It took four months of testing over a span of eight months. Why the hell do people think this would make a good project for a distributed client? I seriously doubt this type of simulation could be made to run on a distributed system. If it were possible, they probably would have used PCs instead of extremely expensive hardware. Besides, didn't anyone look at those extremely cool shots of the datacenter?
Apparently they focus different energy on different universities. I'm not going to name any specifics, but some universities I've seen don't do any development on Windows. Only Solaris and some Linux machines. I guess it all depends on where you're at.
Weren't DOT org's (.org) originally intended purely for non-profits? If enough people support the "subscription" model, slashdot should consider going for non-profit status. If enough people value their services, it would be of benefit to all, especially in the way of content. But they would have to give up ALL ads. (But they could still do underwritings...)
I would also like to add that after having used many of the Gnu utilities for so long, I had no idea how spoiled I was by all of the options they give me until I got some exposure to Solaris 8.
Does anyone really like Sun? If you like Sun, you have probably never had to use Solaris 8. Any given Linux distribution is years ahead of that trash. Especially for desktop users.
The process for a two node cluster set up in Win2k is nearly identical to what you have described. Although I think the licensing is a bit different than when you built that cluster... Lowest version you can buy to install a cluster is Win2k Advanced and you also need the enterprise version of SQL 2k to make it work. When you are looking at a failover type setup (active-passive), you are still looking at a fairly sizeable chunk of change to put this together. If you are frugal, the hardware is gonna cost you $30,000. For licensing, you are looking at very least an additional $30,000, likely more. I was appalled to see how much this was actually going to cost.
And yes, you have to follow many sets of conflicting directions to a friggin' "T" or else it won't work. And do yourself a favor and firewall the hell out of the boxes. Installing service packs on such a mission critical set up just doesn't appeal to me for some reason.
By the way, you say: "...If I were to do it again, I think I would not have used MSCS, but instead have two distinct systems that had some kind of data replication software."
I have no idea how you intend to accomplish this with a database, utilizing a MS solution, but I'd certainly like to hear about it!
The reason I use Linux instead of Win2k is productivity. I can get more done faster with Linux than I can in Windows. Win2k runs deathly slow on my 700 MHz laptop with 128MB RAM and a 12 GB laptop hard drive. This is a fairly typical configuration for a machine that is around a year old. I absolutely hate booting into win2k because of how long it takes to start up. It takes no less than six minutes, possibly longer to get to my desktop where I can actually do some work. With Slackware, it takes all of about thirty seconds from the time I hit the power button until I am checking my e-mail. I don't like the bloated overhead. I do not enjoy the soothing noise as my hard drive churns and Windows uses 150MB of memory before I have done ANYTHING productive. Opening Internet Explorer or "My Computer" takes no less than 23 seconds the first time of the day. After using Slackware regularly for many many months now, I can definitely tell how much slower Win2k is all around and how much less responsive it is to everything I try to do, usually involving development of web pages and basic tasks like e-mail and web browsing. Couple that with a strong sense of being limited in what I can do and I just am no longer a happy windows user. Started out with DOS back in the day and have seen the full progression of Windows and its general trend towards more and more mainstream users, A.K.A. AOL'ers.
And as for Mozilla crashing, I haven't seen Mozilla crash in a LONG time. The pages render near flawlessly these days too.
Where are my moderator points when I need them? Seriously, make a case backed up by something tangible or do not spew forth at all.
By the way, I didn't say I disagreed...
Re:Haven't you heard of a lil thing called OpenBOO
on
LinuxBIOS Gains Steam
·
· Score: 1
The author of the article makes specific mention of IEEE-1275 and makes the point that it is IIRC, defunct. Sounds to me like there is space for an additional system.
I am not sure this is very practical for win9x/winXP home users, as they do not even know they are "logged in". Especially for 9x, since it isn't multiuser and doesn't enforce even the most basic security policy. XP home users don't have to put in a password to log on, and if I am correct, have administrative privileges.(feel free to correct me if I'm wrong here) Not sure what the best solution here is though, since MS users are less than security conscious anyhow and typically don't patch their computers. (How many systems have you seen that had virus defs that were 2 years old?)
You say they were Hueys, someone else said they were Cobras, at least we all agree they weren't Apaches. Damn, how about some nomenclatures? CH-46, CH-53, AH-64, etc... I do agree these weren't gunships. A nice armored cargo helicopter seems more likely for this kind of mission and is what you tend to see more of, since they were intending to carry troops to make the extraction. (You can fit a lot more troops in a nice CH-46.)
It was a signal flare, which he should have used in the field when the helicopter was flying by overhead, not firing a rifle that wouldn't be heard over the noise of the helicopter.
They were following his foot trail. And they would obviously have known he was running and that there was somebody there with him.
As for the sniper scene on the damn, no respectable sniper with that quality of a weapon would have taken a shot in the offhand. All he had to do was lay down, get into a stable position, calculate his range and windage, and slowly squeeze the trigger. Now how hard is that when your prey doesn't know you are there and is sitting stationary? That scene was a bit annoying.
As for the sensational and unrealistic Hollywood effects of the movie, would you really have preferred the realistic scenario in which the highly disciplined pilot doesn't stray from his flight path, doesn't get shot down, and gets out of the military several days after Christmas after being bent over one last time by his CO?
I couldn't understand why they made such a big deal over using his "real name" over the airways. There aren't too many radio transmissions that go out over military radio that aren't encrypted. Especially a fancy pilot's hand-held "I just got shot down" radio. That seemed a bit silly.
Being "brutally far north" is actually a pretty awesome experience. And as for the space suit, you almost have to wear one of those up there anyhow... I would surely like to see the arctic again. It would be nice to do something for such an expedition too.
Last I checked, kdb is already in the kernel. Or do you mean the whole thing? I installed that patch at one time, but find most of the functionality I need with the one currently in the kernel.
Here is the easiest way I have found to deal with an RPM that refuses to install due to a lack of dependencies:
rpm -i --nodeps --noscripts X.rpm
This fixes RPM everytime. (Even if you do something silly like upgrade glibc on redhat 6.2, only to discover that you had better downgrade REALLY fast...)
If it is possible to implement the Hurd with any underlying microkernel, isn't it essentially possible to provide Hurd functionality with a macro-kernel as well? Layer your Hurd-servers around the macro-kernel as user space apps, disable much of the macro-kernel functionality, still suffer from not doing stuff in kernel space. This is possible isn't it? Wouldn't this allow for the "enhanced" security model which is at the heart of the Hurd? In other words, why does one have to re-invent the wheel to accomplish what the Hurd does, as well as work along side of existing systems, and not require a "compatibility" layer? The interviewer admits the Hurd suffers from a scalability standpoint. Also, isn't it possible to provide the security contexts the Hurd offers through projects for the Linux kernel, such as what NSA is working on with SE-Linux and SIDs? (known in the article as "tokens to perform specific tasks".) This is perhaps a different solution to the problem, but the result appears to be the same...
Slashdot Mirror
You have 26 rows.
It took four months of testing over a span of eight months. Why the hell do people think this would make a good project for a distributed client? I seriously doubt this type of simulation could be made to run on a distributed system. If it were possible, they probably would have used PCs instead of extremely expensive hardware. Besides, didn't anyone look at those extremely cool shots of the datacenter?
There isn't a single internal modem I have ever used that is as dirt simple to set up as an external, Hayes compatible modem.
Apparently they focus different energy on different universities. I'm not going to name any specifics, but some universities I've seen don't do any development on Windows. Only Solaris and some Linux machines. I guess it all depends on where you're at.
Weren't DOT org's (.org) originally intended purely for non-profits? If enough people support the "subscription" model, slashdot should consider going for non-profit status. If enough people value their services, it would be of benefit to all, especially in the way of content. But they would have to give up ALL ads. (But they could still do underwritings...)
I would also like to add that after having used many of the Gnu utilities for so long, I had no idea how spoiled I was by all of the options they give me until I got some exposure to Solaris 8.
Does anyone really like Sun? If you like Sun, you have probably never had to use Solaris 8. Any given Linux distribution is years ahead of that trash. Especially for desktop users.
The process for a two node cluster set up in Win2k is nearly identical to what you have described. Although I think the licensing is a bit different than when you built that cluster... Lowest version you can buy to install a cluster is Win2k Advanced and you also need the enterprise version of SQL 2k to make it work. When you are looking at a failover type setup (active-passive), you are still looking at a fairly sizeable chunk of change to put this together. If you are frugal, the hardware is gonna cost you $30,000. For licensing, you are looking at very least an additional $30,000, likely more. I was appalled to see how much this was actually going to cost.
And yes, you have to follow many sets of conflicting directions to a friggin' "T" or else it won't work. And do yourself a favor and firewall the hell out of the boxes. Installing service packs on such a mission critical set up just doesn't appeal to me for some reason.
By the way, you say:
"...If I were to do it again, I think I would not have used MSCS, but instead have two distinct systems that had some kind of data replication software."
I have no idea how you intend to accomplish this with a database, utilizing a MS solution, but I'd certainly like to hear about it!
If only all states would follow suit, we might actually have something.
Check the kernel archives. This has been discussed for a long time. Bitkeeper is not new.
The reason I use Linux instead of Win2k is productivity. I can get more done faster with Linux than I can in Windows. Win2k runs deathly slow on my 700 MHz laptop with 128MB RAM and a 12 GB laptop hard drive. This is a fairly typical configuration for a machine that is around a year old. I absolutely hate booting into win2k because of how long it takes to start up. It takes no less than six minutes, possibly longer to get to my desktop where I can actually do some work. With Slackware, it takes all of about thirty seconds from the time I hit the power button until I am checking my e-mail. I don't like the bloated overhead. I do not enjoy the soothing noise as my hard drive churns and Windows uses 150MB of memory before I have done ANYTHING productive. Opening Internet Explorer or "My Computer" takes no less than 23 seconds the first time of the day. After using Slackware regularly for many many months now, I can definitely tell how much slower Win2k is all around and how much less responsive it is to everything I try to do, usually involving development of web pages and basic tasks like e-mail and web browsing. Couple that with a strong sense of being limited in what I can do and I just am no longer a happy windows user. Started out with DOS back in the day and have seen the full progression of Windows and its general trend towards more and more mainstream users, A.K.A. AOL'ers.
And as for Mozilla crashing, I haven't seen Mozilla crash in a LONG time. The pages render near flawlessly these days too.
Where are my moderator points when I need them? Seriously, make a case backed up by something tangible or do not spew forth at all.
By the way, I didn't say I disagreed...
The author of the article makes specific mention of IEEE-1275 and makes the point that it is IIRC, defunct. Sounds to me like there is space for an additional system.
I am not sure this is very practical for win9x/winXP home users, as they do not even know they are "logged in". Especially for 9x, since it isn't multiuser and doesn't enforce even the most basic security policy. XP home users don't have to put in a password to log on, and if I am correct, have administrative privileges.(feel free to correct me if I'm wrong here) Not sure what the best solution here is though, since MS users are less than security conscious anyhow and typically don't patch their computers. (How many systems have you seen that had virus defs that were 2 years old?)
You say they were Hueys, someone else said they were Cobras, at least we all agree they weren't Apaches. Damn, how about some nomenclatures? CH-46, CH-53, AH-64, etc... I do agree these weren't gunships. A nice armored cargo helicopter seems more likely for this kind of mission and is what you tend to see more of, since they were intending to carry troops to make the extraction. (You can fit a lot more troops in a nice CH-46.)
It was a signal flare, which he should have used in the field when the helicopter was flying by overhead, not firing a rifle that wouldn't be heard over the noise of the helicopter.
They were following his foot trail. And they would obviously have known he was running and that there was somebody there with him.
As for the sniper scene on the damn, no respectable sniper with that quality of a weapon would have taken a shot in the offhand. All he had to do was lay down, get into a stable position, calculate his range and windage, and slowly squeeze the trigger. Now how hard is that when your prey doesn't know you are there and is sitting stationary? That scene was a bit annoying.
As for the sensational and unrealistic Hollywood effects of the movie, would you really have preferred the realistic scenario in which the highly disciplined pilot doesn't stray from his flight path, doesn't get shot down, and gets out of the military several days after Christmas after being bent over one last time by his CO?
I couldn't understand why they made such a big deal over using his "real name" over the airways. There aren't too many radio transmissions that go out over military radio that aren't encrypted. Especially a fancy pilot's hand-held "I just got shot down" radio. That seemed a bit silly.
You have to be voted into the club if you want to write your own anitvirus software. Otherwise, it is illegal. At least in the U.S. it is.
Don't open executables from e-mail as admin on your servers! How hard is that?! A little bit of discipline will save you much pain.
Being "brutally far north" is actually a pretty awesome experience. And as for the space suit, you almost have to wear one of those up there anyhow... I would surely like to see the arctic again. It would be nice to do something for such an expedition too.
Last I checked, kdb is already in the kernel. Or do you mean the whole thing? I installed that patch at one time, but find most of the functionality I need with the one currently in the kernel.
Here is the easiest way I have found to deal with an RPM that refuses to install due to a lack of dependencies:
rpm -i --nodeps --noscripts X.rpm
This fixes RPM everytime. (Even if you do something silly like upgrade glibc on redhat 6.2, only to discover that you had better downgrade REALLY fast...)
If it is possible to implement the Hurd with any underlying microkernel, isn't it essentially possible to provide Hurd functionality with a macro-kernel as well? Layer your Hurd-servers around the macro-kernel as user space apps, disable much of the macro-kernel functionality, still suffer from not doing stuff in kernel space. This is possible isn't it? Wouldn't this allow for the "enhanced" security model which is at the heart of the Hurd? In other words, why does one have to re-invent the wheel to accomplish what the Hurd does, as well as work along side of existing systems, and not require a "compatibility" layer? The interviewer admits the Hurd suffers from a scalability standpoint. Also, isn't it possible to provide the security contexts the Hurd offers through projects for the Linux kernel, such as what NSA is working on with SE-Linux and SIDs? (known in the article as "tokens to perform specific tasks".) This is perhaps a different solution to the problem, but the result appears to be the same...
adhear != adhere
woulden't != wouldn't
circut != circuit
For God's sake, please use a word processor with a spell checker if you can't spell.