No thanks, rather than pay $999 for a tool that will let me write Windows and Linux programs, I choose to write in a portable subset of C and C++, and abstract out the platform specific things. Then porting is as easy as writing a new implementation of my abstracted platform layer. With GCC this is free, and with Java this is not even necessary.
I believe, although I am not positive, that IL code is translated to native code at load time, then the native code is run. Therefore there is an initial penalty at the beginning, but not a continued one like in Java.
UMass-Amherst has just switched their OS course (CS 377) to be taught in Java. We used to offer it in C, using the NACHOS system. However, for some reason, that nobody seems to know, it is now in Java. This seems to be the worst possible decision. I can understand Algorithms in Java (CS 311), but OS should be a low-level programming class. And you simply cannot teach low-level programming in Java. I'm very torn, because I haven't taken this class yet, and OS is what I want to do with my career, but at the same time, I'm two and a half years towards a double degree in CompSci and Computer Systems Engineering, and don't want to go through a transfer.
The new class is so bad, to give you a few examples, the textbook has nearly no sample code. Probably because Java relys on an underlying JVM, which currently needs an underlying OS...... The class has only 3 labs, and one of them is to write a program that reads in a text file, and outputs a sorted list of the words in it. This assignment should be given in a *data-structures* class, not one on OS.
I could go on for pages here, but instead I'll stop. Aarrrgghhh....
Another great gift for a geek: the TINI
on
Gifts For Geeks
·
· Score: 1
Check out the TINI, a SIMM sized embedded Java device. It has built in Ethernet and serial communications, and can be had for under $100, including the SIMM, and appropriate mounting board. It's also got a built in webserver, and telnet server....and you can write your own Java programs for it, and upload them over the ethernet!!
Personally, I think the FCC should look into speeding up the viability of Voice over IP in every home. Combined with the (eventual) roll-out of IPv6, this would solve the phone-number problem. In fact, IPv6 enabled telephones for the home might be the "killer app" that it takes to push IPv6 over the edge into wide-spread adoption.
As a student at UMass-Amherst, I am slightly concerned about the fact that Massachusetts seems to have lots of money for this. We have been going through a series of budget cuts here on campus. The computer science department has been unable to hire all the faculty it needs, and is considering closing their courses to all non-majors. (Even now, it's nearly impossible to get into a CompSci class). They are starting a program where any CompSci major with a 3.5 GPA can go to grad school here for free as long as they teach CompSci 187 (the intro to data structures class). Our libraries have been cutting back on journal subscriptions. This, we've been told, has all been happening because the state doesn't have money.
There are several high-quality computer labs on campus, open to all hours of the night, and located all over the place. There are some in dorms, every engineering building has at least one, and each department has at least one of their own. Here, students without their own computer can go to do computer related work. I don't see the need to subsidize student laptops. As it becomes harder and harder to get classes, I do however see a need for the state to spend this money on more teachers.
Is that the new Samba-TNG FAQ? It sounds very much like the old (pre-fork) Samba-TNG FAQ. At that point Samba-TNG was actually SAMBA_TNG, a checkout marker from the main Samba CVS server, and it was to become the next version of Samba. This appears that the new (post-fork) Samba-TNG crew checked out that branch and are starting their own project based upon it.
Is that the new Samba-TNG FAQ? It sounds very much like the old (pre-fork) Samba-TNG FAQ. At that point Samba-TNG was actually SAMBA_TNG, a checkout marker from the main Samba CVS server, and it was to become the next version of Samba. This appears that the new (post-fork) Samba-TNG crew checked out that branch and are starting their own project based upon it.
Is that the new Samba-TNG FAQ? It sounds very much like the old (pre-fork) Samba-TNG FAQ. At that point Samba-TNG was actually SAMBA_TNG, a checkout marker from the main Samba CVS server, and it was to become the next version of Samba. This appears that the new (post-fork) Samba-TNG crew checked out that branch and are starting their own project based upon it.
I can't get the Win2K/IE 5.5 "FTP Explorer" to work with Linux (doesn't take password and there's no fucking logs to debug courtesy of Microsoft
If you are trying to use MS IE to FTP to another site that requires anything but anonymous access, simply type in the Address bar:
ftp://username:password@ftp.address
It's far from secure, since your plaintext password is just sitting on your monitor for all to see, but it works when you're at the office late at night and you need to quickly get to the ftp site.
My first 3d accelerator was a Voodoo 2, 8MB of RAM. I remember my reaction was, "oohh....I can't tell where the pixels are anymore. Graphics are good. I will never be without 3Dfx again."
When the Voodoo 3 came out, I being a loyal fan of 3Dfx, rushed out to buy one. At the same time, my friend upgraded his Voodoo 2 to a TNT2. I didn't notice much of a difference at all, except that I could now watch TV on my computer. However, my friend had more colors at a much faster rate than I could get on the Voodoo 3.
When the next generation of chips began being released this summer, I was about to upgrade my computer. So I waited a month or two, and read the reviews of all of them. I really couldn't see anything that would make me choose the Voodoo 5. I really wanted to support 3Dfx, but I just couldn't justify that decision. I ended up upgrading to the ATI Radion 32MB DDR, and I love it. I hope 3Dfx can get their acts back together, as they are one of the more supportive companies toward open sourcers.
At least that's not the name for/dev/fd0 in Linux. Can you imagine the embarassment of having to get your sysadmin to mount your 3-1/2-Inch Floppy D...?
This is not quite what I'm trying to say. I do think that if someone wants to write a project, or start a fork that by all means they should. However, what I was trying to say is that in this particular instance, it may not be so good. Database development is some of the most challenging work out there. It takes special talent to be able to do it well, and this talent is evident in all of these products. However, it seems that we would possibly have a better product if all these gifted developers joined together and made one or two databases instead of four.
A few months ago, a MySQL vs PostgreSQL flamewar would occur occasionally on these boards. Then Borland announced that they would be GPLing Interbase, and we had a MySQL vs PostgreSQL vs Interbase war. Now it appears that in the future we will be having MySQL vs PostgreSQL vs Interbase vs Firebird flamewars. I understand that MySQL is for speed and PostgreSQL is for features, but do we really need 4 opensourced databases? I think the community as a whole would benefit if these projects started working on combining their codebases so we had only two databases, one for speed the other for features. Just my $0.02
Yep, sorry. I misinterpreted what you said, and convoluted it into the usual "Microsoft is the devil so anything I can say bad about them, no matter how unrealisitc goes" posts that appear in this story. I thought you were trying to say that no developers will write software for Linux if there are Microsoft programs available. I still don't wholly agree with you (come on - what percentage of the slashdot crowd will use MS Office for Linux?... not to mention the large group of people who will just want free software), but your point is well taken. I don't think this is going to kill GNumeric, the GIMP, StarOffice, etc though.
If you don't get it: MS wants to stifle the coding spirit that makes Linux what it is today. So, who wants to put M$ apps on Linux? M$ does, that's who.
Again I risk getting marked as flamebait, but I really do think that you are far too much of a conspiracy theorist. This is not what Microsoft wants. And even if it was, what software developer do you know that gets stifled by a word processer and a spreadsheet?
<sarcasm>yep, that's right. I know I'm going to stop writing Linux software because there might be a Microsoft port. I think I heard Miguel, RMS, and ESR say the same. And I know Linus said that he will stop working on the kernel if Microsoft starts porting to Linux. This is the same reason that nobody but Microsoft writes applications for Windows. You know, the Microsoft WordPerfect Suite, Microsoft Mathematica, Microsoft Photoshop, Microsoft AIM, Microsoft Madden Football, Microsoft Diablo, Microsoft Delphi, and lets not forget all that Microsoft shareware.</sarcasm>
If people start moving to linux with MS Office, then MS has a chance to stick around in the 64-bit world
What makes you think that they won't port Windows to the 64 bit world? IIRC I saw somewhere that they have nearly finished the IA-64 port of Windows 2000.
Oh my god! An entire office suite is several million lines of code long! And here I was thinking that StarOffice was going to fit entirely in one screenful of code....
Slashdot Mirror
No thanks, rather than pay $999 for a tool that will let me write Windows and Linux programs, I choose to write in a portable subset of C and C++, and abstract out the platform specific things. Then porting is as easy as writing a new implementation of my abstracted platform layer. With GCC this is free, and with Java this is not even necessary.
I believe, although I am not positive, that IL code is translated to native code at load time, then the native code is run. Therefore there is an initial penalty at the beginning, but not a continued one like in Java.
UMass-Amherst has just switched their OS course (CS 377) to be taught in Java. We used to offer it in C, using the NACHOS system. However, for some reason, that nobody seems to know, it is now in Java. This seems to be the worst possible decision. I can understand Algorithms in Java (CS 311), but OS should be a low-level programming class. And you simply cannot teach low-level programming in Java. I'm very torn, because I haven't taken this class yet, and OS is what I want to do with my career, but at the same time, I'm two and a half years towards a double degree in CompSci and Computer Systems Engineering, and don't want to go through a transfer.
The new class is so bad, to give you a few examples, the textbook has nearly no sample code. Probably because Java relys on an underlying JVM, which currently needs an underlying OS...... The class has only 3 labs, and one of them is to write a program that reads in a text file, and outputs a sorted list of the words in it. This assignment should be given in a *data-structures* class, not one on OS.
I could go on for pages here, but instead I'll stop. Aarrrgghhh....
Check out the TINI, a SIMM sized embedded Java device. It has built in Ethernet and serial communications, and can be had for under $100, including the SIMM, and appropriate mounting board. It's also got a built in webserver, and telnet server....and you can write your own Java programs for it, and upload them over the ethernet!!
Personally, I think the FCC should look into speeding up the viability of Voice over IP in every home. Combined with the (eventual) roll-out of IPv6, this would solve the phone-number problem. In fact, IPv6 enabled telephones for the home might be the "killer app" that it takes to push IPv6 over the edge into wide-spread adoption.
There are several high-quality computer labs on campus, open to all hours of the night, and located all over the place. There are some in dorms, every engineering building has at least one, and each department has at least one of their own. Here, students without their own computer can go to do computer related work. I don't see the need to subsidize student laptops. As it becomes harder and harder to get classes, I do however see a need for the state to spend this money on more teachers.
I know, I'm getting redundant with this post but:
Is that the new Samba-TNG FAQ? It sounds very much like the old (pre-fork) Samba-TNG FAQ. At that point Samba-TNG was actually SAMBA_TNG, a checkout marker from the main Samba CVS server, and it was to become the next version of Samba. This appears that the new (post-fork) Samba-TNG crew checked out that branch and are starting their own project based upon it.
Is that the new Samba-TNG FAQ? It sounds very much like the old (pre-fork) Samba-TNG FAQ. At that point Samba-TNG was actually SAMBA_TNG, a checkout marker from the main Samba CVS server, and it was to become the next version of Samba. This appears that the new (post-fork) Samba-TNG crew checked out that branch and are starting their own project based upon it.
Is that the new Samba-TNG FAQ? It sounds very much like the old (pre-fork) Samba-TNG FAQ. At that point Samba-TNG was actually SAMBA_TNG, a checkout marker from the main Samba CVS server, and it was to become the next version of Samba. This appears that the new (post-fork) Samba-TNG crew checked out that branch and are starting their own project based upon it.
I can't get the Win2K/IE 5.5 "FTP Explorer" to work with Linux (doesn't take password and there's no fucking logs to debug courtesy of Microsoft
If you are trying to use MS IE to FTP to another site that requires anything but anonymous access, simply type in the Address bar:
ftp://username:password@ftp.address
It's far from secure, since your plaintext password is just sitting on your monitor for all to see, but it works when you're at the office late at night and you need to quickly get to the ftp site.
I believe that only people with FreeBSD CVS commit access were able to vote.
How about an IV drip, for constant caffine supply. Who needs to drink a double expresso when that oh-so-vital drug can go right into your veins?
When the Voodoo 3 came out, I being a loyal fan of 3Dfx, rushed out to buy one. At the same time, my friend upgraded his Voodoo 2 to a TNT2. I didn't notice much of a difference at all, except that I could now watch TV on my computer. However, my friend had more colors at a much faster rate than I could get on the Voodoo 3.
When the next generation of chips began being released this summer, I was about to upgrade my computer. So I waited a month or two, and read the reviews of all of them. I really couldn't see anything that would make me choose the Voodoo 5. I really wanted to support 3Dfx, but I just couldn't justify that decision. I ended up upgrading to the ATI Radion 32MB DDR, and I love it. I hope 3Dfx can get their acts back together, as they are one of the more supportive companies toward open sourcers.
At least that's not the name for /dev/fd0 in Linux. Can you imagine the embarassment of having to get your sysadmin to mount your 3-1/2-Inch Floppy D...?
Hmmm.....10% = 2.5 years huh? Then the 90% that's complete must have taken 22.5 years. I don't remember Mozilla in 1978.....
Err...did you mean Cobra? ;-)
This is not quite what I'm trying to say. I do think that if someone wants to write a project, or start a fork that by all means they should. However, what I was trying to say is that in this particular instance, it may not be so good. Database development is some of the most challenging work out there. It takes special talent to be able to do it well, and this talent is evident in all of these products. However, it seems that we would possibly have a better product if all these gifted developers joined together and made one or two databases instead of four.
A few months ago, a MySQL vs PostgreSQL flamewar would occur occasionally on these boards. Then Borland announced that they would be GPLing Interbase, and we had a MySQL vs PostgreSQL vs Interbase war. Now it appears that in the future we will be having MySQL vs PostgreSQL vs Interbase vs Firebird flamewars. I understand that MySQL is for speed and PostgreSQL is for features, but do we really need 4 opensourced databases? I think the community as a whole would benefit if these projects started working on combining their codebases so we had only two databases, one for speed the other for features. Just my $0.02
Yep, sorry. I misinterpreted what you said, and convoluted it into the usual "Microsoft is the devil so anything I can say bad about them, no matter how unrealisitc goes" posts that appear in this story. I thought you were trying to say that no developers will write software for Linux if there are Microsoft programs available. I still don't wholly agree with you (come on - what percentage of the slashdot crowd will use MS Office for Linux? ... not to mention the large group of people who will just want free software), but your point is well taken. I don't think this is going to kill GNumeric, the GIMP, StarOffice, etc though.
Umm..oops. Shoulda been #include <stdio.h>
Umm...you can already cause GPF's on Linux. Try to compile:
#include
int main(int argc, char *argv[])
{
int *mem;
mem = 0;
printf("%d", *mem);
return 0;
}
And see what happens when you run it. Wow, you mean GPF's are a processor feature and not a Windows one. Gee-golly.
Again I risk getting marked as flamebait, but I really do think that you are far too much of a conspiracy theorist. This is not what Microsoft wants. And even if it was, what software developer do you know that gets stifled by a word processer and a spreadsheet?
<sarcasm>yep, that's right. I know I'm going to stop writing Linux software because there might be a Microsoft port. I think I heard Miguel, RMS, and ESR say the same. And I know Linus said that he will stop working on the kernel if Microsoft starts porting to Linux. This is the same reason that nobody but Microsoft writes applications for Windows. You know, the Microsoft WordPerfect Suite, Microsoft Mathematica, Microsoft Photoshop, Microsoft AIM, Microsoft Madden Football, Microsoft Diablo, Microsoft Delphi, and lets not forget all that Microsoft shareware.</sarcasm>
What makes you think that they won't port Windows to the 64 bit world? IIRC I saw somewhere that they have nearly finished the IA-64 port of Windows 2000.
Oh my god! An entire office suite is several million lines of code long! And here I was thinking that StarOffice was going to fit entirely in one screenful of code....