They don't have any such thing. An "implementation of Java" requires class libraries as well as a VM, and IBM (along with every other "independent" commercial VM vendor) uses and depends on Sun's class libraries. Which is a MUCH bigger task than just creating a VM.
The VM may be clean room and open-sourcable, but by itself it's useless; Kaffe and Classpath and gcj and all the other existing Free alternatives are at a much greater level of development than IBM's because they do address the class library issue. IBM would probably have the sense to go with the technically better solution - either graft Classpath or Kaffe's libraries onto their VM and release that, or simply back one of the existing open VMs, depending on how tightly their implementation is bound to Sun's libraries.
However, if IBM was going to do this they could have done it years ago - I've pretty much given up hope that they have any intention of backing any of the Free Java projects.
Not true - GNU Classpath is JDK1.4 compatible in many areas, and pretty much complete at JDK1.2 level with the sole exception of CORBA (which has interesting license issues) and Swing, which is being actively worked on (there's a screenshot of Classpath running a Swing demo that's limited to buttons and checkboxes, and it actually works!).
Sure, cloning Swing and many of the other massive libraries in recent JDK versions is a mammoth undertaking and life would be a great deal easier for everyone if Sun would open up the reference implementation. But don't discount the work of the people who are already doing it - they're further along than you think!
(BTW, the reason those URLs point to kaffe.org is just because my own domain for them expired; kaffe.org graciously agreed to host the files, but the results are independently generated and not biased in favor or against any of kaffe's "competitors". Having said that, Kaffe is another project that's made leaps and bounds recently. There are, in fact, multiple completely Free/Open Source implementations of Java now that can run many high-profile Java apps, including Eclipse and Tomcat)
Easy - those thousands of people who don't know each other also send email *back* to the mailing list. Only a few dummies send email back to the spammers.
For something based on statistics, the difference would likely be very noticeable.
Unfortunately, it still doesn't include libapache-mod-mono which is critical for me:( Running the last packaged libapache-mod-mono with the new mono debs still doesn't work (unsurprisingly, since they're the same debs I installed previously)...
Still, it seems the debian mono packaging team is making much more progress than I had guessed from following their mailing list:)
Sadly, the Debian packages are still in rather a strange state of limbo: You can get them if you can find the magic incantation to add to sources.list, but they're a January 14th snapshot (which isn't too old, but if you're going to do something from 2 weeks ago, why bother with a nightly snapshot in the first place, why not the last stable release?) and don't include libapache-mod-mono. Meanwhile libapache-mod-mono is also available separately but doesn't seem to work with the latest packages of everything else.
Which is making my attempts to port an ASP.NET application to work on mono a fairly frustrating experience, to say the least...
I'm using IKVM in production right now in a real (albeit small) company. And mono plan on releasing 1.0 within the next 6 months, which they will recommend for production use - we're evaluating that for use within our company also.
I have some development tools written in Java which have language-independent output (supporting both Java and C#) but our company primarily uses C# these days. Still, every developer had to install the JDK and a JDBC driver in order to run our compile process, since it invokes these Java tools.
These days, I've used IKVM to compile that code to a.NET exe (and the JDBC driver into a dll). Since all our C# developers have the CLR installed by definition, they don't need to install any additional software. This actually happened a little too late because all the developers already had the Java stuff installed, but it'll help when we hire new developers:)
It is true, though, that I've found that following projects like these has helped my ability to write much more boring code. It's also helped when I'm faced with a problem like "hey, we might have a customer who wants to buy our (.NET) stuff but uses primarily Unix, can we help them?" and I can say "Well, it's not 1.0 yet, but there's this project called Mono..."
I'd open the attachment in a safe way (eg "Save as" in a decent mailer), attempt to figure out what it was doing, and if (as in this case) it was a simple program to send my username and computer name to my company's sysadmin, I'd send a message saying that, eg "HarryPotter" on machine "HOGWARTS" (or "Frodo" on "SHIRE", or "Neo" on "MATRIX", or... you get the idea) had run the virus.
Then I'd wait and see if the sysadmin could figure out how to trace that to me...
While "int" is treated "as if" it were an Object in the CLR, and behaves as if it was a subclass of Object, it really is a value type: It's allocated on the stack, cannot be subclassed, and is passed by value when it's used as a method parameter or return value.
The same goes for all the other built in primitive types (float, decimal, double, long, short, bool, char etc), along with user-defined Structs and Enums. These are all considered "Value Types" in the CLR.
When instances of value types need to be treated as Objects, they are boxed, just like in Java, except that it's transparent and automatic and there's no separate class that represents the boxed type - the distinction Java makes between int and Integer simply doesn't exist in C#. (System.Int32 is just an alias for int and isn't equivalent to Integer at all in any way).
The article clearly states that in Generic C#, the JIT actually creates specialized binary code whenever a generic class is instantiated with a value type argument. It even uses the example of List<int> turning into int[] (remember value types like int can't be subclassed, so the subclass problem doesn't arise).
C# provides the "illusion" of no primitives, but it's certainly not true that everything's always boxed. It just provides a more elegant and flexible model of boxing than Java does.
It doesn't have anything to do with autoboxing in Java, but it has everything to do with boxing in Java versus C#.
Both C# and Java require boxing if you want to put an int into an object variable. They differ in whether that boxing is automatic or not (at least until Java makes it automatic in 1.5), but that's irrelevant: it still happens, so the performance impact is the same in both languages.
However, if you're trying to put that same int into an ArrayList<int>, the two languages are very different.
In Java, ArrayList<int> is really just ArrayList<Object> under the covers, and stores its contents into an array of type Object[], so the boxing operation still needs to happen. Hence, there's a performance hit.
In C#, ArrayList<int> is interpreted as such at the runtime level and the JIT compiler creates native code for it that stores the contents in an array of type int[]. When you store values into that array, no boxing is needed, automatic or otherwise. No performance hit.
Both C# 1.1 and Java 1.4 collections can hold primitive types if you box them. In Java 1.5, you get some extra syntactic sugar (which C# had from day one) but the boxing still happens. In C# 2.0, generic collections can hold primitive types without boxing. That's the difference that he's trying to get across in this article, and why it's specific to generics.
All of this is explained very clearly in the article and it amazes me that so many people can't grasp it...
Oh, and with regard to terminal windows: Microsoft's command prompt has the exact same problem. In their case you have to use Enter (!) to copy the text, or right-click the window title bar (!!) and drill down through two levels of nested menus (!!!) to find the copy option.
IMHO, the best solution would be for Ctrl-X and Ctrl-C not to be passed through to the terminal if any text in the window is selected. Not sure what to do about Ctrl-V though...
Why are so many people ignorant of the fact that Ctrl-C and Ctrl-V work just as well in KDE and GNOME as they do in Windows? Have you never even tried it?
Try this in any recent KDE[1], GNOME or Mozilla application:
Select some text Ctrl-X or Ctrl-C Select some other text Ctrl-V
It's strange - you'd think that if doing it the "X way" (select + middleclick) was so much of a pain to people, they'd actually *try* it the windows way and discover that it works, and be happy.
Personally, I use both: the X way for quickly inserting pasted data, and the windows way if I want to replace something at my destination. Now if only X applications were better at non-textual formats...
Stuart.
[1] I haven't actually tested in KDE but I've heard it was fixed in KDE3 to work just the same.
I thought it was funny how every provider was fighting tooth and nail to oppose these regulations but they still jumped at the chance to advertise the new "feature" as soon as they were forced to provide it.
It's almost as if they knew that customers wanted this and they can make a profit by offering it! What a radical notion!
They can't file against Red Hat because part of SCO's defense against RH's countersuit is that RH have no standing because SCO were never going to sue RH. If they then turn around and sue RH now, they've shot their own defense (and despite appearances, SCO do seem to realize that they need all the defenses they can get at this point...)
I've got you beat for "great stuff missed through sheer bladder-related lack of foresight". I missed (a significant chunk of) the architect scene from Reloaded in exactly the same way you're describing. In a movie with half a dozen ten-minute content-free fight scenes, I had to pick the moment where information is being thrown at you faster than you can process it already.
Like yourself, I found that almost superhuman speed was available to me when so motivated.
However, I didn't actually get to see the bits I missed until about 10 days ago. I've vowed never to drink in a movie theatre ever again.
At the end of the first movie (see the context of the original post) we didn't know there had been previous Neos.
The questions I posed were questions that were open at the end of the first movie, not questions that are open now. I consider at least *those* questions to have been well and truly answered by the sequels.
No? How about "Who is the oracle and how does she know this stuff?"
How about "Who was the previous 'one' and if he was able to remake the matrix as he saw fit, why didn't he free everyone then?"
Sure, they managed to make the movie in such a way that you didn't *think* about these open questions very much, but if you *had* thought of them, they'd certainly have been annoying/unbelievable/aggravating.
Doesn't sound like that to me. Although QA is clearly *one of* the benefits of going more community-oriented, they already had "community QA" in the form of RawHide.
Sounds to me more like what they got is *development* of desktop-based packages so that they can concentrate on the enterprise space.
IRV has the following problems that plurality doesn't:
1) Massively more complicated ballot
2) Violates monotonicity (it is possible to actually cause a candidate to lose by voting FOR them). It should go without saying that this is evil.
3) Despite appearing to solve the spoiler problem, it actually has exactly the same spoiler issue as soon as the "third party" starts becoming competitive.
Consider the case of three parties, A, B and C, where A and B are the "mainstream" parties and C is a third party. Suppose that most of C's supporters prefer B to A, but B's supporters are split evenly between A and C for second choice (which is reasonable if B is the "center" party). For the sake of this example, we'll assume that the three parties are fairly close to equal in first-place votes. This gives the following distribution:
A: 1/3 of the population B,A: 1/6 of the population B,C: 1/6 of the population C,B: 1/3 of the population
Now, as long as C stays in third place, it doesn't affect the race between A and B, but if C ever ends up with more first-place votes than B, it ends up handing the election to A - even though B would still win in a straight race between A and B.
(If it helps you to think of A, B and C as Republican, Democrat and Green, do so. Or as Democrat, Republican, Libertarian).
So in fact IRV is a sap to minority parties without actually helping them become part of the mainstream - because if they ever do, the spoiler problem rears its head again. So I stand by my statement that IRV is in *every* way worse than Plurality, and that switching to IRV would be even worse than the status quo.
Slashdot Mirror
They don't have any such thing. An "implementation of Java" requires class libraries as well as a VM, and IBM (along with every other "independent" commercial VM vendor) uses and depends on Sun's class libraries. Which is a MUCH bigger task than just creating a VM.
The VM may be clean room and open-sourcable, but by itself it's useless; Kaffe and Classpath and gcj and all the other existing Free alternatives are at a much greater level of development than IBM's because they do address the class library issue. IBM would probably have the sense to go with the technically better solution - either graft Classpath or Kaffe's libraries onto their VM and release that, or simply back one of the existing open VMs, depending on how tightly their implementation is bound to Sun's libraries.
However, if IBM was going to do this they could have done it years ago - I've pretty much given up hope that they have any intention of backing any of the Free Java projects.
Not true - GNU Classpath is JDK1.4 compatible in many areas, and pretty much complete at JDK1.2 level with the sole exception of CORBA (which has interesting license issues) and Swing, which is being actively worked on (there's a screenshot of Classpath running a Swing demo that's limited to buttons and checkboxes, and it actually works!).
Sure, cloning Swing and many of the other massive libraries in recent JDK versions is a mammoth undertaking and life would be a great deal easier for everyone if Sun would open up the reference implementation. But don't discount the work of the people who are already doing it - they're further along than you think!
(BTW, the reason those URLs point to kaffe.org is just because my own domain for them expired; kaffe.org graciously agreed to host the files, but the results are independently generated and not biased in favor or against any of kaffe's "competitors". Having said that, Kaffe is another project that's made leaps and bounds recently. There are, in fact, multiple completely Free/Open Source implementations of Java now that can run many high-profile Java apps, including Eclipse and Tomcat)
Easy - those thousands of people who don't know each other also send email *back* to the mailing list. Only a few dummies send email back to the spammers.
For something based on statistics, the difference would likely be very noticeable.
This is great!
:( Running the last packaged libapache-mod-mono with the new mono debs still doesn't work (unsurprisingly, since they're the same debs I installed previously)...
:)
Unfortunately, it still doesn't include libapache-mod-mono which is critical for me
Still, it seems the debian mono packaging team is making much more progress than I had guessed from following their mailing list
Thanks for the heads-up!
Sadly, the Debian packages are still in rather a strange state of limbo: You can get them if you can find the magic incantation to add to sources.list, but they're a January 14th snapshot (which isn't too old, but if you're going to do something from 2 weeks ago, why bother with a nightly snapshot in the first place, why not the last stable release?) and don't include libapache-mod-mono. Meanwhile libapache-mod-mono is also available separately but doesn't seem to work with the latest packages of everything else.
Which is making my attempts to port an ASP.NET application to work on mono a fairly frustrating experience, to say the least...
They should appeal the DMCA order, then, because Martian Rock clearly falls under the domain of the RIAM not the RIAA...
I'm using IKVM in production right now in a real (albeit small) company. And mono plan on releasing 1.0 within the next 6 months, which they will recommend for production use - we're evaluating that for use within our company also.
.NET exe (and the JDBC driver into a dll). Since all our C# developers have the CLR installed by definition, they don't need to install any additional software. This actually happened a little too late because all the developers already had the Java stuff installed, but it'll help when we hire new developers :)
I have some development tools written in Java which have language-independent output (supporting both Java and C#) but our company primarily uses C# these days. Still, every developer had to install the JDK and a JDBC driver in order to run our compile process, since it invokes these Java tools.
These days, I've used IKVM to compile that code to a
It is true, though, that I've found that following projects like these has helped my ability to write much more boring code. It's also helped when I'm faced with a problem like "hey, we might have a customer who wants to buy our (.NET) stuff but uses primarily Unix, can we help them?" and I can say "Well, it's not 1.0 yet, but there's this project called Mono..."
I know what I'd do.
I'd open the attachment in a safe way (eg "Save as" in a decent mailer), attempt to figure out what it was doing, and if (as in this case) it was a simple program to send my username and computer name to my company's sysadmin, I'd send a message saying that, eg "HarryPotter" on machine "HOGWARTS" (or "Frodo" on "SHIRE", or "Neo" on "MATRIX", or... you get the idea) had run the virus.
Then I'd wait and see if the sysadmin could figure out how to trace that to me...
I'm afraid you are wrong.
While "int" is treated "as if" it were an Object in the CLR, and behaves as if it was a subclass of Object, it really is a value type: It's allocated on the stack, cannot be subclassed, and is passed by value when it's used as a method parameter or return value.
The same goes for all the other built in primitive types (float, decimal, double, long, short, bool, char etc), along with user-defined Structs and Enums. These are all considered "Value Types" in the CLR.
When instances of value types need to be treated as Objects, they are boxed, just like in Java, except that it's transparent and automatic and there's no separate class that represents the boxed type - the distinction Java makes between int and Integer simply doesn't exist in C#. (System.Int32 is just an alias for int and isn't equivalent to Integer at all in any way).
The article clearly states that in Generic C#, the JIT actually creates specialized binary code whenever a generic class is instantiated with a value type argument. It even uses the example of List<int> turning into int[] (remember value types like int can't be subclassed, so the subclass problem doesn't arise).
C# provides the "illusion" of no primitives, but it's certainly not true that everything's always boxed. It just provides a more elegant and flexible model of boxing than Java does.
Then take your Java library code, compile it with IKVM running on Mono, and write code in C# using it without being locked into any vendor :)
It doesn't have anything to do with autoboxing in Java, but it has everything to do with boxing in Java versus C#.
Both C# and Java require boxing if you want to put an int into an object variable. They differ in whether that boxing is automatic or not (at least until Java makes it automatic in 1.5), but that's irrelevant: it still happens, so the performance impact is the same in both languages.
However, if you're trying to put that same int into an ArrayList<int>, the two languages are very different.
In Java, ArrayList<int> is really just ArrayList<Object> under the covers, and stores its contents into an array of type Object[], so the boxing operation still needs to happen. Hence, there's a performance hit.
In C#, ArrayList<int> is interpreted as such at the runtime level and the JIT compiler creates native code for it that stores the contents in an array of type int[]. When you store values into that array, no boxing is needed, automatic or otherwise. No performance hit.
Both C# 1.1 and Java 1.4 collections can hold primitive types if you box them. In Java 1.5, you get some extra syntactic sugar (which C# had from day one) but the boxing still happens. In C# 2.0, generic collections can hold primitive types without boxing. That's the difference that he's trying to get across in this article, and why it's specific to generics.
All of this is explained very clearly in the article and it amazes me that so many people can't grasp it...
Oh, and with regard to terminal windows: Microsoft's command prompt has the exact same problem. In their case you have to use Enter (!) to copy the text, or right-click the window title bar (!!) and drill down through two levels of nested menus (!!!) to find the copy option.
IMHO, the best solution would be for Ctrl-X and Ctrl-C not to be passed through to the terminal if any text in the window is selected. Not sure what to do about Ctrl-V though...
Why are so many people ignorant of the fact that Ctrl-C and Ctrl-V work just as well in KDE and GNOME as they do in Windows? Have you never even tried it?
Try this in any recent KDE[1], GNOME or Mozilla application:
Select some text
Ctrl-X or Ctrl-C
Select some other text
Ctrl-V
It's strange - you'd think that if doing it the "X way" (select + middleclick) was so much of a pain to people, they'd actually *try* it the windows way and discover that it works, and be happy.
Personally, I use both: the X way for quickly inserting pasted data, and the windows way if I want to replace something at my destination. Now if only X applications were better at non-textual formats...
Stuart.
[1] I haven't actually tested in KDE but I've heard it was fixed in KDE3 to work just the same.
Branched out a bit from advocating house-elf rights, haven't they?
I thought it was funny how every provider was fighting tooth and nail to oppose these regulations but they still jumped at the chance to advertise the new "feature" as soon as they were forced to provide it.
It's almost as if they knew that customers wanted this and they can make a profit by offering it! What a radical notion!
They can't file against Red Hat because part of SCO's defense against RH's countersuit is that RH have no standing because SCO were never going to sue RH. If they then turn around and sue RH now, they've shot their own defense (and despite appearances, SCO do seem to realize that they need all the defenses they can get at this point...)
I've got you beat for "great stuff missed through sheer bladder-related lack of foresight". I missed (a significant chunk of) the architect scene from Reloaded in exactly the same way you're describing. In a movie with half a dozen ten-minute content-free fight scenes, I had to pick the moment where information is being thrown at you faster than you can process it already.
Like yourself, I found that almost superhuman speed was available to me when so motivated.
However, I didn't actually get to see the bits I missed until about 10 days ago. I've vowed never to drink in a movie theatre ever again.
At the end of the first movie (see the context of the original post) we didn't know there had been previous Neos.
The questions I posed were questions that were open at the end of the first movie, not questions that are open now. I consider at least *those* questions to have been well and truly answered by the sequels.
No no no no no, it said "Between *this* world and the machine world".
ie, between the Matrix and the machine world. The oracle isn't *in* "our" world.
No? How about "Who is the oracle and how does she know this stuff?"
How about "Who was the previous 'one' and if he was able to remake the matrix as he saw fit, why didn't he free everyone then?"
Sure, they managed to make the movie in such a way that you didn't *think* about these open questions very much, but if you *had* thought of them, they'd certainly have been annoying/unbelievable/aggravating.
Would it still be a "me, too" if they put a man on the moon? How about if they established a permanent human presence there? How about on Mars?
Any giant leap ahead has to begin by catching up...
Doesn't sound like that to me. Although QA is clearly *one of* the benefits of going more community-oriented, they already had "community QA" in the form of RawHide.
Sounds to me more like what they got is *development* of desktop-based packages so that they can concentrate on the enterprise space.
"thick sulphurous clouds that completely block out the Sun" - the article
"We don't know who struck first. But we do know that it was humans who darkened the sky." - Morpheus
Venus is the real world, Earth is the Matrix?
(yeah, I probably screwed up the Matrix quote. I'll surrender my geek credentials on my way out...)
I don't see any misspelled word in the sentence you quoted. Could you elaborate?
IRV has the following problems that plurality doesn't:
1) Massively more complicated ballot
2) Violates monotonicity (it is possible to actually cause a candidate to lose by voting FOR them). It should go without saying that this is evil.
3) Despite appearing to solve the spoiler problem, it actually has exactly the same spoiler issue as soon as the "third party" starts becoming competitive.
Consider the case of three parties, A, B and C, where A and B are the "mainstream" parties and C is a third party. Suppose that most of C's supporters prefer B to A, but B's supporters are split evenly between A and C for second choice (which is reasonable if B is the "center" party). For the sake of this example, we'll assume that the three parties are fairly close to equal in first-place votes. This gives the following distribution:
A: 1/3 of the population
B,A: 1/6 of the population
B,C: 1/6 of the population
C,B: 1/3 of the population
Now, as long as C stays in third place, it doesn't affect the race between A and B, but if C ever ends up with more first-place votes than B, it ends up handing the election to A - even though B would still win in a straight race between A and B.
(If it helps you to think of A, B and C as Republican, Democrat and Green, do so. Or as Democrat, Republican, Libertarian).
So in fact IRV is a sap to minority parties without actually helping them become part of the mainstream - because if they ever do, the spoiler problem rears its head again. So I stand by my statement that IRV is in *every* way worse than Plurality, and that switching to IRV would be even worse than the status quo.