But why are they inventing so-called "inefficient" code? Because it's really EFFICIENT -- for developers. It's also easier to maintain.
Java/XML-RPC/etc. are all software inventions that make it easier to develop sophisticated programs.
Actually Java/XML-RPC is kinda stupid. It's aycronimal-pixie dust. I hope Java shops don't get caught up in it too much. It has it's place though.
What's wrong with DataOutputStream and DataInputStream? Their trivial to use and quite elegant. And if you want platform interoperability just write your own DataInputStream/DataOutputStream-like classes that write all values in network byte order like:
Actually, you can just dump the Streamish interface and have your data-holding classes implement similar:
int writeYourDataHere( byte[] dst, int off ) throws IOException {
off += writeInt4(num_apples, dst, off); ...
kind of stuff. You can write very serious networking software using this technique. Network latencies what they are make this as fast as native code. Because it's network byte order you have platform interoperability. It's simple once you write the writeXxx methods and *very* efficient.
...replication will make sure you have the latest copy of the data (as of the last time you were online) that you can work with and modify to your heart's content locally. Then the next time you connect to the net, it gets propped to the cloud...
Well if this is in fact the case I can say with certainty that it will definately not work. If I so much as write myself a little note in Word it's gonna quitely be uploaded to a server in the background! Word docs and pics and stuff are big. The whole thing hinges on ultra-high-bandwidth. MS' cook'n up a receipe for disaster.
I think it would be great if M$ used GPL'd software. It would lend a tremendous amount of credability to the movement and inhearently ease the interoperability between UNIX and Windows. The greatest danger for UNIX and friends is that Windows continues to deliberately and successfully prevent integration with UNIX. Let's face it folks, M$ has the desktop locked up tighter than a dolphins ass(I use Linux mainly but I have nothing against Windows as a client platform). Mac OS X, Be, GNOME, or KDE are still riddled with show stopping issues(the lack of a competitive "Office" like software is the biggest). Besides I don't want a windows clone(I use WindowMaker).
So let them use GPL software. Maybe it will ease it's acceptance within the company and help close the gaps. Then one day the truth will come out and we'll all be better off. You think it will really help M$ be competitive. They don't need it. They have clueless IT managers snowed several times over.
Let'em have it. Can't fight it any more than they can stop the whole OSS movement anyway. It's only a matter of time.
Who, matrox? The matrox install is trivial. You just copy the.o into the modules directory. It's the XFree install thats involved. I think that will always be the case. Most people are better off just waiting for it to be provided in their favorite distro.
Can someone shed more light on the status of the G450 and XFree86 support?
It works. I've been running it for about a month and a half. Works pretty well too. I can do rediculous resolutions and 32bit and all that but I just run 1280x1024 at 24. My machine did lock up *once* but it was when I was just starting to set up the new box and crazily tweeking all sorts of stuff.
It's a pain to install though. That is XF4.0.2 I mean. I suspect XF4.0.3 is no different. Actually this release looks like some relativly minor updates. To install you have to pull down like 25MB in 12 files and run there installer. Good luck getting your old X back though. Cross your fingers and toes for that one(actually it wouldn't be that hard to re-install an older version and might be just a matter of moving some dirs back in place). After that, get the latest drivers from matrox and generate or get a config and help from others on the matrox linux form.
I didn't try 3d accel and DRI and GLX or other acronyms but I've read about people doing all that stuff and dual head, etc. If you like hacking though you'll have a blast!
Rogers also points out that ShareSniffer only locates open shares, it doesn't access them.
This is not true. The only way to determine if the share can be connected to without a password is to try the SMB_COM_SESSION_SETUP_ANDX message with a null password which transpires *after* session establishment or other netbios nameservice and session service operations and *after* dialect negotiation. This would be like checking to see if the door on someones house is locked by walking up and turning the knob and opening it a little.
I think that would be considered just as illegal as walking in a looking around.
Funny thing is that Microsoft renamed the Windows networking protocol to the Common Internet FileSystem (CIFS). Perfect:~)
But the second reason - which you miss - might be more important - fewer errors can be detected at compile time... they would only turn up at runtime, or worse, end up as hard to detect bugs. Moreover, the runtime may fail in someplace other than where the error occured. For example, let's say you have a bunch of "polygon" objects in a linked list, and you mistakenly put a "circle" object in that list as well. Much later, you're traversing the list and expect to find a polygons, but instead you find a circle. Type error!
I've recently been writing a bit of c whereas for the past 3 years it's been totally Java. Now I find in practice that exactly the opposite of what you say is true.
If something compiles in Java it means it's pretty solid. I thuroughly abuse the compiler output by just typing and then solve the compile time errors one by one rather than spending a lot of time thinking about types and such(to an extent).
But if something compiles in c it means exactly nothing. It's much more likely that you reference a pointer that points somewhere you didn't expect than accedentally put a circle object into something that's supposed to take polygon objects. What's the difference between this and a linked list data structure in c that takes void *?
Can anyone connect to any host of microsoft.com? I just got a call from someone who cannot connect to microsoft.com and I can't either. Is something up with that domain? Ok, which one of you script kiddies is responsible for this?
But are the Mozilla libs truly shared? Meaning are they re-entrant or whatever is needed to prevent them from simply being copied in memory? I have a feeling that they are not but I could just be spreading lies.
One of the fundamental units of a Microkernel architecture is a uniform network/port based communcations subsystem.
That's one way of implementing a microkernel, but it's hardly the defining feature.
I don't recall calling it a "defining feature". I called it a "fundamental unit".
I don't know a lot about it
So perhaps you shouldn't be stating opinion as fact, hmm?
a) I didn't state anything as fact. b) you have chopped off the second part of the sentence to remove the context of what I am saying which is that CORBA would not suffice as a primary means of interprocess communication for an operating system.
but it has NOTHING to do with CORBA.
One can implement a microkernel using CORBA as the primary means of IPC.
You _could_ but I'm telling you that it would be unbearably slow.
Certainly CORBA services might run as any other Microkernel service but this would be hurrendously slow to use_as the primary means of inter-proccess communication.
I suppose you can provide us with a link to a paper where this has been thoroughly examined? Or are you just spouting nonsense again?
I'm spouting nonsense?! You obviously know nothing of CORBA. First of all it's founded on sharing objects. These are not simply structs but full blown objects as in Object-Oriented objects. Do you know what it takes to serialize an object for transmission over a network? It's slow. Even if they used shared memory to get around serialization of local objects there would still be many issues related to the dynamic binding of methods and other OO stuff that is invariably slow.
If the ORB is in the kernel (if it's not, then it's not really the primary means of IPC now, is it?) and designed to take advantage of that situation (through the use of virtual memory manipulation and direct access to the address space of both processes) then it should be able to provide adequate performance, and a small performance loss can easily be made up for with a dramatic increase in flexibility.
Even if you don't want to use RPCs, you can still use things like CORBA to formalize the structure of the messages that you pass and abstract the details away from the programmer.
Think about what your saying. You're going to run a CORBA ORB _in_ the kernel as the primary means of communication for an operating system?! I'll beleive it when I see it.
hate to be pessimistic, but with all the failures, layoffs, share prices plummeting...i hope Sun has a big war chest.
Are you kidding? Sun's stock actually went up during most of the plummeting and quite frankly, yes, I'd be willing to bet they have an awsome war chest.
Over a remote X session Nautilus is totally worthless and I'd say a large chuck of Solaris desktop users are doing remote sessions a la eXceed. Can you downgrade that thing to 256 colors? I hope so because over X it takes a LONG time for that thing to paint all those pretty icons. Doesn't it use Mozilla libs for crips sake!
I wish someone would focus on the fundamentals rather than making it look great. Someone should make a file manager that works as well as WINFILE.EXE the old Windows file manager!
I have been writing a Java library that we originally were going to license as GPL. Someone pointed out that the whole project would be mearly an acedemic exercise because users would no doubt be using non-GPL code for just about everything else that might call our library in which case they would be in violation. So we switched to LGPL for genderal industry compatibilty. Would you suggest this is the wisest license for Java software?
Similarly, with LGPL could someone simply change access to constructors and methods to "public" to circomvent package protection and allow them manipulate/extend the library from their own proprietary packages? This would not be suitable reciprocation IMO. If not, LGPL works perfectly for Java because I can excercise the license through package protection:~)
The HURD is apparently going to offer CORBA as the primary means of inter-proccess communication.
This is thuroughly false information. One of the fundamental units of a Microkernel architecture is a uniform network/port based communcations subsystem. I don't know a lot about it but it has NOTHING to do with CORBA. Certainly CORBA services might run as any other Microkernel service but this would be hurrendously slow to use as the primary means of inter-proccess communication.
I agree. XML is too verbose and not ideal in many cases. However I would like to see a little more effort go into making these little persistant stores peppered throughout/etc a little more accessable. Using the file system to organize your data is not a suitable solution. Perhaps one could examine what all these configs have in common and come up with 5-6 standard configuration file formats. Get Apache, Samba, exports, fstab, named.hosts, sendmail.mc,....to all use one of these common formats. Now some would be terrribly gerneral(ie sendmail.mc is just preprocessor input) but others might use tagging...etc. I haven't thought about it enough but certainly there must be some compermise that would be acceptable to the application developers. There's just no effort going on in this area at all right now and there should be.
Never happen. Think about what syslog.conf would look like in XML(beleive me, I tried it).
I agree however that setting and retriveing persistant data stored as human readible conf files via a programmable interaface would be a big plus. The main problem is that the application developers don't give a damn about how programmatically accessable the data in their configs are. Someone would have to write parsers which is a major pain in the ass if the application's requirements change. I think someone would have to convince the applicaiton developers to require that the application itself actually use your programmatically smart configuration file component. Actually I think a lot of applications would welcome this. It takes off a little burden and presumably would cleanup the interface.
Then you could add support for XML separatly so that users have a choice(at a small cost in performance as your duplicating changes in two places).
I must say it's not a bad portrait of the state of Linux affairs at the moment. But it's obviously a one sided view of the situation and is meant to be taken as such. It's kind of a troll basically. It's just a way for the guy to vent behind someone elses name(like that Anonymous Coward guy does here on Slashdot:~). This part is what crosses over the objectivity line though:
I think that sooner or later, these Linux fools will self-destruct without our influence. We'll see who has the Mardi Gras celebration then. In the meantime, I hope they enjoy their 5.6 percent of the desktop. It won't last.
Oh, I think it will last. Even when the zealots wake up and go back to using windows the fact that Linux is a superb piece of software will remain. A LOT of techie people use it for serious business reasons and otherwise and it's only getting better. I'm not impressed with KDE and GNOME. I think the real advances will be in infrastucture. Stuff like browsing into an ftp site by simply cd-ing into/net/someftp.site.com/pub or a library for retrieving/setting all the possible system properties in files like named.hosts, httpd.conf,..etc so that persistance is abstracted from the logic. Then some _real_ configuration tools can be built. I just totally made up two(possibly stupid) examples up off the top of my head but I think it will be clever infrastructure stuff like that that will really make Linux shine. I don't wan't KDE or GNOME. Does anyone find it ironic that the're writing Windows 95 for UNIX?!
you do have to be carefull though. Someone setup a box for g++ and Samba prevented the browse lists from working properly. It became a master browser but couldn't find the domain master browser so some hosts on that subnet could not be found in network neighborhood. I looked at the config and didn't find anything wrong. I asked the Samba guys and didn't get much of an answer. I never heard about it again from the IT guys but after being the local Linux advocate it's a blow to the credability. I won't be advocating the default Samba config at least. Use local master = no !
KidSock
Let's burn our own archives!
on
Deja For Sale
·
· Score: 1
I hope this isn't deja's way of saying "we folded" and are never to be seen again. The world will be a dumber place without them(or at least I sure will be).
With the cost of storage as it is I think it's time to start thinking about setting up a news server on Linux to pull my favortite groups and archive them. I wonder how much frequently I would have to burn a cd if I had say 50 groups.
And finally, this means the class file was screwed up. Actually,
at this point I think it didn't find it at all. It kept defaulting
till it reached the system class loader. The class file was truncated
because it was 0 bytes maybe. I wonder what a Netscape 4.x stack trace
looks like when it tries to load a "truncated" classfile.
java.lang.ClassFormatError: com/hushmail/client/gui/mail/HushApplet (Truncated class file)
Ooo, a native method. But that would be the primordial system class
loader. The last resort. Only if it can't find it. Mmmm.
at java.lang.ClassLoader.defineClass0(Native Method)
at java.lang.ClassLoader.defineClass(ClassLoader.java :486)
Hmm, the secure class loader. I thought all applets lived in a
"sand-box". Maybe needed for signed applets.
at java.security.SecureClassLoader.defineClass(Secure ClassLoader.java:111) ; ; ;
We saw this before to but a different line of code. I think we must
have gone down the stack, back up, and now down again?
at sun.applet.AppletClassLoader.findClass(AppletClass Loader.java:142)
Now we switch to the sun Java plugin class loader(maybe better luck
with that?)
at sun.plugin.security.PluginClassLoader.findClass(Pl uginClassLoader.java:213)
Wait a minute... didn't we just do this?
at java.lang.ClassLoader.loadClass(ClassLoader.java:2 97)
at sun.applet.AppletClassLoader.loadClass(AppletClass Loader.java:108)
And that gets delegated to the system class loader
at java.lang.ClassLoader.loadClass(ClassLoader.java:2 53)
Now we ask the applet specific class loader for the class definition
at sun.applet.AppletClassLoader.loadCode(AppletClassL oader.java:366)
More create applet stuff is ok
at sun.applet.AppletPanel.createApplet(AppletPanel.ja va:579)
at sun.plugin.AppletViewer.createApplet(AppletViewer. java:1111)
Now load the actual applet Java code
at sun.applet.AppletPanel.runLoader(AppletPanel.java: 515)
The panel is the drawing area of the applet to be rendered in the
browser window
at sun.applet.AppletPanel.run(AppletPanel.java:293)
So MotifAppletViewer is commissioned to run applet classes for moz
at sun.plugin.navig.motif.MotifAppletViewer.maf_run(M otifAppletViewer.java:125)
at sun.plugin.navig.motif.MotifAppletViewer.run(Motif AppletViewer.java:121)
Normal run method of the applet Thread
Oh, oh, oh make the laughter stop! This is too funny. But I thought it was the posters intention to post the message in plain text to be funny. I guess not because the BrowseX thing is for real. This is maybe even funnier! Oh, oh, ok(laughter under contrasjdaskidjaiosdj ha ha ha mmm ok).
Slashdot Mirror
But why are they inventing so-called "inefficient" code? Because it's really EFFICIENT -- for developers. It's also easier to maintain.
Java/XML-RPC/etc. are all software inventions that make it easier to develop sophisticated programs.
Actually Java/XML-RPC is kinda stupid. It's aycronimal-pixie dust. I hope Java shops don't get caught up in it too much. It has it's place though.
What's wrong with DataOutputStream and DataInputStream? Their trivial to use and quite elegant. And if you want platform interoperability just write your own DataInputStream/DataOutputStream-like classes that write all values in network byte order like:
Actually, you can just dump the Streamish interface and have your data-holding classes implement similar: kind of stuff. You can write very serious networking software using this technique. Network latencies what they are make this as fast as native code. Because it's network byte order you have platform interoperability. It's simple once you write the writeXxx methods and *very* efficient.Well if this is in fact the case I can say with certainty that it will definately not work. If I so much as write myself a little note in Word it's gonna quitely be uploaded to a server in the background! Word docs and pics and stuff are big. The whole thing hinges on ultra-high-bandwidth. MS' cook'n up a receipe for disaster.
So let them use GPL software. Maybe it will ease it's acceptance within the company and help close the gaps. Then one day the truth will come out and we'll all be better off. You think it will really help M$ be competitive. They don't need it. They have clueless IT managers snowed several times over.
Let'em have it. Can't fight it any more than they can stop the whole OSS movement anyway. It's only a matter of time.
Who, matrox? The matrox install is trivial. You just copy the .o into the modules directory. It's the XFree install thats involved. I think that will always be the case. Most people are better off just waiting for it to be provided in their favorite distro.
It works. I've been running it for about a month and a half. Works pretty well too. I can do rediculous resolutions and 32bit and all that but I just run 1280x1024 at 24. My machine did lock up *once* but it was when I was just starting to set up the new box and crazily tweeking all sorts of stuff.
It's a pain to install though. That is XF4.0.2 I mean. I suspect XF4.0.3 is no different. Actually this release looks like some relativly minor updates. To install you have to pull down like 25MB in 12 files and run there installer. Good luck getting your old X back though. Cross your fingers and toes for that one(actually it wouldn't be that hard to re-install an older version and might be just a matter of moving some dirs back in place). After that, get the latest drivers from matrox and generate or get a config and help from others on the matrox linux form.
I didn't try 3d accel and DRI and GLX or other acronyms but I've read about people doing all that stuff and dual head, etc. If you like hacking though you'll have a blast!
Rogers also points out that ShareSniffer only locates open shares, it doesn't access them.
This is not true. The only way to determine if the share can be connected to without a password is to try the SMB_COM_SESSION_SETUP_ANDX message with a null password which transpires *after* session establishment or other netbios nameservice and session service operations and *after* dialect negotiation. This would be like checking to see if the door on someones house is locked by walking up and turning the knob and opening it a little.
I think that would be considered just as illegal as walking in a looking around.
Funny thing is that Microsoft renamed the Windows networking protocol to the Common Internet FileSystem (CIFS). Perfect :~)
for instance, he's currently working on replacing the Linux kernel configuration system with something considerably better, from all accounts
Where's this. I've looked at his home page and around but cannot find anything on it. Is there a link?
I've recently been writing a bit of c whereas for the past 3 years it's been totally Java. Now I find in practice that exactly the opposite of what you say is true.
If something compiles in Java it means it's pretty solid. I thuroughly abuse the compiler output by just typing and then solve the compile time errors one by one rather than spending a lot of time thinking about types and such(to an extent).
But if something compiles in c it means exactly nothing. It's much more likely that you reference a pointer that points somewhere you didn't expect than accedentally put a circle object into something that's supposed to take polygon objects. What's the difference between this and a linked list data structure in c that takes void *?
Can anyone connect to any host of microsoft.com? I just got a call from someone who cannot connect to microsoft.com and I can't either. Is something up with that domain? Ok, which one of you script kiddies is responsible for this?
But are the Mozilla libs truly shared? Meaning are they re-entrant or whatever is needed to prevent them from simply being copied in memory? I have a feeling that they are not but I could just be spreading lies.
Nautilus could be the default file manager
Doesn't Nautilus require Mozilla libs?
More bloat, yea.
Not any closer to pulling me away from WindowMaker.
But it could not be used as an exploit because the shell environment variable space could not hold enough data to "smash the stack".
One of the fundamental units of a Microkernel architecture is a uniform network/port based communcations subsystem.
That's one way of implementing a microkernel, but it's hardly the defining feature.
I don't recall calling it a "defining feature". I called it a "fundamental unit".
I don't know a lot about it
So perhaps you shouldn't be stating opinion as fact, hmm?
a) I didn't state anything as fact. b) you have chopped off the second part of the sentence to remove the context of what I am saying which is that CORBA would not suffice as a primary means of interprocess communication for an operating system.
but it has NOTHING to do with CORBA.
One can implement a microkernel using CORBA as the primary means of IPC.
You _could_ but I'm telling you that it would be unbearably slow.
Certainly CORBA services might run as any other Microkernel service but this would be hurrendously slow to use_as the primary means of inter-proccess communication.
I suppose you can provide us with a link to a paper where this has been thoroughly examined? Or are you just spouting nonsense again?
I'm spouting nonsense?! You obviously know nothing of CORBA. First of all it's founded on sharing objects. These are not simply structs but full blown objects as in Object-Oriented objects. Do you know what it takes to serialize an object for transmission over a network? It's slow. Even if they used shared memory to get around serialization of local objects there would still be many issues related to the dynamic binding of methods and other OO stuff that is invariably slow.
If the ORB is in the kernel (if it's not, then it's not really the primary means of IPC now, is it?) and designed to take advantage of that situation (through the use of virtual memory manipulation and direct access to the address space of both processes) then it should be able to provide adequate performance, and a small performance loss can easily be made up for with a dramatic increase in flexibility.
Even if you don't want to use RPCs, you can still use things like CORBA to formalize the structure of the messages that you pass and abstract the details away from the programmer.
Think about what your saying. You're going to run a CORBA ORB _in_ the kernel as the primary means of communication for an operating system?! I'll beleive it when I see it.
hate to be pessimistic, but with all the failures, layoffs, share prices plummeting...i hope Sun has a big war chest.
Are you kidding? Sun's stock actually went up during most of the plummeting and quite frankly, yes, I'd be willing to bet they have an awsome war chest.
Over a remote X session Nautilus is totally worthless and I'd say a large chuck of Solaris desktop users are doing remote sessions a la eXceed. Can you downgrade that thing to 256 colors? I hope so because over X it takes a LONG time for that thing to paint all those pretty icons. Doesn't it use Mozilla libs for crips sake!
I wish someone would focus on the fundamentals rather than making it look great. Someone should make a file manager that works as well as WINFILE.EXE the old Windows file manager!
I have been writing a Java library that we originally were going to license as GPL. Someone pointed out that the whole project would be mearly an acedemic exercise because users would no doubt be using non-GPL code for just about everything else that might call our library in which case they would be in violation. So we switched to LGPL for genderal industry compatibilty. Would you suggest this is the wisest license for Java software?
Similarly, with LGPL could someone simply change access to constructors and methods to "public" to circomvent package protection and allow them manipulate/extend the library from their own proprietary packages? This would not be suitable reciprocation IMO. If not, LGPL works perfectly for Java because I can excercise the license through package protection :~)
The HURD is apparently going to offer CORBA as the primary means of inter-proccess communication.
This is thuroughly false information. One of the fundamental units of a Microkernel architecture is a uniform network/port based communcations subsystem. I don't know a lot about it but it has NOTHING to do with CORBA. Certainly CORBA services might run as any other Microkernel service but this would be hurrendously slow to use as the primary means of inter-proccess communication.
I agree. XML is too verbose and not ideal in many cases. However I would like to see a little more effort go into making these little persistant stores peppered throughout /etc a little more accessable. Using the file system to organize your data is not a suitable solution. Perhaps one could examine what all these configs have in common and come up with 5-6 standard configuration file formats. Get Apache, Samba, exports, fstab, named.hosts, sendmail.mc, ....to all use one of these common formats. Now some would be terrribly gerneral(ie sendmail.mc is just preprocessor input) but others might use tagging ...etc. I haven't thought about it enough but certainly there must be some compermise that would be acceptable to the application developers. There's just no effort going on in this area at all right now and there should be.
Never happen. Think about what syslog.conf would look like in XML(beleive me, I tried it).
I agree however that setting and retriveing persistant data stored as human readible conf files via a programmable interaface would be a big plus. The main problem is that the application developers don't give a damn about how programmatically accessable the data in their configs are. Someone would have to write parsers which is a major pain in the ass if the application's requirements change. I think someone would have to convince the applicaiton developers to require that the application itself actually use your programmatically smart configuration file component. Actually I think a lot of applications would welcome this. It takes off a little burden and presumably would cleanup the interface.
Then you could add support for XML separatly so that users have a choice(at a small cost in performance as your duplicating changes in two places).
I must say it's not a bad portrait of the state of Linux affairs at the moment. But it's obviously a one sided view of the situation and is meant to be taken as such. It's kind of a troll basically. It's just a way for the guy to vent behind someone elses name(like that Anonymous Coward guy does here on Slashdot:~). This part is what crosses over the objectivity line though:
I think that sooner or later, these Linux fools will self-destruct without our influence. We'll see who has the Mardi Gras celebration then. In the meantime, I hope they enjoy their 5.6 percent of the desktop. It won't last.
Oh, I think it will last. Even when the zealots wake up and go back to using windows the fact that Linux is a superb piece of software will remain. A LOT of techie people use it for serious business reasons and otherwise and it's only getting better. I'm not impressed with KDE and GNOME. I think the real advances will be in infrastucture. Stuff like browsing into an ftp site by simply cd-ing into /net/someftp.site.com/pub or a library for retrieving/setting all the possible system properties in files like named.hosts, httpd.conf, ..etc so that persistance is abstracted from the logic. Then some _real_ configuration tools can be built. I just totally made up two(possibly stupid) examples up off the top of my head but I think it will be clever infrastructure stuff like that that will really make Linux shine. I don't wan't KDE or GNOME. Does anyone find it ironic that the're writing Windows 95 for UNIX?!
KidSock
Why don't you just install printtool? Probably defunct this 'cups' you speak of but if you can't print anyway why not try it.
you do have to be carefull though. Someone setup a box for g++ and Samba prevented the browse lists from working properly. It became a master browser but couldn't find the domain master browser so some hosts on that subnet could not be found in network neighborhood. I looked at the config and didn't find anything wrong. I asked the Samba guys and didn't get much of an answer. I never heard about it again from the IT guys but after being the local Linux advocate it's a blow to the credability. I won't be advocating the default Samba config at least. Use local master = no !
KidSock
I hope this isn't deja's way of saying "we folded" and are never to be seen again. The world will be a dumber place without them(or at least I sure will be).
With the cost of storage as it is I think it's time to start thinking about setting up a news server on Linux to pull my favortite groups and archive them. I wonder how much frequently I would have to burn a cd if I had say 50 groups.
KidSock
READ THIS MESSAGE FROM THE BOTTOM UP
And finally, this means the class file was screwed up. Actually,
a :486)
e ClassLoader.java:111)
s Loader.java:142)
l uginClassLoader.java:213)
... didn't we just do this?
2 97)s Loader.java:108)
2 53)
L oader.java:366)
a va:579). java:1111)
: 515)
M otifAppletViewer.java:125)f AppletViewer.java:121)
Normal run method of the applet Thread
at this point I think it didn't find it at all. It kept defaulting
till it reached the system class loader. The class file was truncated
because it was 0 bytes maybe. I wonder what a Netscape 4.x stack trace
looks like when it tries to load a "truncated" classfile.
java.lang.ClassFormatError: com/hushmail/client/gui/mail/HushApplet (Truncated class file)
Ooo, a native method. But that would be the primordial system class
loader. The last resort. Only if it can't find it. Mmmm.
at java.lang.ClassLoader.defineClass0(Native Method)
at java.lang.ClassLoader.defineClass(ClassLoader.jav
Hmm, the secure class loader. I thought all applets lived in a
"sand-box". Maybe needed for signed applets.
at java.security.SecureClassLoader.defineClass(Secur
; ; ;
We saw this before to but a different line of code. I think we must
have gone down the stack, back up, and now down again?
at sun.applet.AppletClassLoader.findClass(AppletClas
Now we switch to the sun Java plugin class loader(maybe better luck
with that?)
at sun.plugin.security.PluginClassLoader.findClass(P
Wait a minute
at java.lang.ClassLoader.loadClass(ClassLoader.java:
at sun.applet.AppletClassLoader.loadClass(AppletClas
And that gets delegated to the system class loader
at java.lang.ClassLoader.loadClass(ClassLoader.java:
Now we ask the applet specific class loader for the class definition
at sun.applet.AppletClassLoader.loadCode(AppletClass
More create applet stuff is ok
at sun.applet.AppletPanel.createApplet(AppletPanel.j
at sun.plugin.AppletViewer.createApplet(AppletViewer
Now load the actual applet Java code
at sun.applet.AppletPanel.runLoader(AppletPanel.java
The panel is the drawing area of the applet to be rendered in the
browser window
at sun.applet.AppletPanel.run(AppletPanel.java:293)
So MotifAppletViewer is commissioned to run applet classes for moz
at sun.plugin.navig.motif.MotifAppletViewer.maf_run(
at sun.plugin.navig.motif.MotifAppletViewer.run(Moti
at java.lang.Thread.run(Thread.java:484)
Oh, oh, oh make the laughter stop! This is too funny. But I thought it was the posters intention to post the message in plain text to be funny. I guess not because the BrowseX thing is for real. This is maybe even funnier! Oh, oh, ok(laughter under contrasjdaskidjaiosdj ha ha ha mmm ok).
KidSock