I'm sorry, but you've got to be fucking kidding me.
Is this kind of statement really necessary?
NASA knows explicitly what the final deployment platform will look like, from hardware up to OS and available software binaries. It's part of the all-encompassing and overwhelming specification process used when creating a new government (well, NASA/military) project.
NASA has a huge range of deployment platforms, and not all are of military grade hardware/embedded. There are matters of security systems etc.
For what NASA is doing, what they need is a language that is well-understood (Ada most definitely is), and Java doesn't fall into that category yet...C++, maybe, but not Java.
I worked alongside people who were developing Ada in the early 80s, so I know how it is used. I am also aware of the attempts at proof-of-code-correctness that were made at the time. This is extremely difficult with source, and much work in avionics is done by analysing and tracing the resultant machine code. This is why using a language such as Java that has a specified, hardware independent, set of byte codes is so useful. Java is a far smaller language than C++, specifically tailored for safety and security. It also includes built-in standard capabilities for real-time concurrency much like Ada.
"Could it be that NASA is finally giving up on Ada and embracing the safety, reliability, and simplicity of Java?"
BWHAHAHAHAHA!!!
Oh, man. I needed a good laugh today.
Aside from the compulsory Slashdot Java FUD, it's really not a joke. Java has a big advantage in that the the bytes codes produced can be verified, and so the program tested, without any concerns of the final deployment platform. This is a major advantage for an organisation like NASA which most likely has a wide range of hardware on which software is deployed.
Re:If thats your "bigger question"
on
NASA Goes SourceForge
·
· Score: 2, Insightful
If the code is open source, that means ANYBODY can work on it, improve it, or find and squash bugs. If one person makes a mistake somebody else will correct it, If somebody tries to do something harmful to the code. *several* other people will instantly remove the malware.
Just because anybody CAN work on code and deal with bugs, doesn't mean anybody WILL. There is no evidence that bugs in any given OS projects are 'instantly' removed.
As for accountability? Why do we always have to have some poor soul to point the finger at? why do we have to make any one person in particular accountable?
Because if you are going to use a product for any serious use it is customary to look for guarantees that the product is fit for use. Some open source projects have sufficient reputation that they they are trusted in most areas without any such legal or commercial guarantees (such as the Linux kernel). OS in itself is no guarantee of quality.
Would you leap into and drive a free car built by someone you don't know just because they are willing to show you the blueprints and parts list?
If the project fails then we have ALL failed, and, friend, the accountability will be found in the fact that we WILL improve upon the code, we WILL learn from our mistakes, and we WILL prove that Open Source (free) software IS the best way to spend tax payers money when it comes to computer programming:: period
The key to the best way to produce software is to have skilled and motivated developers. The Open Source-ness is not always relevant.
Considering our common ancestor was so far back, what's more likely: parallel and independent development of speech in more recent years, or a singular development WAY back in the day?
Independent development, I would think. The main article here is wrong. The same circuitry is NOT present in humans. As the original article says, the circuitry has a 'human counterpart' - most likely independently evolved.
Not that I mind - I hate how they refuse to adopt native widgets, instead pushing their own cheap imitations that lack all the keyboard shortcuts, font anti-aliasing, and solid feel that native widgets have
Swing can have any keyboard shortcuts you like. Also, in Java 5.0, you can globally turn on antialiasing of all fonts in an application.
Swing can have a solid feel, as it can have any feel you like! There are a huge range of looks and feels.
I have to say good riddance.
You can say what you like, but Java use is growing, and has already overtaken C++ use in most job indicators.
Or DVRs, or hardware firewalls, or pocket PCs, or any of the myriad things people run Linux on. The Linux kernel, even GNU Linux, has a wide variety of uses. Solaris is specializing in a few areas, the main one, of course, being Sun hardware.
This isn't true. Solaris has been running on such systems for a very long time. It is widely used on embedded systems.
Unix WAS about competition. It has been many, many years since that was so.
This just isn't true. You are talking about some very specific technical details of different versions of Unix, but for the average developer they are of little consequence. GNU is almost everywhere, even on Solaris, and cross-compiling is largely trivial.
If Sun were after a monopoly, they would not have put so much work into open source projects such as Open Office. They would not SHIP Linux!
I remember the attempt to provide an VM abstraction from which to compile a program to, and then to compile back to a vendors system. And what happened? The vendors did not want it(In particular, HP, SGI, and Sun fought it). So even that avenue closed.
It's here now. It's called Java. It has a VM, and at runtime it produces native code for a vendors system. It is supported by HP, SGI and Sun (and IBM, and many, many others).
And yes, IMHO, the Unix market is not just sick, but on a death bed. It will do the slow spiral that many other OSes did.
As illustrated by the million or so downloads of Solaris 10?
Fair enough... Well, then either your memory is affected by CRT radiations, or you are the most positive-minded unix user i ever met/read.:)
No, I am simply a general-purpose developer. I have no interest in writing system code or hardware-dependent stuff.
It is not difficult (especially these days, with GNU tools) to write general commercial C/C++ code that can port very quickly between different Unix versions.
How many proprietary UNIX flavors can one run on Sparc hardware besides Solaris?
That is not the point. You have a choice to compile your code for Solaris on Sparc, or HP/UX on whatever, or Linux, or AIX... that is what open systems are about.
We were discussing the openness of the software, not the hardware, as the latter is hidden by the OS.
3. Java. I don't think Sun has made much money from Java, and it's been a huge distraction.
It has been one of the best things they have ever done. They have made a considerable amount of money from J2EE licensing and J2ME.
Sun should have made Java an open specification like, err, EVERY OTHER FRIGGING LANGUAGE EVER MADE,
Java is an open specification. Anyone can implement it, and many do. Your 'every other language' can't possibly include Visual Basic - highly popular, and totally closed.
instead of fighting idiotic lawsuits with MS (who were in the right for a change).
When their clear intention was to kill Java by removing its portability?
It took me 2 hours to figure out the right command magic yesterday to get a local queue to a remote LPD. (Yeah I know you can install Cups on it, I'm just saying, *by default*)
Solaris provides a highly standard implementation of Unix System V printing. The administration of this can be found in a few minutes via Google.
Nonsense. Sun helped pioneer competition in the OS market with their strong backing of open systems. Unix is about competition, as it is about providing implementation of open standards, which make the version of Unix you use a matter of choice, and not something you are tied to.
Why would anyone using Linux go for a closed, proprietary Unix flavor?
Because most of what is done on such systems uses the open, non-proprietary features.
Unix (and similar systems such as Linux) has been such a success over the years because they implement open standards: TCP/IP networking, POSIX, X-Windows etc. This use of open standards and APIs explains why it is so much easier to port programs between different versions of Unix than to other OSes.
To say that Solaris is a 'closed, proprietary Unix flavor' is self-contradictory. Unix is a set of open standards. What is proprietary is the implementation. If you use GNU tools on Solaris, you can even avoid most of that. Commercial Unix users usually don't care about whether or not the kernel source is available; all they care about is the quality of implementations and price.
Not so, they are very much directly competing in the same market for the same customers.
'Linux' covers a wide range of distributions, which are tailored for a wide range of uses. The primary competitor for Solaris is RedHat Enterprise Linux. This is not the same as 'Linux' in general. For example, Solaris is not aimed at the desktop, like Mandrake.
According to http://www.lpl.arizona.edu/impacteffects/
(The impact effect calculator). It wouldn't cause widespread devastation, at least not on a global scale. There would be a lot of damage within 100 km of the impact, but not much vapourised rock or ejecta, so there would not be much effect beyond that. Very bad for those close to the impact, but no significant effect globally.
Visual Basic is not BASIC. Delphi isn't Pascal. Both are proprietary languages with exactly zero portability; as their platforms fade, the code will fade with them.
Visual Basic certainly IS Basic - anyone who has used a typical dialect
Delphi IS Pascal. You can type pretty standard pascal
Basic has never had portability, no matter what the dialect. On the other hand, Delphi conforms pretty closely to Object Pascal, and there are many implementations to which most of the code can be ported.
C remains the canonical language for interoperability, purely because every existing commercially viable language (like every new language, for at least two decades to come) provides a way to call out to C functions.
That is not interoperability - that is low-level library functions. Most interoperability these days is by high-level APIs and protocols: RPC, CORBA, RMI, DCOM, COM+, SOAP etc. These have nothing to do with C - servers and clients can be implemented in any language.
Most jobs posted on job sites are fakes, filled with the current popular buzzwords. (This is easily verified by distributing a stellar resume and seeing how many callbacks you get.) When the next fad pops up, the fake listings will use its buzzwords instead.
Surveys of what developers are actually using bear out the conclusion of the dominance of Java. Check out its growth in Sourceforce - so much for 'fads'.
Actual Java jobs that pay more than scale for web page designers remain rare.
No. Most (real) Java jobs are for J2EE, which is providing serious general commercial infrastructure including behind web apps. Very few people are using C or C++ for this kind of work anymore on new projects.
C++ book sales are still increasing.
But how much in comparison with Java,.NET, Python etc?
I have to agree, the largest failing of Linux (and Java for that matter) is a good set of fonts.
Set the JAVA_FONTS environment variable and your Java app can pick up and use any fonts on your system, no matter what OS you are on. There is no limit to what fonts you can use.
There are ISO standards for a few languages. Those languages have a chance to be viable for a while.
Very few languages fade from viability - check the significant number of current COBOL jobs.
Ten years from now when Java and C# are as nearly forgotten as BASIC and Pascal are now,
Forgotten? Have you any idea how popular Visual Basic is? Delphi (Object Pascal) is still widely used.
C will remain an important language, at least for interoperability purposes,
C is a terrible language for interoperability as it is highly system dependent. Far better are interpreted or JITed bytecode languages, which can have guaranteed compatibility on different platforms. What matters is the ease of use of the interfaces that are exposed, such as CORBA or SOAP.
and C++ will still be well known and heavily used.
Most job sites now show Java more widely used that C++. C++ certainly won't go away, but it's use as a general-purpose language is slowly fading.
and there is the problem. windows is windows, you mention two different desktops.
You can pick one and install just that one.
i understand your POV, however, remember the average user - more than one choice/look/feel is confusing.
I think this is a myth! After all, there are a large variety of controls and styles presented to users from web browsers and multimedia players, and there are very few reports of problems with these.
Not true for every linux app - where some forms don't have context menus, some do. In fact most don't I've found - take xfig for example.
I didn't say every Linux app: I was specifying either KDE or GNOME under linux. If you stick to apps that use those interfaces, you do get a consistent UI.
I'd love to write code that runs on Unix, but by the time I've paid for a development environment and installed it, finished hunting down and installing all the extra packages I need to get something that has parity with even the weakest Linux install, then finished grabbing the source and manually recompiling the packages (and dependencies) that were so far out of date to be useless...
By that time, I've likely realized that I don't give a shit about Unix.
Well, HP-UX, at least.
Well, exactly. The last sentence kind of invalidates the rest. You can't sum up all Unix from just one version.
For example, for Solaris there is www.sunfreeware.com: Loads of GNU packages, binary and source, all up-to-date, packaged up for Solaris, and ready to install.
If you need more than vi/emacs for development there is NetBeans or Eclipse - both high quality IDEs, both can be used for a range of languages, including Java and C++.
I think this is devolving into an argument about mere semantics.
That only happens if you create contradictory statements, such as 'I have experiences but not qualia'. If you have experiences, and the quality of the sensations changes, you are experiencing qualia. By definition.
But the point is that theoretically, a thorough scientific investigation of the brain could do exactly what you're asking: explain what a person's experience of color is.
No, because sensation is subjective. You can't even be sure that anyone else shares the same sensation as you given the same situation, so how can you explain the experience, when you are not even sure what it is you are explaining?
To think that it would is applying a standard to the brain that not applied to anything else. If you applied that same standard to everything, everything would seem mysterious and disconnected.
I think you might see where I am leading! Why is that 'brains' experience?
Slashdot Mirror
I'm sorry, but you've got to be fucking kidding me.
Is this kind of statement really necessary?
NASA knows explicitly what the final deployment platform will look like, from hardware up to OS and available software binaries. It's part of the all-encompassing and overwhelming specification process used when creating a new government (well, NASA/military) project.
NASA has a huge range of deployment platforms, and not all are of military grade hardware/embedded. There are matters of security systems etc.
For what NASA is doing, what they need is a language that is well-understood (Ada most definitely is), and Java doesn't fall into that category yet...C++, maybe, but not Java.
I worked alongside people who were developing Ada in the early 80s, so I know how it is used. I am also aware of the attempts at proof-of-code-correctness that were made at the time. This is extremely difficult with source, and much work in avionics is done by analysing and tracing the resultant machine code. This is why using a language such as Java that has a specified, hardware independent, set of byte codes is so useful. Java is a far smaller language than C++, specifically tailored for safety and security. It also includes built-in standard capabilities for real-time concurrency much like Ada.
"Could it be that NASA is finally giving up on Ada and embracing the safety, reliability, and simplicity of Java?"
BWHAHAHAHAHA!!!
Oh, man. I needed a good laugh today.
Aside from the compulsory Slashdot Java FUD, it's really not a joke. Java has a big advantage in that the the bytes codes produced can be verified, and so the program tested, without any concerns of the final deployment platform. This is a major advantage for an organisation like NASA which most likely has a wide range of hardware on which software is deployed.
If the code is open source, that means ANYBODY can work on it, improve it, or find and squash bugs. If one person makes a mistake somebody else will correct it, If somebody tries to do something harmful to the code. *several* other people will instantly remove the malware.
:: period
Just because anybody CAN work on code and deal with bugs, doesn't mean anybody WILL. There is no evidence that bugs in any given OS projects are 'instantly' removed.
As for accountability? Why do we always have to have some poor soul to point the finger at? why do we have to make any one person in particular accountable?
Because if you are going to use a product for any serious use it is customary to look for guarantees that the product is fit for use. Some open source projects have sufficient reputation that they they are trusted in most areas without any such legal or commercial guarantees (such as the Linux kernel). OS in itself is no guarantee of quality.
Would you leap into and drive a free car built by someone you don't know just because they are willing to show you the blueprints and parts list?
If the project fails then we have ALL failed, and, friend, the accountability will be found in the fact that we WILL improve upon the code, we WILL learn from our mistakes, and we WILL prove that Open Source (free) software IS the best way to spend tax payers money when it comes to computer programming
The key to the best way to produce software is to have skilled and motivated developers. The Open Source-ness is not always relevant.
Considering our common ancestor was so far back, what's more likely: parallel and independent development of speech in more recent years, or a singular development WAY back in the day?
Independent development, I would think. The main article here is wrong. The same circuitry is NOT present in humans. As the original article says, the circuitry has a 'human counterpart' - most likely independently evolved.
Not that I mind - I hate how they refuse to adopt native widgets, instead pushing their own cheap imitations that lack all the keyboard shortcuts, font anti-aliasing, and solid feel that native widgets have
Swing can have any keyboard shortcuts you like. Also, in Java 5.0, you can globally turn on antialiasing of all fonts in an application.
Swing can have a solid feel, as it can have any feel you like! There are a huge range of looks and feels.
I have to say good riddance.
You can say what you like, but Java use is growing, and has already overtaken C++ use in most job indicators.
I don't get it. If sun was responsible for 64-bit computers being commonplace, how come Java doesn't support 64 bits?
Who says it doesn't?
Go to javasoft.com, and check the downloads. You can get Java for 64-bit Linux on AMD and 64-bit Solaris on Sparc and AMD.
Or DVRs, or hardware firewalls, or pocket PCs, or any of the myriad things people run Linux on. The Linux kernel, even GNU Linux, has a wide variety of uses. Solaris is specializing in a few areas, the main one, of course, being Sun hardware.
This isn't true. Solaris has been running on such systems for a very long time. It is widely used on embedded systems.
Unix WAS about competition. It has been many, many years since that was so.
This just isn't true. You are talking about some very specific technical details of different versions of Unix, but for the average developer they are of little consequence. GNU is almost everywhere, even on Solaris, and cross-compiling is largely trivial.
If Sun were after a monopoly, they would not have put so much work into open source projects such as Open Office. They would not SHIP Linux!
I remember the attempt to provide an VM abstraction from which to compile a program to, and then to compile back to a vendors system. And what happened? The vendors did not want it(In particular, HP, SGI, and Sun fought it). So even that avenue closed.
It's here now. It's called Java. It has a VM, and at runtime it produces native code for a vendors system. It is supported by HP, SGI and Sun (and IBM, and many, many others).
And yes, IMHO, the Unix market is not just sick, but on a death bed. It will do the slow spiral that many other OSes did.
As illustrated by the million or so downloads of Solaris 10?
Fair enough... Well, then either your memory is affected by CRT radiations, or you are the most positive-minded unix user i ever met/read. :)
No, I am simply a general-purpose developer. I have no interest in writing system code or hardware-dependent stuff.
It is not difficult (especially these days, with GNU tools) to write general commercial C/C++ code that can port very quickly between different Unix versions.
How many proprietary UNIX flavors can one run on Sparc hardware besides Solaris?
That is not the point. You have a choice to compile your code for Solaris on Sparc, or HP/UX on whatever, or Linux, or AIX... that is what open systems are about.
We were discussing the openness of the software, not the hardware, as the latter is hidden by the OS.
3. Java. I don't think Sun has made much money from Java, and it's been a huge distraction.
It has been one of the best things they have ever done. They have made a considerable amount of money from J2EE licensing and J2ME.
Sun should have made Java an open specification like, err, EVERY OTHER FRIGGING LANGUAGE EVER MADE,
Java is an open specification. Anyone can implement it, and many do. Your 'every other language' can't possibly include Visual Basic - highly popular, and totally closed.
instead of fighting idiotic lawsuits with MS (who were in the right for a change).
When their clear intention was to kill Java by removing its portability?
It took me 2 hours to figure out the right command magic yesterday to get a local queue to a remote LPD. (Yeah I know you can install Cups on it, I'm just saying, *by default*)
Solaris provides a highly standard implementation of Unix System V printing. The administration of this can be found in a few minutes via Google.
"which make the version of Unix you use a matter of choice, and not something you are tied to."
You must be young...
On the contrary. I have been using Unix since the 70s, so I know what I am talking about.
Plain and simple, because they fear competition.
Nonsense. Sun helped pioneer competition in the OS market with their strong backing of open systems. Unix is about competition, as it is about providing implementation of open standards, which make the version of Unix you use a matter of choice, and not something you are tied to.
Why would anyone using Linux go for a closed, proprietary Unix flavor?
Because most of what is done on such systems uses the open, non-proprietary features.
Unix (and similar systems such as Linux) has been such a success over the years because they implement open standards: TCP/IP networking, POSIX, X-Windows etc. This use of open standards and APIs explains why it is so much easier to port programs between different versions of Unix than to other OSes.
To say that Solaris is a 'closed, proprietary Unix flavor' is self-contradictory. Unix is a set of open standards. What is proprietary is the implementation. If you use GNU tools on Solaris, you can even avoid most of that. Commercial Unix users usually don't care about whether or not the kernel source is available; all they care about is the quality of implementations and price.
Not so, they are very much directly competing in the same market for the same customers.
'Linux' covers a wide range of distributions, which are tailored for a wide range of uses. The primary competitor for Solaris is RedHat Enterprise Linux. This is not the same as 'Linux' in general. For example, Solaris is not aimed at the desktop, like Mandrake.
According to
http://www.lpl.arizona.edu/impacteffects/
(The impact effect calculator). It wouldn't cause widespread devastation, at least not on a global scale. There would be a lot of damage within 100 km of the impact, but not much vapourised rock or ejecta, so there would not be much effect beyond that. Very bad for those close to the impact, but no significant effect globally.
Visual Basic is not BASIC. Delphi isn't Pascal. Both are proprietary languages with exactly zero portability; as their platforms fade, the code will fade with them.
.NET, Python etc?
Visual Basic certainly IS Basic - anyone who has used a typical dialect
Delphi IS Pascal. You can type pretty standard pascal
Basic has never had portability, no matter what the dialect. On the other hand, Delphi conforms pretty closely to Object Pascal, and there are many implementations to which most of the code can be ported.
C remains the canonical language for interoperability, purely because every existing commercially viable language (like every new language, for at least two decades to come) provides a way to call out to C functions.
That is not interoperability - that is low-level library functions. Most interoperability these days is by high-level APIs and protocols: RPC, CORBA, RMI, DCOM, COM+, SOAP etc. These have nothing to do with C - servers and clients can be implemented in any language.
Most jobs posted on job sites are fakes, filled with the current popular buzzwords. (This is easily verified by distributing a stellar resume and seeing how many callbacks you get.) When the next fad pops up, the fake listings will use its buzzwords instead.
Surveys of what developers are actually using bear out the conclusion of the dominance of Java. Check out its growth in Sourceforce - so much for 'fads'.
Actual Java jobs that pay more than scale for web page designers remain rare.
No. Most (real) Java jobs are for J2EE, which is providing serious general commercial infrastructure including behind web apps. Very few people are using C or C++ for this kind of work anymore on new projects.
C++ book sales are still increasing.
But how much in comparison with Java,
I have to agree, the largest failing of Linux (and Java for that matter) is a good set of fonts.
Set the JAVA_FONTS environment variable and your Java app can pick up and use any fonts on your system, no matter what OS you are on. There is no limit to what fonts you can use.
There are ISO standards for a few languages. Those languages have a chance to be viable for a while.
Very few languages fade from viability - check the significant number of current COBOL jobs.
Ten years from now when Java and C# are as nearly forgotten as BASIC and Pascal are now,
Forgotten? Have you any idea how popular Visual Basic is? Delphi (Object Pascal) is still widely used.
C will remain an important language, at least for interoperability purposes,
C is a terrible language for interoperability as it is highly system dependent. Far better are interpreted or JITed bytecode languages, which can have guaranteed compatibility on different platforms. What matters is the ease of use of the interfaces that are exposed, such as CORBA or SOAP.
and C++ will still be well known and heavily used.
Most job sites now show Java more widely used that C++. C++ certainly won't go away, but it's use as a general-purpose language is slowly fading.
and there is the problem. windows is windows, you mention two different desktops.
You can pick one and install just that one.
i understand your POV, however, remember the average user - more than one choice/look/feel is confusing.
I think this is a myth! After all, there are a large variety of controls and styles presented to users from web browsers and multimedia players, and there are very few reports of problems with these.
Not true for every linux app - where some forms don't have context menus, some do. In fact most don't I've found - take xfig for example.
I didn't say every Linux app: I was specifying either KDE or GNOME under linux. If you stick to apps that use those interfaces, you do get a consistent UI.
You aren't too swift are you... almost every windows app has the same menubar, click/right click context menu interface.
And that is not true for Linux/KDE or Linux/GNOME?
I'd love to write code that runs on Unix, but by the time I've paid for a development environment and installed it, finished hunting down and installing all the extra packages I need to get something that has parity with even the weakest Linux install, then finished grabbing the source and manually recompiling the packages (and dependencies) that were so far out of date to be useless...
By that time, I've likely realized that I don't give a shit about Unix.
Well, HP-UX, at least.
Well, exactly. The last sentence kind of invalidates the rest. You can't sum up all Unix from just one version.
For example, for Solaris there is www.sunfreeware.com: Loads of GNU packages, binary and source, all up-to-date, packaged up for Solaris, and ready to install.
If you need more than vi/emacs for development there is NetBeans or Eclipse - both high quality IDEs, both can be used for a range of languages, including Java and C++.
I think this is devolving into an argument about mere semantics.
That only happens if you create contradictory statements, such as 'I have experiences but not qualia'. If you have experiences, and the quality of the sensations changes, you are experiencing qualia. By definition.
But the point is that theoretically, a thorough scientific investigation of the brain could do exactly what you're asking: explain what a person's experience of color is.
No, because sensation is subjective. You can't even be sure that anyone else shares the same sensation as you given the same situation, so how can you explain the experience, when you are not even sure what it is you are explaining?
To think that it would is applying a standard to the brain that not applied to anything else. If you applied that same standard to everything, everything would seem mysterious and disconnected.
I think you might see where I am leading! Why is that 'brains' experience?