Re:Languages for the Java VM...
on
The Future of Java?
·
· Score: 4, Insightful
Rogers Sessions took the time to investigate every single one of the projects on the JVM languages page. He posted his results to the ObjectWatch [objectwatch.com] site. After long research, he found only 8 that were actual implementations of a non-Java programming language for the JVM. Of those 8, in his opinion not one was available or suitable for professional development.
Roger Sessions... now there's an unbiased observer. Once upon a time, Roger was a CORBA spec writer, until he basically got booed out of the CORBA camp for non-performance (this spec is going to be SO good when it's finished... really!) Then he became a COM/DCOM/COM+ apologist for Microsoft. Read through his ObjectWatch newsletters and see if you can spot any unbiased commentary on technologies; you will quickly observe a trend: Microsoft good, non-Microsoft bad. I won't say that Roger is a complete bought-and-paid-for Microsoft shill, but he gets his books published by Microsoft Press... you draw your own conclusions.
As for the JVM's language-neutrality (or lack thereof): so what, big deal.
There is no such thing as a language-neutral platform, be it hardware or virtual. Every computer instruction set that has been developed to date has had biases built in; it's inevitable. The system designer has certain goals, and usually has a target set of capabilities, which may only work in assembly language. Take the X86 instruction set: only a fraction of the X86 instructions are useful to a C compiler. The rest of the instructions can only be used by writing assembly language. The only hardware instruction sets that I can think of that are completely accessible by a higher level language are those that were specifically designed to implement a single language (Forth, LISP), and they make lousy targets for other languages.
The question is: does it matter? There are very good reasons why the JVM doesn't allow pointer arithmetic and pointer casting: the designers wanted a provably secure runtime environment, so they excluded unsafe operations. So you can't implement C/C++ for the JVM; big deal. If you really need those capabilities, you're better off operating outside of the JVM and interacting via JNI or some form of RPC. That said, there are still plenty of languages that are written in Java, compile to Java, or compile directly to Java bytecodes; certainly more than currently target the.NET CLI or any other virtual machine. Use one of those, or write your own.
Yesterday I had mod points, and nothing worth modding. Today I'll just have to agree with you. I realize that Rob & the rest of the crew just want to post what they're interested in reading, but I think allowing a little feedback on the stories themselves from the moderators would appropriate.
This has been a problem with small programming shops for decades, and the answer is always the same: put the source code in escrow, with contractual provisions that take effect if your company folds, which gives the customer access and full rights to the source code. Any reasonably competent lawyer can set this up for you, including secure storage of the source code. All you have to do when you do a delivery to a customer is to cut a source code tape and ship that to the lawyer.
Not easier but rather just another way to do make the same Mouse trap. If we go by code size as being "easier" using parseInt requires a try/catch block being implemented since it throws a NumberFormatException. Both are used in different situations.
The difference is this: do you want to create an object and throw it away, or not? Your way creates a new object and then immediately throws it away; my way doesn't. That's a pretty expensive operation... much more expensive than try/catch. If you want to parse numbers from strings without having to worry about catching exceptions, write a static method in a utility class.
>>but you keep the information where it's supposed to be: with the data class, not with a representational storage class.
I do not understand your terminology. Please give a definition of what you call "data" classes and "storage" classes.
String has been overloaded as a class: it serves two purposes. The first is to store strings, and the second is to store string representations of other types of data (numbers, XML, whatever). When used in the first sense, it's a data class; when used in the second sense it's a representational storage class - it's storing a representation of the data, not a directly useable instance of the data. It's the difference between 5+4 (9) and "5" + "5" (55).
My point was that the String class should know how to manipulate strings, but it shouldn't know about manipulating numbers stored in string format.
>>On C#, it's as easy as:
int a = Int.Parse(some_string);
In Java its pretty easy too: int a = new Integer(some_string).intValue();
Under Java it's even easier that that:
int a = Integer.parseInt( some_string );
Just remember to catch the NumberFormatException.
FWIW, String isn't the right place to do data conversion -- you'll have to change the String class everytime you want to add another data conversion for another type. Java already does this the way it should (and so does C#). Sure, the code isn't as 'pretty' that way, but you keep the information where it's supposed to be: with the data class, not with a representational storage class.
Oh Really? [alternic.org] Are you sure? [opennic.org]
Really. I use OpenNIC for my DNS services, but there are still areas where OpenNIC is in dispute with ICANN (and other alternate DNS providers). For example:.info was only recently approved by OpenNIC, and.biz still hasn't been approved, since there's already 1 or 2 other.biz registrars, besides the ICANN approved vendor.
More importantly, even if you use an alternate DNS provider, you're still stuck with getting your IP addresses from ICANN (by way of your ISP). There's NO Internet-compatible way to get around that.
Ok, so why is this "non-election" issue a problem when it comes to ICANN, but when it comes to the non-elected people that decide what's part of the Linux kernel, then it's ok?
Both are deciding things for a community, and none are elected. Hypocritical to complain about one and not the other, don't you think?
The difference lies in the fact that you can choose what OS/Kernel/distribution you want to use, but you can't choose not to use ICANN if you want to use the Internet. ICANN controls the NANA, which means they control the IP addresses, and they control who manages the TLDs.
If you have a problem with the direction that Linus and the Linux kernel developers are going, you can (1) fork the kernel, (2) use Free/Open/Net BSD, (3) use some variant of Windows, or (4) find something else. If you have a problem with ICANN's policies, you're stuck. They've got a monopoly, and they're using it with an agenda. You may or may not agree with their agenda (I don't), but their proposal for eliminating elected board members means that they're eliminating any hope that you or I could influence their agenda short of completely overthrowing the system.
Even dictators have to pretend to have laws and courts...
A couple of months ago when I got my copy of the 20th anniversary DVD. I enjoyed watching it again (and all of the bonus material), but that doesn't make it a great movie, just a fun one. Great movies have good direction, good acting, a truly gripping story, and can stand the test of time. The Day The Earth Stood Still was a great movie -- when you're watching it, you don't really care about the cheesy SFX because you're caught up in the story. Birth of a Nation was a great movie, even if it seems racist-as-all-get-out today. Compared to great movies, Tron easily becomes recognizable for what it is: a good, fun, unpretencious, but certainly not great, movie.
Who in their right mind would rank Barbarella on a Top 20 list of anything? It wouldn't even rank well on a soft pr0n list. For that matter, Sleeper doesn't belong on the list either, and much as I may have enjoyed Tron in the day, it's not a great movie.
On the other hand, IMHO, the other movies on the list are great movies, and would make reasonable candidates for a Top-20 list, even if you or I wouldn't agree with their ordering. Just keep in mind that Top-X lists are just tools that you can choose to use or ignore them as necessary.
My father works for a company that manufactures cranes and materials handling equipment. A number of years ago, they were doing an extensive installation for Bethlehem Steel. The system was responsible for picking up a hopper that came in on a railcar, lifting it 100 feet in the air, moving it over a blast furnace, dumping the raw materials in the blast furnace, and then putting the hopper back on the railcar again.
The first time that they ran a full system test, the crane picked up a fully loaded hopper, took it straight to the top of the shaft, and then immediately dumped the hopper -- right on top of the railcar. Oops. In case you weren't sure, 30 tons of scrap iron will make one a hell of an impact; it took them over a week, just to pick up the pieces.
Oh, man, I'll give you the not ready part, at least for the main course, but MREs are great , especially when compared to the K-rats before them. I loved MRE's, especially the tabasco sauce. And it is a meal -- each one has 2500 calories if you consume the whole thing.
Having not had K-rats, I'll take your word on the improvement... but MRE's made the worst Army mess hall I ever ate in seem positively appetizing by comparison. They even make White Castle seem appetizing.
It's hard to tell without the context of the quotation, but I would assume 'the Product' to refer to Windows XP or Remote Desktop. The emphasis also seems to be on "unless the Device has a separate license for the Product.". That seems to me to have a significant different implication than we are led to believe by the article.
That would tend to imply, to me (and mostly likely to a lawyer as well), that in order to access any applications running on an XP box remotely, the remote 'client' would (1) have to be XP, and (2) would have to have licenses for each of the applications to be run. I don't know about you, but when I've used VNC to remotely access Windows boxes, it's been because I couldn't run the application in question on my current machine (ie: my client was Solaris, or Linux, or even a Palm Pilot).
Now as for the Microsoft interpretation? I would assume an interpretation consistent with my 24 years of using Microsoft software -- the worst case.
Please raise your hand if you develop software for a living; that is, you support yourself and/or your family by developing software.Now, keep your hand raised if you believe that your company could offer the same software that you helped to create as a free, open-source download and still keep you employed.
Raising my hand, and keeping it raised.
I have spent most of my professional career developing either (a) vertical applications, or (b) government applications. In both of these cases, part of the contract was to deliver the source code of the application to the customer, either immediately, or in escrow. Most of the profits actually come from services: customization, integration, or support. If you're very, very lucky, software sales may cover most of the development costs, if you get enough sales.
That aside, I agree with your point: that there is room, and need, for both open and closed software in the world today. I would merely prefer to increase the demand for open software, so as to reduce the demand for me, as a programmer, to reinvent the wheel.
I'm not sure what to think about the fact that Sun has a "Chief Competitive Officer."
What can I say? It would have been cooler to call him the "Chief Fudmeister", but he wouldn't take the position until Sun changed the job title to three words.
I can write a Perl program that defines a CLR compliant.Net framework class (blessed package). I can then inherit that class in a C# class. Or vice versa. You cannot do that with Java.
Well, you might not be able to do that with Perl and Java today, but you can do it with Python and Java today -- check out Jython, the JVM implementation of Python.
Microsoft spent the effort (money) to implement Perl on CRL; anytime that someone wants to do the work to implement Perl on top of the JVM, you'll be able to do what you want. Given that Parrot development continues, you may yet get Perl for JVM courtesy of Jython and Python and Parrot.
And why does the unit have "sealed lead acid batteries" in it?
The batteries are there to: (1) provide power for you (and the unit) while you're switching hydrogen canisters, and (2) depending upon the design, to even out the line voltage.
[Lecture Mode On]
There are two basic designs for UPSes: continuous and intermittent.
The UPSes that you buy for SOHO use are intermittent -- line voltage feeds a battery circuit (battery charger + batteries + inverter) and goes to a relay, which switches between the battery circuit and the normal line voltage. When line voltage goes off, the relay switches; when line voltage comes back, the relay switches back. While the relay is switching, there will be a short interruption in power, but most AC equipment can handle the (very short) interruption. This type of UPS will also have surge protectors built in to filter out high voltage and spikes, but can't do a lot for brownouts other than switch to batteries.
Continuous UPSes work differently -- the line voltage is used to charge the batteries, which run the inverter, which provides clean, uninterrupted power. No relays, no interruptions, no worry about power spikes or brownouts. Unfortunately, you're continuously charging and draining the batteries -- which significantly reduces the working life of the batteries. This type of UPS requires hot-swappable batteries, and is generally a lot more expensive to purchase and maintain (which explains the popularity of the intermittent UPSes).
[Lecture Mode Off]
From what I read on the site, the AirGen acts like an intermittent UPS -- when line voltage shuts off, the AirGen switches to generated power, and switches back when line voltage returns. The batteries are probably there just to provide the necessary power to start and maintain the generator, and to provide power while you're switch canisters. The AirGen *could* be a continuous UPS, with the fuel-cells supplementing line voltage for charging the batteries, but I doubt it -- everything they've posted on their site points towards the intermittent UPS design.
I can see your argument, but surely if I am contemplating installing software on my machine, I have the right to know what, exactly, it blocks, right?
If you're a customer, that's between you and the company that's supplying the software, but my guess is that in order to install the software, you'll have to agree to a license that says you won't get to know that answer. If you didn't agree to the license, you shouldn't have installed it, yada, yada, UCITA, you gave up your rights.
In any event, when this case did come to court a few years ago, the complaining company bought the company that wrote the cracking software and killed the software.
That was pre-DMCA; in the post-DMCA era, the offending company and all of their employees (or at least the officers and programmers) could be indicted for violating the criminal provisions of the DMCA.
Would reverse engineering software that blocks websites to determine what websites, exactly, are being blocked, qualify as a use that is socially acceptable?
It would to *me*, but I'm not a judge. Censorware authors would certainly disagree, as would too many self-appointed, holier-than-thou guardians of community values (like Virginia State Delegate Richard Black, who recently tried to force this cr*p into the Loudoun County Public Libraries, only to lose on First Amendment grounds in US Federal Court).
IANAL, but I suspect that it wouldn't pass muster because blocking software is typically deployed in as an "opt-in" measure -- it's on computers because someone who's responsible for that computer wants it there. If you don't want to use a computer with blocking software, you can go elsewhere, or sue to have it removed (in the case of computers in libraries). Breaking the encryption on URL lists wouldn't help in court, since you can always sue the software vendor and get their lists via deposition or subponea.
Slashdot Mirror
Roger Sessions ... now there's an unbiased observer. Once upon a time, Roger was a CORBA spec writer, until he basically got booed out of the CORBA camp for non-performance (this spec is going to be SO good when it's finished ... really!) Then he became a COM/DCOM/COM+ apologist for Microsoft. Read through his ObjectWatch newsletters and see if you can spot any unbiased commentary on technologies; you will quickly observe a trend: Microsoft good, non-Microsoft bad. I won't say that Roger is a complete bought-and-paid-for Microsoft shill, but he gets his books published by Microsoft Press... you draw your own conclusions.
As for the JVM's language-neutrality (or lack thereof): so what, big deal.
There is no such thing as a language-neutral platform, be it hardware or virtual. Every computer instruction set that has been developed to date has had biases built in; it's inevitable. The system designer has certain goals, and usually has a target set of capabilities, which may only work in assembly language. Take the X86 instruction set: only a fraction of the X86 instructions are useful to a C compiler. The rest of the instructions can only be used by writing assembly language. The only hardware instruction sets that I can think of that are completely accessible by a higher level language are those that were specifically designed to implement a single language (Forth, LISP), and they make lousy targets for other languages.
The question is: does it matter? There are very good reasons why the JVM doesn't allow pointer arithmetic and pointer casting: the designers wanted a provably secure runtime environment, so they excluded unsafe operations. So you can't implement C/C++ for the JVM; big deal. If you really need those capabilities, you're better off operating outside of the JVM and interacting via JNI or some form of RPC. That said, there are still plenty of languages that are written in Java, compile to Java, or compile directly to Java bytecodes; certainly more than currently target the .NET CLI or any other virtual machine. Use one of those, or write your own.
Yesterday I had mod points, and nothing worth modding. Today I'll just have to agree with you. I realize that Rob & the rest of the crew just want to post what they're interested in reading, but I think allowing a little feedback on the stories themselves from the moderators would appropriate.
Other moderations for stories could include:
This has been a problem with small programming shops for decades, and the answer is always the same: put the source code in escrow, with contractual provisions that take effect if your company folds, which gives the customer access and full rights to the source code. Any reasonably competent lawyer can set this up for you, including secure storage of the source code. All you have to do when you do a delivery to a customer is to cut a source code tape and ship that to the lawyer.
I was going to ask the same thing. Too bad I don't have mod points today.
So an MS clone should be a Source Forgery, right?
The difference is this: do you want to create an object and throw it away, or not? Your way creates a new object and then immediately throws it away; my way doesn't. That's a pretty expensive operation ... much more expensive than try/catch. If you want to parse numbers from strings without having to worry about catching exceptions, write a static method in a utility class.
>>but you keep the information where it's supposed to be: with the data class, not with a representational storage class.I do not understand your terminology. Please give a definition of what you call "data" classes and "storage" classes.
String has been overloaded as a class: it serves two purposes. The first is to store strings, and the second is to store string representations of other types of data (numbers, XML, whatever). When used in the first sense, it's a data class; when used in the second sense it's a representational storage class - it's storing a representation of the data, not a directly useable instance of the data. It's the difference between 5+4 (9) and "5" + "5" (55).
My point was that the String class should know how to manipulate strings, but it shouldn't know about manipulating numbers stored in string format.
I always heard it this way:
Under Java it's even easier that that:
Just remember to catch the NumberFormatException.
FWIW, String isn't the right place to do data conversion -- you'll have to change the String class everytime you want to add another data conversion for another type. Java already does this the way it should (and so does C#). Sure, the code isn't as 'pretty' that way, but you keep the information where it's supposed to be: with the data class, not with a representational storage class.
You can program servers to keep copies of network print jobs.
You can remove floppy and CDR/W drives from desktop machines.
The military does this in secure environments -- only authorized people have access to hardware that can take data off of the network.
But you can configure the mail server to make copies of all messages that pass through the server.
Really. I use OpenNIC for my DNS services, but there are still areas where OpenNIC is in dispute with ICANN (and other alternate DNS providers). For example: .info was only recently approved by OpenNIC, and .biz still hasn't been approved, since there's already 1 or 2 other .biz registrars, besides the ICANN approved vendor.
More importantly, even if you use an alternate DNS provider, you're still stuck with getting your IP addresses from ICANN (by way of your ISP). There's NO Internet-compatible way to get around that.
Typo: change "NANA" to "IANA" in paragraph 2, above.
The difference lies in the fact that you can choose what OS/Kernel/distribution you want to use, but you can't choose not to use ICANN if you want to use the Internet. ICANN controls the NANA, which means they control the IP addresses, and they control who manages the TLDs.
If you have a problem with the direction that Linus and the Linux kernel developers are going, you can (1) fork the kernel, (2) use Free/Open/Net BSD, (3) use some variant of Windows, or (4) find something else. If you have a problem with ICANN's policies, you're stuck. They've got a monopoly, and they're using it with an agenda. You may or may not agree with their agenda (I don't), but their proposal for eliminating elected board members means that they're eliminating any hope that you or I could influence their agenda short of completely overthrowing the system.
Even dictators have to pretend to have laws and courts...
A couple of months ago when I got my copy of the 20th anniversary DVD. I enjoyed watching it again (and all of the bonus material), but that doesn't make it a great movie, just a fun one. Great movies have good direction, good acting, a truly gripping story, and can stand the test of time. The Day The Earth Stood Still was a great movie -- when you're watching it, you don't really care about the cheesy SFX because you're caught up in the story. Birth of a Nation was a great movie, even if it seems racist-as-all-get-out today. Compared to great movies, Tron easily becomes recognizable for what it is: a good, fun, unpretencious, but certainly not great, movie.
Who in their right mind would rank Barbarella on a Top 20 list of anything? It wouldn't even rank well on a soft pr0n list. For that matter, Sleeper doesn't belong on the list either, and much as I may have enjoyed Tron in the day, it's not a great movie.
On the other hand, IMHO, the other movies on the list are great movies, and would make reasonable candidates for a Top-20 list, even if you or I wouldn't agree with their ordering. Just keep in mind that Top-X lists are just tools that you can choose to use or ignore them as necessary.
My father works for a company that manufactures cranes and materials handling equipment. A number of years ago, they were doing an extensive installation for Bethlehem Steel. The system was responsible for picking up a hopper that came in on a railcar, lifting it 100 feet in the air, moving it over a blast furnace, dumping the raw materials in the blast furnace, and then putting the hopper back on the railcar again.
The first time that they ran a full system test, the crane picked up a fully loaded hopper, took it straight to the top of the shaft, and then immediately dumped the hopper -- right on top of the railcar. Oops. In case you weren't sure, 30 tons of scrap iron will make one a hell of an impact; it took them over a week, just to pick up the pieces.
Having not had K-rats, I'll take your word on the improvement ... but MRE's made the worst Army mess hall I ever ate in seem positively appetizing by comparison. They even make White Castle seem appetizing.
I've done my time in the US military, and the biggest lie I was told, during my enlisted service, was ...
MRE = Meal, Ready to Eat
That would tend to imply, to me (and mostly likely to a lawyer as well), that in order to access any applications running on an XP box remotely, the remote 'client' would (1) have to be XP, and (2) would have to have licenses for each of the applications to be run. I don't know about you, but when I've used VNC to remotely access Windows boxes, it's been because I couldn't run the application in question on my current machine (ie: my client was Solaris, or Linux, or even a Palm Pilot).
Now as for the Microsoft interpretation? I would assume an interpretation consistent with my 24 years of using Microsoft software -- the worst case.
Raising my hand, and keeping it raised.
I have spent most of my professional career developing either (a) vertical applications, or (b) government applications. In both of these cases, part of the contract was to deliver the source code of the application to the customer, either immediately, or in escrow. Most of the profits actually come from services: customization, integration, or support. If you're very, very lucky, software sales may cover most of the development costs, if you get enough sales.
That aside, I agree with your point: that there is room, and need, for both open and closed software in the world today. I would merely prefer to increase the demand for open software, so as to reduce the demand for me, as a programmer, to reinvent the wheel.
What can I say? It would have been cooler to call him the "Chief Fudmeister", but he wouldn't take the position until Sun changed the job title to three words.
Well, you might not be able to do that with Perl and Java today, but you can do it with Python and Java today -- check out Jython, the JVM implementation of Python.
Microsoft spent the effort (money) to implement Perl on CRL; anytime that someone wants to do the work to implement Perl on top of the JVM, you'll be able to do what you want. Given that Parrot development continues, you may yet get Perl for JVM courtesy of Jython and Python and Parrot.
The batteries are there to: (1) provide power for you (and the unit) while you're switching hydrogen canisters, and (2) depending upon the design, to even out the line voltage.
[Lecture Mode On]
There are two basic designs for UPSes: continuous and intermittent.
The UPSes that you buy for SOHO use are intermittent -- line voltage feeds a battery circuit (battery charger + batteries + inverter) and goes to a relay, which switches between the battery circuit and the normal line voltage. When line voltage goes off, the relay switches; when line voltage comes back, the relay switches back. While the relay is switching, there will be a short interruption in power, but most AC equipment can handle the (very short) interruption. This type of UPS will also have surge protectors built in to filter out high voltage and spikes, but can't do a lot for brownouts other than switch to batteries.
Continuous UPSes work differently -- the line voltage is used to charge the batteries, which run the inverter, which provides clean, uninterrupted power. No relays, no interruptions, no worry about power spikes or brownouts. Unfortunately, you're continuously charging and draining the batteries -- which significantly reduces the working life of the batteries. This type of UPS requires hot-swappable batteries, and is generally a lot more expensive to purchase and maintain (which explains the popularity of the intermittent UPSes).
[Lecture Mode Off]
From what I read on the site, the AirGen acts like an intermittent UPS -- when line voltage shuts off, the AirGen switches to generated power, and switches back when line voltage returns. The batteries are probably there just to provide the necessary power to start and maintain the generator, and to provide power while you're switch canisters. The AirGen *could* be a continuous UPS, with the fuel-cells supplementing line voltage for charging the batteries, but I doubt it -- everything they've posted on their site points towards the intermittent UPS design.
If you're a customer, that's between you and the company that's supplying the software, but my guess is that in order to install the software, you'll have to agree to a license that says you won't get to know that answer. If you didn't agree to the license, you shouldn't have installed it, yada, yada, UCITA, you gave up your rights.
In any event, when this case did come to court a few years ago, the complaining company bought the company that wrote the cracking software and killed the software.That was pre-DMCA; in the post-DMCA era, the offending company and all of their employees (or at least the officers and programmers) could be indicted for violating the criminal provisions of the DMCA.
It would to *me*, but I'm not a judge. Censorware authors would certainly disagree, as would too many self-appointed, holier-than-thou guardians of community values (like Virginia State Delegate Richard Black, who recently tried to force this cr*p into the Loudoun County Public Libraries, only to lose on First Amendment grounds in US Federal Court).
IANAL, but I suspect that it wouldn't pass muster because blocking software is typically deployed in as an "opt-in" measure -- it's on computers because someone who's responsible for that computer wants it there. If you don't want to use a computer with blocking software, you can go elsewhere, or sue to have it removed (in the case of computers in libraries). Breaking the encryption on URL lists wouldn't help in court, since you can always sue the software vendor and get their lists via deposition or subponea.