It is all too easy for provocateurs to do as they please, as these actions are anonymous. But the media go on reporting on this ascribing these actions to "a movement in support of Wikileaks". We don't know that, they don't know that.
Yeah, I'm sure Operation Avenge Assange is completely unrelated to Wikileaks.
Thank you for voicing one of my pet peeves. The crack / hack distinction is very recent and no-one had problems with the term 'hack' the way it used to be used. Besides which, the fucking language does not belong to geeks and the rest of the English speaking world will call it whatever the fuck they like despite what some pimply-assed nerd on Slashdot thinks.
Note that I was not talking about a general-purpose Java replacement, but about something that's good enough for client-side UI. I'm not at all convinced that you need a VM for that. You definitely don't need a VM for "consistent memory model".
You do need to be able to tell pointers from data and it's the fact that C/C++ *cannot* do that that prevents them from implementing GC especially well. The GC collectors for C/C++ are fairly poor and not much better than simple reference counting (which is inefficient and not cache-friendly). The consistent memory model is also the reason that Threading and locking mechanisms used for concurrency can be implemented consistently in Java.
As for VM efficiency - yes, HotSpot is better, but in practice this is offset by inefficient coding promoted and even forced by Java itself (lack of user-defined value types, lack of generics of primitives, virtual dispatch by default etc). End result is that JVMs have to be better to match the same perf, because e.g. they need to do escape analysis to stack-allocate throw-away Point objects, while in.NET it is inherent in Point being a value type.
Utter rubbish. That's pure FUD. I was looking through the results of the Google AI challenge where someone was trying to see wether the LISP programmers were all the best programmers because the competition was won by a LISP programmer. It turned out that, in general, the LISP programmers were pretty good, the C programmers were statistically the best, but the Java programmers ran the whole gammut from best to worst. There were well more Java programmers in the top 10 than any other language.
What this shows is that you can write excellent code in Java or you can write crap. The language does not make either outcome a certainty.
At this point, I'm going to accuse you of sharing the anti-Java language bigotry that pervades slashdot. The things you've said are common myths about Java but simply aren't true in practice. I've used Java for 3D graphics, Ray Tracing and Computer Vision / Image manipulation tasks. I've also done these things in C (which for many years was my favourite tool) and I can assure you that Java performs perfectly well by comparison. The only consistent issue with Java vs C or C++ is that Java will generally use more memory for the same tasks.
Let's look at some of your point:
lack of user-defined value types
Yes, Java lacks structs. So what? That's not an efficiency issue, it's a stylistic issue. Besides which, if you need to marshal data for interacting with a C interface (such as OpenGL) which uses structs, you can allocate memory directly via bytebuffers and marshal the data there.
virtual dispatch by default
That's good OO design. HotSpot effectively inlines methods which aren't overloaded anyway. And, if you explicitly want a method to run as quick as possible and to never be used polymorphically - you use the final keyword. From an efficiency point of view, this is NO different to C or C++ where 'inline' is a hint to the compiler.
Regarding "simple language" - many would argue that Java is too simplistic, especially in 2010. It may have been adequate in 1996, but language design evolves.
It was designed to make the creation of rock solid software easier than it is in C, C++. In that task, it succeeded admirably. It's a great tool for making large, maintainable software systems and minimising complexity. I wouldn't use it for writing printer drivers, hello world programs or scripts, but for most other systems programming it's a pretty good tool.
As for C++ being what it is - it's perfectly possible to use only the subset of the language such that you get something very close to Java. That's precisely what Qt does.
I'm a long time C++ programmer. I like MY C++ code. I hate workin
Also, I like being able to write code on a PC and give it to a Mac user and know it will run. I also used to do PC / Unix development on a Linux desktop. Not having to cross compile or compile on multiple different machines is a win.
*sigh* Slashdot and the anti-java community are continuously getting it wrong regarding the value proposition offered by Java. Look, I hate Oracle as much as anyone else out there, and I'd move from Java if I could find something else that works the same way but there;s NOTHING out there to move to. A replacement needs:
(1) A performant VM. The efficient and portable GC, Concurrency and threading features all rely on the consistent memory model afforded by the VM. The.NET CLR is not as efficient (HotSpot eats the CLR's lunch)
(2) A language that is reasonable simple, clean and unlikely to blow your foot off (c.f. C++)
(3) A huge suite of libraries all the way from free and open source to enterprise grade commercial offerings.
To my mind, a language designed with an open definition and an open source reference implementation targeting a performant VM would be excellent and a reasonable replacement to Java. But that language and VM do not exist, as of yet. I mean, there's Parrot, but that's nowhere near fast enough.
Ok, I know replying to your own posts is probably a bit dodgy - but WTF is with the modding down? The OP states that people don't run anything but Java on JVMs, I point out 5 big JVM based languages and I get modded down?
Is the truth an unpleasant thing to hear these days?/. used to have some credibility.
What's more, and this is continously overlooked in the hysteria coming out of the US - Assange is NOT the leaker. The leak came from the US. This is an internal US issue. Wikileaks is merely publishing information that it's been given, so it is not primarily liable.
Exactly. Ayn Rand hated people and society. She was a very ugly person. People on the right choosing her as a hero are marking out their own stupidity or ignorance.
A much better right wing hero would be Adam Smith, but even he isn't as looney as most of people who claim to look up to his writings (after all, Smith says we should judge a society by how it treats its weakest members). Ayn Rand is more your 'club them all to death and get it over with' kind of thinker. No surprise so many on the American Right look up to her.
Ayn Rand was a fucking idiot. Atlas Shrugged does not espouse a philosophy - it's a polemic. There's a reason that Philosophers don't take her work seriously.
OpenGL is standard Java through a JSR who's number I can't currently recall.
I hardly think the Java memory switches are onerous. Certainly compared to DOS memory management - if you set the maximum heap size to something large, Java will just use whatever it needs with no dramas at all. I'm not sure that qualifies as hard.
It *was* good for the Open Source community. The Linux community, however, continually treated Sun like they were shit all while Sun were contributing more to the community than any other Commercial entity. Nothing Sun did was ever good enough for them.
You can, however, fault them for not thinking: "this is a math problem, I wonder if those 'mathematicians' might have already come up with a solution to this problem".
You were claiming that high taxes on production destroyed economies. I have just given you five counterexamples. That is the (sole) point I was making.
If you base your thinking about economics on moral/socialistic/Marxist ideology, then you may view my comments as flamebaiting, but if you divorce your thinking from magic, you'll see that I am only talking about optimization of economic pressures. I am talking about living within your means. I am talking about the real economy, which is the economy of production.
What's really funny about what you've just said is that by following the reasoning that the "real economy" is the "economy of production" you're actually quite in line with Marx's philosophy: for Marx, the power in the economy came from controling the means of production.
But anyway, that's just a side comment.
Greece is not Scandinavia, but only 20 years ago you could make the same argument about it, yet it cooked its books to get accepted into the European Union and everybody seems to know today that Greece is only able to function as a borrower, who is not productive at all and all of its social obligations are killing it.
The Scandinavian economies are going to go the way of Greece eventually, they are living today on the borrowed time and money.
Again, this is ignoring facts. Scandanavian economies *do* actually produce very well (possible exception: Norway which has high living standards mainly because it's floating on gas and oil)
So Greece fell over - it's nothing like the Scandanvian economies and you can't just assert that because Greece fell over, they will too. You're saying this, presumably, to back up your central argument that taxing production destroyes economies, but your comparison makes no sense at all. Therefore, it's a poor argument, therefore you're left with your unsupported asserion that taxes are evil, therefore you sound dogmatic to me. I'm prepared to listen to argument to the contrary, but I've heard nothing from you that sounds anything like a valid argument to support your claim.
You appear to fervently believe what you're saying, but I don't find your ferver a convincing argument.
Slashdot Mirror
I imagine that NASA have a fair idea about that
It is all too easy for provocateurs to do as they please, as these actions are anonymous. But the media go on reporting on this ascribing these actions to "a movement in support of Wikileaks". We don't know that, they don't know that.
Yeah, I'm sure Operation Avenge Assange is completely unrelated to Wikileaks.
No it's not - it's quite elegant in that it expresses the idea of activism via electronic means in one word.
Thank you for voicing one of my pet peeves. The crack / hack distinction is very recent and no-one had problems with the term 'hack' the way it used to be used. Besides which, the fucking language does not belong to geeks and the rest of the English speaking world will call it whatever the fuck they like despite what some pimply-assed nerd on Slashdot thinks.
Note that I was not talking about a general-purpose Java replacement, but about something that's good enough for client-side UI. I'm not at all convinced that you need a VM for that. You definitely don't need a VM for "consistent memory model".
You do need to be able to tell pointers from data and it's the fact that C/C++ *cannot* do that that prevents them from implementing GC especially well. The GC collectors for C/C++ are fairly poor and not much better than simple reference counting (which is inefficient and not cache-friendly). The consistent memory model is also the reason that Threading and locking mechanisms used for concurrency can be implemented consistently in Java.
As for VM efficiency - yes, HotSpot is better, but in practice this is offset by inefficient coding promoted and even forced by Java itself (lack of user-defined value types, lack of generics of primitives, virtual dispatch by default etc). End result is that JVMs have to be better to match the same perf, because e.g. they need to do escape analysis to stack-allocate throw-away Point objects, while in .NET it is inherent in Point being a value type.
Utter rubbish. That's pure FUD. I was looking through the results of the Google AI challenge where someone was trying to see wether the LISP programmers were all the best programmers because the competition was won by a LISP programmer. It turned out that, in general, the LISP programmers were pretty good, the C programmers were statistically the best, but the Java programmers ran the whole gammut from best to worst. There were well more Java programmers in the top 10 than any other language.
What this shows is that you can write excellent code in Java or you can write crap. The language does not make either outcome a certainty.
At this point, I'm going to accuse you of sharing the anti-Java language bigotry that pervades slashdot. The things you've said are common myths about Java but simply aren't true in practice. I've used Java for 3D graphics, Ray Tracing and Computer Vision / Image manipulation tasks. I've also done these things in C (which for many years was my favourite tool) and I can assure you that Java performs perfectly well by comparison. The only consistent issue with Java vs C or C++ is that Java will generally use more memory for the same tasks.
Let's look at some of your point:
lack of user-defined value types
Yes, Java lacks structs. So what? That's not an efficiency issue, it's a stylistic issue. Besides which, if you need to marshal data for interacting with a C interface (such as OpenGL) which uses structs, you can allocate memory directly via bytebuffers and marshal the data there.
virtual dispatch by default
That's good OO design. HotSpot effectively inlines methods which aren't overloaded anyway. And, if you explicitly want a method to run as quick as possible and to never be used polymorphically - you use the final keyword. From an efficiency point of view, this is NO different to C or C++ where 'inline' is a hint to the compiler.
Regarding "simple language" - many would argue that Java is too simplistic, especially in 2010. It may have been adequate in 1996, but language design evolves.
It was designed to make the creation of rock solid software easier than it is in C, C++. In that task, it succeeded admirably. It's a great tool for making large, maintainable software systems and minimising complexity. I wouldn't use it for writing printer drivers, hello world programs or scripts, but for most other systems programming it's a pretty good tool.
As for C++ being what it is - it's perfectly possible to use only the subset of the language such that you get something very close to Java. That's precisely what Qt does.
I'm a long time C++ programmer. I like MY C++ code. I hate workin
No exception handling is a killer.
Also, I like being able to write code on a PC and give it to a Mac user and know it will run. I also used to do PC / Unix development on a Linux desktop. Not having to cross compile or compile on multiple different machines is a win.
*sigh* Slashdot and the anti-java community are continuously getting it wrong regarding the value proposition offered by Java. Look, I hate Oracle as much as anyone else out there, and I'd move from Java if I could find something else that works the same way but there;s NOTHING out there to move to. A replacement needs:
(1) A performant VM. The efficient and portable GC, Concurrency and threading features all rely on the consistent memory model afforded by the VM. The .NET CLR is not as efficient (HotSpot eats the CLR's lunch)
(2) A language that is reasonable simple, clean and unlikely to blow your foot off (c.f. C++)
(3) A huge suite of libraries all the way from free and open source to enterprise grade commercial offerings.
To my mind, a language designed with an open definition and an open source reference implementation targeting a performant VM would be excellent and a reasonable replacement to Java. But that language and VM do not exist, as of yet. I mean, there's Parrot, but that's nowhere near fast enough.
Go is awful. Have actually looked at the design? It's hardly a step forward from C++
Ok, I know replying to your own posts is probably a bit dodgy - but WTF is with the modding down? The OP states that people don't run anything but Java on JVMs, I point out 5 big JVM based languages and I get modded down?
Is the truth an unpleasant thing to hear these days? /. used to have some credibility.
He's got balls and he's prepared to stand up for what he believes in.
What makes you an expert on his ego?
What's more, and this is continously overlooked in the hysteria coming out of the US - Assange is NOT the leaker. The leak came from the US. This is an internal US issue. Wikileaks is merely publishing information that it's been given, so it is not primarily liable.
Well, it's posted on the net. It MUST be true!
Exactly. Ayn Rand hated people and society. She was a very ugly person. People on the right choosing her as a hero are marking out their own stupidity or ignorance.
A much better right wing hero would be Adam Smith, but even he isn't as looney as most of people who claim to look up to his writings (after all, Smith says we should judge a society by how it treats its weakest members). Ayn Rand is more your 'club them all to death and get it over with' kind of thinker. No surprise so many on the American Right look up to her.
Ayn Rand was a fucking idiot. Atlas Shrugged does not espouse a philosophy - it's a polemic. There's a reason that Philosophers don't take her work seriously.
The mistake they made was that they forgot (or didn't know how) to monetize the open source solutions they had. .
Ok, Pet Peeve time... "Monetize" does NOT mean 'make money out of something'
It is the act of creating currency - literally minting coins or printing notes or banknotes. It can also mean establishing a national currency.
There are hundreds of way to say "profit from", "make money from", "exploit financially", but pretty much one way to talk about monetizing currency.
So QUIT misusing "MONETIZE"!!!
And get off my lawn, dammit!
http://en.wiktionary.org/wiki/monetization
OpenGL is standard Java through a JSR who's number I can't currently recall.
I hardly think the Java memory switches are onerous. Certainly compared to DOS memory management - if you set the maximum heap size to something large, Java will just use whatever it needs with no dramas at all. I'm not sure that qualifies as hard.
It *was* good for the Open Source community. The Linux community, however, continually treated Sun like they were shit all while Sun were contributing more to the community than any other Commercial entity. Nothing Sun did was ever good enough for them.
I know that I haven't heard anything about people running anything other than Java programs on a JVM
You mean, other than Scala, JRuby, Clojure, Groovy, Jython, Rhino and many many more?
There is no better free alternative to Java.
You can, however, fault them for not thinking: "this is a math problem, I wonder if those 'mathematicians' might have already come up with a solution to this problem".
Shhhh! You'll upset the groupthink
So what? We were talking about the relationship between tax and economic strength. What does culture have to do with that?
What the hell does that have to do with what we were talking about?
Saying that Scandinavian countries have the strongest economy is bunk, they are not exporting as much as China or Singapore or Germany for example.
It is not 'bunk'. I didn't say "Biggest" I said "Strongest". i.e.: their GDP per capita is higher than the US (every Scandavian economy)
See here:
http://www.google.com/publicdata?ds=wb-wdi&met=ny_gdp_pcap_cd&tdim=true&dl=en&hl=en&q=world+gdp+per+capita#met=ny_gdp_pcap_cd&idim=country:USA:NOR:DNK:CHE:SWE:FIN&tdim=true
You were claiming that high taxes on production destroyed economies. I have just given you five counterexamples. That is the (sole) point I was making.
If you base your thinking about economics on moral/socialistic/Marxist ideology, then you may view my comments as flamebaiting, but if you divorce your thinking from magic, you'll see that I am only talking about optimization of economic pressures. I am talking about living within your means. I am talking about the real economy, which is the economy of production.
What's really funny about what you've just said is that by following the reasoning that the "real economy" is the "economy of production" you're actually quite in line with Marx's philosophy: for Marx, the power in the economy came from controling the means of production.
But anyway, that's just a side comment.
Greece is not Scandinavia, but only 20 years ago you could make the same argument about it, yet it cooked its books to get accepted into the European Union and everybody seems to know today that Greece is only able to function as a borrower, who is not productive at all and all of its social obligations are killing it.
The Scandinavian economies are going to go the way of Greece eventually, they are living today on the borrowed time and money.
Again, this is ignoring facts. Scandanavian economies *do* actually produce very well (possible exception: Norway which has high living standards mainly because it's floating on gas and oil)
So Greece fell over - it's nothing like the Scandanvian economies and you can't just assert that because Greece fell over, they will too. You're saying this, presumably, to back up your central argument that taxing production destroyes economies, but your comparison makes no sense at all. Therefore, it's a poor argument, therefore you're left with your unsupported asserion that taxes are evil, therefore you sound dogmatic to me. I'm prepared to listen to argument to the contrary, but I've heard nothing from you that sounds anything like a valid argument to support your claim.
You appear to fervently believe what you're saying, but I don't find your ferver a convincing argument.