It's not just flammability. Notice the part that says "All New Materials"? This is to set it apart from say, the chairs my cheap-ass employer bought, where the tag is bright yellow and says "All Second-Hand Material", but, reassuringly, "Contents Sterilized".
Yes, DC-to-daylight is a lot of spectrum representing a lot of bandwidth. But EM spectrum with specific characteristics is very limited. VHF is 30-300MHz, period. If your application needs the characteristics of VHF transmission and reception, that's where ya gotta be, like it or not.
A trip to the NYC Transit Museum is very enlightening when you start looking at the hardware required to get all that current to the right place (think of what a 2000 amp circuit breaker would look like).
Crank up the MID? Where's this? Do that and you'll have the crowd complaining about the tinny sound.
Every time a DJ gets into my effects rack (and they have some ingenious ways of doing it despite everything I do to keep them out once my PA is tuned up) they leave the EQ looking like a smiley face, because they think it maxes the boom-chicks.
What most DJs (who aren't also sound techs) don't know is that the real kick from the bass is not at the low end, it's in the midbass. So when they have the 20Hz and 40Hz sliders at +12 it ends up sounding like the Cerwin-Vegas in a white suburban homeboy's lowered Hyundai.
The best solution to ear-ringing is a reasonably quiet chill room.
Or earplugs. I don't want to get off on a rant here, but earplugs are the best 50 cents you can ever spend. Earplugs have the advantage that they drop the overall sound (preventing the *permanent* damage you can get from just one night in front of the speaker stacks) to let you get close enough to the bass bins for a nice comfortable rib-cage massage, but they do so in a way that someone can talk to you in only a slightly louder than normal voice close to your ear and it's perfectly intelligible.
The fire is burning toward him at 1mph, but the smoke is blowing toward him at 2mph. Therefore all he has to so is asphyxiate himself on the smoke, and then he won't burn to death...
OK, this may be a Stupid Question, but, the announcement said that, "As always, Debian GNU/Linux systems can be upgraded painlessly, in place, without any forced downtime."
I cannot imagine an economic system which will make you happy.
That's because you're thinking in extremes. My ideal, irrespective of implementation, has mechanisms in place to minimize corruption and opportunism (!= ambition).
RSI is a physiological consequence of too much typing
What part of "programming for a living" didn't you understand?
I never said that the amorality of corporations was right or wrong. What we have to consider is that because corproations are amoral, they are in theory free to do whatever they deem necessary to achieve their goals. If those actions are harmful to society, then society should have the capability to reign in those harmful actions.
The free market is not an effective deterrent against corporations' ill behavior, nor does it encourage good behavior. Not because it is unwilling to, but because there is no such thing as a free market. Unfettered markets always become less free, until they are controlled by a monopoly or a collusive oligopoly, which will remain in place until it is disrupted. This disruption may be anything from new technology to revolution. When that happens the process restarts.
With each iteration, the competitors get more and more creative in the ways they compete. The first time around, it may be as simple as "the best product at the lowest price". Next time, it's "the perceived best product at the lowest price". Then "the most convenient product at maybe the lowest price". Then "whatever everybdy else is buying because it's now impossible to tell whether it's the best product or not". Then, it's "whatever everybody else is buying because we're all getting our information about what to buy from the same source which is really the supplier of what we're going to buy, and they're playing on either our hopes, our fears, or our desires to that end." That's where Orwell steps in.
And that's where I step out. You're welcome to have the last word.
I get to choose with which corporations to interact
My point is that absent some sort of outside enforcement of a reasonable level of safety and quality of products, choosing your corporation is like choosing between hanging or the firing squad.
Whether I get RSI by programming for a living implies that, at least in our economic system, I'm overwhelmingly likely to be doing so for a corporation. That corporation, by definition an amoral entity, doesn't care about that except where it may hurt their profits.
When the Challenger shuttle blew up, it wasn't corporate greed that made it do it, was it?
Yes, in fact it was.
The explosion was caused by a set of decisions, some of which being motivated by fear of bad publicity for the for-profit contractors, leading up to the launch taking place even though it was known that the air temperature was out of spec for the O-rings. They knew it and went ahead anyway. I guess it didn't save Morton-Thiokol as much money as they thought it would.
I'll say it again: Corporations are amoral entities, beholden only to their executives and shareholders. A corporation will kill if it is profitable to do so, until and unless the shareholders, through their votes, agree to give up some of their profits in order to be so-called "good corporate citizens".
Many markets are controlled by oligopolies. Think of how many industries have a "big three". They can collude to form de facto monopolies (the inability to prevent which is why libertarianism fails). If the products they produce are considered essential to normal living, then we get them precisely on their terms. It is in their interest for us not to believe we've been forced to grab our ankles in this regard, so part of their strategy is to get us to be grateful for what they're giving us (which, as another poster pointed out, is more Huxleyan than Orwellian).
So, I still stand by my original claim that modern-day corporations are more Orwellian than modern-day governments.
Another part of the problem is that in a lot of the South, education is funded out of sales and income taxes rather than property taxes. So, when the economy goes downhill (I was going to say "south" but...), they lose revenue, and the schools go broke. Up Nawth, we fund our schools primarily from property taxes, which don't change with the economic weather.
Not sure about the safety of the product? Don't buy it. Think your job is dangerous? Quit it.
Grow up. Interaction with corporations is unavoidable unless you completely drop out of the economy. Even then, every habitable area is geographically close to/downwind from/downstream from corporate facilities enough that their activities may still affect you.
Am I to stop buying canned food for fear it's contaminated with botulism? Am I to stop programming for a living for fear that some new form of RSI will disable me decades from now?
Were the workers at Johns-Manville supposed to Just Know that the armloads of raw asbestos they were hauling around would eventually shred their lungs? Were the people who lived around the Union Carbide plant in Bhopal supposed to Just Know that there was seriously bad stuff that could leak out of the plant? Were the Alaskan fishermen supposed to Just Know that some drunken captain might run his tanker aground in Prudhoe Bay?
In all cases the corporations knew these things could happen. The hazards of asbestos were documented in the 20's and 30's, Union Carbide knew the Bhopal plant was in poor repair, and Exxon knew of Hazelwood's drinking problem.
To get back to the original topic, it is truly Orwellian that these companies were able to do these things then hide behind their PR firms and bankruptcy lawyers to get away with relatively small slaps on their wrists.
I guess we should be glad that at least J-M, UC and Exxon didn't sue those people for anything and do them REAL harm.
If you know as much history as you're trying to convince me you do, then you appear to be proactively ignoring a lot of it.
I've never seen film at 48fps, but I have seen Showscan (60fps) and the realism is awesome.
The first time I saw it, a guy came out on stage beforehand to explain what Showscan was and how it worked. It wasn't until they did a slow wipe from him to the beginning of the actual content that we all realized we'd been had, that he *was* the first part of the film.
This was at a Showbiz Pizza Place, the precursor to Chuck-E-Cheese, circa 1983.
I don't have time to answer all your points (because my employer, unlike my government, forbids me from this sort of activity), but I do have time to address one:
(3) The very worst thing that corporations can do to you is sue you into bankrupcy.
Don't know much history, do you?
The very worst thing a corporation can do is kill you, either by selling you a deadly (or fatally defective) product, or by giving you a hazardous job, or by poisoning your environment.
1984 was set in a supposedly advanced western society, and was targeted at readers in western democracies, not Singapore, Burma or Nicaragua.
Looked at from today's perspective (as the original article does), where most of these nations have constitutions that have teeth in them, it's clear that most of these governments are not capable of deploying widespread down-to-the-common-civilian "room 101"-style law enforcement, nor are they capable of convincing their entire populations that, for example, "We have never been at war with Iraq; we have always been at war with Saudi Arabia".
To be able to do that would require outright revolution or a steady long-term erosion of consitutional rights (which in turn would require a steady stream of McCarthys and Ashcrofts).
I stand by my original argument that on the whole, corporations are much more capable of 1984-style control than governments are. Read your employer's policy handbook, and tell me how easy that would be to get codified into civil law.
The point of 1984 was not so much that there would be technology sufficient to implement totalitarianism (which as others have pointed out, we have today). The main thing was that "whoever controls the past controls the future".
That's why I fear Big Media aggregation. When news, history and other public information gets disseminated from fewer and fewer sources, it's going to be more and more tempting for those sources to use that information power to their own ends. Consider the term "Disneyfication." Also:
Ketchup is a vegetable.
Global warming? It's not true, and besides, there's nothing you can do about it.
Corporations are not bound by the pesky constitutions that kept governments from doing what Orwell predicted.
That's not reliable. Say you create a source tree that you intend for others to use as a backdrop for their builds. You build it, filling it full of compiled objects. Then I come along, set up my build tree, VPATHing against yours, and I rebuild. Any source I haven't changed won't be rebuilt because there's an up-to-date compiled object in your tree that I can reuse. So far so good.
But suppose I've changed something in my build environment. The most obvious example is a different compiler (maybe even at the same pathname if I'm on a different machine and building against your NFS-mounted tree). Most tools, including gmake, Sun make, and (IIRC) nmake, will not detect the change, and will incorrectly not rebuild.
So, it's good enough for most development situations, but unless you have VERY tight control over developers' build environments, you're still vulnerable. (One example of tight control: put all the build tools in a well-known place where developers can't change anything, then chroot to it to do the build. Also, have the build scripts completely discard the user's environment, e.g., no more PATH=$PATH:...)
To be fair, there are some similar traps with ClearCase, but they're fewer in number, and relatively easy to work around.
If you're looking to change SCM systems on a large project, your best bet is the comprehensive configuration management toolset report from Ovum (flyer here.) It's jaw-droppingly expensive (currently US$3400), but you'll get that back in time saved when doing your evaluation, because you can prepare from it your short list of vendors you want to bring in for demos and eval licenses.
It's not that CVS doesn't scale to big projects, it's that CVS doesn't scale to big PROCESS. Think of the scripting you'd have to do to get to CMM level 2.
CVS is a version control system (which, to be fair, is all it ever claimed to be), not a configuration management system.
A lot of systems have preop and postop triggers, where the operation runs only if the preop trigger succeeds, and the postop trigger only runs of the operation succeeds.
I'd like to see two other types. The first is a postop-fail trigger that runs only if the operation fails, and the second is an "alternate operation" trigger: "You're about to do something stupid. Would you like to do the right thing instead?"
Clearmake is slower than nmake (and traditional make and GNU make, and...) because none of those other tools make what you build available for reuse by other users, nor do they reuse what others have done. In ClearCase terminology this is called "wink-in" and it can speed builds up by an order of magnitude while decreasing overall disk usage dramatically.
Moreover, none of the traditional make tools store any metadata about what they've built (nmake does though). ClearCase stores a "configuration record" consisting of all the file system objects that contributed to creating the derived object, as well as the command that did the compilation (e.g., "cc -o foo.o -DSTUFF=bother -I/foo/blat/include foo.c").
Clearmake always incurs more overhead than traditional make tools when it comes to deciding whether to rebuild something or not, but once it knows the answer to that, a wink-in can be almost instantaneous (on a good day, I can get 20 winkins per second).
Other wins are that Clearmake figures out all the header file dependencies on its own--bye bye "make depend", and a change in the invoked command, not just the source files, is detected and triggers a rebuild.
OTOH, it is hobbled by its need to be compatible with older make variants; I'd KILL for a "clearnmake" because nmake is a much more powerful tool for directing compilation. Switching to something other than clearmake loses all the winkin capability.
I'd also like something that Does The Right Thing with Java. Make's model has always been, "rebuild only when necessary", but it relies heavily on C/C++'s decoupling of interface from implementation. Since java doesn't have this, clearmake loses, and you end up going to ant or javamake.
Speaking of javamake, it's worth noting that the tool is an outgrowth of the author's postdoc research to determine what needs to be rebuilt due to a single change to a single java source file--it's that nontrivial of a problem.
Slashdot Mirror
I tried this method to make one grain of rice in a 14" roasting pan. Came out soggy.
[How OT is this???]
It's not just flammability. Notice the part that says "All New Materials"? This is to set it apart from say, the chairs my cheap-ass employer bought, where the tag is bright yellow and says "All Second-Hand Material", but, reassuringly, "Contents Sterilized".
I am not making this up.
Yes, but by 2012, they won't be "high-end" tuners, any more than four-head hi-fi is a "high end" VCR now, even though it was ten years ago.
Yes, DC-to-daylight is a lot of spectrum representing a lot of bandwidth. But EM spectrum with specific characteristics is very limited. VHF is 30-300MHz, period. If your application needs the characteristics of VHF transmission and reception, that's where ya gotta be, like it or not.
This has to be the first time a Slashdot post has been referred to *in the original article*!
...that the referenced article is from one of the very few places whose popups aren't blocked by Mozilla's popup killer.
A trip to the NYC Transit Museum is very enlightening when you start looking at the hardware required to get all that current to the right place (think of what a 2000 amp circuit breaker would look like).
Crank up the MID? Where's this? Do that and you'll have the crowd complaining about the tinny sound.
Every time a DJ gets into my effects rack (and they have some ingenious ways of doing it despite everything I do to keep them out once my PA is tuned up) they leave the EQ looking like a smiley face, because they think it maxes the boom-chicks.
What most DJs (who aren't also sound techs) don't know is that the real kick from the bass is not at the low end, it's in the midbass. So when they have the 20Hz and 40Hz sliders at +12 it ends up sounding like the Cerwin-Vegas in a white suburban homeboy's lowered Hyundai.
The best solution to ear-ringing is a reasonably quiet chill room.
Or earplugs. I don't want to get off on a rant here, but earplugs are the best 50 cents you can ever spend. Earplugs have the advantage that they drop the overall sound (preventing the *permanent* damage you can get from just one night in front of the speaker stacks) to let you get close enough to the bass bins for a nice comfortable rib-cage massage, but they do so in a way that someone can talk to you in only a slightly louder than normal voice close to your ear and it's perfectly intelligible.
The fire is burning toward him at 1mph, but the smoke is blowing toward him at 2mph. Therefore all he has to so is asphyxiate himself on the smoke, and then he won't burn to death...
OK, this may be a Stupid Question, but, the announcement said that, "As always, Debian GNU/Linux systems can be upgraded painlessly, in place, without any forced downtime."
How do you upgrade the kernel without a reboot?
That's because you're thinking in extremes. My ideal, irrespective of implementation, has mechanisms in place to minimize corruption and opportunism (!= ambition).
RSI is a physiological consequence of too much typing
What part of "programming for a living" didn't you understand?
I never said that the amorality of corporations was right or wrong. What we have to consider is that because corproations are amoral, they are in theory free to do whatever they deem necessary to achieve their goals. If those actions are harmful to society, then society should have the capability to reign in those harmful actions.
The free market is not an effective deterrent against corporations' ill behavior, nor does it encourage good behavior. Not because it is unwilling to, but because there is no such thing as a free market. Unfettered markets always become less free, until they are controlled by a monopoly or a collusive oligopoly, which will remain in place until it is disrupted. This disruption may be anything from new technology to revolution. When that happens the process restarts.
With each iteration, the competitors get more and more creative in the ways they compete. The first time around, it may be as simple as "the best product at the lowest price". Next time, it's "the perceived best product at the lowest price". Then "the most convenient product at maybe the lowest price". Then "whatever everybdy else is buying because it's now impossible to tell whether it's the best product or not". Then, it's "whatever everybody else is buying because we're all getting our information about what to buy from the same source which is really the supplier of what we're going to buy, and they're playing on either our hopes, our fears, or our desires to that end." That's where Orwell steps in.
And that's where I step out. You're welcome to have the last word.
My point is that absent some sort of outside enforcement of a reasonable level of safety and quality of products, choosing your corporation is like choosing between hanging or the firing squad.
Whether I get RSI by programming for a living implies that, at least in our economic system, I'm overwhelmingly likely to be doing so for a corporation. That corporation, by definition an amoral entity, doesn't care about that except where it may hurt their profits.
When the Challenger shuttle blew up, it wasn't corporate greed that made it do it, was it?
Yes, in fact it was.
The explosion was caused by a set of decisions, some of which being motivated by fear of bad publicity for the for-profit contractors, leading up to the launch taking place even though it was known that the air temperature was out of spec for the O-rings. They knew it and went ahead anyway. I guess it didn't save Morton-Thiokol as much money as they thought it would.
I'll say it again: Corporations are amoral entities, beholden only to their executives and shareholders. A corporation will kill if it is profitable to do so, until and unless the shareholders, through their votes, agree to give up some of their profits in order to be so-called "good corporate citizens".
Many markets are controlled by oligopolies. Think of how many industries have a "big three". They can collude to form de facto monopolies (the inability to prevent which is why libertarianism fails). If the products they produce are considered essential to normal living, then we get them precisely on their terms. It is in their interest for us not to believe we've been forced to grab our ankles in this regard, so part of their strategy is to get us to be grateful for what they're giving us (which, as another poster pointed out, is more Huxleyan than Orwellian).
So, I still stand by my original claim that modern-day corporations are more Orwellian than modern-day governments.
Another part of the problem is that in a lot of the South, education is funded out of sales and income taxes rather than property taxes. So, when the economy goes downhill (I was going to say "south" but...), they lose revenue, and the schools go broke. Up Nawth, we fund our schools primarily from property taxes, which don't change with the economic weather.
I wish I knew more about its history. I hope the inventors made a boatload of money from it before (and hopefully after) it became a Playtex product.
I think we've bought at least a dozen of these, two for our child (original, and wide opening) and the rest as gifts.
Grow up. Interaction with corporations is unavoidable unless you completely drop out of the economy. Even then, every habitable area is geographically close to/downwind from/downstream from corporate facilities enough that their activities may still affect you.
Am I to stop buying canned food for fear it's contaminated with botulism? Am I to stop programming for a living for fear that some new form of RSI will disable me decades from now?
Were the workers at Johns-Manville supposed to Just Know that the armloads of raw asbestos they were hauling around would eventually shred their lungs? Were the people who lived around the Union Carbide plant in Bhopal supposed to Just Know that there was seriously bad stuff that could leak out of the plant? Were the Alaskan fishermen supposed to Just Know that some drunken captain might run his tanker aground in Prudhoe Bay?
In all cases the corporations knew these things could happen. The hazards of asbestos were documented in the 20's and 30's, Union Carbide knew the Bhopal plant was in poor repair, and Exxon knew of Hazelwood's drinking problem.
To get back to the original topic, it is truly Orwellian that these companies were able to do these things then hide behind their PR firms and bankruptcy lawyers to get away with relatively small slaps on their wrists.
I guess we should be glad that at least J-M, UC and Exxon didn't sue those people for anything and do them REAL harm.
If you know as much history as you're trying to convince me you do, then you appear to be proactively ignoring a lot of it.
I've never seen film at 48fps, but I have seen Showscan (60fps) and the realism is awesome.
The first time I saw it, a guy came out on stage beforehand to explain what Showscan was and how it worked. It wasn't until they did a slow wipe from him to the beginning of the actual content that we all realized we'd been had, that he *was* the first part of the film.
This was at a Showbiz Pizza Place, the precursor to Chuck-E-Cheese, circa 1983.
(3) The very worst thing that corporations can do to you is sue you into bankrupcy.
Don't know much history, do you?
The very worst thing a corporation can do is kill you, either by selling you a deadly (or fatally defective) product, or by giving you a hazardous job, or by poisoning your environment.
1984 was set in a supposedly advanced western society, and was targeted at readers in western democracies, not Singapore, Burma or Nicaragua.
Looked at from today's perspective (as the original article does), where most of these nations have constitutions that have teeth in them, it's clear that most of these governments are not capable of deploying widespread down-to-the-common-civilian "room 101"-style law enforcement, nor are they capable of convincing their entire populations that, for example, "We have never been at war with Iraq; we have always been at war with Saudi Arabia".
To be able to do that would require outright revolution or a steady long-term erosion of consitutional rights (which in turn would require a steady stream of McCarthys and Ashcrofts).
I stand by my original argument that on the whole, corporations are much more capable of 1984-style control than governments are. Read your employer's policy handbook, and tell me how easy that would be to get codified into civil law.
The point of 1984 was not so much that there would be technology sufficient to implement totalitarianism (which as others have pointed out, we have today). The main thing was that "whoever controls the past controls the future".
That's why I fear Big Media aggregation. When news, history and other public information gets disseminated from fewer and fewer sources, it's going to be more and more tempting for those sources to use that information power to their own ends. Consider the term "Disneyfication." Also:
Ketchup is a vegetable.
Global warming? It's not true, and besides, there's nothing you can do about it.
Corporations are not bound by the pesky constitutions that kept governments from doing what Orwell predicted.
Gbut gis git gan goriginal gidea? Kor kdid kthey kget kit kfrom ksomeplace kelse?
That's not reliable. Say you create a source tree that you intend for others to use as a backdrop for their builds. You build it, filling it full of compiled objects. Then I come along, set up my build tree, VPATHing against yours, and I rebuild. Any source I haven't changed won't be rebuilt because there's an up-to-date compiled object in your tree that I can reuse. So far so good.
But suppose I've changed something in my build environment. The most obvious example is a different compiler (maybe even at the same pathname if I'm on a different machine and building against your NFS-mounted tree). Most tools, including gmake, Sun make, and (IIRC) nmake, will not detect the change, and will incorrectly not rebuild.
So, it's good enough for most development situations, but unless you have VERY tight control over developers' build environments, you're still vulnerable. (One example of tight control: put all the build tools in a well-known place where developers can't change anything, then chroot to it to do the build. Also, have the build scripts completely discard the user's environment, e.g., no more PATH=$PATH:...)
To be fair, there are some similar traps with ClearCase, but they're fewer in number, and relatively easy to work around.
If you're looking to change SCM systems on a large project, your best bet is the comprehensive configuration management toolset report from Ovum (flyer here.) It's jaw-droppingly expensive (currently US$3400), but you'll get that back in time saved when doing your evaluation, because you can prepare from it your short list of vendors you want to bring in for demos and eval licenses.
It's not that CVS doesn't scale to big projects, it's that CVS doesn't scale to big PROCESS. Think of the scripting you'd have to do to get to CMM level 2.
CVS is a version control system (which, to be fair, is all it ever claimed to be), not a configuration management system.
A lot of systems have preop and postop triggers, where the operation runs only if the preop trigger succeeds, and the postop trigger only runs of the operation succeeds.
I'd like to see two other types. The first is a postop-fail trigger that runs only if the operation fails, and the second is an "alternate operation" trigger: "You're about to do something stupid. Would you like to do the right thing instead?"
And EVERY operation must be triggerable.
Clearmake is slower than nmake (and traditional make and GNU make, and...) because none of those other tools make what you build available for reuse by other users, nor do they reuse what others have done. In ClearCase terminology this is called "wink-in" and it can speed builds up by an order of magnitude while decreasing overall disk usage dramatically.
Moreover, none of the traditional make tools store any metadata about what they've built (nmake does though). ClearCase stores a "configuration record" consisting of all the file system objects that contributed to creating the derived object, as well as the command that did the compilation (e.g., "cc -o foo.o -DSTUFF=bother -I/foo/blat/include foo.c").
Clearmake always incurs more overhead than traditional make tools when it comes to deciding whether to rebuild something or not, but once it knows the answer to that, a wink-in can be almost instantaneous (on a good day, I can get 20 winkins per second).
Other wins are that Clearmake figures out all the header file dependencies on its own--bye bye "make depend", and a change in the invoked command, not just the source files, is detected and triggers a rebuild.
OTOH, it is hobbled by its need to be compatible with older make variants; I'd KILL for a "clearnmake" because nmake is a much more powerful tool for directing compilation. Switching to something other than clearmake loses all the winkin capability.
I'd also like something that Does The Right Thing with Java. Make's model has always been, "rebuild only when necessary", but it relies heavily on C/C++'s decoupling of interface from implementation. Since java doesn't have this, clearmake loses, and you end up going to ant or javamake.
Speaking of javamake, it's worth noting that the tool is an outgrowth of the author's postdoc research to determine what needs to be rebuilt due to a single change to a single java source file--it's that nontrivial of a problem.