Firstly, there are plenty of non-Unix CLI based systems, but none of their suppliers are stupid enough to claim that the browser is part of the OS.
Secondly, Microsoft doesn't use existing standards. It's HTML rendering has been always buggy, especially in terms of CSS handling. Developers have know this, and had to code to deal with it, for years.
Thirdly, you use the phrase 'nothing to do with the Kernel'. That is exactly my point - its not, and never has been, a core feature of any operating system.
How could Explorer possibly be a core part of Windows in embedded systems?
In the second, as far as I'm concerned, IE is a key component of Windows.
Strange how every other operating system that has been developed in the long history of the IT industry has managed to operate without the presence of a HTML renderer as a 'key component'.
What you're left with may be an OS in your opinion, but it's not in mine
So all those GUI-less Linux/UNIX installations used by ISPs to host websites and app servers are not 'operating systems'?
You've oversimplified things by just saying Oracle is fast. If given the right hardware, query time outweighs connection time, and the databases are extremely huge, Oracle performs well. Otherwise, it's too resource intensive to use reasonably
I have found exactly the opposite: having used oracle from version 7, I have seen it run very nicely on positively archaic machines (ancient sparc systems), being robust, fast, and handling bizzare page-length SQL queries, with sub-selects and unions, that MySQL would not go near.
Newer Oracles are even better: 9 was a big step forward. Not resource intensive at all. I have Oracle 10 on a 256MB 2GHz AMD and it runs like a dream; just as fast as MySQL, even with lots of lightweight queries. Its not using that much of the memory - I have heavyweight Java IDEs running at the same time.
Older oracles did indeed try and be resource hogs. The trick with those is to install what you need and no more, and go into the resource specifications during setup (memory and disk use) and simply tell them to cut back.
I have been running Oracle 8 in 128MB on Solaris for years. If you are having trouble in 1GB, something is wrong.
Yep understand that but you answered because oracle is fast I was merely point out that oracle is not always fast, actually if not set up right it is dead slow. I said that towards end of post. I tend to crap on sometimes so my meaning gets lost:)
Well, you must understand I have been in IT for a very, very long time. I used to consider a 25MHz 486 fast!
When people talk about multi-gigahertz machines (i.e. a typical home pc these days) as 'low end' it makes me feel very old and bitter:) I feel I have to defend such hardware....
For large businesses, Microsoft is king. There are a few corporate giants that run Lotus, but most are MS Office + Exchange based.
All you are saying here is that Microsoft is king of windows-based collaboration servers. What about file servers, print servers, app servers, database servers? These are the areas in which Linux (and UNIX) use is huge. There is even a significant Netware presence.
If Microsoft were a transport company, they would be selling thousands of overpriced bicycles, comparing this to truck sales numbers, and calling themselves King of the Road.
Oracle may be fast in comparision to other enterprise databases (Db2, sqlserver etc....) but for some applications/organisations it is just far to over the top.
Absolutely, but that was not the question. It was why use Oracle on low-end hardware, not why use Oracle at all.
That is not true. Well it made be true for you and your company, depending on what deal you guys cut, but not for everyone.
From the oracle website:
"All software downloads are free, and each comes with a development license that allows you to use full versions of the products only while developing and prototyping your applications."
Perhaps you are working on updating an already deployed system? Sounds like it.
Oracle is not fast. Not not fast. Speed isn't Oracle's game. Data integrity is. The point is that you sacrifice speed for things like real atomicity.
Oracle has phenomenal speed, and superb query and index optimisation. Its even faster if you give it raw access to disk. The point of paying a lot for a system like Oracle is you don't sacrifice speed for atomicity. You don't just get speed, you get scalable speed.
Solaris as a dev platform... to put it politely... not the best out there. For a long time there's been no decent C++ compiler, their IDE is so-so, and for compilation speeds,
I'm not sure I understand you here, but there seems to be a confusion between operating system and development tools. There is no such thing as a 'Solaris IDE', any more than there is a 'Linux IDE'. Sure, on most Linux distros you can choose to install GNU C++, but also you might not. The development system is 'GNU/Linux'. You can also set up 'GNU/Solaris' by downloading GNU C++ and all other GNU stuff for solaris from www.sunfreeware.com for years. You don't even need to compile - the software is packaged ready for Solaris.
Because Oracle is fast. Very, very fast. Not only is it fast, but it has serious database features. Its like putting a $30,000 engine in a Yugo.
Also, Oracle allows you use their database for development and prototyping for free. You don't need to pay for a license, or for high-end hardware to host the system, until you are ready to deploy.
Its not whether you do or don't. Its about whether you might. Its about making sensible choices to allow a project to expand, evolve and develop as specifications change. Selecting limited and restricted development tools and platforms is just not good practice.
So going to Java I see more what's missing (not yet duplicated) than what is there.
I think there is a pretty good reason for this. My reading of the history of Java was that what Sun really wanted was Smalltalk, but negotiations failed, so instead they devised a language that was supposed to be somewhat like Smalltalk, and with a minimal C-style syntax. The core principle was that the language should be safe: secure enough to be distributed over devices and networks and it should not just allow, but *enforce*, certain coding standards: hence compulsory exception handling. They wanted to avoid what many considered the mistakes of C++, which was adding too many features without understanding the implications for making code cryptic and potentially unsupportable; this is why they are so resistant to MI and operator overloading, and why they have been very cautious about templates.
I think this is why so many coders really hate Java: it is not a 'quick hack' language, and it can be verbose and tedious to code sometimes. My view is that having seen how the use of a 'do anything' language like C/C++ for general purpose (rather than system) use wrecked so many software projects over the past 15 years, I find the wide use of Java to be a significant and healthy step forward in the IT industry. I would have preferred a more dynamic and interactive language, like LISP or Smalltalk, but the IT industry is reactionary, and the C-ish syntax was probably a major factor in Java's success.
My view is that you are hugely exaggerating what 'business' programming is; I feel you are saying that its anything but technical uses?
I feel a far better comparison is with Pascal. Java is far more like Pascal - Turing complete, general purpose, but not what you would use to write device drivers!
For some strange reason, Java enthusiasts (I' talking other people, not you) keep insisting with almost religious fervor that their platform is the one true solution to everything software
I know - this seems to be a strange aspect of developers! But - I think there is a point: There are few things messier than a mixed-language project or an organisation trying to build up a repository of re-useable code in multiple languages.
I'm sure you have many stories yourself of "100% Pure Java(tm)" code that led straight to a javacore.txt file. Actually no, never - not once, and I have been using Java from when it first came out, but I can see what you are saying.
Personally, I feel Java is kind of unique.... it has so many simultaneous benefits that I haven't seen in any other language in close to 30 years of development experience. (Smalltalk came close, but blew it, which was a damn shame: I adored VisualWorks):
Its free. Its (these days) fast. It really, really is portable in the vast majority of cases. Better still, its portable *after you have compiled it*. It has a working cross-platform GUI (you may think its crap, but at least its there). It has a decent multi-threading model. It is supported by multiple vendors. Its got a reasonable object model.
Nothing else comes close, as far as I can tell, for *general purpose* (not just 'business') coding. Its not perfect, and of course there are bugs, but its a pretty awesome idea.
I could go and cover all the points. To make a start.... (but before you reply, please read my final point)
I am a Smalltalk developer, and have been for 20 years. I love the language, but poor performance and stupid licencing of commercial implementations cut back on my use.
Java 1.5 is out beta now, full next month.
Your points about sockets and files on different platforms are irrelevant: With Java you code general handling of possible timeouts. Who cares about specific machines?
(As an example, there is not a single line of code in major Java network apps, such as Tomcat, that says 'if we are on solaris, do the following..').
I could talk about serial port and USB device use in Java.
I could go on about the nature of the word 'legacy'.
I could describe the many uses to which I have put reflection.
and on and on....
My final point: this is all irrelevant to the argument! The point that was made is that Java is like COBOL. This is a plain silly comparison. For goodness sake, you know full well its silly. Show me the COBOL games, the COBOL network apps, the COBOL applets, the COBOL midlets, the COBOL portable GUI apps. Just because Java doesn't suit *you* does not make it poor language for general developers in a wide range of situations. Have a look at freshmeat - see Java being used for collaboration, games, multimedia, science. Java is not like COBOL!
After further thought, I think we are talking at cross purposes here. Number of processors that can be attached to a system image is not at all related to the scalability of the operating system. The operating system itself (functions such as filesystem and memory access) may show no effective speedup after the addition of a relatvely small number of processors. This number of processors is a lot smaller for Linux than for Solaris (although the 2.6 kernel is far better). Those additional processors can be used by user software, often using specialised multiprocessing facilities such as PVM.
Not according to SGIs specs. They don't have a single image setup that can run more than 256 processors, and to be honest, that is not going to work very well unless you have application-specific tailored code.
The same goes for an Altix. It's a single OS image,
In most cases, it is not a single image, its a cluster of nodes. You can specify up to 256 processors per image on some architectures, but the usual recommendation is about 16max.
multiple inheritance, a free-as-in-speech runtime environment, or a best-of-breed desktop application of any kind.
Yes it does - the eclipse development environment. But who cares? The desktop environment is a absolutely tiny part of the market. The main market for everything the web, where Java rules.
Precisely. Java and COBOL are both used for business applications and not much else, when compared to the alternatives.
No. Java mobile games alone dwarfs almost all other software in terms of deployments. Java is also becoming widely used as a numerical and engineering language: C/C++ and other just did not cut it as a replacement for Fortran. Java does.
Java has the sweet spot of just enough OO-ism without too many confusing abstractions plus a big enough standard library to cover 95% of most uses.
Rubbish. Java has far more OO features than C++. Multiple inheritance was abandoned long ago by mainstream OO language designers as a flawed idea. The interface/contract system, as used by Java, is considered superior. There is no (standard and guaranteed) ability to dynamically load classes and inspect their properties in C++. I can't take a.so compiled on an Intel architecture and load it on PowerPC. I can, of course, do that with a Java.class.
I guess Java fails because its library does not cover 100% of uses. Well, in that case, all languages and development systems ever shipped are failures. Personally, I think 95% is pretty good for a standard library.
But once you leave the business logic arena, Java has a very long way indeed to catch up.
A bizzare statement. I have used Java for business, numerical simulation, websites, mobile device programming, multi-threaded networking and device interfacing. My code is fast and I need to make not a single line change when switching between platforms. Can C++ do that? Of course not! Its C++ that has a long way to go before it can do it, assuming it ever can.
Java is backward because it doesn't do curses? Big deal! That is like saying that Linux is backward because it doesn't provide DOS as standard. Both are legacy.
Cobol does not have an object-oriented structure, templates, sophisticated threading and thread-safe collection classes, network and socket management, exceptions, a fast IO library, reflection, dynamic class loading, portable bytecode, a standard cross-platform client GUI library, standard libraries for encryption, and compression, customisable security management, cross-platform portability of compiled code, a standard framework for clustered server deployment (J2EE), built-in unicode support, standard libraries for high-performance cross-platform graphics (Java2D), a standard for web interface development (JSF), multimedia handling facilities and a very widely-supported standard for deployment on portable devices (J2ME).
Apart from that, its very much like COBOL, in that both languages are widely deployed, robust, reliable and used by some of the largest companies to handle critical data.
I'm also a fan of tennis... just not very good at it myself. I think there is a parallel here.
To imply Linux is dying because doesn't have the market share for residential OSs, is as ridiculous as your logic.
To imply that a language (Java) is dying when it has the largest market share, and is the most in-demand language in the world, is a strange and unusual definition of the term 'dying'.
Java is dead. Sun is grasping for air with this one.
And IBM, Apple, HP... those sad companies must all be mistaken. GNU Java is obviously a total waste of time as well.
It's slow
About the speed of C++ these days. I'd love to see what fast language you are using?
it's old Yeah - a full ten years since the first version. Gosh! We had better give up on C++ and C then... if Java is old, they are senile.
it's not an open language.
Yeah, its such a bad closed idea that IBM, HP, Apple etc. are so foolish to have implemented it.
It was a great idea, and the intentions were good, but Sun fucked it up.
Yeah. It only ended up as one of the main development languages for server software, and got used everywhere from mainframes, applets, app servers, and client applications. Its use as the primary language for the enormous mobile device market (especially games) and its presence on millions of mobile phones is obviously nothing but a minor passing fad. All the colleges teaching Java as a key skill for students must be informed!
Its such a disaster.... and its shocking no-one else has noticed. Good thing you did!
I eagerly await your next few posts:
"Windows - Microsoft's failed idea" "The web - it may be everwhere but it's doomed" "Intel - the chip that no-one used."
On a slightly differeny note I do really get tired of all the Sun bashing that goes on.
Oh come on. Being nothing more than a pioneering and innovative open systems company with decades of experience, if Sun refuses to listen to the wise suggestions of thousands of open source zealots, many of whom have at least a year or two of software experience, what can it expect?
Sun has always been a hardware company, and so they have always had control over the hardware their OS runs on.
No. Sun is also a software company, hence Solaris, and the earlier SunOS. The Unixes have always run on a range of architectures: In the 80s they were offering SunOS on Sparc, Motorola and Intel processors.
They have never yet had to deal with the tangled mess that is PC hardware, with conflicts, obscure devices, and all the rest of it.
Well, yes, they have. They have been producing Solaris x86 for a while now.
And the one thing missing most from the Sun Java desktop is Java
Eh? There are lots of Java apps and utilities on Java Desktop, and, of course, Java itself! Pre-installed, and integrated with Open Office and Mozilla.
And the future is multiprocessors...
Your "beyond 100" is nothing to a 512-way SGI Altix running Linux. There are 1024-way systems under development. Sun can't touch this.
Even the Altix doesn't touch this - its a supercluster system. The suggested configuration for individual nodes is pretty small - a max of 12 or 16 processors. You can specify up to 256 processors, but its unlikely to scale well unless you are using specialised application code.
Slashdot Mirror
Every other corresponds to *nix.
A short history lesson for you:
'Every other' includes....
Multix. RSX-11. VMS. CP/M. RT/OS. VxWorks.
VM/CMS. MS-DOS. DR-DOS. Netware. ICL/george.
S/390.
And hundreds of others.
Hint: before you post about CLI-based OSes... and anything else.... do some research.
Firstly, there are plenty of non-Unix CLI based systems, but none of their suppliers are stupid enough to claim that the browser is part of the OS.
Secondly, Microsoft doesn't use existing standards. It's HTML rendering has been always buggy, especially in terms of CSS handling. Developers have know this, and had to code to deal with it, for years.
Thirdly, you use the phrase 'nothing to do with the Kernel'. That is exactly my point - its not, and never has been, a core feature of any operating system.
How could Explorer possibly be a core part of Windows in embedded systems?
In the second, as far as I'm concerned, IE is a key component of Windows.
Strange how every other operating system that has been developed in the long history of the IT industry has managed to operate without the presence of a HTML renderer as a 'key component'.
What you're left with may be an OS in your opinion, but it's not in mine
So all those GUI-less Linux/UNIX installations used by ISPs to host websites and app servers are not 'operating systems'?
You've oversimplified things by just saying Oracle is fast. If given the right hardware, query time outweighs connection time, and the databases are extremely huge, Oracle performs well. Otherwise, it's too resource intensive to use reasonably
I have found exactly the opposite: having used oracle from version 7, I have seen it run very nicely on positively archaic machines (ancient sparc systems), being robust, fast, and handling bizzare page-length SQL queries, with sub-selects and unions, that MySQL would not go near.
Newer Oracles are even better: 9 was a big step forward. Not resource intensive at all. I have Oracle 10 on a 256MB 2GHz AMD and it runs like a dream; just as fast as MySQL, even with lots of lightweight queries. Its not using that much of the memory - I have heavyweight Java IDEs running at the same time.
Older oracles did indeed try and be resource hogs. The trick with those is to install what you need and no more, and go into the resource specifications during setup (memory and disk use) and simply tell them to cut back.
I have been running Oracle 8 in 128MB on Solaris for years. If you are having trouble in 1GB, something is wrong.
Yep understand that but you answered because oracle is fast I was merely point out that oracle is not always fast, actually if not set up right it is dead slow. I said that towards end of post. I tend to crap on sometimes so my meaning gets lost :)
:) I feel I have to defend such hardware....
Well, you must understand I have been in IT for a very, very long time. I used to consider a 25MHz 486 fast!
When people talk about multi-gigahertz machines (i.e. a typical home pc these days) as 'low end' it makes me feel very old and bitter
For large businesses, Microsoft is king. There are a few corporate giants that run Lotus, but most are MS Office + Exchange based.
All you are saying here is that Microsoft is king of windows-based collaboration servers. What about file servers, print servers, app servers, database servers? These are the areas in which Linux (and UNIX) use is huge. There is even a significant Netware presence.
If Microsoft were a transport company, they would be selling thousands of overpriced bicycles, comparing this to truck sales numbers, and calling themselves King of the Road.
Oracle may be fast in comparision to other enterprise databases (Db2, sqlserver etc....) but for some applications/organisations it is just far to over the top.
Absolutely, but that was not the question. It was why use Oracle on low-end hardware, not why use Oracle at all.
That is not true. Well it made be true for you and your company, depending on what deal you guys cut, but not for everyone.
From the oracle website:
"All software downloads are free, and each comes with a development license that allows you to use full versions of the products only while developing and prototyping your applications."
Perhaps you are working on updating an already deployed system? Sounds like it.
Oracle is not fast. Not not fast. Speed isn't Oracle's game. Data integrity is. The point is that you sacrifice speed for things like real atomicity.
Oracle has phenomenal speed, and superb query and index optimisation. Its even faster if you give it raw access to disk. The point of paying a lot for a system like Oracle is you don't sacrifice speed for atomicity. You don't just get speed, you get scalable speed.
Solaris as a dev platform... to put it politely... not the best out there. For a long time there's been no decent C++ compiler, their IDE is so-so, and for compilation speeds,
I'm not sure I understand you here, but there seems to be a confusion between operating system and development tools. There is no such thing as a 'Solaris IDE', any more than there is a 'Linux IDE'. Sure, on most Linux distros you can choose to install GNU C++, but also you might not. The development system is 'GNU/Linux'. You can also set up 'GNU/Solaris' by downloading GNU C++ and all other GNU stuff for solaris from www.sunfreeware.com for years. You don't even need to compile - the software is packaged ready for Solaris.
You could run Oracle on low-end hardware but why?
Because Oracle is fast. Very, very fast. Not only is it fast, but it has serious database features. Its like putting a $30,000 engine in a Yugo.
Also, Oracle allows you use their database for development and prototyping for free. You don't need to pay for a license, or for high-end hardware to host the system, until you are ready to deploy.
Its not whether you do or don't. Its about whether you might. Its about making sensible choices to allow a project to expand, evolve and develop as specifications change. Selecting limited and restricted development tools and platforms is just not good practice.
So going to Java I see more what's missing (not yet duplicated) than what is there.
I think there is a pretty good reason for this. My reading of the history of Java was that what Sun really wanted was Smalltalk, but negotiations failed, so instead they devised a language that was supposed to be somewhat like Smalltalk, and with a minimal C-style syntax. The core principle was that the language should be safe: secure enough to be distributed over devices and networks and it should not just allow, but *enforce*, certain coding standards: hence compulsory exception handling. They wanted to avoid what many considered the mistakes of C++, which was adding too many features without understanding the implications for making code cryptic and potentially unsupportable; this is why they are so resistant to MI and operator overloading, and why they have been very cautious about templates.
I think this is why so many coders really hate Java: it is not a 'quick hack' language, and it can be verbose and tedious to code sometimes. My view is that having seen how the use of a 'do anything' language like C/C++ for general purpose (rather than system) use wrecked so many software projects over the past 15 years, I find the wide use of Java to be a significant and healthy step forward in the IT industry. I would have preferred a more dynamic and interactive language, like LISP or Smalltalk, but the IT industry is reactionary, and the C-ish syntax was probably a major factor in Java's success.
I think I see where you are coming from.
My view is that you are hugely exaggerating what 'business' programming is; I feel you are saying that its anything but technical uses?
I feel a far better comparison is with Pascal. Java is far more like Pascal - Turing complete, general purpose, but not what you would use to write device drivers!
For some strange reason, Java enthusiasts (I' talking other people, not you) keep insisting with almost religious fervor that their platform is the one true solution to everything software
I know - this seems to be a strange aspect of developers! But - I think there is a point: There are few things messier than a mixed-language project or an organisation trying to build up a repository of re-useable code in multiple languages.
I'm sure you have many stories yourself of "100% Pure Java(tm)" code that led straight to a javacore.txt file.
Actually no, never - not once, and I have been using Java from when it first came out, but I can see what you are saying.
Personally, I feel Java is kind of unique.... it has so many simultaneous benefits that I haven't seen in any other language in close to 30 years of development experience. (Smalltalk came close, but blew it, which was a damn shame: I adored VisualWorks):
Its free.
Its (these days) fast.
It really, really is portable in the vast majority of cases. Better still, its portable *after you have compiled it*.
It has a working cross-platform GUI (you may think its crap, but at least its there).
It has a decent multi-threading model.
It is supported by multiple vendors.
Its got a reasonable object model.
Nothing else comes close, as far as I can tell, for *general purpose* (not just 'business') coding. Its not perfect, and of course there are bugs, but its a pretty awesome idea.
I could go and cover all the points. To make a start.... (but before you reply, please read my final point)
I am a Smalltalk developer, and have been for 20 years. I love the language, but poor performance and stupid licencing of commercial implementations cut back on my use.
Java 1.5 is out beta now, full next month.
Your points about sockets and files on different platforms are irrelevant: With Java you code general handling of possible timeouts. Who cares about specific machines?
(As an example, there is not a single line of code in major Java network apps, such as Tomcat, that says 'if we are on solaris, do the following..').
I could talk about serial port and USB device use in Java.
I could go on about the nature of the word 'legacy'.
I could describe the many uses to which I have put reflection.
and on and on....
My final point: this is all irrelevant to the argument! The point that was made is that Java is like COBOL. This is a plain silly comparison. For goodness sake, you know full well its silly. Show me the COBOL games, the COBOL network apps, the COBOL applets, the COBOL midlets, the COBOL portable GUI apps. Just because Java doesn't suit *you* does not make it poor language for general developers in a wide range of situations. Have a look at freshmeat - see Java being used for collaboration, games, multimedia, science. Java is not like COBOL!
After further thought, I think we are talking at cross purposes here. Number of processors that can be attached to a system image is not at all related to the scalability of the operating system. The operating system itself (functions such as filesystem and memory access) may show no effective speedup after the addition of a relatvely small number of processors. This number of processors is a lot smaller for Linux than for Solaris (although the 2.6 kernel is far better). Those additional processors can be used by user software, often using specialised multiprocessing facilities such as PVM.
Not according to SGIs specs. They don't have a single image setup that can run more than 256 processors, and to be honest, that is not going to work very well unless you have application-specific tailored code.
The same goes for an Altix. It's a single OS image,
In most cases, it is not a single image, its a cluster of nodes. You can specify up to 256 processors per image on some architectures, but the usual recommendation is about 16max.
Java doesn't have templates,
.so compiled on an Intel architecture and load it on PowerPC. I can, of course, do that with a Java .class.
Yes it does, in JAVA 1.5
multiple inheritance, a free-as-in-speech runtime environment, or a best-of-breed desktop application of any kind.
Yes it does - the eclipse development environment. But who cares? The desktop environment is a absolutely tiny part of the market. The main market for everything the web, where Java rules.
Precisely. Java and COBOL are both used for business applications and not much else, when compared to the alternatives.
No. Java mobile games alone dwarfs almost all other software in terms of deployments. Java is also becoming widely used as a numerical and engineering language: C/C++ and other just did not cut it as a replacement for Fortran. Java does.
Java has the sweet spot of just enough OO-ism without too many confusing abstractions plus a big enough standard library to cover 95% of most uses.
Rubbish. Java has far more OO features than C++. Multiple inheritance was abandoned long ago by mainstream OO language designers as a flawed idea. The interface/contract system, as used by Java, is considered superior. There is no (standard and guaranteed) ability to dynamically load classes and inspect their properties in C++. I can't take a
I guess Java fails because its library does not cover 100% of uses. Well, in that case, all languages and development systems ever shipped are failures. Personally, I think 95% is pretty good for a standard library.
But once you leave the business logic arena, Java has a very long way indeed to catch up.
A bizzare statement. I have used Java for business, numerical simulation, websites, mobile device programming, multi-threaded networking and device interfacing. My code is fast and I need to make not a single line change when switching between platforms. Can C++ do that? Of course not! Its C++ that has a long way to go before it can do it, assuming it ever can.
Java is backward because it doesn't do curses? Big deal! That is like saying that Linux is backward because it doesn't provide DOS as standard. Both are legacy.
Comparing Java to COBOL is almost exactly right.
Cobol does not have an object-oriented structure, templates, sophisticated threading and thread-safe collection classes, network and socket management, exceptions, a fast IO library, reflection, dynamic class loading, portable bytecode, a standard cross-platform client GUI library, standard libraries for encryption, and compression, customisable security management, cross-platform portability of compiled code, a standard framework for clustered server deployment (J2EE), built-in unicode support, standard libraries for high-performance cross-platform graphics (Java2D), a standard for web interface development (JSF), multimedia handling facilities and a very widely-supported standard for deployment on portable devices (J2ME).
Apart from that, its very much like COBOL, in that both languages are widely deployed, robust, reliable and used by some of the largest companies to handle critical data.
and I may dislike java
I think we picked that up.
but I'm a huge fan for common sense.
I'm also a fan of tennis... just not very good at it myself. I think there is a parallel here.
To imply Linux is dying because doesn't have the market share for residential OSs, is as ridiculous as your logic.
To imply that a language (Java) is dying when it has the largest market share, and is the most in-demand language in the world, is a strange and unusual definition of the term 'dying'.
Java is dead. Sun is grasping for air with this one.
And IBM, Apple, HP... those sad companies must all be mistaken. GNU Java is obviously a total waste of time as well.
It's slow
About the speed of C++ these days. I'd love to see what fast language you are using?
it's old
Yeah - a full ten years since the first version.
Gosh! We had better give up on C++ and C then... if Java is old, they are senile.
it's not an open language.
Yeah, its such a bad closed idea that IBM, HP, Apple etc. are so foolish to have implemented it.
It was a great idea, and the intentions were good, but Sun fucked it up.
Yeah. It only ended up as one of the main development languages for server software, and got used everywhere from mainframes, applets, app servers, and client applications. Its use as the primary language for the enormous mobile device market (especially games) and its presence on millions of mobile phones is obviously nothing but a minor passing fad. All the colleges teaching Java as a key skill for students must be informed!
Its such a disaster.... and its shocking no-one else has noticed. Good thing you did!
I eagerly await your next few posts:
"Windows - Microsoft's failed idea"
"The web - it may be everwhere but it's doomed"
"Intel - the chip that no-one used."
On a slightly differeny note I do really get tired of all the Sun bashing that goes on.
Oh come on. Being nothing more than a pioneering and innovative open systems company with decades of experience, if Sun refuses to listen to the wise suggestions of thousands of open source zealots, many of whom have at least a year or two of software experience, what can it expect?
Sun has always been a hardware company, and so they have always had control over the hardware their OS runs on.
No. Sun is also a software company, hence Solaris, and the earlier SunOS. The Unixes have always run on a range of architectures: In the 80s they were offering SunOS on Sparc, Motorola and Intel processors.
They have never yet had to deal with the tangled mess that is PC hardware, with conflicts, obscure devices, and all the rest of it.
Well, yes, they have. They have been producing Solaris x86 for a while now.
And the one thing missing most from the Sun Java desktop is Java
Eh? There are lots of Java apps and utilities on Java Desktop, and, of course, Java itself! Pre-installed, and integrated with Open Office and Mozilla.
Java is dead
.Net must be dead, and Visual Basic and C++ as well. If you scan the job markets, they are all less in demand than Java.
What are you smoking?
if Java is dead, then
And the future is multiprocessors...
Your "beyond 100" is nothing to a 512-way SGI Altix running Linux. There are 1024-way systems under development. Sun can't touch this.
Even the Altix doesn't touch this - its a supercluster system. The suggested configuration for individual nodes is pretty small - a max of 12 or 16 processors. You can specify up to 256 processors, but its unlikely to scale well unless you are using specialised application code.