Year versioning makes sense when trying to sell a product... However, when supporting the product, things are completely inverse. For me, when trying to help a customer with their products, it's more helpful for me to have a major.minor version numbering scheme, as this kind of information is usually easier to get out of a customer, even if they don't get all the details completely correct. For instance, lets assume that Windows hadn't jumped to the year-dating system. If I asked a customer what version of Windows he/she was running and got the reply "4 something" that would probably be enough information for me to find some kind of solution for them, as Windows 4.0 (Win95) and 4.10 (Win98) are largely similar, and there are few enough differences so that one solution usually works on both. It's just less information for the user to remember to tell me. However, this also has to be accompanied by some sanity on the behalf of the software manufacturer to give the software meaningful version numbers (e.g. not increment the major version without reason), and most can't seem to do that. We've all watch Microsoft bloat version numbers... Word for Windows "magically" jumped from version 2.0 to version 6.0. Windows NT started life as 3.1, then jumped to 3.5 and 3.51. And I seriously doubt Microsoft SQL server made it legitimately to version 7. We've also seen the Linux community do it, where Slackware jumped to version 7 for no good reason, and numerous distros started their life at 5.x or 6.x.
The older style major.minor version numbering scheme really gives more information to those who really care. I don't want to have to try and diagnose a program whos version numbers include things like 2000, 2003, 2005, 2008 and 2012 and try to figure out which one had serious changes, and which are just minor updates. Unfortunately, it seems that in the Win32 world, major.minor version numbers are going the way of the Dodo.
This is a really important feature. You never know when the apocalypse will strike and the number 4 will simply cease to exist! Or if 8 and 9 suddenly vanished, and we were all forced to change the entire world to octal!
For an e-commerce server, this could also backfire. If a customer decides to check on their favorite site, and finds it down on Jan 01, 2000, I know I personally would have to think long and hard about dealing with them in the future, as it would seem that they obviously can't handle something as simple at date calculation on a site that would've been fairly recently created. Besides, 98% of the Y2K bugs I've seen patches for are either cosmetic issues or extrememly rare and unlikely conditions (like one Win9x patch that fixed an issue where if Windows was initialising the RTC during bootup at exactly Dec 31, 1999 11:59:59.99PM and the RTC rolled over to 2000 before the function was completed, the year would be reset back to 1999).
If you ask me, the 32-bit integer time problem will be much worse and potentially damaging and much more difficult to fix
Things like top aren't too smart about how much RAM things like X takes. It includes all mapped physical address spaces, as well was regular RAM... That means if you have a 32MB video card, you're looking at most programs reporting you have an X server that's bloated to more than 32MB. For example, in my case:
PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND 4252 root 10 0 44732 2975 872 R 0 3.3 9.2 8:30 XFCom_Rage128
There's no way X is taking 44MB of RAM on my system, in fact, it's probably taking much less than even 12MB
The only thing X can safely be accused of is being slow for graphic intensive screen redraws. Sending everything through pipes has some real disadvantages for things like this, which is why these glx projects are pioneering "direct rendering" and by-passing the X server altogether.
No, that's not what they would do. They rip the MPEG2 stream to disk, and then re-encode it with something like Microsoft's ASF format, or MPEG-1 to make a VCD. Microsoft should be happy with piracy... It's the reason their products are so popular!
Surprising that a search engine isn't included in that list, as it can also be made to link to that site. Or how about a link to a link to a link to a link that has a link to DeCSS? Oh, no, I just realized I linked to altavista, now I'm gonna be sued, too!@#%$
More than a few of those are not in the US, some of them are even in areas of the world where such a lawsuit wouldn't even be given the time of day. Oh, well. Glad I live in Canada, where the legal system is a little more sane... Just little things like the fact the loser of a case has to pay the legal fees, etc. I've personally never seen any of those sleeze lawyer ads for Canadian lawyers... (Although Michigan's Jim "The Hammer" Shapiro's commercials were a little amusing.. He needs to cut back on the caffiene a little)
It seems that very few companies are stupid enough to run IIS on their webservers. It seems that MS IIS's biggest market is small ISPs, and webhosting companies. Hell, Apache has nearly a 60% market saturation, whereas IIS is only about 22%.
Besides, Solaris (on a Sun server) is usually a better solution than any x86 platform (for potential reliability and performance, at least). Arguably the biggest thing holding back the x86 is the beast known as "backwards compatibility". If we didn't have to worry about the legions of programs that would be broken by having MORE than 16 IRQs, not having the CGA CRTCs, doing away with the ISA DMA controllers, not having "shadow" IO ports for things like the IDE controllers, etc. I'd personally love to be able to do away with all these things, but then how'd I be able to play all those old 8088/80286 CGA games?!?
I'd believe that the 2.2 kernel's TCP stack could've broken some things. I've had almost 0 luck trying to connect a box running 2.2 to MTS ADSL, whereas 2.0 worked flawlessly (well, before MTS decided to switch to the "better" PPPoE-based ADSL solution).
Also remember that in May, 2.2.x was very immature. It had problems with everything from disk corruption, to a number of drivers not working properly. Things have improved substancially now.
Personally, I prefer PHP3... It has a very 'C'-like syntax that is a bit mixed with Perl syntax. It's also much more readable than Perl is (come on, admit it!), and can be compiled to work in Apache, Netscape Server, or Microsoft IIS, as well as a few other less used httpd's. It also supports database connectivity with a large number of different database products (including mSQL, MySQL, MSSQL, ODBC, postgresql, etc)
I always find it a little unfair that IIS's proprietary ISAPI is constantly being compared to the slower CGI model. If you notice carefully, IIS CGI and Stronghold CGI are nearly the same speed. I'd REALLY like to see a performance comparison between something like Apache DSO, PHP, and IIS ISAPI/ASP. I'm betting that the results would be MUCH closer than the ISAPI/CGI comparison.
That's not entirely true. The loader portion has to be in a partition below the 1024 cylinder mark, otherwise, the generic MS MBR can't locate the partition to boot. For most people this isn't a problem, since they have at the very least a FAT partition at the start of the drive. Just try to install (and boot) NT after filling up the first 8 gigs of space on the drive. You'll probably hit one of two errors. Either a nice blue screen 'o death stating something about inaccessable boot device, or an error about unable to find the bootloader. LILO, on the other hand, fits entirely into the MBR, so as long as the kernel image doesn't go beyond the 8GB mark, you're okay.
Once you get the hang of it, compiling your kernel, sound card module, etc is very simple. The soundblaster live! driver is also very simple... Just download from http://opensource.creative.com then just
./configure make make install depmod -a modprobe emu10k1
Five easy steps for the most part, once you get the hang of it. As long as you've installed the linux kernel source code, everything should be in order. With loadable modules, recompiling one's kernel is almost unneeded, except for enabling some advanced options, like "Advanced router", and such. If you want a easy-to-use Linux distro, try Corel Linux 1.0 -- my only complaint is sometimes it tries to be too "smart", and overwrites a configuration file that I've customized, and some features are not quite evident -- such as installing a new hardware driver. The real power of UNIX-style systems is that they're not easy to learn, but easy to operate.
No, bullying is not the problem. But it's one ingredient in the recipe. Another is the fact that it seems that most parents would rather not take responsibility for their children. Instead of accepting some of the responibility for not paying attention to their children's problems, they look to blame everything else -- TV, games, others, etc.
I've heard nearly all these inane excuses why the parents didn't pick up on the problem, and all I can say is that they're merely excuses. You can't expect me to believe that your job, and your pretty little house in the middle of the suburbs with 51" TV is more important than your own children. Do you REALLY think a child could sneak in massive quantities of weapons, and the parents WOULDN'T notice?
This is just another scapegoat so that the people who should be paying attention have yet another excuse for why they weren't doing anything about the problem before it got lethal. It's about time that people take responsibility for their own actions (or inactions) instead of pointing fingers at everyone else. The kids who pulled the trigger were responsible for killing, that's simple. But the parents, educators, and even the entire community is guilty of ignoring the problems until they escalate into something horrible.
Slashdot Mirror
Year versioning makes sense when trying to sell a product... However, when supporting the product, things are completely inverse. For me, when trying to help a customer with their products, it's more helpful for me to have a major.minor version numbering scheme, as this kind of information is usually easier to get out of a customer, even if they don't get all the details completely correct. For instance, lets assume that Windows hadn't jumped to the year-dating system. If I asked a customer what version of Windows he/she was running and got the reply "4 something" that would probably be enough information for me to find some kind of solution for them, as Windows 4.0 (Win95) and 4.10 (Win98) are largely similar, and there are few enough differences so that one solution usually works on both. It's just less information for the user to remember to tell me. However, this also has to be accompanied by some sanity on the behalf of the software manufacturer to give the software meaningful version numbers (e.g. not increment the major version without reason), and most can't seem to do that. We've all watch Microsoft bloat version numbers... Word for Windows "magically" jumped from version 2.0 to version 6.0. Windows NT started life as 3.1, then jumped to 3.5 and 3.51. And I seriously doubt Microsoft SQL server made it legitimately to version 7. We've also seen the Linux community do it, where Slackware jumped to version 7 for no good reason, and numerous distros started their life at 5.x or 6.x.
The older style major.minor version numbering scheme really gives more information to those who really care. I don't want to have to try and diagnose a program whos version numbers include things like 2000, 2003, 2005, 2008 and 2012 and try to figure out which one had serious changes, and which are just minor updates. Unfortunately, it seems that in the Win32 world, major.minor version numbers are going the way of the Dodo.
This is a really important feature. You never know when the apocalypse will strike and the number 4 will simply cease to exist! Or if 8 and 9 suddenly vanished, and we were all forced to change the entire world to octal!
For an e-commerce server, this could also backfire. If a customer decides to check on their favorite site, and finds it down on Jan 01, 2000, I know I personally would have to think long and hard about dealing with them in the future, as it would seem that they obviously can't handle something as simple at date calculation on a site that would've been fairly recently created. Besides, 98% of the Y2K bugs I've seen patches for are either cosmetic issues or extrememly rare and unlikely conditions (like one Win9x patch that fixed an issue where if Windows was initialising the RTC during bootup at exactly Dec 31, 1999 11:59:59.99PM and the RTC rolled over to 2000 before the function was completed, the year would be reset back to 1999).
If you ask me, the 32-bit integer time problem will be much worse and potentially damaging and much more difficult to fix
Things like top aren't too smart about how much RAM things like X takes. It includes all mapped physical address spaces, as well was regular RAM... That means if you have a 32MB video card, you're looking at most programs reporting you have an X server that's bloated to more than 32MB. For example, in my case:
PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
4252 root 10 0 44732 2975 872 R 0 3.3 9.2 8:30 XFCom_Rage128
There's no way X is taking 44MB of RAM on my system, in fact, it's probably taking much less than even 12MB
The only thing X can safely be accused of is being slow for graphic intensive screen redraws. Sending everything through pipes has some real disadvantages for things like this, which is why these glx projects are pioneering "direct rendering" and by-passing the X server altogether.
No, that's not what they would do. They rip the MPEG2 stream to disk, and then re-encode it with something like Microsoft's ASF format, or MPEG-1 to make a VCD. Microsoft should be happy with piracy... It's the reason their products are so popular!
Surprising that a search engine isn't included in that list, as it can also be made to link to that site. Or how about a link to a link to a link to a link that has a link to DeCSS? Oh, no, I just realized I linked to altavista, now I'm gonna be sued, too!@#%$
More than a few of those are not in the US, some of them are even in areas of the world where such a lawsuit wouldn't even be given the time of day. Oh, well. Glad I live in Canada, where the legal system is a little more sane... Just little things like the fact the loser of a case has to pay the legal fees, etc. I've personally never seen any of those sleeze lawyer ads for Canadian lawyers... (Although Michigan's Jim "The Hammer" Shapiro's commercials were a little amusing.. He needs to cut back on the caffiene a little)
It seems that very few companies are stupid enough to run IIS on their webservers. It seems that MS IIS's biggest market is small ISPs, and webhosting companies. Hell, Apache has nearly a 60% market saturation, whereas IIS is only about 22%.
Besides, Solaris (on a Sun server) is usually a better solution than any x86 platform (for potential reliability and performance, at least). Arguably the biggest thing holding back the x86 is the beast known as "backwards compatibility". If we didn't have to worry about the legions of programs that would be broken by having MORE than 16 IRQs, not having the CGA CRTCs, doing away with the ISA DMA controllers, not having "shadow" IO ports for things like the IDE controllers, etc. I'd personally love to be able to do away with all these things, but then how'd I be able to play all those old 8088/80286 CGA games?!?
Yes, and he lost his token.
I'd believe that the 2.2 kernel's TCP stack could've broken some things. I've had almost 0 luck trying to connect a box running 2.2 to MTS ADSL, whereas 2.0 worked flawlessly (well, before MTS decided to switch to the "better" PPPoE-based ADSL solution).
Also remember that in May, 2.2.x was very immature. It had problems with everything from disk corruption, to a number of drivers not working properly. Things have improved substancially now.
Personally, I prefer PHP3... It has a very 'C'-like syntax that is a bit mixed with Perl syntax. It's also much more readable than Perl is (come on, admit it!), and can be compiled to work in Apache, Netscape Server, or Microsoft IIS, as well as a few other less used httpd's. It also supports database connectivity with a large number of different database products (including mSQL, MySQL, MSSQL, ODBC, postgresql, etc)
I always find it a little unfair that IIS's proprietary ISAPI is constantly being compared to the slower CGI model. If you notice carefully, IIS CGI and Stronghold CGI are nearly the same speed. I'd REALLY like to see a performance comparison between something like Apache DSO, PHP, and IIS ISAPI/ASP. I'm betting that the results would be MUCH closer than the ISAPI/CGI comparison.
Such an article actually exists: Q1266 71 - Windows Overwrites Linux Boot Manager
It's just a pity that it's completely inaccurate, and therefore mostly useless.
What's LILOCONFIG?
I suppose it would be even nicer if it was accurate. I've never heard of LILOCONFIG (or liloconfig for that matter).
That's not entirely true. The loader portion has to be in a partition below the 1024 cylinder mark, otherwise, the generic MS MBR can't locate the partition to boot. For most people this isn't a problem, since they have at the very least a FAT partition at the start of the drive. Just try to install (and boot) NT after filling up the first 8 gigs of space on the drive. You'll probably hit one of two errors. Either a nice blue screen 'o death stating something about inaccessable boot device, or an error about unable to find the bootloader. LILO, on the other hand, fits entirely into the MBR, so as long as the kernel image doesn't go beyond the 8GB mark, you're okay.
Once you get the hang of it, compiling your kernel, sound card module, etc is very simple. The soundblaster live! driver is also very simple... Just download from http://opensource.creative.com then just
./configure
make
make install
depmod -a
modprobe emu10k1
Five easy steps for the most part, once you get the hang of it. As long as you've installed the linux kernel source code, everything should be in order. With loadable modules, recompiling one's kernel is almost unneeded, except for enabling some advanced options, like "Advanced router", and such. If you want a easy-to-use Linux distro, try Corel Linux 1.0 -- my only complaint is sometimes it tries to be too "smart", and overwrites a configuration file that I've customized, and some features are not quite evident -- such as installing a new hardware driver. The real power of UNIX-style systems is that they're not easy to learn, but easy to operate.
No, bullying is not the problem. But it's one ingredient in the recipe. Another is the fact that it seems that most parents would rather not take responsibility for their children. Instead of accepting some of the responibility for not paying attention to their children's problems, they look to blame everything else -- TV, games, others, etc.
I've heard nearly all these inane excuses why the parents didn't pick up on the problem, and all I can say is that they're merely excuses. You can't expect me to believe that your job, and your pretty little house in the middle of the suburbs with 51" TV is more important than your own children. Do you REALLY think a child could sneak in massive quantities of weapons, and the parents WOULDN'T notice?
This is just another scapegoat so that the people who should be paying attention have yet another excuse for why they weren't doing anything about the problem before it got lethal. It's about time that people take responsibility for their own actions (or inactions) instead of pointing fingers at everyone else. The kids who pulled the trigger were responsible for killing, that's simple. But the parents, educators, and even the entire community is guilty of ignoring the problems until they escalate into something horrible.