On the other hand, I've worked with about 10 travelstars, all of which were very quiet. I'd complain to Dell that, as the system integrator, they screwed up and picked the wrong travelstar.
I expect that IBM is not having troubles staying alive. I expect it is because IBM is fairly conservative, and attempting to get rid of problems before they do any serious damage.
However, protection is not provided by parents alone. That's why we have an executive branch of government. Furthermore, no society could modify it's executive powers based on the individual whims of each citizen (or pairs of parents, in this case; and the pair might not agree anyway). Instead, the executive branch acts on the whims of society. If we're lucky, those societal whims are averaged out over time, and nothing changes to fast.
Pornography can be covered by free-speech provisions. At least that is what I've gathered from the recent Supreme court decision regarding virtual child pornography, as well as the Miller vs. Someone case the referred to (which they used as a baseline for pornographic materials which are not covered).
The virtual child pornography case was covered on/. w/in the last 24 hours, from which I found a several links to the decision. The page I viewed had links to other relevant porn decisions.
While there is a problem here, I expect that we can all agree to a few basic principles. If the site describes itself as a porn site, it should go to.prn. If a site wants to verify your age because of laws governing sexually explicit material, it should go to.prn.
I expect that should take care of the sites people are most worried about. After that, a very interesting public debate about porn could ensue, with references made to other countries' standards (from a US viewpoint, since that's where the bill is proposed). I expect that the result would be a loosening of American's "official" standards concerning human nakedness.
Right now, such a debate is nearly impossible. Americans don't seem to make a distinction between nakedness, suggestiveness, and flat-out-screaming pornography. Instituting the simple guidelines I listed above would simplify the situation, and hence might allow such a debate to take place.
It wouldn't make it illegal to tape them. It would make it illegal to tape them without their consent. This is very different.
If you really want to protect your children from abuse, you shouldn't hide that fact from your babysitter. Taping with consent would be a deterrent, possibly preventing the abuse in the first place. Taping without consent, if legal, would help you prosecute the abuser, but your children would still have been abused.
I disagree with absolute interpretations of property ownership. Your house resides in some county, and that county has laws governing how your house is built, what resources (gas, water, etc) it can use, etc. Your house sits underneath airspace you don't own. It is next to your neighbor's houses, and what you do with your house might affect them. It resides in a state and a nation with more laws regulating various aspects of your house.
In short, none of us are one-man nation-states, but we Americans often forget that. We are part of a large society, and our individualistic priorities occasionally need to be adjusted to meet our societies needs. If one doesn't wish to be part of our society, then you are a threat to our society and should leave.
How this applies to filming people in your house, I'm not sure. Only a truly vile scumbag would film their babysitter for anything other than child abuse prevention (which would require a fairly paranoid person at any rate). However, the possibility that home surveilance *could* be abused is not sufficient reason to ban it. And we know that there are plenty of closed-circuit monitors in private office buildings, which indicates some acceptance of such surveilance.
However, notifying your babysitter that you are filming them should not be problem -- if the babysitter objects, they can walk away from the job. I don't see what the problem with this is. Normal citizens are not given the powers of police to go undercover and such things. Even private detectives are not allowed to go undercover like police are. Allowing people to film others in private spaces, without notifying the subjects, seems akin to wiretapping without a warrant at first blush.
From a purely travel point of view, I'd rather go to Finland! I've been to Canada plenty of times.
Thanks for the more optimistic view. It's true that I haven't seen every tree replanting in the US, much less the world. I'm glad to know there are counterexamples to my experience.
Heh, this is the first time I've ever felt a tinge of competitiveness with my operating system. Microsoft just makes me mad (even though I expunged MS from my machine years ago, there's always tech support for friends and family). But Apple is now saying "Okay, we see why you like Unix so much, now we're going to up the ante." It's a great feeling.
Imagine you were a soccer player on a good team. It's fun to play against another good team, with both teams believing they've got what it takes. That's how this add makes me feel. Microsoft, on the other hand, reminds me of an awful game against the foul-mouthed and dirty-playing U14 River City Steelers, refereed by two players from the foul-mouthed and dirty-playing U16 River City Steelers (we lost that game, and many of us were hurt badly -- the next game, with regular referees, we shut them out).
1) "Also, *you* can run "netstat | grep "eth0" | sendmail...etc..." but can the average user?"
The point I was trying to make was that it's often easier for a skilled person (tech support) to dictate a command line and say "was anything printed?", than it is to say "click on this and that, oh you don't have that, oh you have that build of Windows, do you see anything like this, okay then click that, now, describe what you see, in detail".
2) "As to the comment about command lines... I disagree. Just to figure out whats up with a net connection you *should* be able to point and click. In win2k I can right click on my network icon to see if I am connected. In that dialog I can click "properties" and setup my TCP settings, etc..."
Again, I was suggesting that it's easier to describe plain text commands and responses over the phone, than it is to describe mouse movements and an array of 1024x768 pixels over the phone. If you know where to click on your build of Windows, that's great; if I know what to type on my command line, that's dandy. The issue I meant to deal with was giving commands and getting responses over the phone.
Of course, if I'm talking to someone with a unix box, it is usually very easy to login and solve problems remotely. Microsoft used to describe these remote administration features in unix systems as "security holes". Now Microsoft is trying to add them to their on operating systems (but Unisys seems to prefer the Free software program VNC;-).
For the example of piping through more versus piping through grep, a big difference is how much stuff the "customer" will have to read back to you. With grep, you can filter in a way which (in many circumstances I face) allows the "customer" to tell you simply "yes, it printed something" or "no, it didn't print anything". If you pipe through more (or less), the "customer" ends up interpreting the results in some way.
About the level of education of front-line tech support staff, and moreover, the general population, I have this quote from my Windows-using father (who quit school at 15 years old, in 1943): "Back when I used DOS, I knew what I was doing and could control the computer. With Windows, I don't feel like I understand my computer at all."
I personally believe that Microsoft is breeding ignorant computer users, and as such has created a huge tech-support problem for themselves and their "partners". But even if Microsoft didn't create these people, there's always the saying "If any idiot can use Windows, any idiot will." For me, the scary part of that truism is how Microsoft (and Unisys) applies it to system administrators: paraprasing, "Unix requires expensive experts but Windows doesn't".
As a fellow human, I would like to encourage you to continue shutting down computers when they're not needed, to save power. =-)
-Paul Komarek
Re:I think this is the same / links to MRAM articl
on
No More Rebooting?
·
· Score: 2
Aha, so this is why the memory systems are funded by DARPA. Having discovered that their Windows NT controlled ships are dead in the water after a system crash, they're trying to make computers that reboot faster!
This is among my primary complaints about Windows: it's difficult to get useful info from the OS. On a unix system you can run netstat piped through grep to look for open connections, as a simple example.
Not to mention that command lines are more suited to verbal communication than are spatial WIMP metaphors. THe main problem with speaking a command line to a Windows user is that they no nothing about command line conventions, so you have to explain every space to them. That would go away if command lines were more common.
Does this remind anyone else of the Parrot (Python + Perl) joke? In particular, the joke sample code they produced?
Quoting from the joke:
# copy stdin to stdout, except for lines starting with #
while left_angle_right_angle:
if dollar_underscore[0] =eq= "#":
continue_next;
}
I don't want to comment on everything you've written, just one statement:
"I read that there are more trees now than 30 years ago through careful management"
I've seen a fair number of replanted areas, and number of trees is not really the issue. The trees I've seen were pathetic toothpicks compared to the trees removed. They were overdense, and tended to break during winter freezes or high winds. You couldn't use them for lumber (well, you might get one 2x4 from each, and I suppose you could chip them), because they're too small. If these are the trees you've read about, then we haven't yet replaced any of what we've taken. It's not clear to me that these overdense tree plantings will ever resemble the forests they replace.
I think that part of the problem is that we might be screwing up the cycle. Removing some CO2 now would fix one symptom, but not the problem. Note that these are guesses, and I'm no expert.
"do two things: Learn a small amount of French (they'll love you for it)"
This is very true in my experience. When I've used my horrible, broken, wrong, pathetic tatters of French learned long ago, the French I've encountered were patient and made a strong effort to understand and help me. I've been told, though, that they really hate it when some American expects them to know English. Sounds fair enough to me.
You want these people to take something home with them. Two days is very little time. Stay focussed and don't go for breadth. It's fine to do a whizzy tour of lots of stuff for the sake of impressing them and whetting their appetite, but keep the rest simple and focussed.
I've screwed up many one-on-one linux|computers|math|whatever teaching sessions by trying to cover too much stuff to quickly. I'm always forgetting to make it clear to the other person *exactly* what my objective is at all times. As a result, they aren't sure what to "take home" with them, and they end up taking almost nothing home besides the fact that I know what I'm doing and they don't.
So stay focussed. Make the curriculum simple. Whizzy tours are good, but don't expect them to take home more than "gee-whiz" from them. Make it clear at all times what your objective is.
Of course, these rules have very little to do with teaching use of unix systems. But it's while teaching these things that I most often forget these rules.
The problematic details to which I was referring occur at the filesystem level. I haven't seen anything suggesting that software or hardware RAID or linux's lvm layer isn't up to it (though it seems like I remember something vague about possible problems with FreeBSD's vinum). It's ext[23] and UFS that seem to have some issues with really large volumes. And even if there weren't any reports, building a terabyte+ filesystem puts you in a small class of users whose needs are not mainstream. The significance of this is the amount of debugging and fiddling you have to do, or stability problems you have to tolerate. Ask any linux/Alpha user what I mean. =-/
Some posts here have suggested that many of the little gotchas have been taken care of, and also that XFS is doing a good job on really large filesystems. This is all good news. However, it is my opinion that until the new IDE stuff in linux 2.5 solidifies and is backported to 2.4 (or released in 2.6!), big filesystems on linux will have "issues" (even if they aren't bugs). I think the next FreeBSD release ought to clear up a lot of this stuff, too.
What I'd really like is for Andre Hedrick to reply to my posts and prove me wrong!
"That's funny. I always thought the definition of an audiophile was someone who listens to the equipment rather than the music.;-)"
"but I've given up the term "audiophile" when I relised it put me in the same class as people who buy BOSE.:)"
Now those are some funny quotes. =-) I only use the term "audiophile" when defending the interests of people with good ears. That said, it's probably better to buy season tickets for the best seats in the house, than to drop enormous sums of money on audio equipment.
I helped a friend build a recording rig, and in the end we gave up on making it quiet and found good headphones. Someday, maybe, I'll try water cooling.
I am an audiophile. That's why I don't listen to music on computers. Not only can't you get truly good, uncompromising sound equipment in a small space filled with electrical noise (i.e. your computer), but computer fans wipe out significant audio detail.
If you really care about sound, you have a special room for listening, and you won't have your PC in there. Why would anyone spend serious money on audio equipment, and then ruin the sound with computer fans, keyboard and mouse clicks? If you really care about sound, you focus on it while you listen -- that means you aren't surfing the web, writing code, and sending email.
All of that said, I'm also a grad student and spend almost all day, every day writing code and sending email. So my "budget" audio equipment (NAD, because I can't afford the "real" stuff or the listening room to go with it) is in the same room as my computers. If I'm lucky (i.e. once a month), I'll have some time at the end of the day to turn the computers off and do some listening.
Not that you were going to rush out and set up your terabyte array, but here's a word of caution anyway. Terabyte filesystems aren't that common among Linux and FreeBSD users (which I know for sure; I'll bet NetBSD users aren't real interested, either...;-) That means they aren't well tested. Furthermore, there are limits to what is supported. For instance, the 160MB drives aren't supported in linux 2.4 yet, and filesystems have maximum sizes.
Before anyone decides to invest heavily in a terabyte+ array thinking they'll access it as a single logical device, I recommend investing some time reading the linux kernel mailing list, or do similar reading for the OS of choice. Along with some friends, I've spent a lot of time looking at these issues lately, and there are a lot of "gotchas" for unwary users (which almost included us).
Why not? Let's repeat your parent post, but with your suggestion added:
"Here's the kicker.... with $5 worth of parts from Radio Shack, a $45 converter for each serial port you need, and an afternoon's ingenuity (even for a non-programming geek), you can still do something USEFUL with rs232c. Gotta love it. Try that with USB"
I don't want to spend $150 or whatever on a motherboard, and then spend another $50 for basic functionality. I'm sure Abit knows their target market, and I'm sure I'm not in it.
I'm glad I'm not the only one wondering why 4 serial ports isn't standard. And spaced far enough apart (or side by side) that you can use 2 25pin adapters simultaneously.
I'm only familiar with the FreeBSD ports system, and then only somewhat. However, I think it comes down to the same problem I have rpm-based distros, too. Thus this may or may not apply to Gentoo -- I'd love to find out the Gentoo is different.
If I want to use a new version of Mozilla, I'll check to see what the latest "officially" released version. "official" versions allow simple dependency management and file tracking. If the an official version isn't yet available, I'll wait for it because I don't want to "pollute" my filesystem with binaries and such that my packaging system's database/whatever doesn't know about.
What I like about Sorcerer and Lunar Penguin is that the packaging system is really just a bunch of scripts that help you "roll your own" packages. Sure, there's a library of pre-made packages, but changing them only requires a text editor (rpm is the worst offender here, because of their decision to use cpio). In a matter of seconds (i.e. before the urge to get the latest mozilla wears off =-), I can change what version of mozilla the package represents. Just as good, since a package description is just a small directory, I can copy an existing package, make the minor updates, and have a different package.
I'm hoping that you're going to write back and explain that adding a new "package description" (or whatever) to Gentoo is so easy that it only takes seconds and has no real learning curve (again, I'm comparing it to Sorcerer and Lunar Penguin).
Slashdot Mirror
On the other hand, I've worked with about 10 travelstars, all of which were very quiet. I'd complain to Dell that, as the system integrator, they screwed up and picked the wrong travelstar.
-Paul Komarek
I expect that IBM is not having troubles staying alive. I expect it is because IBM is fairly conservative, and attempting to get rid of problems before they do any serious damage.
-Paul Komarek
However, protection is not provided by parents alone. That's why we have an executive branch of government. Furthermore, no society could modify it's executive powers based on the individual whims of each citizen (or pairs of parents, in this case; and the pair might not agree anyway). Instead, the executive branch acts on the whims of society. If we're lucky, those societal whims are averaged out over time, and nothing changes to fast.
-Paul Komarek
Pornography can be covered by free-speech provisions. At least that is what I've gathered from the recent Supreme court decision regarding virtual child pornography, as well as the Miller vs. Someone case the referred to (which they used as a baseline for pornographic materials which are not covered).
/. w/in the last 24 hours, from which I found a several links to the decision. The page I viewed had links to other relevant porn decisions.
The virtual child pornography case was covered on
-Paul Komarek
While there is a problem here, I expect that we can all agree to a few basic principles. If the site describes itself as a porn site, it should go to .prn. If a site wants to verify your age because of laws governing sexually explicit material, it should go to .prn.
I expect that should take care of the sites people are most worried about. After that, a very interesting public debate about porn could ensue, with references made to other countries' standards (from a US viewpoint, since that's where the bill is proposed). I expect that the result would be a loosening of American's "official" standards concerning human nakedness.
Right now, such a debate is nearly impossible. Americans don't seem to make a distinction between nakedness, suggestiveness, and flat-out-screaming pornography. Instituting the simple guidelines I listed above would simplify the situation, and hence might allow such a debate to take place.
-Paul Komarek
It wouldn't make it illegal to tape them. It would make it illegal to tape them without their consent. This is very different.
If you really want to protect your children from abuse, you shouldn't hide that fact from your babysitter. Taping with consent would be a deterrent, possibly preventing the abuse in the first place. Taping without consent, if legal, would help you prosecute the abuser, but your children would still have been abused.
-Paul Komarek
I disagree with absolute interpretations of property ownership. Your house resides in some county, and that county has laws governing how your house is built, what resources (gas, water, etc) it can use, etc. Your house sits underneath airspace you don't own. It is next to your neighbor's houses, and what you do with your house might affect them. It resides in a state and a nation with more laws regulating various aspects of your house.
In short, none of us are one-man nation-states, but we Americans often forget that. We are part of a large society, and our individualistic priorities occasionally need to be adjusted to meet our societies needs. If one doesn't wish to be part of our society, then you are a threat to our society and should leave.
How this applies to filming people in your house, I'm not sure. Only a truly vile scumbag would film their babysitter for anything other than child abuse prevention (which would require a fairly paranoid person at any rate). However, the possibility that home surveilance *could* be abused is not sufficient reason to ban it. And we know that there are plenty of closed-circuit monitors in private office buildings, which indicates some acceptance of such surveilance.
However, notifying your babysitter that you are filming them should not be problem -- if the babysitter objects, they can walk away from the job. I don't see what the problem with this is. Normal citizens are not given the powers of police to go undercover and such things. Even private detectives are not allowed to go undercover like police are. Allowing people to film others in private spaces, without notifying the subjects, seems akin to wiretapping without a warrant at first blush.
-Paul Komarek
From a purely travel point of view, I'd rather go to Finland! I've been to Canada plenty of times.
Thanks for the more optimistic view. It's true that I haven't seen every tree replanting in the US, much less the world. I'm glad to know there are counterexamples to my experience.
-Paul Komarek
Heh, this is the first time I've ever felt a tinge of competitiveness with my operating system. Microsoft just makes me mad (even though I expunged MS from my machine years ago, there's always tech support for friends and family). But Apple is now saying "Okay, we see why you like Unix so much, now we're going to up the ante." It's a great feeling.
Imagine you were a soccer player on a good team. It's fun to play against another good team, with both teams believing they've got what it takes. That's how this add makes me feel. Microsoft, on the other hand, reminds me of an awful game against the foul-mouthed and dirty-playing U14 River City Steelers, refereed by two players from the foul-mouthed and dirty-playing U16 River City Steelers (we lost that game, and many of us were hurt badly -- the next game, with regular referees, we shut them out).
-Paul Komarek
I think I was a little bit misunderstood:
...etc..." but can the average user?"
;-).
1) "Also, *you* can run "netstat | grep "eth0" | sendmail
The point I was trying to make was that it's often easier for a skilled person (tech support) to dictate a command line and say "was anything printed?", than it is to say "click on this and that, oh you don't have that, oh you have that build of Windows, do you see anything like this, okay then click that, now, describe what you see, in detail".
2) "As to the comment about command lines... I disagree. Just to figure out whats up with a net connection you *should* be able to point and click. In win2k I can right click on my network icon to see if I am connected. In that dialog I can click "properties" and setup my TCP settings, etc..."
Again, I was suggesting that it's easier to describe plain text commands and responses over the phone, than it is to describe mouse movements and an array of 1024x768 pixels over the phone. If you know where to click on your build of Windows, that's great; if I know what to type on my command line, that's dandy. The issue I meant to deal with was giving commands and getting responses over the phone.
Of course, if I'm talking to someone with a unix box, it is usually very easy to login and solve problems remotely. Microsoft used to describe these remote administration features in unix systems as "security holes". Now Microsoft is trying to add them to their on operating systems (but Unisys seems to prefer the Free software program VNC
For the example of piping through more versus piping through grep, a big difference is how much stuff the "customer" will have to read back to you. With grep, you can filter in a way which (in many circumstances I face) allows the "customer" to tell you simply "yes, it printed something" or "no, it didn't print anything". If you pipe through more (or less), the "customer" ends up interpreting the results in some way.
About the level of education of front-line tech support staff, and moreover, the general population, I have this quote from my Windows-using father (who quit school at 15 years old, in 1943): "Back when I used DOS, I knew what I was doing and could control the computer. With Windows, I don't feel like I understand my computer at all."
I personally believe that Microsoft is breeding ignorant computer users, and as such has created a huge tech-support problem for themselves and their "partners". But even if Microsoft didn't create these people, there's always the saying "If any idiot can use Windows, any idiot will." For me, the scary part of that truism is how Microsoft (and Unisys) applies it to system administrators: paraprasing, "Unix requires expensive experts but Windows doesn't".
-Paul komarek
As a fellow human, I would like to encourage you to continue shutting down computers when they're not needed, to save power. =-)
-Paul Komarek
Aha, so this is why the memory systems are funded by DARPA. Having discovered that their Windows NT controlled ships are dead in the water after a system crash, they're trying to make computers that reboot faster!
-Paul Komarek
This is among my primary complaints about Windows: it's difficult to get useful info from the OS. On a unix system you can run netstat piped through grep to look for open connections, as a simple example.
Not to mention that command lines are more suited to verbal communication than are spatial WIMP metaphors. THe main problem with speaking a command line to a Windows user is that they no nothing about command line conventions, so you have to explain every space to them. That would go away if command lines were more common.
-Paul Komarek
Does this remind anyone else of the Parrot (Python + Perl) joke? In particular, the joke sample code they produced?
Quoting from the joke:
# copy stdin to stdout, except for lines starting with #
while left_angle_right_angle:
if dollar_underscore[0] =eq= "#":
continue_next;
}
-Paul Komarek
print dollar_underscore;
I don't want to comment on everything you've written, just one statement:
"I read that there are more trees now than 30 years ago through careful management"
I've seen a fair number of replanted areas, and number of trees is not really the issue. The trees I've seen were pathetic toothpicks compared to the trees removed. They were overdense, and tended to break during winter freezes or high winds. You couldn't use them for lumber (well, you might get one 2x4 from each, and I suppose you could chip them), because they're too small. If these are the trees you've read about, then we haven't yet replaced any of what we've taken. It's not clear to me that these overdense tree plantings will ever resemble the forests they replace.
-Paul Komarek
I think that part of the problem is that we might be screwing up the cycle. Removing some CO2 now would fix one symptom, but not the problem. Note that these are guesses, and I'm no expert.
-Paul Komarek
"do two things: Learn a small amount of French (they'll love you for it)"
This is very true in my experience. When I've used my horrible, broken, wrong, pathetic tatters of French learned long ago, the French I've encountered were patient and made a strong effort to understand and help me. I've been told, though, that they really hate it when some American expects them to know English. Sounds fair enough to me.
-Paul Komarek
You want these people to take something home with them. Two days is very little time. Stay focussed and don't go for breadth. It's fine to do a whizzy tour of lots of stuff for the sake of impressing them and whetting their appetite, but keep the rest simple and focussed.
I've screwed up many one-on-one linux|computers|math|whatever teaching sessions by trying to cover too much stuff to quickly. I'm always forgetting to make it clear to the other person *exactly* what my objective is at all times. As a result, they aren't sure what to "take home" with them, and they end up taking almost nothing home besides the fact that I know what I'm doing and they don't.
So stay focussed. Make the curriculum simple. Whizzy tours are good, but don't expect them to take home more than "gee-whiz" from them. Make it clear at all times what your objective is.
Of course, these rules have very little to do with teaching use of unix systems. But it's while teaching these things that I most often forget these rules.
-Paul Komarek
The problematic details to which I was referring occur at the filesystem level. I haven't seen anything suggesting that software or hardware RAID or linux's lvm layer isn't up to it (though it seems like I remember something vague about possible problems with FreeBSD's vinum). It's ext[23] and UFS that seem to have some issues with really large volumes. And even if there weren't any reports, building a terabyte+ filesystem puts you in a small class of users whose needs are not mainstream. The significance of this is the amount of debugging and fiddling you have to do, or stability problems you have to tolerate. Ask any linux/Alpha user what I mean. =-/
Some posts here have suggested that many of the little gotchas have been taken care of, and also that XFS is doing a good job on really large filesystems. This is all good news. However, it is my opinion that until the new IDE stuff in linux 2.5 solidifies and is backported to 2.4 (or released in 2.6!), big filesystems on linux will have "issues" (even if they aren't bugs). I think the next FreeBSD release ought to clear up a lot of this stuff, too.
What I'd really like is for Andre Hedrick to reply to my posts and prove me wrong!
-Paul Komarek
"That's funny. I always thought the definition of an audiophile was someone who listens to the equipment rather than the music. ;-)"
:)"
"but I've given up the term "audiophile" when I relised it put me in the same class as people who buy BOSE.
Now those are some funny quotes. =-) I only use the term "audiophile" when defending the interests of people with good ears. That said, it's probably better to buy season tickets for the best seats in the house, than to drop enormous sums of money on audio equipment.
I helped a friend build a recording rig, and in the end we gave up on making it quiet and found good headphones. Someday, maybe, I'll try water cooling.
-Paul Komarek
I am an audiophile. That's why I don't listen to music on computers. Not only can't you get truly good, uncompromising sound equipment in a small space filled with electrical noise (i.e. your computer), but computer fans wipe out significant audio detail.
If you really care about sound, you have a special room for listening, and you won't have your PC in there. Why would anyone spend serious money on audio equipment, and then ruin the sound with computer fans, keyboard and mouse clicks? If you really care about sound, you focus on it while you listen -- that means you aren't surfing the web, writing code, and sending email.
All of that said, I'm also a grad student and spend almost all day, every day writing code and sending email. So my "budget" audio equipment (NAD, because I can't afford the "real" stuff or the listening room to go with it) is in the same room as my computers. If I'm lucky (i.e. once a month), I'll have some time at the end of the day to turn the computers off and do some listening.
-Paul Komarek
Not that you were going to rush out and set up your terabyte array, but here's a word of caution anyway. Terabyte filesystems aren't that common among Linux and FreeBSD users (which I know for sure; I'll bet NetBSD users aren't real interested, either...;-) That means they aren't well tested. Furthermore, there are limits to what is supported. For instance, the 160MB drives aren't supported in linux 2.4 yet, and filesystems have maximum sizes.
Before anyone decides to invest heavily in a terabyte+ array thinking they'll access it as a single logical device, I recommend investing some time reading the linux kernel mailing list, or do similar reading for the OS of choice. Along with some friends, I've spent a lot of time looking at these issues lately, and there are a lot of "gotchas" for unwary users (which almost included us).
-Paul Komarek
Why not? Let's repeat your parent post, but with your suggestion added:
"Here's the kicker.... with $5 worth of parts from Radio Shack, a $45 converter for each serial port you need, and an afternoon's ingenuity (even for a non-programming geek), you can still do something USEFUL with rs232c. Gotta love it. Try that with USB"
I don't want to spend $150 or whatever on a motherboard, and then spend another $50 for basic functionality. I'm sure Abit knows their target market, and I'm sure I'm not in it.
-Paul
I'm glad I'm not the only one wondering why 4 serial ports isn't standard. And spaced far enough apart (or side by side) that you can use 2 25pin adapters simultaneously.
-Paul Komarek
I'm only familiar with the FreeBSD ports system, and then only somewhat. However, I think it comes down to the same problem I have rpm-based distros, too. Thus this may or may not apply to Gentoo -- I'd love to find out the Gentoo is different.
If I want to use a new version of Mozilla, I'll check to see what the latest "officially" released version. "official" versions allow simple dependency management and file tracking. If the an official version isn't yet available, I'll wait for it because I don't want to "pollute" my filesystem with binaries and such that my packaging system's database/whatever doesn't know about.
What I like about Sorcerer and Lunar Penguin is that the packaging system is really just a bunch of scripts that help you "roll your own" packages. Sure, there's a library of pre-made packages, but changing them only requires a text editor (rpm is the worst offender here, because of their decision to use cpio). In a matter of seconds (i.e. before the urge to get the latest mozilla wears off =-), I can change what version of mozilla the package represents. Just as good, since a package description is just a small directory, I can copy an existing package, make the minor updates, and have a different package.
I'm hoping that you're going to write back and explain that adding a new "package description" (or whatever) to Gentoo is so easy that it only takes seconds and has no real learning curve (again, I'm comparing it to Sorcerer and Lunar Penguin).
-Paul Komarek