wants patents these days is to use them as legal ammo. I doubt oracle has any interest in collecting license fees. The sole purpose of buying sun seems to be to attack google with their IP... for what purpose I don't know.
Even after Jobs's death, rolling boulders downhill at Google just for the lulz would be precisely Ellison's style. He has nothing to lose and potentially a lot to gain.
If true, then it is a breach of fiduciary duty towards Oracle's shareholders. Publicly traded companies aren't supposed to do things "for the lulz."
In my opinion, this is essentially illegal dumping
While the default is Google search, I know of at least two instances of carriers/handset makers changing it out. Samsung had a Verizon Android version of the Galaxy S with Bing search and Motorola had some Droid models with Yahoo search.
Windows resellers were forbidden from doing this.
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
now that you've spent 5 minutes on google tell me whats more relevant vs a blip on the radar, android or "enterprise java applications"?
I'd still say enterprise apps. Mobile is a big market, but I think enterprise is bigger.
all the people rolling their eyes and saying "good luck making any enterprise level apps with a maintainable code base should ask themselves if they've EVER even considered the possibility of using something else.
The thing Java has going for it is the huge and mature stack floating around it. Massive libraries, all of which seemlessly(ish) integrate with each other and your own code.
That's the big issue I've run into doing "big app development" with other languages.. and why I think Java still rules in this domain. Using c++ when it's all your own code.. fantastic! Using a couple 3'rd party libraries.. starts to get ugly but still workable, and you can always create a few adapters if it's really bad. Needing many 3'rd party libraries though and it turns into a nightmare. Not only do you run up against dependancy and configuration issues.. but the code just becomes a mess of different coding styles or a pile of quickly written adapters that become a maintenance headache (and budget killer).
Just having a consistent coding convention that everyone sticks to puts Java ahead of most in long term maintenance. Combined with the set of defacto libraries and tendancy for verbosity and solid OO design, I can look at Java code I've never seen and immediately get a good feel for what's going on. Compared to a large c++ app which can take a substantial amount of time to grok.
ask themselves if they've EVER even considered the possibility of using something else.
I'll agree there are some single-hammer types out there.. but in most cases I've seen many options considered for a project. Usually the decision comes down to more then just technical merits of the toolset anyway .. cost and availability of experts and licensing costs for required tools/tech and stability of said tools/tech are usually way above "what will perform the best".
The API's were deliberately left proprietary by the very people now the subject of Oracles suit - precisely because they did not want to open up mobile Java (via SE or ME).
It's crazy that Google didn't settle when the two Larrys were locked in a room for a day last year.
The truth is Google knows that having to license Android devices will kill the advertising goose that lays the golden egg. In the process these guys don't care that millions of Java developers could have wasted their time learning a dying technology.
Oracle may be greedy, but the real evil villains in this situation are Google.
I actually like Java, but swing is still and always has been a terrible user experience. Stuff that should have been fixed in the 90's is still broken (file chooser dialog) and then there's stuff that they seem to have partially implemented then given up on (system tray support).
Qt has always been relatively niche but it's still hanging around and I think it'd make a great tool for Google, they should make Nokia an offer (a cheap one :) )
They need to look at all mistakes made with mobile platforms for other phones though. Remember that iPhone development was originally to be javascript, but the performance (or whatever) wasn;t good enough so they scrapped it in favour of native code.
WebOS was apparently a great API, but there were concerns about performance too.
Now Microsoft are talking about native code too with their C++ renaissance, citing performance and resource use concerns with .NET (so expect WinPhone8 to be native, c++ development?)
So what should Google do..... look to the others make their NDK into the first-class dev kit. TBH they should have done this from the start, so all those symbian and WinPhone 6.5 devs could reuse all the c++ code they'd built up.
JIT.
Oracle does a lot of stupid things, they're your typically corporation who would shoot themselves in the foot if it had an immediate benefit. Everyone in the company is driven hard by what I've heard is a tyrannical executive team, from my vantage point they have a case of PRS (performance review syndrome).
Not knowing otherwise, I'd say they bought Sun just to destroy it. First they kill OpenOffice, then Solaris, then ZFS, and now their working on Java. Very short sighted, but in my opinion not has bad as IBM's failure to see the value in Sun and more importantly the implications of a merger with one of our largest competitors.
Actually, they offered 0.5% only if they lost, so they would not need to spend time arguing how much.
The amazing thing is that people still write new desktop apps in Java...
You mean like these guys?
Yeah, but that's counterfactual history, it's like wondering what scientific advances could have been produced instead with the billions spent on stockpiling nukes.
If they don't use them to sue people, what would be the point of having them.
The nuclear weapons analogy is very appropriate: You're not supposed to have to use them. If you actually end up litigating a patent, something has gone terribly wrong.
The problem is this: If you have a valid patent and you want to use it for exclusion like patents are intended (like pharmaceutical companies do), you don't end up in court, because your competitors know you have a valid patent and don't bother infringing it, or stop when you tell them to.
But that isn't what happens in the tech industry. Instead, everyone has a huge pile of overly broad and obvious patents which everyone else is infringing (and only because none of them should ever have been issued), and the cost of litigating that many patents is almost always prohibitive. The consequence is that no one can use them for exclusion, because as soon as you file a lawsuit you get one back and it's mutually assured destruction. At the same time, you still have to have a huge patent arsenal in order to deter all the other companies from going to you for a shake down using a huge pile of questionable patents that would almost always cost more to litigate and invalidate than license. In this case the problem was that Oracle was vastly overvaluing the patents -- they were claiming $6B in damages at the start of all this. Now it looks like if they win it's going to end up being more like something less than $50M. Which is almost certainly less than the amount Oracle is having to spend in legal fees.
The sole purpose of buying sun seems to be to attack google with their IP... for what purpose I don't know.
I don't know if that's really it. I think part of it is that there are a very large number of old, conservative, high-spending Oracle customers who use Sun hardware, and if Sun dies then those customers are going to be looking for a new vendor, and in the process they could end up being sold an Oracle competitor's database. So Oracle staged a Sun bailout. They just happened to end up with Java in the process.
The thing is, Java means something different to Oracle than it ever did to Sun. The original point of Java was to stop people from writing apps in Visual Basic or against the Win32 API which then wouldn't run on Solaris and SPARC -- Java was "write once, run anywhere" so you could write your app for whatever you have now and then Sun could come in at some point and pitch some hardware to you and it would still run your software.
Oracle is instead looking at it as a licensing opportunity. Lots of people are using Java, Oracle wants money. The problem is that their patents are crap and claiming copyright on an API is ridiculous. It's like claiming a copyright on the bolt pattern in a piece of industrial equipment so that no competitors can make replacement parts. It's purely functional, and copyright only covers expression, not function. Functionality is the domain of patents.
Yeah, yeah, yeah. Oracle Java isn't open source.
Straight from wikipedia:
On November 13, 2006, Sun released much of Java as free and open source software, (FOSS), under the terms of the GNU General Public License (GPL). On May 8, 2007, Sun finished the process, making all of Java's core code available under free software/open-source distribution terms, aside from a small portion of code to which Sun did not hold the copyright.
oh, I don't know. If Oracle manage to kill Java entirely due to this lawsuit then we might all be winners. :-) :-)
Are you another dot net fan?
Does the word "enterprise" ring a bell?
First reasonable comment on this topic. Most people comparing java to python sound like they've only written hello world in both and have total lack of understanding of how things work. The whole comparison is a joke. If python is going to make it as a main language for a mobile platform it will never going to be CPython everyone if familiar with.
Purely compiled language is fine, but in this case Android isn't going to be that portable. ARM only looks like the only platform, but in fact some chinese manufacturers use MIPS for Android devices and no one is going to compile their app for MIPS just because there are a dozen of devices out there they've never heard of, and even if they will fragmentation.
Android platform is BIG. And it works quite OK, actually.
Can't agree with that entirely, it doesn't work at all in most cases, but it's true, that is not something you'd like to throw away, and that is what you WILL have to do if you change the core language/technology, because all the apps would become incompatible.
Umm...aren't all languages ultimately compiled?
Umm wait lol...aren't all languages ultimately compiled?
Does google even have any direct revenue for android?
Jason
Not in pure form directly. It is completely free.
They encourage developers to download the source SDK (all 8.1 GB of it). I have it on my machine and take updates all the time. This doesn't include the Android kernel, but this is also freely available for download. In fact, the Android mods to Linux have [just recently] been added back to the mainline kernel tree.
All of the "front facing" [to app developers] APIs are common. If you're a small shop app developer, you're good to go.
But, handset manufacturers may need some custom mods to the kernel to provide optimal performance on their platforms or support a unique device they have. They can [and some do] make the mods themselves. They can maintain their own source deltas [which is easy enough to do because everything is maintained by git].
If, however, you want professional developer support [beyond filing a bug tracking report], Google will [probably] provide that--for a fee (or some other revenue sharing arrangement). Since most telcos and handset/tablet manufacturers are risk averse, they will usually have such an arrangement.
Like a good neighbor, fsck is there
Please learn the meaning of the word literally.
I mean google might be concerned that not many people know it, but Apple took the exact same punt with objective C, but ultimately objective C's strengths as a rapid development platform won over a lot of coders who might otherwise be spooked away from it.
It wasn't really Apple that took that punt, it was NeXT, sometime around 1986. And yes, ObjC's dynamic, smalltalk-like OOP runtime helped a lot with designing a powerful and straightforward framework for application development. But the other main selling point was that ObjC was fully backward compatible with C, so all of your existing C code could be used. And these days, even C++ can be used. And C/C++ basically runs the world, even today. It has huge momentum. ObjC(++) at its core was just a small runtime on top of their libc. So if you wanted, you could just use (and continue writing) the backend/business code of your app in C/C++ and only write the UI (the part that, in many cases, has to be rewritten anyway) in ObjC. Such a thing can make switching to a new programming language ten times less risky.
over 50x as much RAM as C and over 2000x as much RAM as Pascal.
You HAVE TO be on some serious acid. I could bet money on that.
are you insane? what percentage of people have mobile devices? what percentage of those now have smartphones? what smartphone platform... wait no... what mobile phone platform of any variety has the highest market share?
now that you've spent 5 minutes on google tell me whats more relevant vs a blip on the radar, android or "enterprise java applications"?
Fawk! Another one! On acid! At /.! In the same thread!
First off, I like Go and while it takes some getting used to, I think it could become a fantastic language. That said, I don't think it is ready for something like Android yet.
oh lord no. .NET is pretty much the a mixup of Visual Basic and Java. It uses the worst aspects of VB like references and the worst parts of Java, like garbage collection and horrible memory use and performance.
At least MS has realised its not quite good enough.
If Oracle sues your company (eg Google) if you use their product, why on earth would any startup or new project use Java as a platform? I understand if you are already locked in migrating might be prohibative. Don't they want new users and thus new revenu streams? I am obviously missing something fundamental here.
Go actually has some very serious garbage collection issues on 32 bit platforms right now. This may be part of the reason that Google hasn't been pushing it for Android (yet.)
http://groups.google.com/group/golang-nuts/browse_thread/thread/ab1971bb9459025d/745df87ac82d3690
By acquiring an injunction against Google's implementation of Android, Oracle could accomplish a major market disruption that would get loads of everyone's attention: an end to the sales of Android devices, and possibly, the bricking of millions of existing Android devices.
Such a monumental disruption for consumers might be just about enough to get the public, and politicians, to admit that software patents are inappropriate, and the legal minefield they create is not good for American businesses.
Oracle could have a lot to gain from the death of software patents, as they could develop many new products without regard to existing patents. This is the argument put forth by Daniel Eran Dilger when the suit was first filed. http://www.roughlydrafted.com/2010/08/14/how-oracle-might-kill-googles-android-and-software-patents-all-at-once/
I find Dilger's analysis of the industry to be particularly enlightening, as he is a master of both the technology and the history of these platforms.
As other commenters have noted, Larry Ellison seems like the kind of executive who would undertake a case like this for personal reasons. If it is true that he dislikes the situation of software patents, the case will proceed with them as Oracle's target.
Further, as a less ambitious interpretation, there is the argument that Google forked Java simply to avoid paying the licensing fee. IANAL, but that's what the law is there to prevent. We need to stop Google from stealing everyone else's properties and getting away with it simply because "people like Google search and Gmail too much".
oh lord no. .NET is pretty much the a mixup of Visual Basic and Java. It uses the worst aspects of VB like references and the worst parts of Java, like garbage collection and horrible memory use and performance.
At least MS has realised its not quite good enough.
Garbage Collection = bad? Ah, then you must be a C/C++ programmer :)
BTW, do you know the latest standard for C++ includes garbage collection?
They can argue FRAND in that case because Moto has actually signed a bunch of disclaimers when they submit their patents to the standard org. I very much doubt you can argue FRAND on a random technology by claiming that it is a "de facto standard".
Fair enough. That doesn't mean Google won't try [and may be successful]. This may be strengthened slightly by Sun's previous two separate attempts to submit Java as a standard to JTC1 and ECMA.
Oracle is playing a dangerous game [for themselves] on several fronts.
The reason that Java became popular was the "write-once-run-anywhere" [WORA] and that Oracle [nee Sun] would provide access for any platform. If Oracle wins, this collapses and many independent developers would see the need to reconsider their strategy in light of the fact that Oracle [on a whim] could deny access to a platform that is the developer's primary market.
The implications go far beyond Google. If Oracle wins, this would imperil not just Google, but telcos, handset makers, and tens of thousands of software developers. Legalities aside, the court will be aware of the potential widespread economic chaos that might ensue and temper its judgement accordingly. The court might refuse the injunction and compel Oracle to offer a license [Google might have to pay the $30M damages + license fees but it would be free to pursue Dalvik/Android]
Further, if Oracle prevails, the federal government might view Oracle as a "sole source supplier". In other words, no government contract would be able to use Java in it. For example, because Apple is considered to be a sole source supplier, the FDA will not approve any medical system that uses Apple/Mac technology. That's why you always see PC's (or Sun's) at your local doctor's office. Or, if Java has found its way into certain gov't systems, the federal gov't may seize/nullify the patents under national security grounds.
Also, I believe Google has filed with the USPTO for a reexamination on the patents, arguing they are obvious or there is prior art, etc. Personally, I'm not currently up on what patents are being asserted. But, as a computer engineer, I'm hard pressed to see what could patentable in the JVM as machine architecture simulators/emulators have been around since the 1960's.
In the 1980's, when [AT&T] Unix was the only variant around, they were controlling it and didn't want a formal standard. That's how POSIX came about. Using [court tested] "clean room" techniques, they were able to come up with a standard that gave rise to other implementations (e.g. minux and linux). That's why linux is called POSIX-compliant and not Unix-compliant. This could happen for Java (HP had a clean room Java implementation for embedded systems in the 90's).
In the late 1990's when Microsoft was creating a Windows specific variant of Java [mainly to eviscerate Java], Sun took them to court and got a preliminary injunction. The appeal [which MS won] was that the punishment did not fit the crime. In other words, a breach of contract should not be punished by means of an injunction. No doubt this ruling will be cited in the current case.
Google didn't clean room the Dalvik JVM for the same purpose as MS. The Sun/Java JVM assumes, more or less, a fairly powerful machine (e.g. a PC/Mac/mainframe, etc.). It is too slow/bloated for "low power" (e.g. CPU speed, memory, disk space) handset/tablet. Likewise for the Java Runtime Environment (JRE). It's meant to be a one-size-fits-all kitchen sink approach. Way too big to fit on a small nimble device, yet it would need additional handset specific classes and the generic classes that have no use in a handset would need to be trimmed.
Weaning off Java might be as herculean a task as the U.S. converting to the metric system. At worst, Google might have to call it something other than "Java". But, end users kn
Like a good neighbor, fsck is there
The problem is that there's more than Java at stake. Oracle also has a bunch of patents pertaining to VMs in general, especially JIT-compiling. From what I've heard, those are broad enough that e.g. Microsoft licenses them to cover .NET - and .NET is fairly different, implementation-wise, from HotSpot. Now Oracle is arguing that those same patents also apply to Dalvik's JIT. If that's still on the table in this lawsuit, dodging it would be much harder than just changing languages.
I still have no god damn idea why Oracle is doing this other than amazing short sightedness.
The conventional analysis is that Oracle needs some of Google's key database patents to be able to scale Oracle Database much beyond it's current state. They only bought Sun to get Java to clobber Google over the head with, so they'd enter into a patent-cross-licensing deal.
Presumbaly Google has this figured out and either intends to drag this out until Oracle is no longer relevant, to teach Oracle a lesson and make them pay through the nose to license their patents, or perhaps take ownership of Java in exchange for the patent license.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Maybe they are not that crazy, and didn't settle because the Law says that APIs aren't subjected to copyright. But who knows what they were thinking...
Rethinking email
yep, GC = it's ok to use up as much memory as you like, without thinking about how to clear it up. It also teaches you that all objects are solely comprised of memory structures, when they're often containing other things that need clearing up. you end up with poorly performing overall systems, occasional memory leaks, and lazy programmers.
All in all, GC is a bad thing for computing. so much so that even MS has implemented smart pointers for some resources, and put in the IDispose pattern and then put in the using construct to make object lifetimes deterministic! If they went to all that trouble you know that is was a bad design decision in the first place.
Me, I've done a lot of things. currently doing C# and not enjoying it as much as I am told I should.
It would be cheaper?
The problem is that there's more than Java at stake. Oracle also has a bunch of patents pertaining to VMs in general, especially JIT-compiling. From what I've heard, those are broad enough that e.g. Microsoft licenses them to cover .NET - and .NET is fairly different, implementation-wise, from HotSpot. Now Oracle is arguing that those same patents also apply to Dalvik's JIT. If that's still on the table in this lawsuit, dodging it would be much harder than just changing languages.
VM's go back to the 60's. Java was released in 1995. But, perl has a VM and it was created in 1987.
To me, JIT-compiling isn't much different than a multistage regular compiler (e.g. gcc--released in 1987). The first stage ("front end") parses the language and produces an intermediate meta description (e.g. register transfer language or RTL). This is passed off to the next stage which does optimization and code generation. The RTL is very much like a VM. Thus, all the papers on compiler optimization techniques that have been published for regular compilers serve as a basis for prior art against JIT. I vaguely recall some entity shipping the RTL files and having the target system do the final stage dynamically. That's an example of WORA/JIT prior art.
MS licenses things because it has its own patent portfolio. It's easier to license and MS doesn't want to set a precedent of patent busting because that might give others such ideas. Also, no doubt MS has some patents that they have cross-licensed to Oracle. "So, why rock the boat"?
But, as I mentioned previously, when Sun went after MS and wanted an injunction, MS fought back. Since Oracle wants an injunction, Google will fight back and I suspect that regardless of the trial outcome, the patent busting effort will continue unless Oracle sobers up.
Oracle seems intent on alienating just about everybody. HP is upset that Oracle is dropping support for its database products on Itanium architectures (of which HP has a lot). If Oracle keeps this up, they're going to become the subject of an antitrust investigation. Also, I suspect people (e.g. programmers) will retaliate by designing out Oracle products.
If Oracle does win, and Android is killed, it will anger telcos. Oracle might wake up one day with telephone/internet service to its corporate headquarters cut off :-) So, Oracle, who's your daddy now? :-)
Like a good neighbor, fsck is there
I think the biggest hurdle would be switching the IT infrastructure to windows and then being committed to sticking with that choice for ever.
See Mono
Mono allows you to use visual studio to develop Linux applications. It does not help you run applications written for windows on Linux, because so much of the API has not been ported yet, or doesn't have an equivalent on Linux at all.
I'm well aware of that thanks, it doesn't mean QT will have a future when Nokia folds, or is subsumed by Microsoft.
Who do you think pays for the vast majority of QT development at present? Maintaining a cross-platform toolkit is very hard and takes constant effort, which is why QT has been so valuable for desktop apps, Nokia has no incentive to do that, and MS has negative incentive to do that, and it's not an exciting problem that would attract lots of open-source developers. Nokia has tried to push QT toward mobile development fast (along with all their other failed efforts like maemo, etc, going in a hundred directions at once), but now they have settled on Windows Phone, QT will be left to drift like Symbian has been, and funding gradually cut. Maybe they'll try to sell it again - that would be the best possible outcome for QT.
Those are the reasons I think QT is a zombie, or will be one soon - feel free to differ with that analysis, but code quality and utility are not the biggest factors in it - this is not a technical judgement of QT (which I have used and found to be adequate), but a political judgement of where the software landscape will be in 5-10 years, and the likelihood of QT being a viable ecosystem which supports the effort put in. It'd be nice if corporate politics didn't come into it, but I'm afraid it does.
They need to look at all mistakes made with mobile platforms for other phones though. Remember that iPhone development was originally to be javascript, but the performance (or whatever) wasn;t good enough so they scrapped it in favour of native code.
I think that Apple was visionary here in trying to choose web apps but just could not execute in time, and therefore took the native route as a shortcut (their mistake was using native internally, and then telling developers to use clearly inferior (at the time) web apps, instead of improving web apps enough so that they could use them internally). If they could have pulled it off properly (i.e. if they had known and prepared for the gold-rush that iOS would become, they had no idea), they would be in a much better place right now, and prepared for the inevitable move back to the web as people discover mobile apps are not the centre of the universe or a new paradigm.
Games are a different ball-game completely - those should be completely native and using opengl for the interface too (often they don't rely on the OS for UI stuff), but for the vast majority of apps, all they need is to be able to access native code as required for speed/libraries - on top of that they could be built in any sort of language, and using the web for interfaces would be a great fit for google - they already have some of the best web development people, and could leverage that for a phone UI.
Personally, what I think Google should do/should have done is develop a simple C library (C++ simply is not necessary for this, and if they choose to use it can be hidden behind a C API) which provides hooks for everything the phone does, and use this as a basis for any and all native interfaces they wish to provide, along with as the basis for a webUI which uses html and css for layout and some other language (Would prefer not JS or Java but there are many options) for code. Crucially they would have to port all their apps to this and lead by example, and improve it enough that performance was perfect.
That would give them a path to transition off Java, appeal to devs looking to use code x-platform, a huge market of devs who know web development, and let them respond quickly to changes in the development scene by adding in new top-level languages as they become popular. Instead they are now stuck with Java (which is widely used in corporates, but *not* by developers of consumer apps or on iOS) and in a fight with the owners of their language runtime - that's not going to end well for anyone.
but ultimately objective C's strengths as a rapid development platform won over a lot of coders who might otherwise be spooked away from it.
Walking into Brad Cox' office was all it took for me to distrust ObjC
First reasonable comment on this topic. Most people comparing java to python sound like they've only written hello world in both and have total lack of understanding of how things work.
Have you written any complex applications in Python? If so, what problems did you run into? If not, why do you feel you are qualified to comment on whether it would work for complex projects?