Yep, and applications still crash sometimes, too. I've even had the OS itself go poorly-responsive or even casters-up a few times in the past half-decade.
I'm not complainging about Linux reliability, but...
I understand the reasons why cost metrics are poor, and your mentioned difficulties of making things better. But I wonder if we took the amount of brainpower that has been applied to spam filters and apply some of that to fixing cost metrics, we might be better off.
Isn't the real issue that we have poor cost metrics and service agreements on the Internet?
The cost of sending spam is virtually invisible to the spammer.
That's it, plain and simple. As long as 'one single potential customer' responds, it's 'Step 3: PROFIT!' (I know the spammer paid for the bulk mailer program, and pays for his/her rotating ISP fees, but the costs are effectively near zero.) But that spam wasn't really free, we all pay for it. Does anyone have a good estimate of how much Internet bandwidth gets wasted on spam? Or how much recipient time gets wasted downloading it, even if it is then automatically filtered and thrown away?
Today I pay a flat fee for my cable modem, and it carries certain Terms Of Service. Same back when I was on dialup. There is supposed to be an aggregate bandwidth cap on cable, but I never worry about it except when I'm going to download a new set of ISOs. There were dialup time limits as well, but in that case they were mostly worried about camping on the modem pool, not bandwidth usage.
Is it possible to come up with a more refined billing model that could effectively shift costs back to spammers without killing mailing lists?
In the theoretical purity of a free and open marketplace where informed consumers had a choice, I would agree with you.
OTOH, in a marketplace where there is every attempt to shove products down our throats and make us pay for things we didn't necessarily intend to buy, deny us the ability to choose, and misinform, I fear I must disagree.
I wish I could agree with you more.
I wish I didn't feel our legislature (and executive branch) was so 0wn3d.
Our government responds to campaign finance, and the lion's share of that is done by large corporations and other aggregates that want to make sure that THEIR rights come first.
Most people don't understand enough about computers to understand how completely OUR rights in this realm have been trampled, already.
Did you use 'noatime' for filesystems on the CF drive? Obviously this won't do squat for the issue of frequent writes to/var, but it should eliminate the need to read the OS into tmpfs.
Thanks for the reference. I've read the whole "White Hart," which is obviously where I picked the story up. For a while, it spawned the whole "future bar" genre in science fiction as a collecting place for short stories.
I was only going to put out a spoiler if someone couldn't identify the story/book. At least for a while we still have public libraries where this can most likely be borrowed.
A scientist searching for the "perfect tune," included various discussions about how good tunes stuck with you, and possible effects when the "perfect tune" was found.
I'll stop there, and not spoil.
Unfortunately I can remember neither author nor title, but perhaps someone else will. If nothing pops up, respond and I can give a spoiler.
This one is more reputable, I believe credited to Arthur C. Clarke.
It was a short story about an Earth-to-Moon (orbit-to-orbit) space race, in the spirit of the Kremer prize. The spacecraft were propelled by ion engines, which were energized by Whimshurst-type machines, which were powered by...
bicycles.
The racers pedaled their way to the moon, the pedals effectively powering the ion engines that drove them. The race took several days, with the right stuff added in for absurd athletics, rest breaks, minimal life-support, race security, etc.
No doubt someone here will do the math that I never bothered trying to do. One of these days, maybe I will.
that we should ignore all of those silly Opteron and Athlon 64 announcements in the past six months, because next year Intel will announce something that will blow them all away, and lead us all back to the One True Processor Roadmap.
Does this qualify as a pre-announcement, that just happens to be overlapping a competitor's introduction? I seem to remember that several decades ago, another three-letter company got in a decade-long heap of trouble for just that type of behavior. (Amoung others, but then there are more stories of things Intel has done to keep AMD 'present, but weak.')
Last I knew, we all participated in the private sector. Stupidity in the private sector bleeds over just like it does in the public sector.
Then too, there's the old song: "Well, I'm changing my name to Chrysler..."
"When they hand a million grand out, I'll be standing with my hand out. I'll get mine."
Too often, the cries at the stupidity of the public sector are simply giving up. IMHO, both forms of stupidity are equally bad, especially when unpunished. (by votes or by the marketplace)
Sure, volume manufacturing would make shuttle tile material cheaper, but it's a question of whether it's worth doing. For most earthbound applications, weight doesn't matter much, and size/volume may not be *that* important. So what if firebrick has to be five times thicker, twenty times heavier, and still doesn't do as good a job. It's a wood stove. (or blast furnace, or kiln, or whatever) Even in volume manufacturing, I doubt shuttle tiles would approach firebrick closely enough to be worth it.
Western civilization arose and became dominant through innovation - "Standing on the shoulders of giants." The way patent and copyright laws are going in the West, the giants not only no longer want anyone standing on their shoulders, they don't even want anyone casting similar shadows or reaching for the same goals.
Look to China, and expect them to walk a fine line between sufficient copyright and patent protection that we will still trade with them, yet avoid the sheer lunacy we're seeing now. I wouldn't be surprised if copyright and patent issues force dual-design, in some cases to separate internally acceptable from exportable. Through the next century China's domestic market will be the next boom area, and I doubt they're going to let Western copyright and patent silliness stop them from modernizing, even if it does prevent some exports.
We're imposing legal morbidity on our technology. Those who don't will have an edge over us.
An early adult dream, as opposed to the late childhood dream of an orbiting factory, was to build a woodstove lined with shuttle tiles. The idea was to preheat incoming air and burn with very little heat loss other than deliberate extraction. That would allow the firebox to stay at the right temperature for clean burning, even at low burn rates.
Anyway, shuttle tiles aren't cheap, and it isn't just because they're on the shuttle built to government specifications. Shuttle tile material is expensive to fabricate, and I've heard indications that it's really not a very nice thing to have around. Not quite like asbestos, but in that general direction.
Before picking up a piece of aerogel, I'd like to hear more about its biological properties. I can easily see bits and pieces flaking off and floating through the air, to be inhaled or ingested.
In my high school years, (early 70s) my dream was to set up an orbiting factory making foam steel, flat (radial gradient density glass) lenses, and hollow ball bearings. Of course bubbloy apparently doesn't need to be manufactured in zero G, and my (or anyone else's) factory isn't up there.
IIWIC, (If I Were In Charge) I'd declare a tax holiday on 50+ mile content in order to promote greater space access. I might exclude satellite communications relays, since that industry is already well developed and I'm trying to foster new development. OTOH maybe not, since more development opportunities might arise.
To answer your question directly, cost is probably the biggest reason. But there's more...
"programmers have to start thinking of the disk as a sequential device rather than a random access device."
"I think we'd all be better off when solid state, non-mechanical disks become commonplace."
Now that you mention it, I don't think so. In reality, there's a lot time spent dissing the latency (and even bandwidth) of DRAM. (any flavor) That's why caching is being elevated to a fine art form. That's why Intel is introducing L3 cache in their new "Extreme" gaming chip.
The reality is that the next level up is never as "good" as the level below, only denser and cheaper. Caching helps get around this, and good old Single Level Store extended the concept by treating the disk as the real (bit) address space and main memory as a cache of that.
Even without Single Level Store already disk paging and caching mechanisms, head elevator algorithms, I/O schedulers and the like already do some of this 'sequentializing' for us, behind the covers.
IMHO, we're getting where he wants to go, through the back door....the other answer to your question, by the time solid-state disk prices are low enough, access methods will have shifted enough and memory hierarchy grown enough that the effects simply won't be what you'd expect given the technology, today.
If this is Lindows, you are probably not running sshd. Furthermore, you should not do so.
To do damage, you need two things. First, you need to get into the machine, and second, you need to elevate your privilege to root. So to get nasty, you either need an exploit for a service running as root, or you need a second exploit to become root. Especially in recent years, most daemons have tried very hard to avoid running as root, forcing the need to find two bugs instead of one.
Lindows runs the ordinary user as root. I don't know if they have a better policy for running daemons as non-root - I certainly hope so. But running ordinarily as root removes one line of defense against crackers.
Add to that the fact that Lindows is aimed at the inexperienced market, so these systems *may* be less likely to be correctly patched.
I don't think we're saying anything different. I wasn't calling nature immature, just us.
Your line, "the capacity for our species to deal with this conflict and stop it from reaching a species-destroying level," says everything I want to say. I agree with you, in this. My point of tying it to space access is that, ready access to levels of power needed for everyday space access can cause greater destruction in 'immature' hands.
IMHO we need to make more fundamental changes to our nature. Simply putting us in more cages either means that some cages wipe themselves out (Hmmm, maybe this IS evolution, maybe its not a bad idea, after all.) or we manage to invent interplanetary war and have the cages fighting each other - no net gain. Back to the self-destructing cages - that is still my fundamental change to our nature, just through a different and more lethal means.
But if I could put 1000kg up in orbit for $500 of pure energy, I'd be sending up a heck of a lot more than 1000kg. First off, I'd take my family for a vacation.
Even if it is finite and ordinary, given the technology to get into orbit for even 50% efficiency compared to energy cost, the business would expand . We'd still be producing massive amounts of energy, moving even more massive amounts of things to orbit. (Still mostly just people down from orbit, at least until we get asteroid mining running.)
It's still a concentration of energy, but more than that, getting that efficiency to 50% represents whole new technologies. The possiblities for abuse are fantastic. Discussions on tethers and elevators touch on accidents, and it's pretty much decided that proper design can minimize the risk, even if an accident happens.
But imagine if tether technology was used to deliberatily design a weapon.
I'm in favor of all of this stuff, but IMHO we HAVE to mature, as a species, as we gain access to it.
Your 35 cars worth of energy are focused into 1kg of mass. Not much gets done with 1kg, except perhaps a projectile. To even begin doing something reasonable, you need to be talking 1000kg, or more. So the kinetic energy of 35,000 of your example cars is focused into less mass than one of them. Then even 1000kg is on the light side, when we talk of wishing to get real industry and life started above the atmosphere.
You're right, in that the numbers are finite and manageable. But they're still much bigger than the amounts of energy we normally throw around on Earth. It might be more interesting to compare orbital energy to earthmoving machinery.
Slashdot Mirror
Yep, and applications still crash sometimes, too. I've even had the OS itself go poorly-responsive or even casters-up a few times in the past half-decade.
I'm not complainging about Linux reliability, but...
To deny the possibity of a crash is arrogant.
It's good that the FireGL drivers are in the box. The next questions are:
How about nVidia drivers, are they in-the-box?
Is this stuff in the downloadable ISOs?
What kernel is used? Support for nForce2 has really picked up on recent kernels. (post 2.4.20)
(Another A7N8X Dlx owner with ATI video)
Not even sshd?
/sbin/sendmail will be there for crash reporting?
(or how about chrond/atd/sgifam or those other desktop daemons?)
How about programs that make the assumption that either port25 or
I understand the reasons why cost metrics are poor, and your mentioned difficulties of making things better. But I wonder if we took the amount of brainpower that has been applied to spam filters and apply some of that to fixing cost metrics, we might be better off.
Isn't the real issue that we have poor cost metrics and service agreements on the Internet?
The cost of sending spam is virtually invisible to the spammer.
That's it, plain and simple. As long as 'one single potential customer' responds, it's 'Step 3: PROFIT!' (I know the spammer paid for the bulk mailer program, and pays for his/her rotating ISP fees, but the costs are effectively near zero.) But that spam wasn't really free, we all pay for it. Does anyone have a good estimate of how much Internet bandwidth gets wasted on spam? Or how much recipient time gets wasted downloading it, even if it is then automatically filtered and thrown away?
Today I pay a flat fee for my cable modem, and it carries certain Terms Of Service. Same back when I was on dialup. There is supposed to be an aggregate bandwidth cap on cable, but I never worry about it except when I'm going to download a new set of ISOs. There were dialup time limits as well, but in that case they were mostly worried about camping on the modem pool, not bandwidth usage.
Is it possible to come up with a more refined billing model that could effectively shift costs back to spammers without killing mailing lists?
In the theoretical purity of a free and open marketplace where informed consumers had a choice,
I would agree with you.
OTOH, in a marketplace where there is every attempt to shove products down our throats and make us pay for things we didn't necessarily intend to buy, deny us the ability to choose, and misinform,
I fear I must disagree.
I wish I could agree with you more.
I wish I didn't feel our legislature (and executive branch) was so 0wn3d.
Not just Executive, but Legislative, as well.
Our government responds to campaign finance, and the lion's share of that is done by large corporations and other aggregates that want to make sure that THEIR rights come first.
Most people don't understand enough about computers to understand how completely OUR rights in this realm have been trampled, already.
Did you use 'noatime' for filesystems on the CF drive? Obviously this won't do squat for the issue of frequent writes to /var, but it should eliminate the need to read the OS into tmpfs.
I think I read that one, too. But the bicycle/ion engine story seemed more appropriate to the topic.
Thanks for the reference. I've read the whole "White Hart," which is obviously where I picked the story up. For a while, it spawned the whole "future bar" genre in science fiction as a collecting place for short stories.
I was only going to put out a spoiler if someone couldn't identify the story/book. At least for a while we still have public libraries where this can most likely be borrowed.
Science fiction story reference...
A scientist searching for the "perfect tune," included various discussions about how good tunes stuck with you, and possible effects when the "perfect tune" was found.
I'll stop there, and not spoil.
Unfortunately I can remember neither author nor title, but perhaps someone else will. If nothing pops up, respond and I can give a spoiler.
This one is more reputable, I believe credited to Arthur C. Clarke.
...
It was a short story about an Earth-to-Moon (orbit-to-orbit) space race, in the spirit of the Kremer prize. The spacecraft were propelled by ion engines, which were energized by Whimshurst-type machines, which were powered by
bicycles.
The racers pedaled their way to the moon, the pedals effectively powering the ion engines that drove them. The race took several days, with the right stuff added in for absurd athletics, rest breaks, minimal life-support, race security, etc.
No doubt someone here will do the math that I never bothered trying to do. One of these days, maybe I will.
that we should ignore all of those silly Opteron and Athlon 64 announcements in the past six months, because next year Intel will announce something that will blow them all away, and lead us all back to the One True Processor Roadmap.
Does this qualify as a pre-announcement, that just happens to be overlapping a competitor's introduction? I seem to remember that several decades ago, another three-letter company got in a decade-long heap of trouble for just that type of behavior. (Amoung others, but then there are more stories of things Intel has done to keep AMD 'present, but weak.')
Last I knew, we all participated in the private sector. Stupidity in the private sector bleeds over just like it does in the public sector.
Then too, there's the old song:
"Well, I'm changing my name to Chrysler..."
"When they hand a million grand out, I'll be standing with my hand out. I'll get mine."
Too often, the cries at the stupidity of the public sector are simply giving up. IMHO, both forms of stupidity are equally bad, especially when unpunished. (by votes or by the marketplace)
Just remember mantra...
The private sector does everything right.
The government does everything wrong.
Idiocy is the exclusive right of the government sector.
Anyone want a bridge?
Sure, volume manufacturing would make shuttle tile material cheaper, but it's a question of whether it's worth doing. For most earthbound applications, weight doesn't matter much, and size/volume may not be *that* important. So what if firebrick has to be five times thicker, twenty times heavier, and still doesn't do as good a job. It's a wood stove. (or blast furnace, or kiln, or whatever) Even in volume manufacturing, I doubt shuttle tiles would approach firebrick closely enough to be worth it.
Much as I might like the idea.
Here it is, and I'm not even kidding.
Western civilization arose and became dominant through innovation - "Standing on the shoulders of giants." The way patent and copyright laws are going in the West, the giants not only no longer want anyone standing on their shoulders, they don't even want anyone casting similar shadows or reaching for the same goals.
Look to China, and expect them to walk a fine line between sufficient copyright and patent protection that we will still trade with them, yet avoid the sheer lunacy we're seeing now. I wouldn't be surprised if copyright and patent issues force dual-design, in some cases to separate internally acceptable from exportable. Through the next century China's domestic market will be the next boom area, and I doubt they're going to let Western copyright and patent silliness stop them from modernizing, even if it does prevent some exports.
We're imposing legal morbidity on our technology. Those who don't will have an edge over us.
An early adult dream, as opposed to the late childhood dream of an orbiting factory, was to build a woodstove lined with shuttle tiles. The idea was to preheat incoming air and burn with very little heat loss other than deliberate extraction. That would allow the firebox to stay at the right temperature for clean burning, even at low burn rates.
Anyway, shuttle tiles aren't cheap, and it isn't just because they're on the shuttle built to government specifications. Shuttle tile material is expensive to fabricate, and I've heard indications that it's really not a very nice thing to have around. Not quite like asbestos, but in that general direction.
Before picking up a piece of aerogel, I'd like to hear more about its biological properties. I can easily see bits and pieces flaking off and floating through the air, to be inhaled or ingested.
In my high school years, (early 70s) my dream was to set up an orbiting factory making foam steel, flat (radial gradient density glass) lenses, and hollow ball bearings. Of course bubbloy apparently doesn't need to be manufactured in zero G, and my (or anyone else's) factory isn't up there.
IIWIC, (If I Were In Charge) I'd declare a tax holiday on 50+ mile content in order to promote greater space access. I might exclude satellite communications relays, since that industry is already well developed and I'm trying to foster new development. OTOH maybe not, since more development opportunities might arise.
To answer your question directly, cost is probably the biggest reason. But there's more...
...the other answer to your question, by the time solid-state disk prices are low enough, access methods will have shifted enough and memory hierarchy grown enough that the effects simply won't be what you'd expect given the technology, today.
"programmers have to start thinking of the disk as a sequential device rather than a random access device."
"I think we'd all be better off when solid state, non-mechanical disks become commonplace."
Now that you mention it, I don't think so. In reality, there's a lot time spent dissing the latency (and even bandwidth) of DRAM. (any flavor) That's why caching is being elevated to a fine art form. That's why Intel is introducing L3 cache in their new "Extreme" gaming chip.
The reality is that the next level up is never as "good" as the level below, only denser and cheaper. Caching helps get around this, and good old Single Level Store extended the concept by treating the disk as the real (bit) address space and main memory as a cache of that.
Even without Single Level Store already disk paging and caching mechanisms, head elevator algorithms, I/O schedulers and the like already do some of this 'sequentializing' for us, behind the covers.
IMHO, we're getting where he wants to go, through the back door.
Not only does it give your name and address, but it includes handy links to Yahoo Maps and Mapquest to show where you live.
If this is Lindows, you are probably not running sshd. Furthermore, you should not do so.
To do damage, you need two things. First, you need to get into the machine, and second, you need to elevate your privilege to root. So to get nasty, you either need an exploit for a service running as root, or you need a second exploit to become root. Especially in recent years, most daemons have tried very hard to avoid running as root, forcing the need to find two bugs instead of one.
Lindows runs the ordinary user as root. I don't know if they have a better policy for running daemons as non-root - I certainly hope so. But running ordinarily as root removes one line of defense against crackers.
Add to that the fact that Lindows is aimed at the inexperienced market, so these systems *may* be less likely to be correctly patched.
I don't think we're saying anything different. I wasn't calling nature immature, just us.
Your line, "the capacity for our species to deal with this conflict and stop it from reaching a species-destroying level," says everything I want to say. I agree with you, in this. My point of tying it to space access is that, ready access to levels of power needed for everyday space access can cause greater destruction in 'immature' hands.
IMHO we need to make more fundamental changes to our nature. Simply putting us in more cages either means that some cages wipe themselves out (Hmmm, maybe this IS evolution, maybe its not a bad idea, after all.) or we manage to invent interplanetary war and have the cages fighting each other - no net gain. Back to the self-destructing cages - that is still my fundamental change to our nature, just through a different and more lethal means.
But if I could put 1000kg up in orbit for $500 of pure energy, I'd be sending up a heck of a lot more than 1000kg. First off, I'd take my family for a vacation.
Even if it is finite and ordinary, given the technology to get into orbit for even 50% efficiency compared to energy cost, the business would expand . We'd still be producing massive amounts of energy, moving even more massive amounts of things to orbit. (Still mostly just people down from orbit, at least until we get asteroid mining running.)
It's still a concentration of energy, but more than that, getting that efficiency to 50% represents whole new technologies. The possiblities for abuse are fantastic. Discussions on tethers and elevators touch on accidents, and it's pretty much decided that proper design can minimize the risk, even if an accident happens.
But imagine if tether technology was used to deliberatily design a weapon.
I'm in favor of all of this stuff, but IMHO we HAVE to mature, as a species, as we gain access to it.
Your 35 cars worth of energy are focused into 1kg of mass. Not much gets done with 1kg, except perhaps a projectile. To even begin doing something reasonable, you need to be talking 1000kg, or more. So the kinetic energy of 35,000 of your example cars is focused into less mass than one of them. Then even 1000kg is on the light side, when we talk of wishing to get real industry and life started above the atmosphere.
You're right, in that the numbers are finite and manageable. But they're still much bigger than the amounts of energy we normally throw around on Earth. It might be more interesting to compare orbital energy to earthmoving machinery.