Re: newb: running Java service?
Tom--
If your users are running as applets, why not run your server as a servlet?
Does it need to hold state information across the entire userset, or is
each user interaction stateless?
What I'm suggesting is something like this: create a number of different
servlets to perform each task ("AddFooServlet", "RemoveFooServlet", etc.),
and have the applet simply open up URL connections (with appropriate
parameters sent as URL parameters) to the servlet of the appropriate type.
The servlet performs its thing, hands back any data, and quits. The data
(either direction) could even be URL-friendly string representations of
Serialized data, if necessary.
I can hear the OO purists preparing their flamethrowers now, so let me
issue the one argument that will make all this non-OO work worthwhile:
PERFORMANCE. Holding state is wasteful, so long as the various operations
can be held in a stateless manner (or state is held within the back-end
database, which is usually easier than you'd imagine), since state means
that object is now bound to a single client, and cannot service other
clients--this means you will need to hold 1 object on the server for every
client currently using the system, EVEN IF THE CLIENT ISN'T CALLING YOU AT
THE MOMENT.
With a stateless model, which works well with HTTP servers anyway (and
would allow for HTML-only thin client layers if the requirement came down
the pike, which it usually does, in my experience), a given servlet can be
unloaded when client "a" is still thinking, or service clients "b", "c" or
"d" as necessary. With a stateful model, 4 objects would need to be held
(one for a, b, c and d) until each client dropped out.
Whenever you start thinking about distributed objects, push for
statelessness as much as possible; if that's not feasible, push the state
someplace else, like the back-end database.
In your particular case, a servlet could take the SQL (or whatever you're
handing the back-end, you didn't specify) and pass it directly through to
the database. The servlets can open one (or more, if you want to support
some sort of pooling mechanism, or need to differentiate by
userid/passwords) connection to the database on startup, close it when the
servlet is unloaded, and use the same connection over and over to avoid
having to make that connect each time.
Opinions welcome, well-thought-out arguments welcome, flames > NUL. Hope
this helps.
At 10:15 PM 5/4/99 -0400, Tom Roche wrote:
>First: I hope this, being more of a Linux question rather than a Java
>question, isn't off-topic: if so, please excuse.
>
>Second: feel free to correct any and all errors in conception or
>usage you see--I know all too little about Linux, Unix, and system
>administration. (But I hope to devote more time to this over the
>summer.)
>
>I'm working on a Java-based website, running on Apache on a Linux
>box. It has a backend that connects users (applets) to the
>database. For testing, I've been logging in and running a script that
>runs my backend classes. Now I want the backend to run as a service,
>i.e. without a user logged in (like FTP, etc).
>
>Is "service" the proper term to use here? ("daemon"?)
>
>What is "the best way" to do this?
>
>I believe the best way to do this is via inetd: is this correct?
>
>If so, what would be the proper settings for inetd.conf?
>
>Your assistance is appreciated, [EMAIL PROTECTED]
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: newb: running Java service?
>> Another option is to use servlets, the converse of "applets" but embedded >> on the server side. Apache supports JServ (see java.apache.org) which >> runs standalone alongside the web server. Servlets let you do fancy >> things which you may want in the future. In particular you could pool >> your database connections and reuse them from servlet to servlet. > >Caveat on pooling database connections in servlets: > >You need to make sure those database connections are thread safe. Since >there is only one instance of a servlet running, you don't want threads >stealing resources from each other. > This caveat holds for *any* stateless machine that can be executed concurrently (as servlets are, at least within any self-respecting servlet engine, anyway). >IMHO though, servlets are very useful, and are my favorite way of doing >servers in java. Even if you don't plan to access them through a web >browser. > And once you do that, is your web server really *just* a web server, or a particularly over-specialized form of generic application server.? :) How about this: instead of running a web server, run a generic application server (EJB, CORBA, who cares) that has a "servlet" (sorry to reuse the term) that listens on port 80, understands HTTP, and hands back HTML resources? Six of one, half-dozen of the other, you might argue, but most application servers are starting to go the clustered/load-balancing/fault-tolerance route, and if your application server is an EJB server, your HTTP SessionBean can always get swapped out when there's no HTTP requests Try doing *that* with Apache. :) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Swing never waits?
>I started up the SwingSet demo and noticed how slow it was. I ran >top and noticed that even when I was doing nothing (i.e. no mouse >moves, no selections, just sitting there) the demo was eating up >16 MB of memory and 50-60% of my cpu. It seems as if the virtual >machine never waits (at least when running this demo). > Because Swing is a fully lightweight library, it can't rely on any of the operating-system-specific constructs underneath the JVM to more efficiently make use of the CPU. Instead, it spins off a daemon thread to (basically) poll the OS for the mouse position and information and broadcast that information to the Swing JComponents under the mouse pointer at that moment. Inefficient, yes, but that's what "Write Once, Run Anywhere" buys you--necessary inefficiency sacrificed on the altar of portability. :) I've never measured the CPU occupation of an AWT app, though; might be interesting to run an AWT example and see if it takes up the same amount of CPU. If not, then Swing's obviously doing things in the background that an AWT app's not. (Maybe just filtering the events through all the event listeners?) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Swing never waits?
>This sounds suspicious. Is this conjecture or based on analyzing the Swing
code?
>There's always an AWT window under a top-level Swing window, and I would
expect
>Swing to be using AWT events to be tracking input devices.
>
A little conjecture, a little analysis. Swing uses the AWT event queue,
true, but because all lightweight components must exist within an AWT
toplevel component, Swing has to take over from there and dispatch all
messages from thence forth. Whereas a native Win32 app (even if it's not
optimized and/or makes use of such whizzy-bang nifties like tooltips) can
"park" itself in the CPU when the application is disabled, a Java AWT/Swing
app can't necessarily do that. See below for what I mean.
>Not that there aren't a lot of CPU-sucking details buried in Swing (or
>lightweight components in general)... I just question whether this is one of
>them.
>
One of the details in Win32 AWT is that in the native Win32 event loop
(buried within the native implementation of the Win32 java.awt.Toolkit
implementation), the following calls are made:
/*
* Create the one-and-only toolkit window. This window isn't
* displayed, but is used to route messages to this thread.
*/
m_toolkitHWnd = CreateToolkitWnd("theAwtToolkitWindow");
ASSERT(m_toolkitHWnd != NULL);
/*
* Setup a GetMessage filter to watch all messages coming out of our
* queue from PreProcessMsg().
*/
m_hGetMessageHook = ::SetWindowsHookEx(WH_GETMESSAGE,
(HOOKPROC)GetMessageFilter,
0, GetCurrentThreadId());
Both of these are horribly inefficient; I'm not sure why Java/AWT needs to
route all messages through a single window (probably has to do with native
threads, come to think of it), nor why they need a message hook. One of the
properties, however, of hooks is that you will be called for each and every
message that comes through the system, whether you were intended for it or
not.
Anybody with more Swing/AWT internals knowledge than I care to comment?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: newb: running Java service?
>Looks like we're going more into what the original poster bargained for =) > Uh what was the original question again? :) >EJB is nice I certainly agree, but as alx points out: EJB servers are >either a) at least $10,000 per installation or b) under beta testing or >development. But they are certainly the way to go in the future if you >want to do any seriously heavy server-side development in Java. > Agreed. But I'd watch prices come back down as a thousand different vendors suddenly jump into the market with low-range to super-high-range EJB servers. What's more, you can usually download these eval versions for development to get a good idea of what EJB development is like without shelling out a penny. (Gotta love this Open Source movement, eh?) :) >I'm writing a course on EJB so if you want any pointers to resources on >them I'd be happy to share. > Course for whom? >And there are open source projects in this area too. > www.EJBHome.com (now a division of Iona) is one, I know of (I think) one or two others as well -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
My turn to ask.... :)
I've got a problem running JDK 1.2v1 on my RedHat 5.2 install, but only AWT stuff. I know this has crossed this forum before, but I'm kinda asking for somebody to hold my hand on this and walk me through what I need to do to make this work. I'm trying to run the Notepad sample from the demos/jfc directory. Error I get back is: uncaught exception: java.lang.UnsatisfiedLinkError: /usr/jdk1.2/jre/lib/i386/lib fontmanager.so: libstdc++-libc6.0-1.so.2: cannot open shared object file: No such file or directory java.lang.UnsatisfiedLinkError: /usr/jdk1.2/jre/lib/i386/libfontmanager.so: libstdc++-libc6.0-1.so.2: cannot open shared object file: No such file or directory at java.lang.ClassLoader$NativeLibrary.load(Native Method) at java.lang.ClassLoader.loadLibrary0(Compiled Code) at java.lang.ClassLoader.loadLibrary(Compiled Code) at java.lang.Runtime.loadLibrary0(Compiled Code) at java.lang.System.loadLibrary(Compiled Code) at sun.security.action.LoadLibraryAction.run(Compiled Code) at java.security.AccessController.doPrivileged(Native Method) at sun.awt.font.NativeFontWrapper.(NativeFontWrapper.java:41) at sun.awt.X11GraphicsEnvironment.initDisplay(Native Method) at sun.awt.X11GraphicsEnvironment.(X11GraphicsEnvironment.java:61) at java.lang.Class.forName0(Native Method) at java.lang.Class.forName(Compiled Code) at java.awt.GraphicsEnvironment.getLocalGraphicsEnvironment(Compiled Code) at java.awt.Font.initializeFont(Compiled Code) at java.awt.Font.(Compiled Code) at javax.swing.plaf.metal.DefaultMetalTheme.(Compiled Code) at javax.swing.plaf.metal.MetalLookAndFeel.createDefaultTheme(Compiled Code) at javax.swing.plaf.metal.MetalLookAndFeel.getDefaults(Compiled Code) at javax.swing.UIManager.setLookAndFeel(Compiled Code) at javax.swing.UIManager.setLookAndFeel(Compiled Code) at javax.swing.UIManager.initializeDefaultLAF(Compiled Code) at javax.swing.UIManager.initialize(Compiled Code) at javax.swing.UIManager.maybeInitialize(Compiled Code) at javax.swing.UIManager.getUI(Compiled Code) at javax.swing.JPanel.updateUI(Compiled Code) at javax.swing.JPanel.(Compiled Code) at javax.swing.JPanel.(Compiled Code) at javax.swing.JRootPane.createGlassPane(Compiled Code) at javax.swing.JRootPane.(Compiled Code) at javax.swing.JFrame.createRootPane(Compiled Code) at javax.swing.JFrame.frameInit(Compiled Code) at javax.swing.JFrame.(Compiled Code) at Notepad.main(Compiled Code) Obviously, it's the top line that's causing the problem--where do I find/build the "libstdc++-libc6.0-1.so.2" file? I've used Linux before (back in the 0.99pl13 days), but not recently; I've used g++ before, but not since 2.6.x something. (I'm a little out of date with Linux these days.) I know Joe Burks posted a fix on how to get this to work, but it didn't work for me; I'm sure it's some Linux-configuration thing that's keeping me down. Help? Oh, and off-topic: Does anybody have a good reference on Linux shared libs these days? It's changed significantly since I last dabbled in Linux-C/C++ development. Ted Neward Dorado Software -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: My turn to ask.... :)
>> I've got a problem running JDK 1.2v1 on my RedHat 5.2 install, but only AWT >> stuff. I know this has crossed this forum before, but I'm kinda asking for >> somebody to hold my hand on this and walk me through what I need to do to >> make this work. > >With a preamble like that, I'd have expected a harder problem :-). > I'll try harder next time, just for you. :) >> I'm trying to run the Notepad sample from the demos/jfc directory. Error I >> get back is: >> >> uncaught exception: java.lang.UnsatisfiedLinkError: >> /usr/jdk1.2/jre/lib/i386/lib >> fontmanager.so: libstdc++-libc6.0-1.so.2: cannot open shared object file: >> No such file or directory > >This is one of those platform-specific things that will be resolved for >the real release. You've got a shared lib somewhere that the JVM needs; >it's called something like /usr/lib/libstdc++.so.2.8.0 (or something >similar, depending on what version you have installed). You can, until >the final JVM comes out, create a symbolic link that should solve it for >you: > > ln -s libstdc++.so.2.8.0 /usr/lib/libstdc++-libc6.0-1.so.2 > ^^ >(substitute the real name of the library you have on your system). > OK, but now it's kicking out all these "font specified in font.properties not found" messages, although that could be because I'm actually running it over the wire on my NT box here at work. Don't suppose you know how I can correct those either, do you? (I'm running eXceed for NT here, running the app on my Linux box at home.) :) Thanks! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Question
>> I was thinking about this discussion of servlets and application servers >> and taking into consideration that I know nothing about EJB, but have some >> minor experience with an app server,... >> >> So I pose this question to my java compatriots with experience in other >> realms of server side java. Why would I use an app server or some other >> technology? What does EJB give me? Couldn't I use EJB with servlets as >> opposed to an EJB centric appserver? > >In a nutshell (speaking as an "expert" who's read a few dozen more EJB >documentation pages than you have :-), EJB is good for centralizing your beans >on large, high-capacity server(s) that offer the sort of performance, >scalability, and reliability you need in an enterprise. The beans live, >serverside, in containers that handle a lot of the plumbing issues -- so, for >example, the beans can implement the business logic and let the container worry >about how to connect to databases. You connect to beans from the client side >with ordinary RMI calls. > As another "expert" who's working for a company doing a LOT of development based on EJBeans, I'd say the above is correct, to a degree. I believe that the EJB community isn't quite yet sure what they've created, and are still fishing for the patterns they need to do efficient, effective EJB development. (My opinion is that EJBs and its bean-managed-persistence/container-managed-persistence is fundamentally flawed and will be the most highly rewritten part of the EJB spec over time.) Think "web server for generic applications" and you're not too far off the mark. Yes you can put your business logic into beans, but it doesn't have to be, and in fact may not want it to be there, depending on scalability desires. I'm not an expert; I don't think ANYBODY is with something this new. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Stream
Use FileInputStream or FileOutputStream. At 09:09 AM 5/6/99 +0200, Ozer Irfan wrote: >Hello. >How open a file for read/write ? > >Thanks > >Irfan > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: A little help
Your import statement (in your teste code) must be wrong; otherwise, java
would pick it up.
To determine if its a CLASSPATH issue, try this:
javax.swing.tree.DefaultTreeModel model = new
javax.swing.tree.DefaultTreeModel(root);
That'll tell you if you left out the import, got the import wrong, or your
CLASSPATH is screwed up somehow. (Honestly, though, if the CLASSPATH were
screwed, you'd be getting a lot more errors than this one--I'm betting my
money on the import statement.)
At 01:42 PM 5/6/99 +0100, Mario Jorge Nunes Filipe wrote:
>Pavel Tolkachev wrote:
>>
>> Hello Mario,
>>
>> Try to use this code snippet:
>>
>> // -cut here--
>> import javax.swing.JTree;
>> import javax.swing.tree.DefaultMutableTreeNode;
>> import javax.swing.tree.DefaultTreeModel;
>> ...
>> DefaultMutableTreeNode root = new DefaultMutableTreeNode("Root");
>> DefaultMutableTreeNode c1 = new DefaultMutableTreeNode("Child1");
>> DefaultMutableTreeNode c2 = new DefaultMutableTreeNode("Child2");
>> DefaultMutableTreeNode gc1 = new DefaultMutableTreeNode("Grandchild1");
>> DefaultMutableTreeNode gc2 = new DefaultMutableTreeNode("Grandchild2");
>> DefaultMutableTreeNode ggc1 = new
>> DefaultMutableTreeNode("GrandGrandChild1");
>> DefaultMutableTreeNode c3 = new DefaultMutableTreeNode("Child3");
>>
>> root.add(c1);
>> root.add(c2);
>> root.add(c3);
>> c1.add(gc1);
>> c1.add(gc2);
>> gc1.add(ggc1);
>>
>> DefaultTreeModel model = DefaultTreeModel(root);
>>
>> JTree tree = new JTree(model);
>> ...
>> // -cut here--
>>
>> This is just the idea (to give you some starting point). To make
>> applets, you have to provide the availability of Swing to the target
>> browser somehow. Anyway to make something useful with JTree you will
>> *have* to read the documentation and to do many experiments. Tutorial is
>> the good starting point.
>
> Hi Pavel
>
> thanks for your help but now I have a different problem (actually
>sumething i had already faced before). Watch this:
>
>13:38:01$ javac teste.java
>teste.java:42: Method
>DefaultTreeModel(javax.swing.tree.DefaultMutableTreeNode) not found in
>class teste.
>DefaultTreeModel model = DefaultTreeModel(root);
> ^
>1 error
>
>I'm using the swingall.jar wich came with netbeans 2.1. Is there a
>better one? Where ?
>
>Thanks
>--
>Mario Filipe
>[EMAIL PROTECTED]
>http://neptuno.sc.uevora.pt/~mjnf
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: JARs and performance...
Robert-- >Trying to optimize JAR performance. We have a JAR that is 700K using >JDK1.1.7 and are looking for ways to improve the performance when using >it. At this time we cannot migrate to Java2 and utilize the >JArURLConnection. Does anyone have any suggestions? We would like to >split the JAR into two JARs: 1 for GUI and one for CORE components but >have run into difficultly due JAR loading at the 1.1 level. Additinoally, >we have looked into client side install with class date checking sort of a >Smart Update... > >From the context of what you're saying, I'm guessing that you're running a Java application and not an applet on the client. What I can suggest is that you write a custom ClassLoader that uses java.net.URL to connect to a given web server, check dates (against the .jar in your local path), download the .jar if necessary, then proceed with normal class loading (which is probably your "SmartUpdate" idea in a nutshell). Are your jars compressed? JDK 1.1 had a problem with compressed .jars, as I recall, but I'm not 100% sure of that. Alternatively, I don't see any reason why you couldn't pull the source for URLClassLoader out of 1.2, put it into your own "java2" package, and use it within java 1.1. You'd have to make a few source modifications to adjust for the changed ClassLoader rules in 1.2, but that shouldn't be rocket science. >Any past experience in this area would be helpful... > I haven't done this precise thing, but I've done a lot of work with ClassLoaders; I'll be happy to help as best I can. Hope any of this rambling makes sense or helps out -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: JARs and performance...
>> From the context of what you're saying, I'm guessing that you're running a >> Java application and not an applet on the client. What I can suggest is >> that you write a custom ClassLoader that uses java.net.URL to connect to a >> given web server, check dates (against the .jar in your local path), >> download the .jar if necessary, then proceed with normal class loading >> (which is probably your "SmartUpdate" idea in a nutshell). > >I've been looking for a way to provide users of my software with a >"SmartUpdate" feature. Your idea sounds great, the only problem is when >your want to save the new JAR in place of the old one: under Unix it >should work, but under windows the original JAR is locked and it is >impossible to squash it with the new one. > >Would you have any idea on how to replace a JAR in-place and circumvent >file locking under some platforms? This problem has been bugging me for >more than a year. > Hmm Are you willing to go to native code for your client bootstrap app? You can use JNI Invocation to create the java VM, inside the Java app drop the .jar into a new directory (call it "incoming" or something), then copy the .jar over to the Java directory when the java VM exits. Beyond that, I'm not sure--it really depends on how the Win32 JNI code is opening the file for use--if they're using exclusive locking, then I think you're SOL unless somebody knows of a Win32 API call that can work around or break an exclusive lock. Give me a bit (can't at the moment--busy), and I'll see if I can't figure out how they're doing it under the hood. Send me a reminder in a day or two if you don't hear from me first. >Conscience is what hurts when everything else feels so good. > I LIKE this line! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: java versus c++ or perl
Oh, Lord, what, you WANT a flame war? Look, here's my bottom line: With Power Comes Complexity. Perl is simpler than Java, Java is simpler than C++. With that simplicity you lose a corresponding amount of power. But the end result is that it doesn't matter--you can accomplish just about any project in just about any language. C++ will be fastest, Java will come in second (due to the presence of JITs and the like), and Perl will be last. The "best" language to use is the one that (a) everybody feels comfortable with, (b) has the capacity to easily do what the project demands, both now and in the future, (c) produces maintainable code that, when everybody on the project leaves and Management wants an update later, can easily be understood by the poor schmuck who inherits it all. My personal preference is for C++ and Java, since (a) those are the two languages I know the best, and (b) I'm adept enough with either one that I can produce reusable code that can be used in a variety of situations. If I had to choose one vs. the other, I'd lean towards Java, simply because it's a nice compromise between a high-level language like Perl and an object-oriented language like C++--not too much power sacrificed to get some nice simplicity. Java also has the advantage that Perl (AFAIK--I'm not even a Perl newbie) lacks in that Java has JNI, which would allow you to "call down" to C++ code if necessary. C++ code can also create a Java VM within it (again via JNI), so using one doesn't exclude the other, and I think Perl can also be fired up within a C++ process via library calls--why not use all three? At 02:08 PM 5/6/99 -0700, [EMAIL PROTECTED] wrote: >we are having a discussion here at work about the deployment of a web based >database, > >i think jdbc is the choice othere think perl or c++ is the way to go, any >thoughts/links etc? > >thanks > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: JARs and performance...
>I have thought about pulling the class out of 1.2, but was not certain of >the affects. Additionlly, I thought of just rewrting their code in 1.1.7. >Either way that is an option. > >The Jar will consist of classes the will be run as an >application/applet/serialized bean. But for the performance part the >applet is the big concern. Such a big jar, start off with big problems. > True, but I would argue that the actual performance of downloading the entire jar onto the local VM each time the app starts is probably not that big of a hit, and once it's there, there's no more need to go across the wire. >Do you think the 1.1 security model will hamper a ClassLoader like that? > I don't think so, unless you're running as an applet--the AppletSecurityManager disallows ClassLoaders of all forms. > >-Bob > > >On Thu, 6 May 1999, Ted Neward wrote: > >> Robert-- >> >> >Trying to optimize JAR performance. We have a JAR that is 700K using >> >JDK1.1.7 and are looking for ways to improve the performance when using >> >it. At this time we cannot migrate to Java2 and utilize the >> >JArURLConnection. Does anyone have any suggestions? We would like to >> >split the JAR into two JARs: 1 for GUI and one for CORE components but >> >have run into difficultly due JAR loading at the 1.1 level. Additinoally, >> >we have looked into client side install with class date checking sort of a >> >Smart Update... >> > >> >From the context of what you're saying, I'm guessing that you're running a >> Java application and not an applet on the client. What I can suggest is >> that you write a custom ClassLoader that uses java.net.URL to connect to a >> given web server, check dates (against the .jar in your local path), >> download the .jar if necessary, then proceed with normal class loading >> (which is probably your "SmartUpdate" idea in a nutshell). >> >> Are your jars compressed? JDK 1.1 had a problem with compressed .jars, as I >> recall, but I'm not 100% sure of that. >> >> Alternatively, I don't see any reason why you couldn't pull the source for >> URLClassLoader out of 1.2, put it into your own "java2" package, and use it >> within java 1.1. You'd have to make a few source modifications to >> adjust for the changed ClassLoader rules in 1.2, but that shouldn't be >> rocket science. >> >> >Any past experience in this area would be helpful... >> > >> I haven't done this precise thing, but I've done a lot of work with >> ClassLoaders; I'll be happy to help as best I can. >> >> Hope any of this rambling makes sense or helps out >> >> >> -- >> To UNSUBSCRIBE, email to [EMAIL PROTECTED] >> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >> > > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
JDK 1.2-pre2 on RedHat 5.2
.0.so => libtcl8.0.so libvgagl.so.1 => libvgagl.so.1.3.0 libvga.so.1 => libvga.so.1.3.0 libslang.so.0 => libslang.so.0.99.38 libreadline.so.3 => libreadline.so.3.0 libhistory.so.3 => libhistory.so.3.0 libpq.so.1 => libpq.so.1.1 libpq++.so.1 => libpq++.so.1 libpgtcl.so.1 => libpgtcl.so.1 libecpg.so.1 => libecpg.so.1.1 libp2c.so.1 => libp2c.so.1.2.0 libnewt.so.0.30 => libnewt.so.0.30 libpanel.so.3.0 => libpanel.so.1.9.9e libncurses.so.3.0 => libncurses.so.1.9.9e libmenu.so.3.0 => libmenu.so.1.9.9e libform.so.3.0 => libform.so.1.9.9e libncp.so.1 => libncp.so.1.0 libungif.so.3 => libungif.so.3.0.0 libtiff.so.3 => libtiff.so.3.4 libstdc++.so.2.8 => libstdc++.so.2.8.0 libpng.so.2 => libpng.so.2.1.0 libjpeg.so.6 => libjpeg.so.6.0.1 libjpeg.so.62 => libjpeg.so.62.0.0 librle.so.1 => librle.so.1.0.0 libppm.so.1 => libppm.so.1.0.0 libpnm.so.1 => libpnm.so.1.0.0 libpgm.so.1 => libpgm.so.1.0.0 libpbm.so.1 => libpbm.so.1.0.0 libfbm.so.1 => libfbm.so.1.0.0 libstdc++.so.2.7.2 => libstdc++.so.2.7.2.8 libg++.so.2.7.2 => libg++.so.2.7.2.8 libgdk_imlib.so.1 => libgdk_imlib.so.1.8.0 libImlib.so.1 => libImlib.so.1.8.0 libguile.so.2 => libguile.so.2.0.0 libgtk.so.1 => libgtk.so.1.0.6 libgdk.so.1 => libgdk.so.1.0.6 libgpm.so.1 => libgpm.so.1.13 libgmp.so.2 => libgmp.so.2.0.2 libglib.so.1 => libglib.so.1.0.6 libgimpui.so.1 => libgimpui.so.1.0.1 libgimp.so.1 => libgimp.so.1.0.1 libgck.so.1 => libgck.so.1.0.0 libgdbm.so.2 => libgdbm.so.2.0.0 libgd.so.1 => libgd.so.1.2 libf2c.so.0 => libf2c.so.0.22 libexpect5.26.so => libexpect5.26.so libcrack.so.2 => libcrack.so.2.7 libBLT.so.2 => libBLT.so.2.4 libopcodes-2.9.1.0.15.so.0 => libopcodes-2.9.1.0.15.so.0.0.0 libbfd-2.9.1.0.15.so.0 => libbfd-2.9.1.0.15.so.0.0.0 libamu.so.1 => libamu.so.1.0.1 libz.so.1 => libz.so.1.1.3 libpanel.so.4 => libpanel.so.4.2 libncurses.so.4 => libncurses.so.4.2 libmenu.so.4 => libmenu.so.4.2 libform.so.4 => libform.so.4.2 /lib: libproc.so.1.2.6 => libproc.so.1.2.6 libdl.so.1 => libdl.so.1.9.5 ld-linux.so.1 => ld-linux.so.1.9.5 libpam_misc.so.0 => libpam_misc.so.0.64 libpam.so.0 => libpam.so.0.64 libpwdb.so.0 => libpwdb.so.0.55 libuuid.so.1 => libuuid.so.1.1 libss.so.2 => libss.so.2.0 libext2fs.so.2 => libext2fs.so.2.4 libe2p.so.2 => libe2p.so.2.3 libcom_err.so.2 => libcom_err.so.2.0 libtermcap.so.2 => libtermcap.so.2.0.8 libutil.so.1 => libutil-2.0.7.so libresolv.so.2 => libresolv-2.0.7.so libpthread.so.0 => libpthread-0.7.so libnss_nis.so.1 => libnss_nis-2.0.7.so libnss_files.so.1 => libnss_files-2.0.7.so libnss_dns.so.1 => libnss_dns-2.0.7.so libnss_db.so.1 => libnss_db-2.0.7.so libnss_compat.so.1 => libnss_compat-2.0.7.so libnsl.so.1 => libnsl-2.0.7.so libm.so.6 => libm-2.0.7.so libdl.so.2 => libdl-2.0.7.so libdb.so.2 => libdb-2.0.7.so libcrypt.so.1 => libcrypt-2.0.7.so libc.so.6 => libc-2.0.7.so libBrokenLocale.so.1 => libBrokenLocale-2.0.7.so ld-linux.so.2 => ld-2.0.7.so Let's state my particulars up front--I'm a complete Linux newbie. (I installed *everything* onto the system, just to try and avoid having to add things in later. OK, OK, I learned my lesson, next time I reinstall I'll be a bit saner about it.) I tried this once before with JDK 1.2-pre1, and had some problem (can't remember what now). Anybody got a good link I can run off to, or (better yet, from my perspective) willing to hand-hold me through this? The FAQ on the blackdown.org site doesn't seem to have my problem listed there, and the archives didn't turn up anything useful, either. Advance. Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [java-linux] Needing help integrating ORBacus with Java-Linux
The theory in CORBA is this: you always use the particular IDL-to- compiler that came with that ORB. For example, it would absolutely make no sense whatsoever to use Orbix's IDL-to-C++ compiler if you're using Visibroker as the ORB, as certain ORB-specific features (generated, perhaps by the Orbix IDL compiler) would never work inside of Visibroker. Having said that, I suppose it's possible to write 100% "portable" IDL-generated code that could work with any ORB; I've never tried. I'd argue that the first step in getting these two to interoperate is to make sure you use the ORBacus IDL compiler with the ORBacus code. I know that OOC offers precompiled versions of their idl-to-java compiler on their website, just in case you don't want to download the C++ ORB and build it yourself. I'd really start there and see what happens next. Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message- From: Bryce McKinlay <[EMAIL PROTECTED]> To: Nathan Meyers <[EMAIL PROTECTED]> Cc: [EMAIL PROTECTED] <[EMAIL PROTECTED]>; [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Monday, August 09, 1999 10:25 PM Subject: Re: [java-linux] Needing help integrating ORBacus with Java-Linux >Nathan Meyers wrote: > >> Bryce McKinlay wrote: >> > >> > If you're using ORBacus, you should use ORBacus's jidl compiler >> > (www.orbacus.com). You can not use the classes generated by Sun's IDL compiler >> > with ORBacus (or vice-versa). >> >> After diving into the list archives, I find a confusing mix of messages >> about this situation. Could you clarify? Are .java files generated by >> jidl3.1.3 always, sometimes, or never usable with JDK1.2's CORBA >> implementation? > >I know from personal experience that classes generated by Sun's IDL compiler don't >work correctly with the ORBacus CORBA runtime (maybe they do in some circumstances, >but you definatly don't want to rely on it). I guess it is possible that going the >other way (using ORBacus jidl with Sun runtime) does work - I've never tried it. My >impression was that you shouldn't try to do this, however. > >regards > > [ bryce ] > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Error in starting Lookup Service on Linux(RH6.0,i386)
This bug originated, AFAIK, because MS doesn't ship with TCP/IP installed as a network protocol on any of its OSs; Linux shouldn't have that problem, AFAIK. If you can "ping localhost" or "ping " from the local prompt, you've got TCP/IP installed. The "error setting options" of the java.net.SocketException intrigues me--I wonder what options Jini/Java is trying to set that Linux's TCP/IP stack doesn't like? One other question--are you certain rmid is running? It's not silently exiting due to some exception that you don't see? Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message- From: Adil Atilgan <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>; [EMAIL PROTECTED] <[EMAIL PROTECTED]>; JINI-USERS <[EMAIL PROTECTED]> Date: Wednesday, August 11, 1999 8:30 AM Subject: Re: Error in starting Lookup Service on Linux(RH6.0,i386) >Check your Network Status. > >"If you are running on a computer that was not connected to a network, there >is a particularly annoying bug (feature?) of RMI under Windows for which you >must compensate or you will get an ActivationException " > > > >-Original Message- >From: Nagaraj S.B <[EMAIL PROTECTED]> >To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>; >JINI-USERS <[EMAIL PROTECTED]> >Date: Friday, August 13, 1999 1:43 PM >Subject: Error in starting Lookup Service on Linux(RH6.0,i386) > > >>Hi, >>Following is the procedure followed by me for starting Jini Lookup >>service. >>#java -jar /home/jeocs/jini1_0/lib/tools.jar -port 8080 >>#rmid >>#java -jar >>-Djava.security.policy=/home/jeocs/jini1_0/example/lookup/policy.all >>/home/jeocs/jini1_0/lib/reggie.jar http://BSLHOST:8080\reggie-dl.jar >>/home/jeocs/jini1_0/example/lookup/policy.all /home/jeocs/jini1_0/log1 >>public >> >>Then it is giving following error after 4-5 secs. >> >>java.rmi.activation.ActivateFailedException: failed to activate object; >>nested exception is: >> java.rmi.activation.ActivationException: exception in object >>constructor; nested exception is: >> java.net.SocketException: error setting options >>java.rmi.activation.ActivationException: exception in object >>constructor; nested exception is: >> java.net.SocketException: error setting options >>java.net.SocketException: error setting options >> at >>sun.rmi.transport.StreamRemoteCall.exceptionReceivedFromServer(Compiled >>Code) >> at sun.rmi.transport.StreamRemoteCall.executeCall(Compiled Code) >> at sun.rmi.server.UnicastRef.invoke(Compiled Code) >> at >>sun.rmi.server.Activation$ActivatorImpl_Stub.activate(Compiled Code) >> at java.rmi.activation.ActivationID.activate(Compiled Code) >> at sun.rmi.server.ActivatableRef.activate(Compiled Code) >> at sun.rmi.server.ActivatableRef.invoke(Compiled Code) >> at com.sun.jini.reggie.RegistrarImpl_Stub.getServiceID(Compiled >>Code) >> at com.sun.jini.reggie.CreateLookup.create(Compiled Code) >> at com.sun.jini.reggie.CreateLookup.create(Compiled Code) >> at com.sun.jini.reggie.CreateLookup.main(Compiled Code) >> >>- >>What does it mean ?. Where can I get details about these errors ?. >>Thanks in advance, >>Nagaraj S.B. >> >> >>-- >>To UNSUBSCRIBE, email to [EMAIL PROTECTED] >>with a subject of "unsubscribe". Trouble? Contact >>[EMAIL PROTECTED] > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Why won't java 1.2 load an external swing package
Note that when they say, under "How the Java launcher finds Extension classes", "There is no option provided for changing the location of the extension directory." This is technically untrue; there's no *documented* way, but if you specify "-Djava.ext.dirs=...", you can specify a list of directories to use as the extensions directory or directories. Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message- From: Daniel Barclay <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Cc: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Friday, August 27, 1999 11:24 AM Subject: Re: Why won't java 1.2 load an external swing package > >> From: Ondrej Popp <[EMAIL PROTECTED]> > > > >> I tried adding .jar files before rt.jar in the CLASSPATH >> in the hope that the java runtime would use those instead of the >> integrated ones. However this does not seem to work at all. Does anyone >> know why? > >JDK 1.2 uses multiple class paths. See "How Classes are Found" at > http://java.sun.com/products//jdk/1.2/docs/tooldocs/findingclasses.html > >Daniel > > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Where/When is a new release? (Big News!)
I thought I downloaded pre-v3 just a few weeks ago; "java -version" still
reports itself as v2, though.
BTW, there's a libsunwjit.so file in my java distribution, but it's not
marked executable; do I have to make it executable myself to take advantage
of it? ("java" keeps telling me there's no JIT installed.)
Ted Neward
Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here
http://www.javageeks.com/~tneward
"I don't even speak for myself; my wife won't let me." --Me
-Original Message-
From: Rachel Greenham <[EMAIL PROTECTED]>
To: Mark Wielaard <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Tuesday, September 07, 1999 6:22 AM
Subject: Re: Where/When is a new release? (Big News!)
>Mark Wielaard wrote:
>>
>> Hi,
>>
>> I saw that the status page says that there have been some big
improvements
>> in the recent Blackdown port.
>>
>> <http://www.blackdown.org/java-linux/jdk1.2-status/jdk1.2-status.html>
>>
>> > Big news!
>> > We've made significant progress with the problems plaguing the native
>> > threads implementation. Some core parts of the native threads library
>> > have been reimplemented to better utilize Linux threads. We also
believe
>> > we have discovered whyinterrupts were being "lost", and understand how
>> > to work around the problems.
>>
>> But I can not find this new version. Is there a new version already?
>> We have an application that uses JDK1.2 and currently doesn't run on the
>> Blackdown JDK because of Thread/Socket/Interrupt issues so we really like
>> to test this new version.
>
>Odd, I thought pre-v2 was the version that was already out. I think
>that's what I've been running for months...
>
>java version "1.2"
>Classic VM (build Linux_JDK_1.2_pre-release-v2, green threads,
>nojit)
>
>Hmm. Someone's made a boo-boo.
>
>--
>Rachel
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sometimes easy things are hard/impossible in Java
"Box" component? Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message- From: Nathan Meyers <[EMAIL PROTECTED]> To: Kontorotsui <[EMAIL PROTECTED]> Cc: Java-Linux List <[EMAIL PROTECTED]> Date: Thursday, September 09, 1999 11:37 AM Subject: Re: Sometimes easy things are hard/impossible in Java >Kontorotsui wrote: >> >> After extensive experience with Java GUI, mostly by using Swing, I wonder why >> there are hard tasks which can be accomplished very easily and easy ones which >> look almost impossible. >> >> Here are two examples. >> >> I have a grid with 3 buttons in the first row and 2 buttons in the third, I >> wanted to place the third row buttons centered, like this: >> >> XX XX XX >> >> XX XX XX >> >>XXX XXX >> >> but no layout manager allows to do it in a single panel, I had to do several >> panels. > >I've found Swing's Box component great for this sort of job. Yes, you >have to create multiple boxes (this example would require 3 horizontal >boxes inside a vertical box), but they're easy to use, low-overhead, and >the "glue" and "strut" components make it dead-easy to construct this >sort of layout. > >Nathan > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: mutlicast sockets
Somebody made mention, in the JavaSpaces list, that if you want to do
JavaSpaces (which uses Multicast sockets, as I understand it), that you have
to have multicast support compiled into the kernel.
It's hearsay and rumor, but maybe it'll help?
Ted Neward
Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here
http://www.javageeks.com/~tneward
"I don't even speak for myself; my wife won't let me." --Me
-Original Message-
From: Wright Robert B Civ 96 CG/SCWDE <[EMAIL PROTECTED]>
To: '[EMAIL PROTECTED]' <[EMAIL PROTECTED]>
Date: Friday, September 10, 1999 7:24 AM
Subject: mutlicast sockets
>i'm getting a "java.net.SocketException: No such device" when executing
this
>code:
>
>MulticastSocket socket = new MulticastSocket();
>InetAddress address = InetAddress.getByName("230.0.0.1");
>socket.joinGroup(address);
>
>through jdb i discovered that the joinGroup method is throwing the
>exception. i may not have my linux box configured correctly. any ideas?
>
>thanks,
>
>rob
>
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: package problem
"java" takes the *class* name of the class to execute, not the *file* name
of the classfile; try this:
java foobar.Foo
instead.
Ted Neward
Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here
http://www.javageeks.com/~tneward
"I don't even speak for myself; my wife won't let me." --Me
-Original Message-
From: Roland Silver <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Saturday, September 11, 1999 5:57 PM
Subject: package problem
>I have a problem with packages that has me stumped. I've boiled it
>down to two simple classes, defined in Foo.java and Bar.java; both
>are in the directory /home/rollo/Java/foobar on my i386 Linux machine.
>
>//Foo.java
>package foobar;
>import java.util.*;
>public class Foo {
> public Foo() {
> Bar bar = new Bar();
> }
> public static void main(String[] args) {
> System.out.println("Foo here");
> }
>}
>
>//Bar.java
>package foobar;
>public class Bar {
> public Bar() {
> Foo foo = new Foo();
> }
>}
>
>Current directory is /home/rollo/Java/foobar, and
>CLASSPATH specifies the following three paths:
> /usr/local/jdk117_v3/lib/classes.zip
> /home/rollo/TIJ/exercises
> /home/rollo/Java
>
>The command
> javac Foo.java
>compiles OK, as does
> javac Bar.java
>putting Foo.class and Bar.class in the foobar directory, but
> java Foo
>complains:
> Can't find class Foo
>
>I'd really appreciate it if someone can help me with this problem!
>* What am I doing wrong?
>* How do I fix it?
>
>I am running Blackdown JDK version 1.1.7-V3 on an i386 machine under
>Red Hat Linux 6.0.
>
>-- Roland Silver <[EMAIL PROTECTED]>
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Class not found
CLASSPATH is assumed to be "." unless another CLASSPATH is given; even then, classes/jars in the Extensions directory are automatically present. Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message- From: Thomas M. Sasala <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Cc: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Wednesday, September 15, 1999 6:10 AM Subject: Re: Class not found > > With JDK 1.2, you don't need to specify the classpath >when you run an app. That might be your probelm. > > >Christian Posse wrote: >> >> Hi, >> >> sorry for maybe asking an obvious question. I have seen that problem >> posted in >> the past but I can't find the answer to it. >> >> I just upgraded to jdk 1.2pre_v2 (glibc 2.1) from 1.1.7_v3. >> I have RH 6.0. What worked before under 1.1.7 v3 does not work any >> more. >> (The problem is not Kaffee, I did not install it). >> >> set MYCLASSPATH= .: >> java -classpath $MYCLASSPATH com.dir1.dir2.MyApp >> >> Exception in thread "main" java.lang.NoClassDefFoundError: >> com/dir1/dir2/MyApp >> >> any clue would be greatly appreciated. Thanks in advance. >> Christian >> >> -- >> To UNSUBSCRIBE, email to [EMAIL PROTECTED] >> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > >-- >+---+ >+ Thomas M. Sasala, Electrical Engineer [EMAIL PROTECTED] + >+ MRJ Technology Solutionshttp://www.mrj.com + >+ 10461 White Granite Drive, Suite 102(W)(703)277-1714 + >+ Oakton, VA 22124 (F)(703)277-1702 + >+---+ > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Reinstall 1.1.7
Rachel--
CLASSPATH is only set by the launching process *if* no CLASSPATH is set in
the environment. Fortunately, however, the runtime library (rt.jar) and any
extensions (in jre/lib/ext) are managed by a separate ClassLoader, so those
don't need to be on the CLASSPATH--only your "user classes".
(Look at lines 136-138 in "launcher/java.c"; the file comes as part of the
'src.jar' source-code file that comes as part of JDK 1.2:
/* Set default CLASSPATH */
if ((s = getenv("CLASSPATH")) == 0) {
s = ".";
}
)
Ted Neward
Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here
http://www.javageeks.com/~tneward
"I don't even speak for myself; my wife won't let me." --Me
-Original Message-
From: Rachel Greenham <[EMAIL PROTECTED]>
To: Surjan Singh <[EMAIL PROTECTED]>
Cc: Tom Whitcomb <[EMAIL PROTECTED]>; [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
Date: Friday, September 17, 1999 9:58 AM
Subject: Re: Reinstall 1.1.7
>Surjan Singh wrote:
>>
>> I always thought that you _don't_ need the classpath. If you have your
>> PATH variable correctly (for the release you want to use, e.g.
>> /usr/local/jdk1.2/bin) then you don't need to worry about anything else.
>>
>> The only reason you would want your CLASSPATH set, is if you have other
>> Java libraries you want your compiler/java/appletviewer to see.
>>
>> If you want to remove a release of the JDK just remove the folder and
>> its contents.
>
>It's correct that you don't need to set the classpath for JDK1.2. It
>works in a much more sane and sensible way than it did before. If you
>have a CLASSPATH set when you invoke the JVM, it is *added* to the
>default one needed to make Java work. IIRC in JDK1.1 this wasn't the
>case: If you had your own libraries to add, you had to define the whole
>classpath.
>
>--
>Rachel
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [Problem with JNI_CreateJavaVM()]
Um. I thought he was just asking for help. He posted the code that was
breaking, it's a short snippet, and he was fairly clear about the
environment he was using; that's already more than *most* of the posts I've
seen come across this, and other, mailing lists.
Pierre, I'm running RH5.2 and Blackdown 1.2-pre-v2, and the code compiled (I
had to add the "-I" include directives, though--your g++ command-line below
was missing those, but that would yield a compile-time error, and not a
run-time segfault) and ran without a hitch; I'd check to make sure the JDK
installation itself is good, in case there's some configuration issue at
stake here. (Juergen, I didn't need the LD_BIND_NOW directive to get this to
run--is it really required?)
Ted Neward
Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here
http://www.javageeks.com/~tneward
"I don't even speak for myself; my wife won't let me." --Me
-Original Message-
From: William D.Webb <[EMAIL PROTECTED]>
To: Pierre Heroux <[EMAIL PROTECTED]>;
[EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Monday, September 20, 1999 12:27 PM
Subject: Re: [Problem with JNI_CreateJavaVM()]
>Dude, you want us to write the code for you as well??
>
>Pierre Heroux <[EMAIL PROTECTED]> wrote:
>> I have a "Segmentation fault (core dumped)" message when I execute the
>> following source.
>>
>> #include
>> #include
>>
>> int main()
>> {
>> const int nOpts = 4;
>> JavaVMOption options[nOpts];
>> int ii=0;
>> options[ii++].optionString = "-Djava.compiler=NONE";
>> options[ii++].optionString = "-Djava.class.path=.";
>> options[ii++].optionString = "-Djava.library.path=.";
>> options[ii++].optionString = "-verbose:jni,class,gc";
>>
>> JavaVMInitArgs vm_args;
>> vm_args.version = JNI_VERSION_1_2;
>> vm_args.options = options;
>> vm_args.nOptions = nOpts;
>> vm_args.ignoreUnrecognized = JNI_TRUE;
>>
>> JavaVM *jvm;
>> JNIEnv *env;
>> jint res = JNI_CreateJavaVM(&jvm, (void **)&env, &vm_args); /*debugger
>> message : Program received signal SIGSEGV, Segmentation fault.
>> 0x401d5ca8 in main_arena ()*/
>> if(res < 0)
>> {
>> cerr << "Can't create Java VM" << endl;
>> exit(1);
>> }
>> cout << "JNI_CreateJavaVM() is OK" << endl;
>> jvm->DestroyJavaVM();
>> }
>>
>> g++ -g simple.cpp -L/usr/local/jdk/jre/lib/i386
>> -L/usr/local/jdk/jre/lib/i386/classic
>> -L/usr/local/jdk/jre/lib/i386/native_threads -ljava -ljvm -lpthread
>> -lhpi -o Simple
>> Simple
>>
>> I use java version "1.2"
>> Classic VM (build Linux_JDK_1.2_pre-release-v2, native threads, nojit)
>> on RedHat5.2
>>
>>
>> Thanks a lot,
>> Pierre
>>
>>
>>
>>
>> --
>> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>> with a subject of "unsubscribe". Trouble? Contact
>[EMAIL PROTECTED]
>
>
>
>Get free email and a permanent address at http://www.amexmail.com/?A=1
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: "proper" place for the .so native libraries
Does "had good luck" mean without having to modify LD_LIBRARY_PATH? Or
/etc/ld.so.conf?
Ted Neward
Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here
http://www.javageeks.com/~tneward
"I don't even speak for myself; my wife won't let me." --Me
-Original Message-
From: dave madden <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Thursday, September 30, 1999 3:04 PM
Subject: Re: "proper" place for the .so native libraries
> =>From: Brett Smith <[EMAIL PROTECTED]>
> =>...
> =>Where is the "proper" location for the *.so native library files?
>
>I've had good luck putting them under:
>
> .../jdk-1.2/jre/lib/i386/
> appropriate platform here!
>
> =>One JNI tutorial said the following:
> =>LD_LIBRARY_PATH='pwd'
> =>export LD_LIBRARY_PATH
>
>The first problem is that you've used forward-tics instead of
>back-tics. You should have said:
>
>LD_LIBRARY_PATH=`pwd`
>export LD_LIBRARY_PATH
>
>This has a better chance of working, but LD_LIBRARY_PATH is
>notoriously cranky. It's best if you can set things up so you don't
>need it.
>
> =>Also, does the blackdown install set a LD_LIBRARY_PATH? or need one?
if
> =>so, where is it located?
>
>The "java" shell script sets LD_LIBRARY_PATH appropriately when you
>run it; if you had a LD_LIBRARY_PATH set previously, then it
>{pre?app?}pends the directories it needs. (Same with CLASSPATH, I
>think.) I've had trouble with both, and now try my damndest to avoid
>having to set both.
>
>d.
>
>
>--
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Be warned--emails with "LINKS.VBS" are a virus attack
This is not a spam; over the past week, I've received several emails containing the text "Check this" with an attachment, "LINKS.VBS", which several email virus-scanners have flagged as a virus. It's intended as just a warning/FYI; I don't want to perpetuate a myth, but this looks real--Norton/Symantec's product as well as a few others have flagged and caught this. Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Jar file Access to images
Rob-- Check out URLClassLoader, and email me privately if you don't get it from there. URLClassLoader allows you load classes, resources and other "stuff" from .jar, .zip, or subdirectory (depending on the URL(s) you pass in, of course) without modification to your client code. Does that solve the issue? Ted Neward Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message- From: Robert Simmons <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Monday, October 18, 1999 4:36 AM Subject: Jar file Access to images >Greetings, I cross posted this to about 3 newsgroups and havent gotten a >reply yet so I figured I would post here because i need it answered. So dont >flame me for posting a general question. > >I have an application that I intend to package as a jar, with its images in >the Jar. How do I construct a URL object to load those images into IconImage >instances from the LOCAL jar file (ie, the file the program is running in.) >Further, is there a generic way I can have it load from either the >subdirectory, if the app is not run from a jar, such as when Im doing >interim builds during programs, and from the jar in a distribution. > >- TIA >-- Rob > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: performance
No--don't do that. DON'T convince yourself that you need to avoid using a particular language feature "because of its impact on performance". Use the language feature, profile the app, THEN worry about performance. There are dozens of things you can do to better improve app speed (including using something OTHER than Java, like C++ or asm) without sacrificing a key part of the Java language (exception handling, that is). Get it working, get it right, THEN get it fast. --Martin Fowler, "Refactoring" Ted NewardPatterns/C++/Java/CORBA/EJB/COM-DCOM spoken herehttp://www.javageeks.com/~tneward "I don't even speak for myself; my wife won't let me." --Me -Original Message-From: ³Â·å <[EMAIL PROTECTED]>To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>Date: Tuesday, October 19, 1999 8:14 PMSubject: performance hi linux java fan I want to know if the finally will affect the performace of the program I know that if the catch happen the program will affect the performance ,but I don't konw the finally will do what Thanks yours simple fan
Re: easy question
Wow, are *you* in for a mess. :) OK--this is just a suggestion, and half this stuff is being made up on the spot, but this is how I'd structure your directory. Assume your project is called "Project" Project +--src +--cls These two are your Java "source" and "class" directories, respectively. Assuming you're building with Java 2, you can do the following quickie script to build all the java code in the 'src' directory (from the project root) and have the compiled .class files show up in the 'cls' directory: (where $(PROJECT_HOME) = the "project root" directory) find $(PROJECT_HOME) -name '*.java' -print > $(PROJECT_HOME)/.sourcefiles javac -srcpath $(PROJECT_HOME)/src -d $(PROJECT_HOME)/cls @$(PROJECT_HOME)/.sourcefiles Add a $(DEBUG) or other flags-type environment variable to switch debug/nondebug modes on or off; I'd put this into a makefile that also controls the JNI/C compilation (discussed in a second). This script doesn't perform full dependency-checking, so you'll probably want to do a 'rm -r *.class' from the $(PROJECT_HOME) directory, but it's quick and dirty and doesn't require .java-file dependency tracking, at the cost of longer compiles. Adjust as necessary for your own needs. Then, under the same directory, do Project +--c_src +--inc This is, of course, where the JNI/C code goes, and the JNI headers go into the $(PROJECT_HOME)/inc (or 'include', if you prefer) directory. Depending on what #include syntax you prefer (I like '#include <$(PROJECT)/file.h>', as opposed to just '#include "file.h"', so I know where each include file is coming from), you can rename the 'inc' directory to $(PROJECT) and set the compiler's "-I $(PROJECT_HOME)" flag. That way your JNI/C code is explicitly documenting where it's getting the JNI headers from. Do the usual make/makedeps with this, either in the same makefile that handles the java code (ideally), or in a makefile that's fired from the java project's makefile. For the HTML/JavaScript code, I'd do Project +--public_html +--cgi_bin javascript I'm less familiar with how to arrange HTML files in a source-controlled project, since I've not done much with them recently, but since HTML requires no compilation, you're really just trying to create the appropriate HTML structure here, nothing more. For debug/release deployment, I'd set up your makefile/build-script to compile everything, with $(DEBUG) flags turned on or off (depending on whether you do 'make -DDEBUG' or not, for example), and have the generated executables, .jar files and .zip (of the HTML files) files "released" (that is, copied) to one of two directories in the same project directory: Project +--alpha +--release Ideally, you'd be able to then run a script from within each directory to "deploy" the code (to a Web server or simply to another location on your filesystem) as appropriate. As far as RCS goes, either just store the files in a separate location, or use CVS (my preference) and store them on another server--it doesn't affect how the files themselves are laid out on disk, IMHO. It's rough, crude, and could probably use several iterations of refinement. I *more* than welcome comments--I'd love to hear anybody's ideas on so complex an application of "proper build procedures". Ted Neward CTO, WebRaiser ( http://www.webraiser.com ) Java Instructor, DevelopMentor ( http://www.develop.com ) Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward -Original Message- From: Solomon Douglas <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Cc: David Harrison <[EMAIL PROTECTED]> Date: Monday, November 08, 1999 11:42 AM Subject: easy question >I apologize that this isn't a specifically Linux-related question. > >I'm working on a project involving several servlets and other classes, >organized in multiple java packages, and I don't think I've figured >out the ideal way to structure my directories. This is a web-based >application, so in addition to the server-side Java classes, I also >have HTML and javascript files to deal with. Thank goodness there are >no applets (yet), because that would make it even more confusing! I >have the following files: > >- Java source code files and their associated RCS files >- C source code (and RCS) for JNI methods >- the development version of the shared libraries for the JNI methods >- the production version of the shared libraries for the JNI methods >- the development version of the Java class files >- the production version of the Java class files >- static HTML and JavaScript source files (development versi
Re: code question
Absolutely not true! Posting answers (and, yes, being wrong once in a while) is the best way to expand your knowledge! Trust me (from personal experience, and LOTS of it), you'll NEVER forget the answers to the questions you're wrong on! :) Ted Neward CTO, WebRaiser ( http://www.webraiser.com ) Java Instructor, DevelopMentor ( http://www.develop.com ) Author: "Core OWL 5.0", "Advanced OWL 5.0", "Server-Side Java" http://www.manning.com ) Patterns/C++/Java/CORBA/EJB/COM-DCOM spoken here http://www.javageeks.com/~tneward -Original Message- From: Martin Kavalar <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Tuesday, November 16, 1999 6:09 PM Subject: Re: code question >Hey! >Guess it was a big mistake trying to post a reply for the first time. Theres >like 49 answers telling me that im wrong. Sorry bout that! >martin > > On Tue, 16 Nov 1999, >corey wrote: > > Hello! >> > You cant read anything on the local machine with java, just from the server >> > that contains the class files. This is because of secuity. >> > cya >> > Martin >> >> This is not true. As long as the security model in the >> browser is set properly you can read and write files >> to the local drive just as if you were running an >> application. >> >> --Corey >> >> > On Mon, 15 Nov 1999, >> > Yohans Mendoza wrote: > hi all, > >> > > is it possible to open, read and write files froma an applet? >> > > >> > > I know it's off topic, but I really need to know that. >> > > >> > > TIA >> > > >> > > --Yohans >> > > >> > > ~ >> > > Yohans Mendoza Unix Administrator >> > > [EMAIL PROTECTED] Sirius Images Inc. >> > > http://www2.utep.edu/~yohanshttp://www.sirius-images.net >> > > ~ >> > > >> > > >> > > -- >> > > To UNSUBSCRIBE, email to [EMAIL PROTECTED] >> > > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >> > >> > >> > -- >> > To UNSUBSCRIBE, email to [EMAIL PROTECTED] >> > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >> > >> >> /* Corey Brown (WB0RXQ): 20m, 15m, 2m(146.82) 70cm(443.65) */ >> /* RAPID/FASTAR Development District */ >> /* AT&T NCS | [EMAIL PROTECTED] */ >> /* Alpharetta, Ga 30005 | [EMAIL PROTECTED] */ >> /* (770)750-8071 | New rays from an ancient sun (JS) */ >> >> >> -- >> To UNSUBSCRIBE, email to [EMAIL PROTECTED] >> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [Partially OT] A Java parser - part 2
What about EBNF? That's pretty standardized, no? Ted Neward http://www.javageeks.com/~tneward On Sun, 21 Nov 1999, Michael Emmel wrote: > Vincent Risi wrote: > > > Andreas Rueckert wrote: > > > > > Hi! > > > > > > On Thu, 18 Nov 1999 Paul Mclachlan wrote: > > > > > > The next problem occured when I wanted 2 parsers to use the same base classes > > > for the AST. Never managed to do that, and I couldn't patch JavaCC, since there > > > were no sources available. > > > > TLA's are fine if you know what they stand for, what is AST? > > Abstract Syntax Tree : ) > Whats a TLA ? > > Umm yes I was looking at antltr it allows sub classing. > It looks like I like the power of antlr and the syntax of Javacc : ) > It should not be too hard to write a Javacc--> antlr preprocessor. > Then antlr can accept Javacc grammers ... > Anyone intrested > > What would be really nice god forbid is if there was some sort of standard syntax > for > java parser and lexer generators. Life would be cool if you could publish a > "standard" grammer. > > Mike > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Question: JAVA_DEBUG environment variable
Look in the src.jar file, under launcher.java.c; it's coded explicitly in there. launcher/java.c is the standard Sun boilerplate for all of the "native" tools in theJDK (javac, java, ...). Ted Neward http://www.javageeks.com/~tneward On Tue, 23 Nov 1999, Jacob Nikom wrote: > Hi, > > In one of his messages Nathan Meyers mentioned JAVA_DEBUG environment > variable which should be set up for debugging. I looked at debugging > documentation but could not find any mention about JAVA_DEBUG variable. > What kind of value JAVA_DEBUG variable should have? Do I really need it > for java debugging? Is it Blackdown JDK specific variable? > > Thank you, > > Jacob Nikom > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: another possibility wrt the press-release.
I thoroughly agree with Nathan's assessment--it makes no sense to believe that Sun and/or Inprise somehow *want* this bad publicity. Quite frankly, given the nature of the lack of responsiveness of the Blackdown team themselves (no offense, guys, but this is simply personal observation) to various postings on this list (including one of my own, some time back) offering to help fix some bug or another, I'm fully ready to believe the "there was no response" story--it reeks with credibility. Simply pointing fingers and creating conspiracy theories does nothing to advance our position as Linux/Java users. This is the wonderful thing about the capitalist system--if you don't care for Sun's or Inprise's attitude towards us, or any other community, stop using their products. Use a competitor's, and encourage your friends to do so, as well. In time, if enough people agree with you, then either the commercial company will change their attitude, or else they will go bankrupt and die. It's far more likely that they'll do some sort of adjustment to their policies before allowing themselves to go under--witness what's going on with the tobacco companies in this country, as evidence. For my money, I'll download the Inprise product (I have a particular beef with Borland as a company--why'd you kill OWL just before my OWL books came out?--but that's neither here nor there), evaluate it, and see if it works for me. If the Blackdown port is better, great. If the Inprise port is better, great. If I do my own port, and I prefer that, instead, great. Whatever solution solves the problem, is what works for me. Ted Neward Java Instructor, DevelopMentor ( http://www.develop.com ) http://www.javageeks.com/~tneward -Original Message- From: Nathan Meyers <[EMAIL PROTECTED]> To: Paolo Ciccone <[EMAIL PROTECTED]> Cc: Mike Ajemian <[EMAIL PROTECTED]>; [EMAIL PROTECTED] <[EMAIL PROTECTED]>; [EMAIL PROTECTED] <[EMAIL PROTECTED]>; [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Friday, December 10, 1999 8:21 PM Subject: Re: another possibility wrt the press-release. >Paolo Ciccone wrote: >> >> On Fri, Dec 10, 1999 at 02:43:47PM -0500, Mike Ajemian wrote: >> >> > to our postings" crap). The community will suffer as 2 products that >> > should be on the same code base are now completely separate products >> > (which Inprise likely wants since it doesn't dilute their product with >> > an Inprise-Blackdown label.) >> >> Sorry to disappoint your "conspiracy" theory, I really mean it in a >> good way, but we really don't care who provides the JDK. If Blackdown >> tomorrow shows a JDK that outperform ours we will be happy like hell, >> it will make JBuilder and JDataStore run faster. Great ! Same thing if >> it comes from Sun or IBM. > > >I'd like to add, as one who has no connection with Inprise or Sun, that >I continue to be astounded at accusations like this. There's a real >simple story here: Inprise had a critical business need, they approached >Blackdown about collaboration (that's called "due diligence"), Blackdown >didn't return their calls, so Inprise did some work on its own. I don't >understand the need to explain what happened with complicated conspiracy >theories. > >Let's consider an alternate theory. Why has Blackdown visibility been >poor? Why did they fail to respond to Inprise? It's obvious! Blackdown >wants to own the Java/Linux franchise and make its members fantastically >wealthy on a Blackdown IPO. Ridiculous? Of course. But it makes as much >sense as some of the other theories flying around. > >Another insight to consider: Linux is hot! Everything having to do with >Linux is hot. Why in the world would organizations like Sun or Inprise >make any effort to repudiate the Linux community? When Sun failed to >acknowledge Blackdown in its PR, it screwed up. In fact, it screwed up >even more by not making a *big deal* of the Blackdown connection. It is >a PLUS, not a MINUS, to associate your name with the Linux community, >and these bizarre and complicated explanations of why Sun and Inprise >would want to hide that connection just don't wash. > >Finally, consider that the proof is in the pudding. What has come out of >this horrific conspiracy? Well, for the first time, Linux users have a >JDK with a working JPDA, and a JIT that's faster and better than the >long-broken sunwjit. And what do you have to pay to use this new JDK? >Nothing. > >So what happens next? Let's see, Inprise has a working JPDA and a great >JIT, and Blackdown has solved the native threading problems. Maybe it's >time to move past the accusations and conspiracy t
Re: appreciating sun
Well said, all around. Ted Neward Java Instructor, DevelopMentor ( http://www.develop.com ) http://www.javageeks.com/~tneward -Original Message- From: noisebrain <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] <[EMAIL PROTECTED]> Date: Monday, December 13, 1999 1:07 AM Subject: appreciating sun >I find some of the sun-bashing both counter-productive and >slightly embarrassing. > >Personally, I *want* new versions of the jdk to be released >in synch with the Windows and Solaris releases, rather than >a year later. I want Sun to support Linux rather than >think that we're all a bunch of crazed flamers. > >Sun made a mistake, but they have apologized now in several places. > >The "Sun will never be our friend"/"Sun is an evil corporation" >speak strikes me both as irrelevant and as a bit immature. >Despite open source, I doubt that your local supermarket will >be free anytime soon. Corporations will continue to exist. > >Sun is the creator of Java, and they don't have to give it away to >anyone. They certainly don't have to port it to Linux... but they >chose to put some little bit of energy into this. The community >does appreciate Blackdown. Why resent Sun for helping us? > >"We don't need Sun"/use Kaffe/etc. No, I think we do: Java wouldn't >have become a standard without Sun's backing, and Java wouldn't >be worth my mindshare if it wasn't a standard - there are other >more advanced/interesting languages out there, but most people >can't afford to re-invent the wheel in an advanced but obscure language. > Further, Sun is driving Java development, Kaffe isn't. No one >is preventing the open source community from pushing Java ahead, >or from developing a better-than-java successor, etc. Historically >though Open Source seems to do well at re-implementing well established >standards. One of my gripes with Linux is that it has a 70s mindset - >anything that is late 90s (streaming media, videoconferencing, java) >will be part of your packaged RH cd years after it appears on win/mac. > >"Sun doesn't get open source." Well, Sun is a corporation, but >among corporations they seem to get it more than most. McNealy(?) >said last week that the software is becoming free, they're releasing >source to Solaris (albeit under their license), etc. > >Sun has apologized, so has Inprise, let's accept their apologies >for a change? > > > > > >-- >To UNSUBSCRIBE, email to [EMAIL PROTECTED] >with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
