JVM/Java memory footprint

Discussion in 'Java' started by alexcpn@gmail.com, Jan 29, 2007.

  1. Guest

    Hi,
    Though I have some three four years experience in C++ ,I did not have
    that much oppurtunity to work in Java. Currenly I was doing some
    analysis for a very simple CLI and was surprised to come with a memory
    restriction. I found that if I use Java for developing the CLI
    application I will be exhausting the memory of our Application Server
    (AS). Just to mention the architecture ,users (telecom operators) use
    a metaframe server client ( =something like remote desktop) to login
    to the AS and then open a GUI to work on it. With Java I can service
    only about 15 clients with the available memory . I just checked the
    reason for this resource crunch and found that already many Java based
    GUI's are served by the AS and each is taking some 25 MB or more. The
    first thing I thought is that adding more RAM will solve this ( though
    this is not an easy option) . Then I understood that a 32 bit system
    can have just about 3 gb ram for applciations and our AS had already 4
    GB with 1.5 GB VM also configured. Then I thought there might be ways
    to make the JVM shareable. But no. Also tried to use the flags to fine
    tune JVM. No go there also. Irrespective of how much you limit the
    actual limit is in the heap size allocated to the application and not
    to the private bytes of JVM.

    -----------------------------------------
    Data from Jconsole for the Java application

    Memory
    Current Heap Size 4.3 MB
    Max heap size 12. 2 MB ( set by flag)
    Committed memory 5 MB
    Operating System
    Committed VM - 26.7 MB

    Data from Perfmon (for java.exe)

    Private Bytes - 27.7 MB
    ----------------------------------------

    I am part of a large team that developes and maintains a telecom
    network managment software system. Also this software is used all
    around the globe almost in around 190 countries usually by telecom
    service providers. So this is a very real problem that I am speaking
    about. I am getting to love the simplicty of Java and of the great IDE
    Eclipse as much as I love the power of STL; but I am comming face to
    face with the main constraint of Java, its memory footprint and I
    guess thousands of others like me must have faced similar problems.

    If so the next question is what is SUN doing about this. I could not
    find it in their top 25 RFE's or top 25 bugs. (http://bugs.sun.com/
    bugdatabase/top25_bugs.do http://bugs.sun.com/bugdatabase/
    top25_rfes.do )

    1) Is this really not a problem in the outside world then ?

    2) Or is it that with the introduction of 64 bit HW and serveres this
    will be of no significance ?

    Anyway I doubt if companies like ours will adopt 64 bit HW and servers
    because of the costs involved in it ( note this is just a logical
    guess, I have no experience in such decisions)

    Anyway I am now forced to use C++ for the client and use Java on the
    server side. Fortunately most of the Java in the server side is done
    up as EJB's all conatined in the JBoss server thus consuming only one
    JVM there)
    -------
    I would like to get some comments on questions 1 and 2.

    Thanks
    Alex.C.P
     
    , Jan 29, 2007
    #1
    1. Advertising

  2. Daniel Pitts Guest

    On Jan 29, 6:42 am, wrote:
    > Hi,
    > Though I have some three four years experience in C++ ,I did not have
    > that much oppurtunity to work in Java. Currenly I was doing some
    > analysis for a very simple CLI and was surprised to come with a memory
    > restriction. I found that if I use Java for developing the CLI
    > application I will be exhausting the memory of our Application Server
    > (AS). Just to mention the architecture ,users (telecom operators) use
    > a metaframe server client ( =something like remote desktop) to login
    > to the AS and then open a GUI to work on it. With Java I can service
    > only about 15 clients with the available memory . I just checked the
    > reason for this resource crunch and found that already many Java based
    > GUI's are served by the AS and each is taking some 25 MB or more. The
    > first thing I thought is that adding more RAM will solve this ( though
    > this is not an easy option) . Then I understood that a 32 bit system
    > can have just about 3 gb ram for applciations and our AS had already 4
    > GB with 1.5 GB VM also configured. Then I thought there might be ways
    > to make the JVM shareable. But no. Also tried to use the flags to fine
    > tune JVM. No go there also. Irrespective of how much you limit the
    > actual limit is in the heap size allocated to the application and not
    > to the private bytes of JVM.
    >
    > -----------------------------------------
    > Data from Jconsole for the Java application
    >
    > Memory
    > Current Heap Size 4.3 MB
    > Max heap size 12. 2 MB ( set by flag)
    > Committed memory 5 MB
    > Operating System
    > Committed VM - 26.7 MB
    >
    > Data from Perfmon (for java.exe)
    >
    > Private Bytes - 27.7 MB
    > ----------------------------------------
    >
    > I am part of a large team that developes and maintains a telecom
    > network managment software system. Also this software is used all
    > around the globe almost in around 190 countries usually by telecom
    > service providers. So this is a very real problem that I am speaking
    > about. I am getting to love the simplicty of Java and of the great IDE
    > Eclipse as much as I love the power of STL; but I am comming face to
    > face with the main constraint of Java, its memory footprint and I
    > guess thousands of others like me must have faced similar problems.
    >
    > If so the next question is what is SUN doing about this. I could not
    > find it in their top 25 RFE's or top 25 bugs. (http://bugs.sun.com/
    > bugdatabase/top25_bugs.dohttp://bugs.sun.com/bugdatabase/
    > top25_rfes.do )
    >
    > 1) Is this really not a problem in the outside world then ?
    >
    > 2) Or is it that with the introduction of 64 bit HW and serveres this
    > will be of no significance ?
    >
    > Anyway I doubt if companies like ours will adopt 64 bit HW and servers
    > because of the costs involved in it ( note this is just a logical
    > guess, I have no experience in such decisions)
    >
    > Anyway I am now forced to use C++ for the client and use Java on the
    > server side. Fortunately most of the Java in the server side is done
    > up as EJB's all conatined in the JBoss server thus consuming only one
    > JVM there)
    > -------
    > I would like to get some comments on questions 1 and 2.
    >
    > Thanks
    > Alex.C.P


    Hmm, I'm not sure I understand your setup, but it appearch you have a
    single server machine which runs both the Java server process and the
    multiple client processes?

    Usually, Java clients are run on seperate client machines, so the
    number of clients doesn't effect the footprint.

    Anyway, to answer your questions, I've never found this to be a
    problem in any of our environments, so perhaps there is some problem
    with your architecture/design/configuration. I think that this problem
    doesn't have anything to do with the 32 vs 64bit platform, but perhaps
    a misunderstanding somewhere.
    Java does not have a tiny footprint, but last I checked, 2048M(2gig)/
    32Mb = 64. That means that you'd be able to run around 64 seperate JVM
    processes on one machine. This isn't a typical situation, and it still
    doesn't seem to fit your conclusions.

    Maybe it would help my understanding if I know what a "CLI" was. I
    only know it as Clear Interrupt Flag :)

    Hope this helps,
    Daniel.
     
    Daniel Pitts, Jan 29, 2007
    #2
    1. Advertising

  3. Guest

    On Jan 30, 2:56 am, "Daniel Pitts" <>
    wrote:
    > On Jan 29, 6:42 am, wrote:
    >
    >
    >
    > > Hi,
    > > Though I have some three four years experience in C++ ,I did not have
    > > that much oppurtunity to work in Java. Currenly I was doing some
    > > analysis for a very simple CLI and was surprised to come with a memory
    > > restriction. I found that if I use Java for developing the CLI
    > > application I will be exhausting the memory of our Application Server
    > > (AS). Just to mention the architecture ,users (telecom operators) use
    > > a metaframe server client ( =something like remote desktop) to login
    > > to the AS and then open a GUI to work on it. With Java I can service
    > > only about 15 clients with the available memory . I just checked the
    > > reason for this resource crunch and found that already many Java based
    > > GUI's are served by the AS and each is taking some 25 MB or more. The
    > > first thing I thought is that adding more RAM will solve this ( though
    > > this is not an easy option) . Then I understood that a 32 bit system
    > > can have just about 3 gb ram for applciations and our AS had already 4
    > > GB with 1.5 GB VM also configured. Then I thought there might be ways
    > > to make the JVM shareable. But no. Also tried to use the flags to fine
    > > tune JVM. No go there also. Irrespective of how much you limit the
    > > actual limit is in the heap size allocated to the application and not
    > > to the private bytes of JVM.

    >
    > > -----------------------------------------
    > > Data from Jconsole for the Java application

    >
    > > Memory
    > > Current Heap Size 4.3 MB
    > > Max heap size 12. 2 MB ( set by flag)
    > > Committed memory 5 MB
    > > Operating System
    > > Committed VM - 26.7 MB

    >
    > > Data from Perfmon (for java.exe)

    >
    > > Private Bytes - 27.7 MB
    > > ----------------------------------------

    >
    > > I am part of a large team that developes and maintains a telecom
    > > network managment software system. Also this software is used all
    > > around the globe almost in around 190 countries usually by telecom
    > > service providers. So this is a very real problem that I am speaking
    > > about. I am getting to love the simplicty of Java and of the great IDE
    > > Eclipse as much as I love the power of STL; but I am comming face to
    > > face with the main constraint of Java, its memory footprint and I
    > > guess thousands of others like me must have faced similar problems.

    >
    > > If so the next question is what is SUN doing about this. I could not
    > > find it in their top 25 RFE's or top 25 bugs. (http://bugs.sun.com/
    > > bugdatabase/top25_bugs.dohttp://bugs.sun.com/bugdatabase/
    > > top25_rfes.do )

    >
    > > 1) Is this really not a problem in the outside world then ?

    >
    > > 2) Or is it that with the introduction of 64 bit HW and serveres this
    > > will be of no significance ?

    >
    > > Anyway I doubt if companies like ours will adopt 64 bit HW and servers
    > > because of the costs involved in it ( note this is just a logical
    > > guess, I have no experience in such decisions)

    >
    > > Anyway I am now forced to use C++ for the client and use Java on the
    > > server side. Fortunately most of the Java in the server side is done
    > > up as EJB's all conatined in the JBoss server thus consuming only one
    > > JVM there)
    > > -------
    > > I would like to get some comments on questions 1 and 2.

    >
    > > Thanks
    > > Alex.C.P

    >
    > Hmm, I'm not sure I understand your setup, but it appearch you have a
    > single server machine which runs both the Java server process and the
    > multiple client processes?
    >
    > Usually, Java clients are run on seperate client machines, so the
    > number of clients doesn't effect the footprint.
    >
    > Anyway, to answer your questions, I've never found this to be a
    > problem in any of our environments, so perhaps there is some problem
    > with your architecture/design/configuration. I think that this problem
    > doesn't have anything to do with the 32 vs 64bit platform, but perhaps
    > a misunderstanding somewhere.
    > Java does not have a tiny footprint, but last I checked, 2048M(2gig)/
    > 32Mb = 64. That means that you'd be able to run around 64 seperate JVM
    > processes on one machine. This isn't a typical situation, and it still
    > doesn't seem to fit your conclusions.
    >
    > Maybe it would help my understanding if I know what a "CLI" was. I
    > only know it as Clear Interrupt Flag :)
    >
    > Hope this helps,
    > Daniel.


    Hi, Let me clarify - CLI just means the simple command line
    interface :) .Well
    the architecture is like this. We have a Application Server (AS) ,
    which is also a metaframe server. Many operators with 'dumb' terminals
    can establish a seession with the AS using a metaframe client. (say
    like using a glorified version of telnet, or remote desktop or RDP).
    Each time a new operator connects and invokes a java GUI or CLI a new
    instance of Java.exe is started on the AS. Now there is another server
    (say BS) which servers the AS. The business logic of the java clients
    in the AS is served by the EJB components in the BS. Since all these
    are developed as EJB's they run in JBoss container which needs only
    single JVM (Java.exe). Also already there are 17 JVM's already running
    in the AS consuming about 710 MB on avearge .Hope this clarifies

    Regards
    Alex.C.P
     
    , Jan 30, 2007
    #3
  4. Daniel Pitts Guest

    On Jan 29, 9:40 pm, wrote:
    > On Jan 30, 2:56 am, "Daniel Pitts" <>
    > wrote:
    >
    >
    >
    > > On Jan 29, 6:42 am, wrote:

    >
    > > > Hi,
    > > > Though I have some three four years experience in C++ ,I did not have
    > > > that much oppurtunity to work in Java. Currenly I was doing some
    > > > analysis for a very simple CLI and was surprised to come with a memory
    > > > restriction. I found that if I use Java for developing the CLI
    > > > application I will be exhausting the memory of our Application Server
    > > > (AS). Just to mention the architecture ,users (telecom operators) use
    > > > a metaframe server client ( =something like remote desktop) to login
    > > > to the AS and then open a GUI to work on it. With Java I can service
    > > > only about 15 clients with the available memory . I just checked the
    > > > reason for this resource crunch and found that already many Java based
    > > > GUI's are served by the AS and each is taking some 25 MB or more. The
    > > > first thing I thought is that adding more RAM will solve this ( though
    > > > this is not an easy option) . Then I understood that a 32 bit system
    > > > can have just about 3 gb ram for applciations and our AS had already 4
    > > > GB with 1.5 GB VM also configured. Then I thought there might be ways
    > > > to make the JVM shareable. But no. Also tried to use the flags to fine
    > > > tune JVM. No go there also. Irrespective of how much you limit the
    > > > actual limit is in the heap size allocated to the application and not
    > > > to the private bytes of JVM.

    >
    > > > -----------------------------------------
    > > > Data from Jconsole for the Java application

    >
    > > > Memory
    > > > Current Heap Size 4.3 MB
    > > > Max heap size 12. 2 MB ( set by flag)
    > > > Committed memory 5 MB
    > > > Operating System
    > > > Committed VM - 26.7 MB

    >
    > > > Data from Perfmon (for java.exe)

    >
    > > > Private Bytes - 27.7 MB
    > > > ----------------------------------------

    >
    > > > I am part of a large team that developes and maintains a telecom
    > > > network managment software system. Also this software is used all
    > > > around the globe almost in around 190 countries usually by telecom
    > > > service providers. So this is a very real problem that I am speaking
    > > > about. I am getting to love the simplicty of Java and of the great IDE
    > > > Eclipse as much as I love the power of STL; but I am comming face to
    > > > face with the main constraint of Java, its memory footprint and I
    > > > guess thousands of others like me must have faced similar problems.

    >
    > > > If so the next question is what is SUN doing about this. I could not
    > > > find it in their top 25 RFE's or top 25 bugs. (http://bugs.sun.com/
    > > > bugdatabase/top25_bugs.dohttp://bugs.sun.com/bugdatabase/
    > > > top25_rfes.do )

    >
    > > > 1) Is this really not a problem in the outside world then ?

    >
    > > > 2) Or is it that with the introduction of 64 bit HW and serveres this
    > > > will be of no significance ?

    >
    > > > Anyway I doubt if companies like ours will adopt 64 bit HW and servers
    > > > because of the costs involved in it ( note this is just a logical
    > > > guess, I have no experience in such decisions)

    >
    > > > Anyway I am now forced to use C++ for the client and use Java on the
    > > > server side. Fortunately most of the Java in the server side is done
    > > > up as EJB's all conatined in the JBoss server thus consuming only one
    > > > JVM there)
    > > > -------
    > > > I would like to get some comments on questions 1 and 2.

    >
    > > > Thanks
    > > > Alex.C.P

    >
    > > Hmm, I'm not sure I understand your setup, but it appearch you have a
    > > single server machine which runs both the Java server process and the
    > > multiple client processes?

    >
    > > Usually, Java clients are run on seperate client machines, so the
    > > number of clients doesn't effect the footprint.

    >
    > > Anyway, to answer your questions, I've never found this to be a
    > > problem in any of our environments, so perhaps there is some problem
    > > with your architecture/design/configuration. I think that this problem
    > > doesn't have anything to do with the 32 vs 64bit platform, but perhaps
    > > a misunderstanding somewhere.
    > > Java does not have a tiny footprint, but last I checked, 2048M(2gig)/
    > > 32Mb = 64. That means that you'd be able to run around 64 seperate JVM
    > > processes on one machine. This isn't a typical situation, and it still
    > > doesn't seem to fit your conclusions.

    >
    > > Maybe it would help my understanding if I know what a "CLI" was. I
    > > only know it as Clear Interrupt Flag :)

    >
    > > Hope this helps,
    > > Daniel.

    >
    > Hi, Let me clarify - CLI just means the simple command line
    > interface :) .Well
    > the architecture is like this. We have a Application Server (AS) ,
    > which is also a metaframe server. Many operators with 'dumb' terminals
    > can establish a seession with the AS using a metaframe client. (say
    > like using a glorified version of telnet, or remote desktop or RDP).
    > Each time a new operator connects and invokes a java GUI or CLI a new
    > instance of Java.exe is started on the AS. Now there is another server
    > (say BS) which servers the AS. The business logic of the java clients
    > in the AS is served by the EJB components in the BS. Since all these
    > are developed as EJB's they run in JBoss container which needs only
    > single JVM (Java.exe). Also already there are 17 JVM's already running
    > in the AS consuming about 710 MB on avearge .Hope this clarifies
    >
    > Regards
    > Alex.C.P



    Well, that does help explain what you mean :)
    That isn't a typical use of Java technology. Usually you have only a
    handful of processes on one machine, either server processes or client
    processes. Rarely do you have both.
    Perhaps your glorified Telnet would be better implemented as a
    connection to the Java server, rather than a connect to the machine
    the server runs on. It really depends on how 'dumb' your terminals
    are. If they are capable of running a JVM themselves, then I would go
    that approach.

    Also note that there are different types of JVM's out there. At the
    very least, there is a Client JVM and a Server JVM that come
    standard. With Java 1.6, Sun is introducing a method that allows all
    of the JVMs to utilize shared memory for a portion of the footprint
    (standard class files, and probably some other static JVM information)

    Anyway, my suggestion is to try to move the client JVM process from
    application server to the client machine, or at try to find a way to
    have each client JVM process service more than one user.
     
    Daniel Pitts, Jan 30, 2007
    #4
  5. nukleus Guest

    In article <>, "Daniel
    Pitts" <> wrote:
    >
    >
    >On Jan 29, 6:42 am, wrote:
    >> Hi,
    >> Though I have some three four years experience in C++ ,I did not have
    >> that much oppurtunity to work in Java. Currenly I was doing some
    >> analysis for a very simple CLI and was surprised to come with a memory
    >> restriction. I found that if I use Java for developing the CLI
    >> application I will be exhausting the memory of our Application Server
    >> (AS). Just to mention the architecture ,users (telecom operators) use
    >> a metaframe server client ( =something like remote desktop) to login
    >> to the AS and then open a GUI to work on it. With Java I can service
    >> only about 15 clients with the available memory . I just checked the
    >> reason for this resource crunch and found that already many Java based
    >> GUI's are served by the AS and each is taking some 25 MB or more.


    I looked into this footprint issue as in my case, the amount of heap
    required could be very large and that is for each thread running under
    the same instance of an app. Just to run the simpliest application,
    JVM swallows tens of megabytes. And when you run the application that
    could be in a way similar to your situation, literally hundreds of
    megabytes are simply eaten up. I ended up inserting code to free up
    all the huge lists and buffers that are generated on the fly and
    not to preload large buffers with the data, say news article bodies,
    and postpone it to the very last moment, when this data is needed.
    Then, once it is used, I explicitly empty the buffers by removing
    the vector contents or equivalent thereof.

    Just to process a 25000 article archive, you could be loosing
    terabytes, and not even clear where, cause your entire archive
    is less than 50 megs. Where did all those terabytes go?
    I am still trying to figure it out.

    Seemed to help in that i see that the garbage collector is cleaning
    all this stuff and heap grows as lil as possible. But it places
    about major restriction on your code as you have to juggle things
    like a clown in circus, and have to let go of things that were
    already referenced and loaded, as you might want to do for
    some feature operations.

    I do not recall having ANY of these problems with
    all my C++ experience, and here, you have to be aware to the
    point that you have to make sure you declare your local variables
    in a very specific scope. Otherwise, you might be loosing huge
    amounts of memory if your main processing loop is not exited,
    and it might have to run till completion of some major operation
    that could be taking minutes if not hours to complete.

    I, personally, think that the footprint issue is one of MAJOR
    problems with Java's architecture and approach and have reasons
    to believe that it'll never be resolved, in principle.

    The whole idea of JVM is nothing different than the Pascal's
    P machine idea. People wanted to invest millions into the
    hardware chips that run the P machine code directly, and what
    happened at the end? Anybody heard of Pascal?
    Well, dead as it gets.

    And what is such a compelling need for the virtual machine?
    To make it MORE portable?
    But where did all this "portability" go,
    when you can't even build an app, working on the same
    platform and be assured that unless other co-developers,
    clients or customers, using your code, use the same exact
    compiler as you do, they won't be able to even reproduce
    your results unless they switch their entire development
    environment to that which YOU use.

    What kind of joke is this?

    WHO in his clear mind would even bother?
    Which customer, after double clicking on your executable
    and seeing the error message as to JVM, is even going to
    bother reading that error message?

    Who is going to go hunting for the JVM, being subject
    of megabux lawsuits?

    And Sun, Microsoft, Borland, and that is about 99 percent
    of the market share, do just about ANYTING and EVERYTHING
    imaginable on making Java as much incompatible as they
    can manage, and entrap you into their proprietary
    architectures to the point of unbelievable.

    And they hope to achieve what at the end?
    Java becoming more portable and more acceptable?
    How?

    Just look at the very development environment.
    Borland, for example, eats up over 400 megs,
    and you never know what is going to happen when you
    simply use your scroll bar ot look at one of the menus,
    without even ACTIVATING it. What I am seeing is they
    eat up gigabytes just when you select a different source
    file. So, you'd better have a few spair terabytes of
    RAM on your box. Otherwise, you'll be waiting for hours
    with all the swapping going on.

    And this is Borland, and not some shareware.
    Windows XP goes as far, as to require nothing less than
    XP SP2 to even INSTALL thei compiler and development
    environment.

    WHY?

    What for?

    Is it some kernel mode drivers that you need to run
    a compiler and editor? Why do you have to reboot the
    box after installing a compiled and development environment?
    Can you imagine what it means for some major operations?

    At least Borland installs. Thanks for that much.

    Then, on the top of it, once you upgrade to a new version,
    all of a sudden, your perfectly good code does not even
    compile. The same exact code that worked perfectly on their
    own previous version and that compiles on other compilers.

    Where is all that portabilty thing they are talking about?

    Not only that, but it is LITERALLY impossible to figure
    out why. Because you get an error message that says
    essentially: Error 2 + 2 is not equal to four.
    WHAT?

    And they give you an error, indicating that the package,
    declared in one of sorce file does not correspond to
    exactly the same source file in their own error message.
    Not only that, but even if it compiles after you remove
    some perfectly valid declarations, and try to run your
    app, it claims it can not find you MAIN class!!!

    What?

    But HOW did you compile it on the first place?
    WHERE do I find the thing I can change to make sure
    your own compiler and run time environment reconcile?
    Try to push the HELP button?
    Well, what I found out, that just about the LAST thing
    you'd want to do, is to push the help button, as, first
    of all, it'll eat about half of a terabyte of memory,
    and, after waiting for minutes, LITERALLY, you'll see
    the help screen, your eye brows may get covered with
    the snow, as you'd have to read about 50 pages of UTTERLY
    incomprehensible stuff, that has NOTHING to do with your
    situation, and even if you do so, you are not likely
    to get ANY answers at the end. Only more questions
    and more headache.

    Thanks to this group, at least in SOME cases, you can
    ask mere mortals if they know the answer to your simpliest
    problme of all, and, if you are lucky enough, and they
    are in the mood of doing so, they'll give you some
    scetchy answer, that is nothing more than a hint,
    and you'd have to go back to your puter and sweat it
    out to full extent. Kinda sadomasochistict trip.

    And what CAN you expect from those people?
    Not only that, but you will be insulted and rediculed,
    told that in Java, the names of routines MUST be in
    lower case, even though Sun itself has some things
    that start in upper case and some don't.

    Now, what WHAT specific REASON do you have to make
    all the method names in lower case?
    Does it make your code MORE readable?
    - Nope, just the other way around.

    Does it prevent your compiler to compile this thing?
    - Not even in the picture.

    Does it make MORE clear what are you trying to do?
    - Are you kidding?

    So...

    What IS that dictate good for?

    Just read this very group.
    You'd be amazed if you are new to java.

    Those poor "clueless", when they are in just about the
    biggest problem they can imagine at the moment,
    instead of getting assistance,
    get a brainwashing dictates on what you MUST do!

    You see?

    MUST!

    And unless you do so, you are considered of an
    "inferior race"!
    Nothing more.
    Nothing less...

    And they'll be laughing at you like you are some kind
    of idiot.

    And why?

    Can ANY of those "experts" explain WHY is it that
    you MUST use lower case letters in just about ANY
    situation imaginable?

    I, personally prefer to capitalize each part of
    a name and use long names that are self descriptive
    by simply looking at them. Because when you work on
    things and have to juggle about 10 screens if you
    are lucky, you have a breakpoint hit, and need some
    information from the help subusystem to see what
    and why, you have VERY little chance to follow some
    dictates on how YOU name things. You need to be able
    to VERY quickly go through very large amounts of source
    code and by just glancing at ANY name for few millisconds,
    you should be able to understand that is that thing,
    what it does and why is it there on the first place.

    By REQUIRING to use lower case letters in the INITIAL
    section of the name, you simply make them schitsophrenic,
    as the mind have to constantly switch form one thought
    template to another, in millisecond time frames!

    It has to be able to distinguish the first part of the
    word and use a different template to recognize it,
    from the REST of your name, which MUST be capitalized.

    What is this?

    A mad house?

    Are you, people, serious?

    What IS the REASON behind all this insanity?

    Well, somebody in the stone ages, at the very beginning
    of java trip, decided that this is the way to go.
    There was EXACTLY the same trip with Microsoft,
    when they invented a system of naming things,
    a system that allows you from the very name
    to see what kind of object or variable you are using,
    down to the variable type!

    Well, at least THAT made some sense,
    even though I am not sure what is the CURRENT hype
    on the most innoncent thing in the world imaginable.

    And here, you may not even get a reply on your
    question, just because you show these people
    some piece of code that does not fit THEY view
    on Universe and a life force as such.

    Wake up, people.

    WHAT are you doing here on the first place?
    Why do you even bother to answer questions?
    You see, if I am an expert, built according to YOUR
    templates, I don't need your help. I can teach YOU,
    and teach you using REASON, and not some dictates
    as to how the planet Earth is to turn and where
    is the Sun to rise in the morning.

    Is it clear enough?

    I spent DAYS and still can not figure out what is the
    problem on the first place. Just try to push the help
    button. First of all, it'll eat up a terabyte of memory
    and you'd be sitting there mezmerized, waithing for it
    to stop swapping like mad and load all the stuff,
    that is the simpliest thing in the world, just to
    display a single help screen, which shouldn't take
    more than a few milliseconds nowadays.

    I am sitting here with a perfectly good computer
    with the amount of memory enough to load the Universe,
    at least from the virtual memory standpoint,
    and about 90 percent of my time is spend on
    waiting for a stupid help screen to load
    or by merely switching some tab on development
    user interface, before I even DO some action!
    Just to switch to another source file,
    you need to swap the Universe?

    I wonder how many terabytes will it take
    just to start up the next version of their
    development environment.

    Not only that, but when you upgrade to a new version
    and start it up, you are like in a zombie land.
    There is just nothing left of what you know
    and what worked perfectly well in the last version
    of their own product.
    Now, just to create a hello project, displaying a blank
    screen, you'd have to get a PHD at nothing less than
    MIT or major universities of that grade.

    You'll have to spend a couple of years fully adjusting
    to their news version of the Universe, where they
    added another 5 levels of indirection and abstraction
    just to make it more "universal" and more "powerful".

    How many times do you have to reinvent the wheel
    before you realize that you are nothing more than a fool?

    Now, just as you eventually get used to their new
    environment, surely enough, it is outdated and you'd
    have to shelve out kilobux to upgdate to,
    and the whole story of the wheel of Karma
    as told by the Idiot, is to make its next turn.

    Do you, people, have time for all these "improvements"?

    Why is it that it considered nearly illegal,
    to use the AWT package, if you don't even want to have
    anything fancier than that?

    What I have done with AWT, is something I am not sure
    they can do with all their latest and gratest version
    of a super-sex machine. My user interface is designed
    in such a way, that the user does not even realize he
    is using it. It is so natural and so intuitive, that
    you must be UTTERLY clueless to get lost in the mazes
    of parameters giving you so much power that it'll take
    generations to fully appreciate it.

    You can have fields of 3 miles long if you want to,
    which is exactly the case with usenet articls and
    various headers. You can edit them, maximize or shrink
    the dialogs, resize just about anything you can imagine,
    and have just about ALL your diagogs and windows
    available SIMULTANEOUSLY. You want it? You use it.
    You don't want it, no problem, the dafaults make it
    perfectly comfortable.

    At ANY given junction, just hit F1 key, and you'll get
    the EXACT and detailed TO THE POINT help on just about
    ANYTHING under the Sun, and it does not have to swap
    a few terabytes to do so, and you don't have to wait
    for half an hour just because you pushed the "wrong"
    button.

    Why do I need all this super-sex?

    My code should compile on ANY half decent java system
    on ANY platform, going back to the beginning of Java time.
    Even after a generation, it will run like a champ,
    unless those software "kings" totally negate their own
    version of the Universe.

    Why do you think Bill Gates has become what he is?

    Well, because of that lil stupid bios thing that
    sits in your box, and the principles of backward
    compatibility.

    Anyone here ever heard of Zilog, which is one of the oldest
    CPU manufacturers, whose CPU was compatible with intel,
    and only was more powerful.

    What happened to them?

    Well, not even sure there is a trace of them left.
    Why?

    Because with every new version of their CPU,
    they would completely change the instruction set,
    which means that you'd have to throw away all your
    perfectly working copies if you want to get a marginal
    improvement on your performance.

    What happened to Motorola, another major major
    CPU manufacturer?

    The same thing. New version = new instruction set.
    Throw away all you have and shell out a few kilobux
    if you want a 10 percent improvement.

    I am simply amazed Mac is still in business.
    With every new version of a Motorola CPU,
    they'd have to fire all their programmers
    and hire the new ones, cause that version
    was utterly incompatible with a new one.
    You'd have to rewrite the KERNEL itself.
    Nothing less, nothing more.
    Oh, sure, some claim that it is only the compiler
    back and that has to be scrapped.

    But how many gazillions upon gazillons of
    resourses does it waste?

    It is like trying to run,
    and, at the same time, tying yourself with
    a rope to the wall.

    What kinda "progress" are you going to make,
    if you have to return to the exact same place
    in that Wheel Of Karma, just another resolution?

    Do you think you all have time for that?

    Sorry to tell you, but your rears is on fire.
    Haven't you noticed?

    WHY do I have to scrap the AWT, which is what Java
    was to a large extent on the first place?

    Why do they refuse to recognized their own old code,
    that is at the very core of their entire product line?

    See?

    Bill gates can run some programs that are sitting
    in my archive boxes that ran on 8088 CPU,
    more than a generation ago,
    and some of those programs, and even the news reader
    i am using thise very moment, are written so long
    ago, that Sun and all those pure blooded "experts",
    would consider the stone age technology.

    Well...

    May be.

    But you know what?

    THEY RUN.

    And I have seen just about ALL your fanciest gadgets,
    and just about ALL their versions, and STILL I am using
    this newsreader, and, after trying just about anything
    imaginable, still find it about the most comfortable
    thing to work with, even if it crashes on me once in a while,
    even after i have to resize some windows after bringing
    the program up because, for some strange reason, it
    forgets the last session settings in some instances.

    Because it is like a glove in my hand.
    About the easiest, the most unobtrusive,
    the most functional, the most intutivive,
    the most flexible gadget I've ever heard of in terms
    of newsreaders.

    And it will run for generations to come.

    THAT is real progress.
    THAT is technology that is even worth mentioning.
    THAT is the way to go.

    Now...

    I am being laughed at by all these "experts" around here
    because I told them: No javax stuff. AWT is fine for what
    I am doing. Not even interested. Because if you use Borland
    development environment, you have support of some things,
    but when you use the same perfectly working code, say
    for example with the biggest and baddest of them all,
    they do not even recognized ANYTHING beyond AWT, and
    that is not some old and outdated version.

    See?

    Now, my code compiles on just about the oldest bullock
    card and functionality and gui look probably more flexible
    and more powerful than all you fancy gadgets and supersex.

    THAT is what makes a difference between the master
    and a bio-robot, brainwashed to literal oblivion,
    and accepthing just about ANY template they are programmed
    with as the ultimate reality, of which they haven't
    a SLIGHTLEST clue.

    Good enough?

    Now...
    Back to original subject,
    and don't you please try to be "holier" than me,
    blaming me for all the fattest posts you can imagine.
    Because I am not writing for you alone,
    regadless of what kinda "big fish" you think you are.
    Seen fishes of ALL sizes.

    I am writing this for generations to come
    and they'll be able to find this information
    with a single mouse click, and it may save them
    years if not generations as they do not have to
    make the same mistakes again, as they will have
    even less time that you all have. They'll have
    nanoseconds to make major decistions affecting
    the life force as such, and the last thing they
    have time for is the lectures and dictates
    about name capitalization principles, senile aregument
    as it gets. Uttery rigid and inflexible. Utterly dictatorial,
    and a byproduct of constant and unending brainwashing
    to accept the LOWEST common denominator
    as a word of God.

    Nothing less.
    Nothing more...

    Enjoy.

    Now...

    And once you get to that help screen,
    your ear drums will be slightly popping
    as you'll be given 50 pages of UTTERLY incomprehensible
    material and explanations about those transcontinental,
    fully distributed, database supported, utterly abstract
    and super prototyped, globally distributed and
    55 layers deep structures, architectures and designs,
    would be referring to yet anothor thing of the same
    maginitude.

    But, to get an answer on the simpliest things
    in the world, such as how do i get the keyboard events,
    you'd have to graduate from Java Academy and get
    youself a phd, wasting years of your life, trying
    to comprehend just about the simplies things imaginable.

    Just about the ONLY thing I have seen so far with java,
    is that unless you use a 100 percend C code syntax,
    you are in trouble and in a major way. And that goes
    to just about every single architectural element in Java.

    Just try to use ascync sockets in a dedicated socket
    thread. Maaaan, you'd be having a headache for monts,
    trying to figure out the simpliest things possible.
    Never even HEARD of such problems.
    If you push help button, they'll be telling you
    about abstract virtual, supercontinental, globally
    distributed, super prototyped things that you can't
    even comprehend as they seem to have NOTHING to do
    with what you are trying to do on the first place?

    Have you seen their latest super-sophisticate version
    of Sockets, that exist since the time of Unix?
    Well, just don't push that help button,
    as your hair may raise seeing all these "superstructures",
    containg about 5 levels deep of other superstructures,
    just to do abut the simpliest thing imaginable.

    But yes, they are fully and globally distributed
    and you could asses some file, that is sitting on
    your book shelf on the first place, from as far,
    as a planet Mars?

    And for what?

    What if i could care less about those globally distributed
    databases on the first place?

    Could I simply create a socket object in a single line
    of code and not worry about syper-transcontinental
    giga-sucking pipe to the Internet?

    Why do I have to use Boing 747 to go to the store and
    get myself some food?

    You, so called software architects, that I wouldn't
    even let CLOSE to, cause it is nothing more and nothing
    less than a PURE grade headeache at the end, and it
    would take you generations on how to explain to me
    how to push a tab on your super-sophisticated GUI,
    that, I, of an "inferior" race, kind and creed,
    can not simply even BEGIN to comprehend the power of.

    But you know what?
    My stuff is gonna work
    after you are long gone,
    unless Sun, Microsoft, Boralnd and other manufacturers
    SPECIFICALLY and EXPLICITLY sabotage things
    and force you to forever shell out kilobux
    just to get that marginal 10 percent improvement,
    and if you are lucky. Because their bloatware
    will crash your puter when you try to start a simple
    editor, just to find a few words and replace them
    with othe words.

    And bloatware is a compliment,
    as it is nothing less
    and nothing more than a monsterware,
    where these sadists torture you once you start
    their program.

    Has ANYBODY tried to install windows?

    Have you noticed what is going on?

    Well, from the time you push that setup button,
    your computer, that you paid the money for,
    is not yours any more.

    You won't be able to cancel anything if you
    see something funky going on.
    You won't be able to suspend the operation
    and let you have a chance to look at your file system.
    You won't be able to even get help or information,
    as your full screen is taken by their setup program.
    You can't even REBOOT, unless you hit the power off
    key, and if you do that, you may loose half of your
    disk and totally devastate your system.

    Just because you pushed the "wrong" button.
    Willy will be switching your screen on and off,
    rebooting the system several times,
    asking you the questions you might not even heard of,
    and forcing you to do things that do not have to be
    done in the middle of setup, when you ALRREADY lost
    your operating system and your file system is nothing
    more than a mess of various state snapshots.

    ANd there is NOTING you can do about it.

    What do you call this "advanced" software?

    I call it sadistm.
    LITERALLY.

    You have no power of your own computer,
    and for HOURS, and pray the lord that setup
    completes, as, if by ANY chance, there is the most
    minite and most insignificant and inappropriate
    error somewhere, he'll simply freeze your screen,
    and FORCE you to reboot. He won't even give you
    a chance too look at your own file system and,
    quite possibly, get some stupid and utterly insignificant
    driver from your old version, that hasn't been
    updated for about a generation on the first place.

    Nope. You can't do that.
    Why?
    Because microsucks considers it "wrong".
    Have any of you tried to recover from the situation
    of this kind?

    Maaan. just about the BEST approach, by FAR,
    is to completely forget about trying to recover.
    The best way is to simply try to boot from another
    copy of operating system, or even go as far,
    as to use the boot floppies. Then copy all the
    files of any significance to a fifferent partition,
    if you didn't make a backup on the first place,
    which is a MUST if you are dealing with microsux.
    Not even that, but I made a backup of the entire
    partition, and when things went utterly berserk,
    i tried to restore the previous version of the
    operationg system, just to realize that some key
    files and drivers were, for some magic reason,
    abscent in the backup, and that is TOTAL devastation.
    You'll have to spend days if not weeks
    just to recover your whole world.

    Is THIS the way to design and architect software
    and any ware imaginable?

    Why is it taking 10 times less and 100 times simplier
    on that completely free Linux, that is considered to
    be an "old technology". They don't even use C++ in
    most of their packages last time i looked at some.

    And the systems is 10 times more reliable, faster
    leaner and smaller.

    With all those terabux and tens of thousands of programmers,
    you couln't create anything better?

    Just wait till you get the new version of Windows.
    You'd better sell your house to purchase yourself
    a 4 terabyte memory, even if you use a 1k program.

    What I heard about it, is nothing less than virtual
    madness. Not sure if you can run that thing with
    the biggest and baddest of them all, mainframes,
    with a super cooled CPU and a special service
    room just to keep all that cooling equipment in.

    Is THAT the "progress" we are talking about here?

    Progress of WHAT?

    About the fastest one way road to hell?

    Is there ANY operation in Java, beyond declaration
    of the simpliest data types there are, that does not
    use at least 5 layers of abstraction?

    What IS it?

    Not only that, but you can not even rely on the
    standardest and oldest concepts such as arrays and
    vectors and be assured that you have a random access
    to the elements, as that is all "implementation dependent",
    which means that vector you think is nothing more than
    a simple array of objects, may be implemented as a linked
    list, and there is no guarrantee whatsoever on what
    implementation is going to implement it in the way
    it OUGHT to be implemented?

    I had to rewrite MAJOR sections of code because of this
    single thing and, doing it in the development environment,
    is nothing less than an equivalent of disaster,
    as you may indadvertently introduce bugs, you could have
    never imagined possible, even in principles.

    So...

    If you are working on complex enough application,
    you'd have to retest thousands if not tens of thousands
    of various cases, and some of that stuff isn't even
    a subject to automated testing. You'd have to spend a double
    amount of development time, trying to describe all possible
    error conditions, starting with completely wrong user entries
    in hundreds of different parameter fields, choices, etc.,
    going down to network errors, disconnections, files not
    present, corrupt data and on and on and on.

    Where is that paradise they were promising to all?
    Where is that portability, declared left and right
    on just about every corner in your favorite downtown?

    Basically, the more I look at it, the more it seems
    that Java has no future, at least for a generation.
    With the amount of hassles, inter-continental law suits
    on just about the biggest level imaginable, and the very
    JVM architecture does not look like ANYTHING new under Sun,
    and, unless a generation of time passes, and JVM hardware
    chips are wired into your box right off the shelf,
    if people would even remember what that Java thing was
    on the first place, I just do not see it becoming a
    widely acceptable technology that has a stable enough
    foundation to survive the challanges forthcoming.

    It is about the most convoluted design I have ever seen
    with just about as many levels of abstraction one could
    imagine in his wildest dreams, even while doing the simpliest
    things in the world.

    Just out of curiocity, I wonder what would happen,
    if you managed to get a handle on the object and examine
    some method with a kernel level debugger.
    How many millions of references upon references to the
    structures, pointing to other structure does it take
    to add 2 and 2, which takes a single clock cycle
    of your CPU?

    How many terabytes have to be swapped in and out to do so?

    One day I'll do just that and report the results of it,
    just out of curiosity. Because what I am seeing from the
    day one of my Java "encounter", is that it is probably
    wasting 1000 times more resources in terms of memory
    footprints, performance and on and on and on. I just
    see it in the way the very development environment works.

    It crawls like a turtle where it shold be flying like a bird,
    or even sputnik, nowadays. After literally generations of
    development of all sorts of languages and architectures,
    THIS is what we ended up with?

    How many generations will it take for this thing to become
    as a glove, you put on your hand?

    Anybody has a guess?

    To me, this (footprint) is one of the major problems with Java
    and the whole idea of JVM. I have noticed looking at source
    published by various people that they explicitly import only
    those packages that are needed for their app.

    So, instead of saying "import java.awt",
    they'll explicitly call for some sub-package, and that is just
    about the LAST thing I would like to worry about.

    That seemed to be strange, but i am SURE there IS a reason
    why they are doing so. I am not even sure that if you have
    some unused code in your program, that eventually athrophied,
    that thing would not be compiled in and hang as a piece of
    dead wood in your app, eating resources for no reason whatsoever.

    I have seen people here asking help on trying to get a listing
    of all their unused code. For what? That stuff shouldn't
    even be compiled in on the first place. No need to even
    worry about. The reason you have some unused code is because
    you have switched to a new approach, and, having that
    unused code in your source files may have ten times more
    benefits than it creates problems, because it shows you
    in no uncertain terms, what was the previous design and
    allows you to tweak your new design and avoid all the
    pitfalls you might fall into, not realizing there are some
    tricky situations that there were handled by the old version.

    And i see it left and right.
    I have a special religious ceremony,
    where i go to all the source files
    and remove all the code that is commented out in BULK,
    the entire methods need be. Some of it is obviously
    outdated and after seeing the new version run stably
    for sufficient amount of time, you can painlessly remove
    all the code that is commented out, and that code,
    in most cases, is your golden treasure, as you can see
    the most minute permutation in your own thought process
    and you can see how things could be simplified even futher,
    making it clearer, more efficiend and more universal.

    And this kind of stuff you can expect with just about
    every single step in your development process, going down
    to expecting a reasonable behavior when user pushes some
    button, checkbox, or enters text in one of the fields.

    For example, in MY case, there is a need for a user to
    be able to save his current configuration at ANY moment,
    regardless of what frame is he working with at the moment.

    In fact, he can have ALL the frames open at the same time
    to see various parameters, such as files, complex filters,
    global parameters, etc. So... He can even go as far,
    as starting some major operation, and, in real time,
    open up as many parameter frames and even change those
    parameters on the fly, while current operation is in progress,
    which isn't the easiest thing to manage on the first place.

    He can change some parameters that in real time,
    the next object processed would have those parameters
    so there is no need to restart some major operation,
    which is probably the trickiest thing there is,
    especially considering that he can fire up multiple
    treads that do different jobs. Sure, some of it is
    not possible in principle, but quite a lot of it
    is doable, giving the user tremendous power, flexibility,
    ease of use and the most intuitive user interface
    imaginable.

    THAT is what I call architecture.
    THAT is what your stinky piece of software for
    on the first place.

    TO FACILITATE THE USER,
    and not raised the tripple armored walls around him
    and FORCE him to do it the Willy Da Gates way.

    THE USER IS THE MASTER,
    and not you, slaves,
    tied to your desks
    with the golden chains,
    not even able to sneeze
    without getting a "permission"
    from some nazi style sadists,
    whose mind is prototyped with the most
    primitive templates there are,
    not even knowing who he is.
    So lost is he in this biorobotic maze
    of NWO, cooking on your plate this very moment,
    eventually leading, according to official dictates,
    to a global totalitarian system,
    where there are only two classes left:
    The "elite", those blue blooded "Aryan" race,
    and the "slaves", aka the "filth of inferior race".

    THIS IS LITERAL MODEL and i have piles of specific
    reference and globally distributed archives that
    are guaranteed to survive for at least a generation
    unles the very life on the planet Earth ends,
    which is not that inconceivable,
    at least in the current state of affairs.

    Now...

    For that, he has to be able to hit a function key,
    from ANY window, and update his set of parameters or
    reload the news ones if he wishes, and it all has to
    be immediately reflected in ANY and all windows,
    open or closed.

    Realizing that some of those frames may be able to start
    a lenghty process of extracting huge amounts of information,
    filtering it, converting one data representation into others,
    such as generating perfectly valid web pages, with full
    support for style sheets, in line, if so desired,
    this thing could be swallowing hundreds of megabytes.

    With the original event model, that was a direct copy
    of original Unix based model, where you have the eventHandler
    that could process just about any event imaginable
    and report to the higher level wether it actually handled
    the event, so they could see if THEY need to handle it,
    everything was about the simpliest thing in the world.

    And now?
    Well, just about for every single GUI element on every
    single frame of yours, you'd have to hang the event
    handling code. I've spent DAYs on trying to figure out
    how to do it without hanging all the abstract fat on
    every single checkbox, button or a text field.

    Second major problm in MY case, is that the user, and I am
    dealing with ordinary users, not some huge corporations, if he
    would even bother to get your program on the first place,
    expects to simply double click on an executable
    (Windows environment), and see how program works,
    which is what nearly 100 percent of present software products do.

    But that is not possible with this JVM architecture.
    He would have to download 5 times as much code as the application
    itself needs, to get the JVM AND install it on his machine, and
    that is about the LAST thing he would want to bother about.
    In vast majority of cases, people do not even know what Java is,
    and could not even bother with installing all these "virtual machines",
    just because of security risks alone.

    People claim that Java is "safe", that you can't write files
    and on and on and on, but I don't believe a single
    word of it. I happened to work on one project at HP in the
    silicon valley doing some kernel level stuff,
    and there was one guy who worked in the kernel security
    department developing security related code.

    He told me of some security issues with some major programs
    on the market. He could totally take over your machine even
    though all you were doing on the net is web browsing, which
    seems to be one of the most innocent things there is.

    I also developed a related code for a firewall/monitor tool
    that allows customers and providers to visually monitor their
    traffic, create various rule sets for specific instances.
    And, what i have seen is that vast majority of people do not
    even suspect that they are literally open to all the hackers
    out there, just by simply using their web browser, I am not
    even talking about specific ports and protocols they have open
    on their systems. There isn't a single session when I login
    and use internet, when I am not attacked, and some of those
    attacks originate at the very ISP provider facilities
    from what I am seeing, and automatically probe your system
    for vulnerabiltis as soon as you log in.
    Things like that. I can hardly remember a SINGLE time
    when i logged in that there were not a MAJOR security probe
    performed on my machine.

    So, to claim "java is a safe as heaven" in terms of security,
    you need to be a 5 year old kid. I have personally seen some
    stuff, you couldn't have even IMAGINED in your WILDEST dreams.
    Nothin new to it.

    Yes, Java does have its attractive points and IS more abstract
    and platform independent than C++ can ever be, especially if
    you consider all the MS stuff. But, at the same time,
    what can be done to you and your machine,
    just by you enabling some java related features,
    is nothing less than circus miracles.

    But, what I find is that it is WAY less portable as people
    keep claiming. In fact, I am not sure there exists a
    development environment that even cares to bother about
    portability issues. Borland does it in its ways, and if you
    are not careful, you'll wire-in some of their proprietary
    things on the simpliest level, such as layouts, and, if
    you are not a fool, when you try to build your app, the
    other compilers would tell you they never heard of such layout
    or things of this nature.

    So, that claimed portability is vastly exagerrated.
    Borland is not compatible with MS stuff, and those, in their
    turn, do not even support the javax packages, and, instead,
    do just about all they can to trap you into their proprietary
    approach, and, in some cases, you might not even noticed
    that you have already used some of their stuff, until you try
    to rebuild that app under a different compiler.

    Simple things like toString() method, that has a proprietary
    microsoft's equivalent ToString(), which comes earlier
    in a collating sequence. So, when you use some object,
    and hit the dot key to expand it, the FIRST thing that
    is presented to you is Microsoft's version, and you might
    not even noticed it as it all happens in a semi-automatic way.
    I have fallen into this exact trap, and not once.
    The only thing that saved my skin is to try to rebuild it
    using a different development environment and I had to waste
    some time fixing some of those UTTERLY unnecessary screwups.

    So, all this looks like walking on a mine field and takes
    away so much energy on worrying about all these compatibility
    issues, it is simply hard to believe.

    In case of code I am working on right now, I spent at least
    70 if not 90 percent of my time on all these issues. On every
    single hour I spend actually developing stuff, I spent sometimes
    DAYS to figure out the solutions to the simpliest problems
    imaginable.

    Just to do a GUI layout of a modern app, it ALREADY took me a week,
    and the stuff is so standard and I use so little of those
    "sophisticated" features, that all that I am wasting my time on,
    could be done in few hours, if not minutes, under any environment
    I am aware of.

    Quite often I wonder what kind of mud I've got myself into
    with this whole Java thing. I could do the same thing in C++
    and be done with it monhs ago, without even bothering about
    screen resolutions, font sizes and on and on and on.

    The Java footprint is a footprint of a monster and its
    overhead is well beyond obscene.

    So...

    Hope that helps YOUR situation :---}

    Good luck.

    >> The
    >> first thing I thought is that adding more RAM will solve this ( though
    >> this is not an easy option) . Then I understood that a 32 bit system
    >> can have just about 3 gb ram for applciations and our AS had already 4
    >> GB with 1.5 GB VM also configured. Then I thought there might be ways
    >> to make the JVM shareable. But no. Also tried to use the flags to fine
    >> tune JVM. No go there also. Irrespective of how much you limit the
    >> actual limit is in the heap size allocated to the application and not
    >> to the private bytes of JVM.
    >>
    >> -----------------------------------------
    >> Data from Jconsole for the Java application
    >>
    >> Memory
    >> Current Heap Size 4.3 MB
    >> Max heap size 12. 2 MB ( set by flag)
    >> Committed memory 5 MB
    >> Operating System
    >> Committed VM - 26.7 MB
    >>
    >> Data from Perfmon (for java.exe)
    >>
    >> Private Bytes - 27.7 MB
    >> ----------------------------------------
    >>
    >> I am part of a large team that developes and maintains a telecom
    >> network managment software system. Also this software is used all
    >> around the globe almost in around 190 countries usually by telecom
    >> service providers. So this is a very real problem that I am speaking
    >> about. I am getting to love the simplicty of Java and of the great IDE
    >> Eclipse as much as I love the power of STL; but I am comming face to
    >> face with the main constraint of Java, its memory footprint and I
    >> guess thousands of others like me must have faced similar problems.
    >>
    >> If so the next question is what is SUN doing about this. I could not
    >> find it in their top 25 RFE's or top 25 bugs. (http://bugs.sun.com/
    >> bugdatabase/top25_bugs.dohttp://bugs.sun.com/bugdatabase/
    >> top25_rfes.do )
    >>
    >> 1) Is this really not a problem in the outside world then ?
    >>
    >> 2) Or is it that with the introduction of 64 bit HW and serveres this
    >> will be of no significance ?
    >>
    >> Anyway I doubt if companies like ours will adopt 64 bit HW and servers
    >> because of the costs involved in it ( note this is just a logical
    >> guess, I have no experience in such decisions)
    >>
    >> Anyway I am now forced to use C++ for the client and use Java on the
    >> server side. Fortunately most of the Java in the server side is done
    >> up as EJB's all conatined in the JBoss server thus consuming only one
    >> JVM there)
    >> -------
    >> I would like to get some comments on questions 1 and 2.
    >>
    >> Thanks
    >> Alex.C.P

    >
    >Hmm, I'm not sure I understand your setup, but it appearch you have a
    >single server machine which runs both the Java server process and the
    >multiple client processes?
    >
    >Usually, Java clients are run on seperate client machines, so the
    >number of clients doesn't effect the footprint.
    >
    >Anyway, to answer your questions, I've never found this to be a
    >problem in any of our environments, so perhaps there is some problem
    >with your architecture/design/configuration. I think that this problem
    >doesn't have anything to do with the 32 vs 64bit platform, but perhaps
    >a misunderstanding somewhere.
    >Java does not have a tiny footprint, but last I checked, 2048M(2gig)/
    >32Mb = 64. That means that you'd be able to run around 64 seperate JVM
    >processes on one machine. This isn't a typical situation, and it still
    >doesn't seem to fit your conclusions.
    >
    >Maybe it would help my understanding if I know what a "CLI" was. I
    >only know it as Clear Interrupt Flag :)
    >
    >Hope this helps,
    >Daniel.
    >
     
    nukleus, Jan 30, 2007
    #5
  6. Lew Guest

    nukleus wrote:
    [major top-posted rant snipped]

    wow.

    - Lew
     
    Lew, Jan 30, 2007
    #6
  7. Daniel Pitts Guest

    On Jan 30, 7:12 am, Lew <> wrote:
    > nukleus wrote:
    >
    > [major top-posted rant snipped]
    >
    > wow.
    >
    > - Lew


    Wow indeed.

    Yes, Java has a different approach to memory management than C++. It
    is common to see a larger footprint per JVM than you would in a C++
    application. Keep in mind though, that while the memory is allocated
    in the JVM, much of it is garbage collectable, and will be collected
    before more memory is allocated.

    Also, what was all that irrelevant chatter about kernel security and
    knowing a guy who can hack into your computer while you browse the
    web?

    nukleus wrote:
    > He told me of some security issues with some major programs
    > on the market. He could totally take over your machine even
    > though all you were doing on the net is web browsing, which
    > seems to be one of the most innocent things there is.


    Actually, browsing the web is not one of the most innocent things.
    Its one of the most dangerous, and most people should know that. Its
    safer if you know what you're doing, just like those of us smart
    enough not to lick the sidewalk. Sure, you could get infected, but
    you can also protect yourself in so many ways. The only time I got a
    virus was from a CD out of a programming book. I've never gotten a
    virus or a worm, and I don't run antivirus software.

    > I also developed a related code for a firewall/monitor tool
    > that allows customers and providers to visually monitor their
    > traffic, create various rule sets for specific instances.
    > And, what i have seen is that vast majority of people do not
    > even suspect that they are literally open to all the hackers
    > out there, just by simply using their web browser, I am not
    > even talking about specific ports and protocols they have open
    > on their systems. There isn't a single session when I login
    > and use internet, when I am not attacked, and some of those
    > attacks originate at the very ISP provider facilities
    > from what I am seeing, and automatically probe your system
    > for vulnerabiltis as soon as you log in.
    > Things like that. I can hardly remember a SINGLE time
    > when i logged in that there were not a MAJOR security probe
    > performed on my machine.


    What does this have to do with Java anyway?
    No one said connecting to the internet was safe. No one said walking
    outside was safe either. I have to say, I don't get security probes
    very often, perhaps its a *feature* of your ISP?
     
    Daniel Pitts, Jan 30, 2007
    #7
  8. Alex Hunsley Guest

    nukleus wrote:
    [snip lots of rant]
    > Thanks to this group, at least in SOME cases, you can
    > ask mere mortals if they know the answer to your simpliest
    > problme of all, and, if you are lucky enough, and they
    > are in the mood of doing so, they'll give you some
    > scetchy answer, that is nothing more than a hint,
    > and you'd have to go back to your puter and sweat it
    > out to full extent. Kinda sadomasochistict trip.


    If you don't like the help here, you could demand back the money you
    paid for it...

    > And what CAN you expect from those people?


    Once people get over their self-entitlement trip (the one that have such
    a thing), people often find help. More so if they're willing learn a bit
    about the group and follow some suggestions.

    > Not only that, but you will be insulted and rediculed,
    > told that in Java, the names of routines MUST be in
    > lower case, even though Sun itself has some things
    > that start in upper case and some don't.
    >
    > Now, what WHAT specific REASON do you have to make
    > all the method names in lower case?
    > Does it make your code MORE readable?
    > - Nope, just the other way around.



    > Does it prevent your compiler to compile this thing?
    > - Not even in the picture.
    >
    > Does it make MORE clear what are you trying to do?
    > - Are you kidding?
    >
    > So...
    >
    > What IS that dictate good for?


    It's called convention. Following a widely held convention aids quick
    understanding of something (by most people).

    > Just read this very group.
    > You'd be amazed if you are new to java.
    >
    > Those poor "clueless", when they are in just about the
    > biggest problem they can imagine at the moment,
    > instead of getting assistance,
    > get a brainwashing dictates on what you MUST do!


    Helping people to make themselves more understandable isn't a bad
    thing. And quite a few people do get help here (of the sort you're
    mainly thinking of).

    > You see?
    >
    > MUST!
    >
    > And unless you do so, you are considered of an
    > "inferior race"!
    > Nothing more.
    > Nothing less...
    >

    [snip]
    > And here, you may not even get a reply on your
    > question, just because you show these people
    > some piece of code that does not fit THEY view
    > on Universe and a life force as such.


    Get over your sense of entitlement. People help here for nothing. They
    don't get paid. The least they are entitled to is to decide what *sort*
    of advice they give, and who they reply to.
    And telling people why they're not having much success getting replies
    counts as helping them.

    > Wake up, people.
    >
    > WHAT are you doing here on the first place?
    > Why do you even bother to answer questions?
    > You see, if I am an expert, built according to YOUR
    > templates, I don't need your help. I can teach YOU,


    You think there is only one sort of problem? There are many sorts of
    problems possible. There are many ways code can be 'broken'.


    > and teach you using REASON, and not some dictates
    > as to how the planet Earth is to turn and where
    > is the Sun to rise in the morning.


    Reason tells us that following convention aids understanding.

    > Is it clear enough?
    >
    > I spent DAYS and still can not figure out what is the
    > problem on the first place. Just try to push the help
    > button. First of all, it'll eat up a terabyte of memory
    > and you'd be sitting there mezmerized, waithing for it
    > to stop swapping like mad and load all the stuff,
    > that is the simpliest thing in the world, just to
    > display a single help screen, which shouldn't take
    > more than a few milliseconds nowadays.

    [snip more]

    Just a little comment in the middle of that rant for you.... carry on!
     
    Alex Hunsley, Feb 1, 2007
    #8
  9. Lew Guest

    Alex Hunsley wrote:
    > Just a little comment in the middle of that rant for you.... carry on!


    Don't feed the trolls.

    - Lew
     
    Lew, Feb 1, 2007
    #9
  10. Alex Hunsley Guest

    Lew wrote:
    > Alex Hunsley wrote:
    >> Just a little comment in the middle of that rant for you.... carry on!

    >
    > Don't feed the trolls.


    I don't think he/she/it it *is* a troll. Just misguided. I'm not losing
    any sleep over it...


    >
    > - Lew
     
    Alex Hunsley, Feb 1, 2007
    #10
  11. nukleus Guest

    In article <45c221d6$0$764$>, Alex Hunsley
    <> wrote:
    >nukleus wrote:


    >[snip lots of rant]


    Lots of stamp?

    >If you don't like the help here, you could demand back the money you
    >paid for it...


    I DEMAND IT!!!!!

    >> And what CAN you expect from those people?


    >Once people get over their self-entitlement trip


    What is "self-entitlement trip"?

    Is it from those "help" files trip?

    >(the one that have such
    >a thing), people often find help.


    I am not sure you understand what I am saying,
    and this is not an issue with this group only.
    It is the issue just about everywhere you look.

    The issue is this:

    When someone needs assistance,
    it is OBVIOUS they do not know enough,
    pretty much by definition.

    Now, instead of telling them "go look up some
    other pile of you know what by now", why not
    take your time and explain the tricky parts
    of the issue. Do you realize that your articles
    are in my archives going back at least half a
    year and they are available globally via Google
    for the VERY least.

    And, being as smart as your very first sentence
    states, talking about "rants", do you have
    enough active neurons in your cockpit to comprehend
    that if you give an in-depth response,
    not only will you help one particular individual,
    but you'll cover the issue FOREVER, be it only
    YOUR own version. Then some other "expert"
    can do just the same and on the same thread,
    and, by the time we are thru with that thread,
    you gave a globally distributed version
    that finally talks human language and covers the
    issue.

    Do I need to chew more, mr. smart?

    And I have seen what kinda help people provide.
    They are like hoarding something and are just
    sending people from one pile of you know what
    by now, to another.

    And the thing could be finished literally taking
    a few minutes of your time. Finished COMPLETELY
    pretty much.

    Then...

    You could make your own archive of your own
    geniououous revelations, and simply repost it
    need be, if the same issue arises again.
    You could also make your supersexy web page
    and show how great you REALLY are by simply
    pushing a button and automatically generating
    the supersexy web pages and an index of your
    great discoveries.

    And what are you doing instead?

    Trying to make a FOOL out of me?
    Trying to tell me that you, high priests,
    are not going to give me your opinion on
    something that is the issues for me right now?

    Well...

    Tough.

    Because the opinions I've got so far,
    were nothing more that pointers,
    which means...

    Which means my time could be better spent
    by simply going directly to those piles of garbage
    they call documentation and dig up their source
    code, project files, etc. and finally realizing
    where that screw up was, just like i did with
    Borland JBuilter the other day.

    It turns out that their own project configuration
    files do not reconcile with sources. I do not wish
    to waste another sentence on telling you where and how.

    Get the drift?

    Oki, doki.

    Zo...

    Whats next fer breakfast?

    >More so if they're willing learn a bit
    >about the group and follow some suggestions.


    I know enough about usenet to the point that
    you look to me like clueless with your
    whining. And about what?

    You have a "problem" with me personally?
    Are you going to teach me how the planets turn?

    This is EXACTLY what i was telling you in my post.
    You see, you have your own ideas of "manners"
    and how people SHOULD!!!! behave
    when they come to the royal highness of your kind
    and creed. And you will be TEACHING them something
    that has to do NOTHING with what they came here for
    on the first place.

    Instead of addressing a specific issue,
    you will be engaging in this rotten guilt instillation
    trip, like you are nothing less than a Son of God,
    having all the answers to how life force operates.

    Sorry, I dont know you enough.
    But I can push a couple of buttons here.
    Hang on, the breakpoint is hit...

    :---}

    Just kidding, just to show you your relative significance
    in the scheme of things.

    So...

    You just pick the subject and i will produce you
    all your great deeds on this group within a minute
    and post it here in bulk if you wish. AUTOMATICALLY.
    You see?

    Do you know me to make these childish judgements.

    Some of you, arrogant "experts" with pious faces
    are even telling me they are getting fed up
    because my posts are too long.

    Oh, oh, oh, oh, oh.

    Do you, experts, know the simpliest usenet principle?
    Well, if you don't like it, don't read it.
    Don't follow up on it, and just forget about it.

    I don't need your morality lectures,
    cause, first of all, i do not expect ANYTHING of
    real value from you on the first place.
    So...
    I am not dependent on you like on some oxygen
    suppliers.

    Simple enough?

    And then?

    >> Not only that, but you will be insulted and rediculed,


    Isn't this the case?
    Do you want me to produce all your priestly
    brainwashing statements?

    >> told that in Java, the names of routines MUST be in
    >> lower case, even though Sun itself has some things
    >> that start in upper case and some don't.
    >>
    >> Now, what WHAT specific REASON do you have to make
    >> all the method names in lower case?
    >> Does it make your code MORE readable?
    >> - Nope, just the other way around.

    >
    >> Does it prevent your compiler to compile this thing?
    >> - Not even in the picture.
    >>
    >> Does it make MORE clear what are you trying to do?
    >> - Are you kidding?
    >>
    >> So...
    >>
    >> What IS that dictate good for?

    >
    >It's called convention. Following a widely held convention aids quick
    >understanding of something (by most people).


    ALL biorobots have these kinds of answers.
    But this is not an answer of ANY kind,
    because there is no reasoning in it,
    there is only an rears licking ideology
    of the herd.

    One more time: During the WWII, 99.99 percent
    of German people, screemed their heads of in
    that nazi histeria, shouting:

    Seig hail!
    Seig heil!

    You understand?

    Can you connect these two together
    without me chewing it for you?

    I bet not.

    So, i'll explain to the high priests of your kind
    and grade:

    WHATEVER YOU AGREE WITH THE HERD,
    DOES NOT IMPLY THAT IT MAKES ANY SENSE.

    Sorry, I don't have letters of larger size.

    Now, when Jesus went to school, and the alphabet
    was introduced to pupils, the teacher said:
    this is alpha, this is betta, and this is...

    And Jesus said: Sire, before you get to beta,
    you need to explain what is alpha. What is it for,
    why is it there on the first place, what makes
    it different and things like that.

    You understand?
    Can you connect it now?

    Translating: You need to use YOUR brain
    to figure out what is what and who is who,
    instead of ligning up in a long line and
    licking i hope you know what in front of you.

    Zo...

    You did not provide the very REASON
    WHY this "convention" has ANY validity whatsoever.
    All you said is just you all agree.
    Not a problem at all.
    You can agree that the Earth if flat.
    But you know how much difference all your
    agreements make to me?

    Well, just a waste of time typing all this.
    THAT is how much.

    I STILL do not have that "help" you are promising.
    And help not to me, but to YOURSELF,
    mr. confused, who thinks he is to teach others
    how to breathe and how to walk.

    >> Just read this very group.
    >> You'd be amazed if you are new to java.
    >>
    >> Those poor "clueless", when they are in just about the
    >> biggest problem they can imagine at the moment,
    >> instead of getting assistance,
    >> get a brainwashing dictates on what you MUST do!

    >
    > Helping people to make themselves more understandable isn't a bad
    >thing.


    Bad thing.
    Good thing.

    That is how you are prototyped.
    A simple, primitive template.
    "Coca cola is good. Drink coca cola".

    Is THAT the level we are talking about here?

    WHO are you to help others "to make themselves more
    understandable"?

    To WHOSE ears?

    Is it just because your closed mind can not tune
    into the wavelength of those "clueless" and talk
    THEIR language, by ANY chance?

    When i read that so called documentation, and,
    accidentally, look into their own source files,
    i, suddenly, realize that this so called documentation
    came directly from their comments before the methods
    in their own source files, which means...

    What do you think it means, mr. smart?

    Well...

    What it means to ME, is that this can not POSSIBLY
    be documentation to provide to those "cluless".
    It is rather SPECIFICATION.
    Do you see the difference between specification
    and documentation?

    Dear mr. smart, the difference is simple enough
    even for the most clueless of all to comprehend.

    Specifications are written for the experts
    and for those, working in the next department,
    so they know exactly what this thing does.
    It is a language of coded symbology, when one
    definition refers to another and that one,
    yet to another.

    But to someone who does not know all these
    technical intricacies, just about ALL this
    so called documentation produces is HEADACHE.

    You understand, mr. smart?

    Because they had a SIMPLE question in the beginning,
    and ended up being forced to learn the whole java
    arthitecture. But ALL they wanted to know is how
    to cath that stupid keyboard key.

    Documentation, in its turn, also needs to be different.
    I happen to have had a discussion with one of the
    prominent people in the field, with a phd in education,
    who was a vice president of one of successful companies
    in Silicon Valley, USA, developing computerised educational
    products, and I asked him a simple question:

    Why is it the the pupil has to be forced to read the
    entire book cover to cover instead of looking at things
    in that book at HIS paste and according to HIS interests
    and individual tendencies.

    And the answer was?

    Well, we did a study on this subject
    and came to a conclusion
    that this is very ineffective method.
    It takes MORE time to learn this way.

    And you know what I say?

    I say: screw your "research".

    The material needs to be presented on at least
    three different levels.

    1. Intruductory, for those "clueless",
    who don't even have an idea what they are asking.
    They are just beginning, and you can not give
    them the design specs, written by high priests.
    They'll go nuts. That is ALL you can hope to achive.

    2. Once you are no longer "clueless",
    and become interested in getting as much bang
    for your buck in some specific issue,
    the material can be expanded and made more detailed,
    more in depth.

    3. But once you know this thing nearly in and out,
    there comes the ULTIMATE level,
    when you start questioning the very roots of that
    specific subject.

    THAT is my education system.

    So...

    When i ask you about layout,
    which i NEVER had to deal with on the first place,
    what kind of "help" would you provide?

    Go read some OTHER pile of dogshit
    they call documentation?

    Then what do I need you for?

    You are just about as good as a plain corrupt priest.
    That is all.

    Nothing more.
    Nothing less.

    And you know what I do to those priests with pious faces
    of "holier than thouh"?

    Guess, mr. smart...

    And then?

    >And quite a few people do get help here (of the sort you're
    >mainly thinking of).


    Oh, so you are not merely mr. smart,
    but a subtle poisonous snake?

    Do you think i can not read your insult?

    One more time, at this junction, i am not even DREAMING
    of getting help, which is what you meant, mr. pervert.

    Seconly, what you meant, is that now your blind mind
    has already been committed to some things.

    You are saying that I did not formulate my questions
    correctly, in enough detail and precision?

    Sire, your single sentence is so full of poison,
    my screen got wet here. Lemme whipe it off.

    Oki doki.

    And then?

    >> You see?
    >>
    >> MUST!


    Is this true, by ANY chance?

    >> And unless you do so, you are considered of an
    >> "inferior race"!
    >> Nothing more.
    >> Nothing less...


    >[snip]


    Why did not you follow up on it?
    Isn't it EXACTLY what you are doing this very moment?

    >> And here, you may not even get a reply on your
    >> question, just because you show these people
    >> some piece of code that does not fit THEY view
    >> on Universe and a life force as such.

    >
    >Get over your sense of entitlement. People help here for nothing. They
    >don't get paid. The least they are entitled to is to decide what *sort*
    >of advice they give, and who they reply to.


    Clear to a five year old.
    And then?

    >And telling people why they're not having much success getting replies
    >counts as helping them.


    Uhu.
    So, instead of talking THEIR language,
    which is what REAL masters do,
    you want THEM to talk YOUR language,
    of which they know nothing to begin with.

    Zo...

    What kind of an excersize is this?

    Lemme expand it one level.

    It is an excersize in SADISM,
    no matter what kind of sticker you all agree
    to put on it.

    Yes, you can give your opinion, your advice,
    share your experience, but if you take away their
    freedon to express themselves in the way that is
    more natural to THEM, and at THIS particular junction,
    forcing them to jump up to YOUR "high" level
    before you even BOTHER to give them that stinky hint,
    then...

    Pretend you skipped about 5 paragraphs here.
    What a pitty. Juicy material it could be.

    >> Wake up, people.
    >>
    >> WHAT are you doing here on the first place?
    >> Why do you even bother to answer questions?
    >> You see, if I am an expert, built according to YOUR
    >> templates, I don't need your help. I can teach YOU,


    >You think there is only one sort of problem? There are many sorts of
    >problems possible. There are many ways code can be 'broken'.


    I am saying that no matter what kind of problem
    comes my way, and not matter how complex it is,
    i'll find my way out of this maze,
    and can expect VERY little from others
    as most of those "professionals", just about ANY
    place you look, are simply hoarding some
    "secret trick" and would never tell it to anyone,
    even if you paid them. Because THAT is how they
    maintain their "value" in a company. That lil
    nothing is the thing only THEY know, and they go
    as far, as to sabotage the entire projects,
    making it as difficult to others as possible.
    I know this trip in and out. It is not a theory
    of ANY kind.

    Because these people are full of fear of survival,
    and THAT is exactly why they accumulate those things
    and hide them in their bag, as that gives them some
    assurance that they are not going to be kicked on
    their ass as they know something, the others don't
    seem to know.

    Enough?

    Or you want more?

    >> and teach you using REASON, and not some dictates
    >> as to how the planet Earth is to turn and where
    >> is the Sun to rise in the morning.


    >Reason tells us that following convention aids understanding.


    You can twist it in just about ANY way you please,
    but if we address the issues of nothing less than
    reason itself, things are going to get even hotter.
    I promise.

    Did you understand what I meant by reason?
    Reason versus a stupid belief, a result of agreement
    on the lowest common denominator.

    Reason is when you use your mind
    and are not simply blinded by all the prejudice
    they programmed you with.

    Simple as that.

    >> Is it clear enough?


    No?

    We need to chew on this more?

    Why don't you simply tell me what are the
    tricks and boundary conditions of gridbags
    and where are pitfals?


    Or you'd prefer to waste ten times more time
    to make a fool out of me and prove that your
    priesthood is something that has validity to it?

    >> I spent DAYS and still can not figure out what is the
    >> problem on the first place. Just try to push the help
    >> button. First of all, it'll eat up a terabyte of memory
    >> and you'd be sitting there mezmerized, waithing for it
    >> to stop swapping like mad and load all the stuff,
    >> that is the simpliest thing in the world, just to
    >> display a single help screen, which shouldn't take
    >> more than a few milliseconds nowadays.


    >[snip more]


    >Just a little comment in the middle of that rant for you.... carry on!


    This stuff isn't quite rant, mr. smart.
    And it is good you say it, just the way you do.
    Because, from now on, it becomes an inprint.
    And inprint of what kind of "helper" you are.

    And what I see is that you are but a poisonous sadist,
    becase instead of addressing a SINGLE specific issue
    and giving your view on that, you DELIBERATELY
    witheld all your "wisdom"
    and reduced the whole thing to the puppet theatre level.

    Yep, mr. smart.

    Btw, why did you even BOTHER to read that long post
    of mine?
    Are you a pervert, by ANY humble chance?
     
    nukleus, Feb 2, 2007
    #11
  12. nukleus Guest

    In article <>, Lew
    <> wrote:
    >Alex Hunsley wrote:
    >> Just a little comment in the middle of that rant for you.... carry on!

    >
    >Don't feed the trolls.


    Oh, another one of those priests with pious faces,
    with a rubber stamp in the hand,
    making assumptions and fabrications.

    Sire, i wantched you for some time.
    You even happen to get on one of my lists.
    Unfortunately, i'll have to take you off of it.

    Sire, I suspect you consider yourself
    to be the highest priest around here.

    Is it true, by ANY huble chance?

    Is your blood at a boiling temperare ALREADY?
    Would you kindly whipe that foam from your mouth?

    >- Lew
     
    nukleus, Feb 2, 2007
    #12
  13. nukleus Guest

    In article <hNrwh.212507$>, Alex Hunsley
    <> wrote:
    >Lew wrote:
    >> Alex Hunsley wrote:
    >>> Just a little comment in the middle of that rant for you.... carry on!

    >>
    >> Don't feed the trolls.

    >
    >I don't think he/she/it it *is* a troll.


    That fool is simply fabricating his poisonous lies
    in his attempt to get enough ass lickers around
    to start some conspiracy of the herd, yelling:

    Zig hail!
    ZIg heil!

    >Just misguided.


    Oh, that is a pleasant surprise.

    So we even have guides here.
    Thats refreshing.

    >I'm not losing
    >any sleep over it...


    Btw, why would you even bother to follow up
    on this?

    Any particular reason?
    Or just killing time of your life?

    No wonder those kids that come here to ask
    a simple question and, instead of getting an answer,
    get a pile of brainwashing propaganda
    and, simply tell you into your faces:

    "Go **** yourself".

    And split.

    Is THAT the goal of the education system around here?

    Is it some kind of a passport?

    "Either you become like us,
    or you are not going to get a single word
    on the point of your question,
    and we will gather around
    to laugh at you, clueless"?

    Am I missing something here?

    So...

    Are you trying to get enough ass lickers here
    and start a conspiracy of taking over nothing
    less than big-8? Or is this some kind of an
    equivalent of a Bohemian Grove?
    Ever heard?

    I can show you da way to the REAL power.
    You know what I mean?

    What are you sitting here, wasting away for?
    You could be rulers of entire big-8!!!

    You know what big-8 is, by ANY humble chance?

    Good luck,
    and don't forget to have a good hearty lick
    on that fat ass in front of you,
    cause at the moment, HE is closer to that sweet fruit
    you've been craving for your entire lives.

    You, people, are pathetic, you see?
    Because you are wasting 99% of your energy
    on the issues of morality
    in a TECHNICAL group.

    And, in a technical group, if you are creative
    to ANY extent, you don't have that luxury
    of wasting your life on morality,
    because you have too much fun dealing
    with REALITY.

    Makes sense?

    Planet Earth,
    Planet Earth,
    ANYBODY home?

    Oh, that static again?

    How unfortunate,
    you need it all too bad at the moment.

    I almost thoght of promoting some of you
    and making something useful out of this
    lil game you play here,
    but now i'd have to think about it again.

    Zorry to tellya that.

    Btw, have you noticed what is the name of this thread?

    So...

    Do you have any comments on this topic?

    Or my original post does not have enough
    specific issues they blue blooded of your kind
    could even begin to address?

    What do you think about that footprint?

    >> - Lew
     
    nukleus, Feb 2, 2007
    #13
  14. Alex Hunsley Guest

    nukleus wrote:
    > In article <hNrwh.212507$>, Alex Hunsley
    > <> wrote:
    >> Lew wrote:
    >>> Alex Hunsley wrote:
    >>>> Just a little comment in the middle of that rant for you.... carry on!
    >>> Don't feed the trolls.

    >> I don't think he/she/it it *is* a troll.

    >
    > That fool is simply fabricating his poisonous lies
    > in his attempt to get enough ass lickers around
    > to start some conspiracy of the herd, yelling:
    >
    > Zig hail!
    > ZIg heil!


    Sorry Lew, you were right in calling 'troll'.
     
    Alex Hunsley, Feb 2, 2007
    #14
  15. nukleus Guest

    In article <mfxwh.213699$>, Alex Hunsley <> wrote:
    >nukleus wrote:
    >> In article <hNrwh.212507$>, Alex Hunsley
    >> <> wrote:
    >>> Lew wrote:
    >>>> Alex Hunsley wrote:
    >>>>> Just a little comment in the middle of that rant for you.... carry on!
    >>>> Don't feed the trolls.
    >>> I don't think he/she/it it *is* a troll.

    >>
    >> That fool is simply fabricating his poisonous lies
    >> in his attempt to get enough ass lickers around
    >> to start some conspiracy of the herd, yelling:
    >>
    >> Zig hail!
    >> ZIg heil!

    >
    >Sorry Lew, you were right in calling 'troll'.


    You forgot to say: lick, lick.
     
    nukleus, Feb 2, 2007
    #15
  16. Daniel Pitts Guest

    On Feb 1, 5:22 pm, (nukleus) wrote:
    > In article <45c221d6$0$764$>, Alex Hunsley
    >
    > <> wrote:
    > >nukleus wrote:
    > >[snip lots of rant]

    >
    > Lots of stamp?
    >
    > >If you don't like the help here, you could demand back the money you
    > >paid for it...

    >
    > I DEMAND IT!!!!!
    >

    [snip more rant]

    What's the word I'm looking for?
    Oh yeah, *plonk*
     
    Daniel Pitts, Feb 2, 2007
    #16
  17. nukleus Guest

    In article <>, "Daniel
    Pitts" <> wrote:
    >On Feb 1, 5:22 pm, (nukleus) wrote:
    >> In article <45c221d6$0$764$>, Alex Hunsley
    >>
    >> <> wrote:
    >> >nukleus wrote:
    >> >[snip lots of rant]

    >>
    >> Lots of stamp?
    >>
    >> >If you don't like the help here, you could demand back the money you
    >> >paid for it...

    >>
    >> I DEMAND IT!!!!!


    >[snip more rant]


    Yep, considering that your sense of humor
    is on the par with my fridge, what else
    would I expect you to leave in the post?

    What are you foaming at the mouth about,
    mr. smart?

    >What's the word I'm looking for?
    >Oh yeah, *plonk*


    EXACTLY. Just like just about all of those
    high priests eventually do.

    You see, for some strange reason, it is not
    enough for you to simply set your filters
    and be done with it.

    Nope. You have to come here and announce
    to the whole world what an intolerant
    blood boiling idiot you are, while, at the same
    time, thinking it makes you "superior",
    just by the mere fact that you announced
    your foolish "plonk" thing. Do you realize
    how much do I care about your "plonks"?

    And the most amazing thing about this
    is that you are also perverts, inevitably so,
    because most of you, if not all, simply
    use this stupid template of announcing your
    plonk, but in actuality, you are still reading
    those, you claim to have plonked, and I
    have seen PLENTY of those cases. And it goes
    as far, as once in a while, you even follow
    up on the their posts. Because you could be
    made to dance to just about any tune there is,
    and, eventually, can not contain hiding in the back
    and pretending you are not seeing it all.
    Once you are tickled enough, you will appear
    out of nowhere. See?

    Because, first of all, by plonking me, YOU
    are at a disadvantage, cause I can read your
    posts if I want to, but you can not read mine.
    So, from now on, no matter how much fun I can
    have with you, showing your great achievements,
    and laughing my rears of at you, you wont be able
    to even know that you might have become
    but a laughing stock.

    See?

    I do not have, never had, and have no plans
    to "plonk" anyone, even the most outrageous idiots,
    cause I know how to use my "next" button. And quite
    often, those I should have "plonked", turn out to
    be the about the finniest posts one can find.

    Btw, what is the Subject header of this thread?
    Why are you posting all this stuff being UTTERLY
    off topic?

    Are you a troll?

    This is a technical group and you claim to have
    more than but a few broken neurons on line.
    Can't you figure out the simpliest thing of this grade?

    Or you, "experts" around here, think this group
    BELONGS to you and YOU are here to do whatever YOU
    please, regardless of your own netti-quetti-betti-fetti,
    you yourselves claim to follow, or equivalent thereof
    with all your "thou shalt not do this, and thou shalt not
    do that" stuff?

    THOU SHALT NOT POST OFF TOPIC IN A TECHNICAL GROUP.

    THOU SHALT NOT FOLLOW THE POSTS OF THOSE,
    THEY CONSIDER TO BE OF INFERIOR RACE OR CREED.

    THOU SHALT NOT ENGAGE IN INSULTS AND REDICULE
    OF THOSE WHO DO HAVE A VALID POINT.

    You want more?

    Or shell I let you talk to my monkey?

    Now, back to the topic of this thread:

    What i found out is that in order for you to
    run a simple program, which only does things like

    int foo = 1,

    what would all your super-advanced technoly
    of delayed just about everything do?

    Well, if you set a brake point on main(),
    and then start up your app, you may be unpleasantly
    surprised, that before your breakpoint is hit,
    some heavy duty time has to elapse, and that is just
    about the first thing i have noticed when i tried
    to run the code i interited. For some strange reason,
    it would take it a long time to evan come up.
    When I looked at processes and resources,
    i was not so pleasantly surprized to see that by
    the time I select my first choice from the menu,
    before actually doing anything, it swallowed
    about 20 megs. For what?

    Well, it turns out that before you main() gets hit,
    you, royal architects, have to load the entire
    bloatware, called JVM.

    And even though that simpliest program that does not
    even do anything consists of a single CPU instruction,
    taking 1 clock cycle, for some strange reason, you
    have to load the whole universe.

    For what?

    Well, because it is called JVM.
    Bloatware that is.
    It can't use the same concepts that are used
    in virtual memory management, called page fault
    and bring up only those things, it actually needs
    this very moment, it seems.

    True, I haven't investigated this issue to the end,
    and I hardly need to, because I see the run time
    results without even figuring out why is it so
    bloated.

    And I bet you, if you run several java apps,
    you'd be loading the same JVM again and again,
    despite the fact that they can't even talk to
    each other without going outside the scope
    of JVM itself.

    So...

    What is this great breakthru are we talking about here,
    if you don't seem to manage even thouse things that
    were resolved generations ago?

    I was quite surprised that the AWT wasn't even
    thread safe and if you want to use threads,
    you'd better download 50+ megs of the latest
    version of the bloatware, and not only you,
    but ALL your customers and co-developers.

    How could THAT be?

    Are you joking or something?

    So, for the real world applications, especially
    those, we are talking on this specific thread,
    you have to load literally gigabytes of JVM
    alone, one for each instance of your app.

    So, if you are writing apps for the biggest and
    baddest of themn all, and ESPECIALLY in network
    related areas, which Java claimed to be designed
    for on the first place, you better tell your boss
    that you need to order the latest version of
    the mainframe with super cooled CPU and create
    a huge service room just to keep all the cooling
    equipment in.

    You'd have to learn how to walk on your ears
    if you want to be involved in anything big
    related to java.

    So, if he, for some strange reason, thinks that
    they just want to push a few buttons and update
    the global system of whatever they deal with,
    YOU are in trouble. Because he'll probably think
    that it is YOU that is the problem and not that
    bloatware they have chosen to step into.

    Good enough for "on-topic" material?

    Your turn, oh masters of dizastas.

    Cause it looks like we're going to have some fun here.
     
    nukleus, Feb 2, 2007
    #17
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Beatrice Rutger

    JMS memory footprint size

    Beatrice Rutger, Jun 5, 2005, in forum: Java
    Replies:
    0
    Views:
    370
    Beatrice Rutger
    Jun 5, 2005
  2. gbrun
    Replies:
    1
    Views:
    456
    Andrey Kuznetsov
    Feb 19, 2006
  3. Adam Warner

    Memory footprint of a subclass

    Adam Warner, Feb 27, 2006, in forum: Java
    Replies:
    3
    Views:
    655
    Adam Warner
    Feb 27, 2006
  4. Replies:
    9
    Views:
    1,117
    nibblix
    Apr 24, 2006
  5. nick
    Replies:
    58
    Views:
    1,954
    Bart van Ingen Schenau
    Mar 16, 2009
Loading...

Share This Page