Tough? You're probably cooking them too fast. Have you tried preparing one sous-vide ?
Close. Slow cooked in a wet sand pit filled with hot rocks and covered with palm leaves is the regional cooking method. Kind of like a clam bake, without the clams.
It's wait for milestone ClueBat-Zeta. That's the point where every living person on Earth knows that Verizon isn't going to gain world domination by forcing its handset users to buy music and BREW based games through the Verizon store. I'd be amazed if they make much more money on their V.Cast music service than they spend on advertising it.
I'm sick of getting crippled smartphones from Verizon, with GPS disabled, no Java, and no bluetooth profiles other than what is needed to get a hands-free kit. I'm sick of having to registry hack my phone so I can install the applications that *I* want. If selling phones with network service were illegal, I'd be reasonably happy with Verizon because they've got the best coverage at my house. I'd be ecstatic if I could buy wireless bandwidth as a pure commodity, not as some kind of value added package.
Releasing an open phone goes so far against Verizon's corporate culture, I was shocked at the news that they were planning an Android phone. I won't be in line on the first day to buy it though; I'm going to wait to see what they've done to cripple it.
The problem is that we're using the words "competent" and "nice" as if they each had only one meaning, and each applied only in one field.
What you are saying is that you have a competent boss. If he were in an engineering position, he'd be an incompetent engineer, but he's in a management position and he's doing his job competently. He may or may not be "nice", although it's hard to tell the difference in certain kinds of management.
Having managed engineers, I can tell you what they want:
(1) They want to be given assignments on which they can use their special professional skills productively. (2) They want to improve their skills so they can do more ambitious projects. (3) They want to be recognized for their accomplishments. (4) They don't want to deal with issues that they don't have the tools, time or organizational authority to take effective action on.
In other words, they want to be taken care of. This happily is also what you need to do to get the most engineering productivity out of a team. You can use the analogy of managing a stable of elite gladiators if you like. Or you can use the analogy of the kindly and nurturing science teacher who inspired the gifted kids to go into science and engineering. But it's very hard to distinguish "competence" from "niceness" if you are managing a team.
From the management perspective, you have a problem. It's not a matter of square pegs and round holes, it's a more like of everywhere you look you have irregular shaped holes you have to shove people into. Engineers are very malleable within certain set boundaries. You can reasonably ask an electrical engineer to learn something about software, or a software engineer to learn something about biology. I once showed some of my software guys how to use a breadboard to figure out what was needed to interface devices and to figure out whether a serial port was RS-232 or TTL, and they were pleased as punch. Most of them were happy to meet with customers and work with them on requirements, but very, very few were ready to or interested in managing customer relationships. That was tough because the higher ups realized that if they had a team of engineers that could do everything, including managing customers, then *management's* job would be a lot easier. It's easy to have a slogan like "Customer satisfaction is everyone's job." It might be true, but it doesn't mean everyone *does the same parts of the job*.
What is "nice" is situationally dependent. Saying, "you're too heavy" isn't nice when somebody says their life's dream is to climb Everest, but it is nice when somebody is about to bungie jump with bungies that won't keep him from smacking his head on the ground.
An engineer's job is to convert requirements and constraints into analyses and designs. Therefore they value clarity and accuracy more than the average person. They don't want to see problems minimized (probably with the intent to encourage them) or exaggerated (to make them feel good about themselves). They want the most accurate and precise description of the constraints they'll be working with. That translates into a communication style that would be blunt in, say, the sales world, where many managers come from. That in part is the reason why engineers often don't make good customer relationship managers. They'd like to be free to say, "in this particular aspect of the scenario, you'd actually be better off with the competitor's product." Of course most of them have the common sense not to say that kind of thing, and in any case that's not what a customer wants to hear after he's made a commitment to a vendor, but anything less than honesty is less than comfortable for a good engineer.
Writing a program that fits in 8KB is a lot easier when you are using tools and architectures designed to target such memory constrained systems.
First of all, you have an 8 bit processor. Most commands are one byte, most operands are one or two bytes.
Secondly, you don't load everything into RAM. A lot of what you want is on system ROM or (in this case I'm guessing) a game ROM. It's tricky to manage this, true, but the user in those days tolerated delays that didn't interfere with actual game play.
I had an assignment a few years ago to evaluate whether an embedded system that tracked container locations with GPS and sat communications could be modified to send its data encrypted. The processor was an 8 bit PIC and there was something like 10K RAM to play with. My conclusion was that from a programming standpoint it could be done, but that a real cryptographer would have to be involved to vett the software design, given the low entropy of the messages.
I've been programming since the 1970s, professionally since the early 80s.
There are more great programmers today than there was then, and many, many more very good ones, and vastly more really bad ones.
Things were different then. It's very common now to build programs that run for an unpredictable amount of time, handling asynchronous data inputs of an unpredictable nature. That amounts to your typical desktop application or web service. Back then, that described mainly operating systems. The vast majority of programs processed a bounded stream of data to produce easily characterizable outputs: take this set of time sheet records and produce a set of payroll records, etc. Application programming today is what we used to call "systems programming".
There were no application frameworks. Only function libraries. The only "system architecture" we had to deal with was the operating system, and that was by in large easier to understand and master than a typical Java framework. The numbers of abstractions we had to learn to get anything done was small. When we needed an abstraction, we created ones *for our own use* and while we'd have liked them to be usable for other purposes or for other programmers, it wasn't as big a priority as getting the job done.
Yes, we had to deal with resources constraints in RAM, storage and processing power. I've programmed on systems with 4K of RAM, professionally on systems with 16K of RAM. But that's only *one* kind of difficulty. Back then a programmer who could write maintainable code to produce an easily verifiable result with little resource use was a great programmer. That's all it took. That's only a *start* today. To be a great programmer now, you've got to deal with two things that we hardly ever had to worry about: complexity and uncertainty. That's why I have to rate the job of programmer as less satisfying than it used to be; so much more of your time is spent dealing with the fallibility of others rather than dreaming up new ways of doing things.
A top notch programmer now might be expected to produce a *framework* which is supposed to make life easier for some programmers he's never met working on problems he's never seen. How does he know he's got it right? In fact, usually they get it at least a little wrong, often very wrong. If a framework is workable for a modest number of problems and can be fixed with respect to the problems people want to use it for, then it's successful. Framework designer fallibility imposes a burden on programmers downstream, who have to figure out how to hammer the job requirements they're given into the frameworks they've got to do things immensely more complicated than what the vast majority of programmers did back in the early 80s.
I'd say that most programmers today have more skills and knowledge than we had back then, except within the very narrow bounds of the kind of work we were doing. Within those bounds, there are many more programmers today who could do those jobs as well or better than we did, although the proportion of programmers who could do that might be smaller compared to the number of currently active programmers.
As far as 8K of RAM is concerned, I don't think comp sci grads train for that kind of work, but I'd bet you'd find a lot of EE grads interested in embedded systems who could do great things with it.
Not in that shot, but in others it looks more real -- probably they've been photoshopped. At a hundred yards I probably wouldn't be able to tell it from some of the squad automatic weapon.
In any case the bar for "looks real" is pretty low. Some modern pistols look like toys to me, especially the ones that aren't entirely black.
Personally, I don't think the world should be divided into "pro-gun" and "anti-gun" people.
I'm not a shooter, unless you count bows, but I'm all for people being able to own firearms. If we take a hostile stance towards others pleasures, then sooner or later that'll come back to bite us.
On the other hand, I don't have a deep moral revulsion towards limiting ownership of very powerful military style weapons, provided anybody who shows they can operate, store and transport those weapons responsibly can still own and enjoy them. Some people would call me "anti-gun" because of that, but I think of myself as more of an "anti-moron with guns they can't handle" kind of a person.
I also think it should be a lot harder to get a driver's license.
Well, true. Life would be simpler and more civilized if people could be relied upon to exercise common sense and courtesy. For one thing, we wouldn't need posted speed limits for cars. But a small percentage of idiots multiplied by a large number of people mean there are a lot of idiots around, and everybody else spends much of their life working around that fact.
If I was to cut off 0.000020 of your human body, would you laugh at me because that wouldn't even be a hair?
For a 200 pound man, 0.000020 works out to about 1.8 grams. I could remove your pineal gland and still have 1.7 grams left in my budget to take the thin slice out of your spinal cord that kills you.
In systems, small fractions can matter.
The three most common gases in the atmosphere are N2 (780840 ppm), O2 (209460 ppm) and Argon (9340). This means that leaves 360 ppm, of which the preindustrial concentration of 280 ppm was 77%. Why did I choose that denominator? Because N2, O2 and Argon are completely transparent to IR, and are therefore irrelevant in the argument that CO2 is not the most abundant greenhouse gas. The next candidate on the list is Methane at #7, however it is thousands of times less abundant than CO2. Nonetheless it contributes something like 1/3 as much as CO2 does to radiative warming, because it is a much more potent gas. It has also increased during industrial times. Below that we have to go down to N2O and Freon-12, which contribute 1/3 as much as Methane despite being much rarer.
383 ppm (or if you prefer 0.000386) doesn't sound like much, but there is a LOT of atmosphere. The effect of CO2 on climate was well established long before the first global warming paper was published (circa 1960 -- well before it became a political issue). It can be worked out from the concentrations of gas in the atmosphere and the laboratory observable properties of CO2. Its the kind of thing beginning astronomy students work out once they've had integral calculus. I remember reading that the dinosaurs lived in a hotter world because of CO2 concentrations in my post-Sputnik science books, so it's not something Al Gore cooked up to convince us all to go socialist. It was politically correct under Ike too.
Let me remind you how colonialism works. You get to keep whatever you claim and which nobody else can force you to relinquish. All pretence of civilization and legality when it comes to claims of "sovereignty" in cases like this are just that: pretences.
The emptiness of any "legal" claims to sovereignty over Antarctica can readily be seen by the fact that most countries ignore territorial claims and those that do have conflicting claims, yet it makes no difference. Nobody is going to insist on pressing their claims (or forcing others to relinquish their claims) because it's not worth getting into even a diplomatic spat over a "legal" absurdity.
The place where extending territorial claims is going to get nasty is on the other side of the globe -- in the Arctic. Between climate change and energy resources, we might see a shooting war there some time in the next generation.
Well, sure, but the CIR is come hell or high water.
It sounds like you had a frame relay circuit -- or perhaps frame relay was used from Verizon's POP. I've seen similar business circuits that had zero CIR and were perfectly functional. If you aren't doing telephony or video conferencing, you don't need a huge CIR as long as your average throughput is good.
If you want a higher CIR, you just pay for it. It's not like a scam or anything, bandwidth is a commodity. If 16k CIR with a 768k burst rate wasn't sufficient to your needs, then you could buy more. It's not like buying "Internet service" from Comcast and getting non-standard routing practices to strangle your sockets whenever it pleases them, Verizon in this case was up front about what you were getting. I *really* hate Verizon wireless because I have to struggle with them over who gets to control my smartphone, but packaging bandwidth that way doesn't strike me as nefarious.
The more narrowly you frame the question, the easier it is. Progress in semiconductors can be measured in transistors per mm^2. Technological progress as a whole is very difficult to quantify.
I have a simplified model in my head I call "the sphere of knowledge". Inside the sphere is everything we know, outside the sphere is everything we don't know, and the surface of the sphere is what we are currently discovering. If you measure the size of the sphere by the radius, and find the rate of increase is dropping, you might think the rate of progress is decreasing; however the surface area of the sphere which is proportional to the square of the radius might still be increasing at a constant rate, and the volume increasing accelerating. That captures the problem of looking at progress as a whole; it might not look like it is expanding as fast, but if you look at specific areas, the rate of change is profound.
I started in the computer business in the early 80s. It was a time of rapid change, with IC technology making computational power available to many businesses for the first time and even a few home hobbyists. The rate of expansion of computer ownership probably dropped by the early 00s. Yet during the time when "computerization" was slowing down, we saw the widespread adoption of the single most amazing technological innovation of my lifetime, far more significant in my opinion than the moon landing, which I remember watching: the Internet search engine. It's a stupendous leap in locating information, practically unimagined by science fiction even a few years before.
Of course, knowledge really isn't a sphere; it's an irregular region in hyperspace with gaping multidimensional holes in it. That means that entire new fields of knowledge and brand new job descriptions will emerge around tiny seeds of knowledge development. My bets are on biology and biotechnology. Designer drugs, artificial organs and limbs, science tested performance enhancing drugs, bioengineered organisms to perform chemical and environmental tasks, the list of possibilities goes on and on.
I wouldn't count any chickens before they're hatched. SpaceX has had something like two successful launches of their smaller vehicle. While that is a tremendous accomplishment, there is no way that vehicle has the capacity to take over the resupply duties. Their larger vehicle could do the job but it hasn't flown yet. It *may* be able to do the job, but not enough to be relied upon by next year.
If SpaceX has a contract, it amounts to government underwriting of private R&D. That's probably the right thing to do, but it's not proof that space is ready for true private enterprise yet.
That's a good point about economic extrapolations. Platinum is an extremely useful metal, it's also hard to find (.003 ppb of the Earth's crust. Therefore it's very, very expensive, about $1200/ounce. Iron is also very useful, but it is forty million times more abundant in the Earth's crust, and it's cost as scrap (not ore) is $50-$100 per ton. So a single ounce of Pt is worth ten to twenty tons of Fe, or something like 500,000-600,000 as valuable.
In any case, GP is assuming that the diamonds are mined, cleaned, and packed ready to ship.
When it comes to space exploration -- particularly planetary exploration -- mass is king. Let's say there were great deposits of gold on Mars. It wouldn't be worth going there to get it until we'd completely exhausted mines on Earth, because of the cost of moving the mass from Mars to Earth. The same goes for that mainstay of science fiction: asteroid mining. It may be that as the Earth's population reaches the fifty billion mark, iron will be come dear enough to get it from space, but not any time soon.
So the things most worth going to space for are the lightest things. Knowledge. Information.
In fact, if you look at the commercial importance of space, arguably space technology is information technology. What are the things that are worth putting a commercial satellite in orbit for? Relaying communication. Taking pictures of the Earth. Even space tourism is ultimately about sensory information.
I visited Hawaii on business and they put me up at the "Hawaiian Village" at Waikiki. I can tell you that a sufficiently contrived local experience is less informative than a remote exploration. Fortunately, I was there to work so I got to meet Hawaiians without flaming batons being involved.
I just watched the Macgyver pilot with my kids -- I'd never seen the show before because I didn't have a TV when it was on. I was surprised to see that Macgyver apparently lived in the Griffith Observatory building. Maybe he haunts it like the Phantom of the Opera.
I don't understand this hostility to Lisp. I haven't used Lisp in a long time, but I wouldn't, say, turn up a job that required Lisp programming -- in fact I'd probably jump for it.
Yes, there are all those ugly parentheses, but that just reminds you to keep procedure bodies short -- very, very short. It's kind of a mindset. In a way Lisp reminds me of Unix. The great thing about Unix was the "everything is a file" paradigm -- back in the day at least. It reduced the number of interfaces you had to know. That seems to be the philosophy of RESTful web services: applications manipulate the state of Web resources. In Lisp, everything is an s-expression that you process recursively; if you *have* to have one tool in your toolbox, that's a pretty powerful one, and you can make your own abstractions pretty easily. In what other language would implementing a simple interpreter for the language (or at least a simplified but recognizable form of the language) be a fairly basic exercise?
You get about one shot every 2 minutes if you have four guys that know what they are doing, and you burn more than $10 worth of powder for ever shot. And the things are heavy. They will not get far.
Ah, but there's a 50% chance sonny has the DIY allele of the TNKER gene, in which case he'll be making his own black powder.
One guy built the first IC in two months at Texas Instruments, without much help. Several early microprocessors were designed by teams of about 5 people. It took 3,000 people, in grey cubicles in Santa Clara, to design the Pentium Pro/II/III architecture, the first superscalar IC.
Which demonstrates how economical it is to achieve technological advancement by basic research than it is to make it a byproduct of product design.
Superscalar instruction execution was the brainchild of a single man: Seymour Cray. It was common on RISC CPUs in the 80s -- in fact it was a major motivation for RISC designs. If it hadn't been for the basic research which lead to RISC architectures, proven examples of superscalar CPUs outside supercomputing would probably not have existed. Intel would not have introduced the first superscalar microprocessor (the i960), and without that experience Intel certainly would not have risked the effort it took to graft superscalar execution onto its broken but commercially popular x86 architecture.
What's odd is that environmentalists have been saying this for years. At least the academic ones. They've known for years that the human population is not supportable in a low-tech way, and that improving the lives of the vast majority of people on the planet sustainably requires the development of new technology. Seven billion people trying to cut down trees and burn them for fuel would be an ecological disaster.
Perhaps the problem is saying "environmentalist" like it means only one possible thing.
I once listened to a socialist argue for, and an objectivist argue against "the tragedy of the commons", neither apparently realizing one position that might reasonably be taken from the model is that private property is more sustainable than common property. It was rich to hear the socialist argue that a single manager was the only way to manage the commons properly, only to be convincingly rebutted by the objectivist's argument in favor of collective management.
Slashdot Mirror
Tough? You're probably cooking them too fast. Have you tried preparing one sous-vide ?
Close. Slow cooked in a wet sand pit filled with hot rocks and covered with palm leaves is the regional cooking method. Kind of like a clam bake, without the clams.
Firefox has been shown to make it so under not especially exotic conditions.
That's why it's number 1.
It's wait for milestone ClueBat-Zeta. That's the point where every living person on Earth knows that Verizon isn't going to gain world domination by forcing its handset users to buy music and BREW based games through the Verizon store. I'd be amazed if they make much more money on their V.Cast music service than they spend on advertising it.
I'm sick of getting crippled smartphones from Verizon, with GPS disabled, no Java, and no bluetooth profiles other than what is needed to get a hands-free kit. I'm sick of having to registry hack my phone so I can install the applications that *I* want. If selling phones with network service were illegal, I'd be reasonably happy with Verizon because they've got the best coverage at my house. I'd be ecstatic if I could buy wireless bandwidth as a pure commodity, not as some kind of value added package.
Releasing an open phone goes so far against Verizon's corporate culture, I was shocked at the news that they were planning an Android phone. I won't be in line on the first day to buy it though; I'm going to wait to see what they've done to cripple it.
The problem is that we're using the words "competent" and "nice" as if they each had only one meaning, and each applied only in one field.
What you are saying is that you have a competent boss. If he were in an engineering position, he'd be an incompetent engineer, but he's in a management position and he's doing his job competently. He may or may not be "nice", although it's hard to tell the difference in certain kinds of management.
Having managed engineers, I can tell you what they want:
(1) They want to be given assignments on which they can use their special professional skills productively.
(2) They want to improve their skills so they can do more ambitious projects.
(3) They want to be recognized for their accomplishments.
(4) They don't want to deal with issues that they don't have the tools, time or organizational authority to take effective action on.
In other words, they want to be taken care of. This happily is also what you need to do to get the most engineering productivity out of a team. You can use the analogy of managing a stable of elite gladiators if you like. Or you can use the analogy of the kindly and nurturing science teacher who inspired the gifted kids to go into science and engineering. But it's very hard to distinguish "competence" from "niceness" if you are managing a team.
From the management perspective, you have a problem. It's not a matter of square pegs and round holes, it's a more like of everywhere you look you have irregular shaped holes you have to shove people into. Engineers are very malleable within certain set boundaries. You can reasonably ask an electrical engineer to learn something about software, or a software engineer to learn something about biology. I once showed some of my software guys how to use a breadboard to figure out what was needed to interface devices and to figure out whether a serial port was RS-232 or TTL, and they were pleased as punch. Most of them were happy to meet with customers and work with them on requirements, but very, very few were ready to or interested in managing customer relationships. That was tough because the higher ups realized that if they had a team of engineers that could do everything, including managing customers, then *management's* job would be a lot easier. It's easy to have a slogan like "Customer satisfaction is everyone's job." It might be true, but it doesn't mean everyone *does the same parts of the job*.
What is "nice" is situationally dependent. Saying, "you're too heavy" isn't nice when somebody says their life's dream is to climb Everest, but it is nice when somebody is about to bungie jump with bungies that won't keep him from smacking his head on the ground.
An engineer's job is to convert requirements and constraints into analyses and designs. Therefore they value clarity and accuracy more than the average person. They don't want to see problems minimized (probably with the intent to encourage them) or exaggerated (to make them feel good about themselves). They want the most accurate and precise description of the constraints they'll be working with. That translates into a communication style that would be blunt in, say, the sales world, where many managers come from. That in part is the reason why engineers often don't make good customer relationship managers. They'd like to be free to say, "in this particular aspect of the scenario, you'd actually be better off with the competitor's product." Of course most of them have the common sense not to say that kind of thing, and in any case that's not what a customer wants to hear after he's made a commitment to a vendor, but anything less than honesty is less than comfortable for a good engineer.
Writing a program that fits in 8KB is a lot easier when you are using tools and architectures designed to target such memory constrained systems.
First of all, you have an 8 bit processor. Most commands are one byte, most operands are one or two bytes.
Secondly, you don't load everything into RAM. A lot of what you want is on system ROM or (in this case I'm guessing) a game ROM. It's tricky to manage this, true, but the user in those days tolerated delays that didn't interfere with actual game play.
Finally, users didn't expect much. Graphics were mostly simple blitted bitmaps.
I had an assignment a few years ago to evaluate whether an embedded system that tracked container locations with GPS and sat communications could be modified to send its data encrypted. The processor was an 8 bit PIC and there was something like 10K RAM to play with. My conclusion was that from a programming standpoint it could be done, but that a real cryptographer would have to be involved to vett the software design, given the low entropy of the messages.
I've been programming since the 1970s, professionally since the early 80s.
There are more great programmers today than there was then, and many, many more very good ones, and vastly more really bad ones.
Things were different then. It's very common now to build programs that run for an unpredictable amount of time, handling asynchronous data inputs of an unpredictable nature. That amounts to your typical desktop application or web service. Back then, that described mainly operating systems. The vast majority of programs processed a bounded stream of data to produce easily characterizable outputs: take this set of time sheet records and produce a set of payroll records, etc. Application programming today is what we used to call "systems programming".
There were no application frameworks. Only function libraries. The only "system architecture" we had to deal with was the operating system, and that was by in large easier to understand and master than a typical Java framework. The numbers of abstractions we had to learn to get anything done was small. When we needed an abstraction, we created ones *for our own use* and while we'd have liked them to be usable for other purposes or for other programmers, it wasn't as big a priority as getting the job done.
Yes, we had to deal with resources constraints in RAM, storage and processing power. I've programmed on systems with 4K of RAM, professionally on systems with 16K of RAM. But that's only *one* kind of difficulty. Back then a programmer who could write maintainable code to produce an easily verifiable result with little resource use was a great programmer. That's all it took. That's only a *start* today. To be a great programmer now, you've got to deal with two things that we hardly ever had to worry about: complexity and uncertainty. That's why I have to rate the job of programmer as less satisfying than it used to be; so much more of your time is spent dealing with the fallibility of others rather than dreaming up new ways of doing things.
A top notch programmer now might be expected to produce a *framework* which is supposed to make life easier for some programmers he's never met working on problems he's never seen. How does he know he's got it right? In fact, usually they get it at least a little wrong, often very wrong. If a framework is workable for a modest number of problems and can be fixed with respect to the problems people want to use it for, then it's successful. Framework designer fallibility imposes a burden on programmers downstream, who have to figure out how to hammer the job requirements they're given into the frameworks they've got to do things immensely more complicated than what the vast majority of programmers did back in the early 80s.
I'd say that most programmers today have more skills and knowledge than we had back then, except within the very narrow bounds of the kind of work we were doing. Within those bounds, there are many more programmers today who could do those jobs as well or better than we did, although the proportion of programmers who could do that might be smaller compared to the number of currently active programmers.
As far as 8K of RAM is concerned, I don't think comp sci grads train for that kind of work, but I'd bet you'd find a lot of EE grads interested in embedded systems who could do great things with it.
Not in that shot, but in others it looks more real -- probably they've been photoshopped. At a hundred yards I probably wouldn't be able to tell it from some of the squad automatic weapon.
In any case the bar for "looks real" is pretty low. Some modern pistols look like toys to me, especially the ones that aren't entirely black.
Personally, I don't think the world should be divided into "pro-gun" and "anti-gun" people.
I'm not a shooter, unless you count bows, but I'm all for people being able to own firearms. If we take a hostile stance towards others pleasures, then sooner or later that'll come back to bite us.
On the other hand, I don't have a deep moral revulsion towards limiting ownership of very powerful military style weapons, provided anybody who shows they can operate, store and transport those weapons responsibly can still own and enjoy them. Some people would call me "anti-gun" because of that, but I think of myself as more of an "anti-moron with guns they can't handle" kind of a person.
I also think it should be a lot harder to get a driver's license.
Well, true. Life would be simpler and more civilized if people could be relied upon to exercise common sense and courtesy. For one thing, we wouldn't need posted speed limits for cars. But a small percentage of idiots multiplied by a large number of people mean there are a lot of idiots around, and everybody else spends much of their life working around that fact.
If I was to cut off 0.000020 of your human body, would you laugh at me because that wouldn't even be a hair?
For a 200 pound man, 0.000020 works out to about 1.8 grams. I could remove your pineal gland and still have 1.7 grams left in my budget to take the thin slice out of your spinal cord that kills you.
In systems, small fractions can matter.
The three most common gases in the atmosphere are N2 (780840 ppm), O2 (209460 ppm) and Argon (9340). This means that leaves 360 ppm, of which the preindustrial concentration of 280 ppm was 77%. Why did I choose that denominator? Because N2, O2 and Argon are completely transparent to IR, and are therefore irrelevant in the argument that CO2 is not the most abundant greenhouse gas. The next candidate on the list is Methane at #7, however it is thousands of times less abundant than CO2. Nonetheless it contributes something like 1/3 as much as CO2 does to radiative warming, because it is a much more potent gas. It has also increased during industrial times. Below that we have to go down to N2O and Freon-12, which contribute 1/3 as much as Methane despite being much rarer.
383 ppm (or if you prefer 0.000386) doesn't sound like much, but there is a LOT of atmosphere. The effect of CO2 on climate was well established long before the first global warming paper was published (circa 1960 -- well before it became a political issue). It can be worked out from the concentrations of gas in the atmosphere and the laboratory observable properties of CO2. Its the kind of thing beginning astronomy students work out once they've had integral calculus. I remember reading that the dinosaurs lived in a hotter world because of CO2 concentrations in my post-Sputnik science books, so it's not something Al Gore cooked up to convince us all to go socialist. It was politically correct under Ike too.
Let me remind you how colonialism works. You get to keep whatever you claim and which nobody else can force you to relinquish. All pretence of civilization and legality when it comes to claims of "sovereignty" in cases like this are just that: pretences.
The emptiness of any "legal" claims to sovereignty over Antarctica can readily be seen by the fact that most countries ignore territorial claims and those that do have conflicting claims, yet it makes no difference. Nobody is going to insist on pressing their claims (or forcing others to relinquish their claims) because it's not worth getting into even a diplomatic spat over a "legal" absurdity.
The place where extending territorial claims is going to get nasty is on the other side of the globe -- in the Arctic. Between climate change and energy resources, we might see a shooting war there some time in the next generation.
Well, sure, but the CIR is come hell or high water.
It sounds like you had a frame relay circuit -- or perhaps frame relay was used from Verizon's POP. I've seen similar business circuits that had zero CIR and were perfectly functional. If you aren't doing telephony or video conferencing, you don't need a huge CIR as long as your average throughput is good.
If you want a higher CIR, you just pay for it. It's not like a scam or anything, bandwidth is a commodity. If 16k CIR with a 768k burst rate wasn't sufficient to your needs, then you could buy more. It's not like buying "Internet service" from Comcast and getting non-standard routing practices to strangle your sockets whenever it pleases them, Verizon in this case was up front about what you were getting. I *really* hate Verizon wireless because I have to struggle with them over who gets to control my smartphone, but packaging bandwidth that way doesn't strike me as nefarious.
The more narrowly you frame the question, the easier it is. Progress in semiconductors can be measured in transistors per mm^2. Technological progress as a whole is very difficult to quantify.
I have a simplified model in my head I call "the sphere of knowledge". Inside the sphere is everything we know, outside the sphere is everything we don't know, and the surface of the sphere is what we are currently discovering. If you measure the size of the sphere by the radius, and find the rate of increase is dropping, you might think the rate of progress is decreasing; however the surface area of the sphere which is proportional to the square of the radius might still be increasing at a constant rate, and the volume increasing accelerating. That captures the problem of looking at progress as a whole; it might not look like it is expanding as fast, but if you look at specific areas, the rate of change is profound.
I started in the computer business in the early 80s. It was a time of rapid change, with IC technology making computational power available to many businesses for the first time and even a few home hobbyists. The rate of expansion of computer ownership probably dropped by the early 00s. Yet during the time when "computerization" was slowing down, we saw the widespread adoption of the single most amazing technological innovation of my lifetime, far more significant in my opinion than the moon landing, which I remember watching: the Internet search engine. It's a stupendous leap in locating information, practically unimagined by science fiction even a few years before.
Of course, knowledge really isn't a sphere; it's an irregular region in hyperspace with gaping multidimensional holes in it. That means that entire new fields of knowledge and brand new job descriptions will emerge around tiny seeds of knowledge development. My bets are on biology and biotechnology. Designer drugs, artificial organs and limbs, science tested performance enhancing drugs, bioengineered organisms to perform chemical and environmental tasks, the list of possibilities goes on and on.
I wouldn't count any chickens before they're hatched. SpaceX has had something like two successful launches of their smaller vehicle. While that is a tremendous accomplishment, there is no way that vehicle has the capacity to take over the resupply duties. Their larger vehicle could do the job but it hasn't flown yet. It *may* be able to do the job, but not enough to be relied upon by next year.
If SpaceX has a contract, it amounts to government underwriting of private R&D. That's probably the right thing to do, but it's not proof that space is ready for true private enterprise yet.
That's a good point about economic extrapolations. Platinum is an extremely useful metal, it's also hard to find (.003 ppb of the Earth's crust. Therefore it's very, very expensive, about $1200/ounce. Iron is also very useful, but it is forty million times more abundant in the Earth's crust, and it's cost as scrap (not ore) is $50-$100 per ton. So a single ounce of Pt is worth ten to twenty tons of Fe, or something like 500,000-600,000 as valuable.
In any case, GP is assuming that the diamonds are mined, cleaned, and packed ready to ship.
Well, you put your finger on an important point.
When it comes to space exploration -- particularly planetary exploration -- mass is king. Let's say there were great deposits of gold on Mars. It wouldn't be worth going there to get it until we'd completely exhausted mines on Earth, because of the cost of moving the mass from Mars to Earth. The same goes for that mainstay of science fiction: asteroid mining. It may be that as the Earth's population reaches the fifty billion mark, iron will be come dear enough to get it from space, but not any time soon.
So the things most worth going to space for are the lightest things. Knowledge. Information.
In fact, if you look at the commercial importance of space, arguably space technology is information technology. What are the things that are worth putting a commercial satellite in orbit for? Relaying communication. Taking pictures of the Earth. Even space tourism is ultimately about sensory information.
I visited Hawaii on business and they put me up at the "Hawaiian Village" at Waikiki. I can tell you that a sufficiently contrived local experience is less informative than a remote exploration. Fortunately, I was there to work so I got to meet Hawaiians without flaming batons being involved.
I just watched the Macgyver pilot with my kids -- I'd never seen the show before because I didn't have a TV when it was on. I was surprised to see that Macgyver apparently lived in the Griffith Observatory building. Maybe he haunts it like the Phantom of the Opera.
I don't understand this hostility to Lisp. I haven't used Lisp in a long time, but I wouldn't, say, turn up a job that required Lisp programming -- in fact I'd probably jump for it.
Yes, there are all those ugly parentheses, but that just reminds you to keep procedure bodies short -- very, very short. It's kind of a mindset. In a way Lisp reminds me of Unix. The great thing about Unix was the "everything is a file" paradigm -- back in the day at least. It reduced the number of interfaces you had to know. That seems to be the philosophy of RESTful web services: applications manipulate the state of Web resources. In Lisp, everything is an s-expression that you process recursively; if you *have* to have one tool in your toolbox, that's a pretty powerful one, and you can make your own abstractions pretty easily. In what other language would implementing a simple interpreter for the language (or at least a simplified but recognizable form of the language) be a fairly basic exercise?
What weapon do the police have that you can't?
You get about one shot every 2 minutes if you have four guys that know what they are doing, and you burn more than $10 worth of powder for ever shot. And the things are heavy. They will not get far.
Ah, but there's a 50% chance sonny has the DIY allele of the TNKER gene, in which case he'll be making his own black powder.
Which demonstrates how economical it is to achieve technological advancement by basic research than it is to make it a byproduct of product design.
Superscalar instruction execution was the brainchild of a single man: Seymour Cray. It was common on RISC CPUs in the 80s -- in fact it was a major motivation for RISC designs. If it hadn't been for the basic research which lead to RISC architectures, proven examples of superscalar CPUs outside supercomputing would probably not have existed. Intel would not have introduced the first superscalar microprocessor (the i960), and without that experience Intel certainly would not have risked the effort it took to graft superscalar execution onto its broken but commercially popular x86 architecture.
What's odd is that environmentalists have been saying this for years. At least the academic ones. They've known for years that the human population is not supportable in a low-tech way, and that improving the lives of the vast majority of people on the planet sustainably requires the development of new technology. Seven billion people trying to cut down trees and burn them for fuel would be an ecological disaster.
Perhaps the problem is saying "environmentalist" like it means only one possible thing.
I once listened to a socialist argue for, and an objectivist argue against "the tragedy of the commons", neither apparently realizing one position that might reasonably be taken from the model is that private property is more sustainable than common property. It was rich to hear the socialist argue that a single manager was the only way to manage the commons properly, only to be convincingly rebutted by the objectivist's argument in favor of collective management.
So, where am I supposed go for up-to-date, well-informed scare-mongering?