Exactly. There are passive hubs and active hubs. Passive hubs are merely good to connect two passive devices (such as a mouse/keyboard...). Active hubs are also VERY cheap. I don't know what the big deal is about...
I do agree with you. All the tomcat installs I have encoutered were so ugly, it took me 1/2 day just to figure out where the servlets were... That wasn't standalone tomcat installs though, that always was Apache/Tomcat.
I still can't figure out why my favorite webserver is only growing at 24% (377 to 469). Is tomcat that bad? I install it everywhere I can (I probably account for more than 2% of these numbers)!
I still don't understand why a majority of webservers I have found around were configured as Apache+Tomcat, and they would only have static content and a couple of servlets/JSP. What's he point of putting an Apache on the front end in this case?
Anyways, maybe I should switch to something else...;-)
Thank you very much for re-stating what I already stated. Big webservers have a slower growth than small ones. It's always nice to see that someone actually did read my post.
You will notice that I tried not to include anything below 1000, because the validity of the statistical analysis would decrease dramatically.
While IIS is not the leader, it is interesting to note the trend on open-source webservers:
Apache from 22M to 31M (40%) Jetty from 1150 to 3731 (324%) Resin from 24224 to 57113 (235%)
vs. Closed source ones: IIS from 9.7M to 9.6M (-0.1%) Lotus-Domino from 78k to 86k (10%) Oracle from 6629 to 8167 (23%) Weblogic from 5344 to 7844 (46%)
It looks like a. The big boys have a trend that is slower than the small ones b. Open source grows a lot faster.
That says a lot about the dynamic of open source webservers in general, and probably all open source tools to some degree.
Linux got in the way of M$ World Domination(tm). Google also got in the way of M$WD. Strange, that SCO seems to prefer targets that are enemies of Microsoft?
I know! I know! Maybe because every IT firm is an enemy of Microsoft? Oh wait, this would discard the whole Microsoft conspiracy theory...
I, for one, do not believe one second to this conspiracy theory. They would attack any IT company out there, that someone would say: "Hey, they attack an enemy of Microsoft, it must be them behind the hood..."
Have you thought about the possibility that SCO is acting on their own stupidity?
While I agree that the first versions were slow and ugly (JDK 1.2 and before), it is not anymore. There is a lot of good looking look and feel out there and computers have evolved so that it is really acceptable in terms of speed.
Well, I agree that allocation and deallocation can be simpler in GCs. That would be basically because the GC knows what he is doing and have a consistent behavior, where a manual allocator doesn't have a clue what the program is doing in advance.
However, the complexity that is removed from the programmer with a GC - when and what to free - is obviously a job that manual memory managers doesn't care about. Is this complexity adding a huge overhead? I don't necessarily think so, but it might depend on the GC and the program involved. Again, there is no "All purpose" perfect solution. A gc is a good compromise - especially with large projects/lazy programmers - but it might prove to be less efficient for some specific tasks.
I would have liked a memory management section to the test to illustrate this point though. This way we wouldn't have to speculate.
The real puzzlement in this test is that all their tests (except IO) is independant code that is trivial to compile with a JIT. One area all these languages lags behind a native GCC code is memory allocation. The garbage collector, while adding programming comfort, is also adding a huge overhead. Of course, that was conveniently left out. I guess the author wanted to make a point rather than get a real-life test. I mean, in the real life, unless you are talking scientific application, you will deal with a great amount of memory allocation, not a 1000000 loop iteration to compute some kind of meaningless number.
You are so wrong, that it is a shame you posted this one!
AWT has native widgets: Combo box, menu, button, text area, input box, checkbox, etc... Not only primitives.
What you are describing is Swing, not AWT.
Swing relies on the most basic AWT features: Component/Container and drawImage, and re-implement the whole widget sets in Java, relying on these two AWT components.
For real math stick with Fortran, SciLab or Matlab.
Matlab? Please tell me you're kidding. Sure, it's a nice and powerful language, but it is horribly slow.
Well, as stated by the parent, for real math stick with Fortran, SciLab or Matlab. Didn't say anything about performance, did he? You might dislike Matlab for its slowliness, but don't forget it solves a different type of problems than the others. As Always, use the right tool for the job.
All right, now remove all of the Office start-up crap on boot and try again. I don't have that.
Either that, or add your boot-up time to the Word start-up time. Yes, that would make it pretty slow. But I guess you would have to add this time to EVERY application startup, right? How can you start an application without the OS first? Maybe I should add the BIOS boot time too, and the time I spent building the computer, installing the O.S., earning the money to buy it, learning my job to earn this money... Well, finally if we add-up all that, 30 years was about the time it took to start Word after all.
See, some of us don't like having Office suck up all of our RAM when we're not using it. If you think I have ANY KIND of office startup, you are really silly. Would I post a comment about Word startup time if I had Office preloaded? I guess you can find anything on slashdot.... so that wouldn't be so surprising after all.
But no, I don't have Office preloaded on any of my computers.
While I agree with you for point 1, I just strongly disagree about 2&3. 2&3 will obviously be at the charge of the defense (except for the "They must then prove that it is data that they own" part).
Stop spreading FUD, you're becoming more and more like them.
Then you have a problem. I just started Word and here is my config:
1. PIII 700 2. never defragged my HDD, ever. 3. 413 MB of memory used, 256MB of memory installed 4. Network is irrelevant, I don;t know why you felt you had to notice it. Anyways, I also have a 1.1Mbps DSL, 128k upload 5. I have some apps running: Tomcat, seti, cygwin, xfree86, 7xterms, Outlook, Mozilla, 2xIE, MozillaMail, jEdit (bloated java-based text editor), Psi, MSN Messanger, McAfee Guardian, VNC Server, WinAmp, VShield. 6. I have not used word in the last 48 hours. 7. My system is up since several month.
Total: 4 seconds. Started when I clicked the icon in the start menu, ended when I started my first letter. Word 2000 though.
I seriously suggest you check your system for something wrong. Or maybe you roll back your Office 2002 install, if 2002 is the problem.
Wait a second. You're saying that if a user installs some unrelated piece of software and that causes Word to break, that's not Word's fault?
Well, I can build a software that will break any application on any OS. As long as the installer has the right priviledges. And that would be true for 95% of the apps on 95% of the OSes. So well, I guess you would blame ALL APPLICATIONS, because mine is breaking them.
I also suppose that if you were to buy a bicycle and put it in your garage, and that caused your car to stop functioning, you wouldn't blame the car manufacturer? You got the wrong analogies. If I was to buy a bicycle and attach it to the back of my car. Then I would notice that my gas consuption has dramatically increased, then no, I wouldn't blame it on the car manufacturer. Even if my "enhanced car" (ie: a car that could lead me on small trails with the embedded bike) doesn't respect the manufacturers promises anymore.
Unless you have a P75, I don't see what you are talking about. MSWord has always started in less that 3 seconds on my system (PIII 700) and I can tell you that sometimes it is terribly bloated (My system, not Word).
Wait, did I just admit running Windows on slashdot? Bye bye Karma.
Slashdot Mirror
In a few month, they'll have to give everything away to SCO anyways...
Exactly. There are passive hubs and active hubs. Passive hubs are merely good to connect two passive devices (such as a mouse/keyboard...). Active hubs are also VERY cheap. I don't know what the big deal is about...
current programmers had never had any access to protected UNIX or AIX code
Dude, IBM has told everyone a long long time ago that their AIX programmers were developping Linux. It is about too late to retract this statement.
WebLogic is from BEA, you're confusing it with SebSphere. It is a 100% Java J2EE application server.
I do agree with you. All the tomcat installs I have encoutered were so ugly, it took me 1/2 day just to figure out where the servlets were... That wasn't standalone tomcat installs though, that always was Apache/Tomcat.
Tomcat WAS slow when it was given to the Apache foundation by Sun, because it was a reference implementation. This is not true anymore.
As usual, AC is spreading FUD. See here
I still can't figure out why my favorite webserver is only growing at 24% (377 to 469). Is tomcat that bad? I install it everywhere I can (I probably account for more than 2% of these numbers)!
;-)
I still don't understand why a majority of webservers I have found around were configured as Apache+Tomcat, and they would only have static content and a couple of servlets/JSP. What's he point of putting an Apache on the front end in this case?
Anyways, maybe I should switch to something else...
Thank you very much for re-stating what I already stated. Big webservers have a slower growth than small ones. It's always nice to see that someone actually did read my post.
You will notice that I tried not to include anything below 1000, because the validity of the statistical analysis would decrease dramatically.
While IIS is not the leader, it is interesting to note the trend on open-source webservers:
Apache from 22M to 31M (40%)
Jetty from 1150 to 3731 (324%)
Resin from 24224 to 57113 (235%)
vs. Closed source ones:
IIS from 9.7M to 9.6M (-0.1%)
Lotus-Domino from 78k to 86k (10%)
Oracle from 6629 to 8167 (23%)
Weblogic from 5344 to 7844 (46%)
It looks like
a. The big boys have a trend that is slower than the small ones
b. Open source grows a lot faster.
That says a lot about the dynamic of open source webservers in general, and probably all open source tools to some degree.
Linux got in the way of M$ World Domination(tm). Google also got in the way of M$WD. Strange, that SCO seems to prefer targets that are enemies of Microsoft?
I know! I know!
Maybe because every IT firm is an enemy of Microsoft? Oh wait, this would discard the whole Microsoft conspiracy theory...
I, for one, do not believe one second to this conspiracy theory. They would attack any IT company out there, that someone would say: "Hey, they attack an enemy of Microsoft, it must be them behind the hood..."
Have you thought about the possibility that SCO is acting on their own stupidity?
the slow and ugly Swing
While I agree that the first versions were slow and ugly (JDK 1.2 and before), it is not anymore. There is a lot of good looking look and feel out there and computers have evolved so that it is really acceptable in terms of speed.
Well, I agree that allocation and deallocation can be simpler in GCs. That would be basically because the GC knows what he is doing and have a consistent behavior, where a manual allocator doesn't have a clue what the program is doing in advance.
However, the complexity that is removed from the programmer with a GC - when and what to free - is obviously a job that manual memory managers doesn't care about. Is this complexity adding a huge overhead? I don't necessarily think so, but it might depend on the GC and the program involved. Again, there is no "All purpose" perfect solution. A gc is a good compromise - especially with large projects/lazy programmers - but it might prove to be less efficient for some specific tasks.
I would have liked a memory management section to the test to illustrate this point though. This way we wouldn't have to speculate.
The real puzzlement in this test is that all their tests (except IO) is independant code that is trivial to compile with a JIT. One area all these languages lags behind a native GCC code is memory allocation.
The garbage collector, while adding programming comfort, is also adding a huge overhead. Of course, that was conveniently left out. I guess the author wanted to make a point rather than get a real-life test. I mean, in the real life, unless you are talking scientific application, you will deal with a great amount of memory allocation, not a 1000000 loop iteration to compute some kind of meaningless number.
You are so wrong, that it is a shame you posted this one!
AWT has native widgets: Combo box, menu, button, text area, input box, checkbox, etc... Not only primitives.
What you are describing is Swing, not AWT.
Swing relies on the most basic AWT features: Component/Container and drawImage, and re-implement the whole widget sets in Java, relying on these two AWT components.
For real math stick with Fortran, SciLab or Matlab.
Matlab? Please tell me you're kidding. Sure, it's a nice and powerful language, but it is horribly slow.
Well, as stated by the parent, for real math stick with Fortran, SciLab or Matlab. Didn't say anything about performance, did he?
You might dislike Matlab for its slowliness, but don't forget it solves a different type of problems than the others. As Always, use the right tool for the job.
And what should I look for exactly?
All right, now remove all of the Office start-up crap on boot and try again.
I don't have that.
Either that, or add your boot-up time to the Word start-up time.
Yes, that would make it pretty slow. But I guess you would have to add this time to EVERY application startup, right? How can you start an application without the OS first? Maybe I should add the BIOS boot time too, and the time I spent building the computer, installing the O.S., earning the money to buy it, learning my job to earn this money... Well, finally if we add-up all that, 30 years was about the time it took to start Word after all.
See, some of us don't like having Office suck up all of our RAM when we're not using it.
If you think I have ANY KIND of office startup, you are really silly. Would I post a comment about Word startup time if I had Office preloaded? I guess you can find anything on slashdot.... so that wouldn't be so surprising after all.
But no, I don't have Office preloaded on any of my computers.
Skill.
;-) The real skill is in the O.S. ;-) ;-)
That's called a swap file.
While I agree with you for point 1, I just strongly disagree about 2&3. 2&3 will obviously be at the charge of the defense (except for the "They must then prove that it is data that they own" part).
Stop spreading FUD, you're becoming more and more like them.
Then you have a problem. I just started Word and here is my config:
1. PIII 700
2. never defragged my HDD, ever.
3. 413 MB of memory used, 256MB of memory installed
4. Network is irrelevant, I don;t know why you felt you had to notice it. Anyways, I also have a 1.1Mbps DSL, 128k upload
5. I have some apps running: Tomcat, seti, cygwin, xfree86, 7xterms, Outlook, Mozilla, 2xIE, MozillaMail, jEdit (bloated java-based text editor), Psi, MSN Messanger, McAfee Guardian, VNC Server, WinAmp, VShield.
6. I have not used word in the last 48 hours.
7. My system is up since several month.
Total: 4 seconds. Started when I clicked the icon in the start menu, ended when I started my first letter. Word 2000 though.
I seriously suggest you check your system for something wrong. Or maybe you roll back your Office 2002 install, if 2002 is the problem.
Wait a second. You're saying that if a user installs some unrelated piece of software and that causes Word to break, that's not Word's fault?
Well, I can build a software that will break any application on any OS. As long as the installer has the right priviledges. And that would be true for 95% of the apps on 95% of the OSes. So well, I guess you would blame ALL APPLICATIONS, because mine is breaking them.
I also suppose that if you were to buy a bicycle and put it in your garage, and that caused your car to stop functioning, you wouldn't blame the car manufacturer?
You got the wrong analogies. If I was to buy a bicycle and attach it to the back of my car. Then I would notice that my gas consuption has dramatically increased, then no, I wouldn't blame it on the car manufacturer. Even if my "enhanced car" (ie: a car that could lead me on small trails with the embedded bike) doesn't respect the manufacturers promises anymore.
steve knows this history [...] he won't let it happen again
Little have you learned about the history of humanity. You should have said:
steve knows this history [...] he won't let it happen again anytime soon
Unless you have a P75, I don't see what you are talking about. MSWord has always started in less that 3 seconds on my system (PIII 700) and I can tell you that sometimes it is terribly bloated (My system, not Word).
Wait, did I just admit running Windows on slashdot? Bye bye Karma.
Ooops, my bad...