A "conventional" postdoc is working on someone else's research, just like a gradual student would -- except that the postdoc has, er, a Ph.D.
I see a lot of postdocs more or less staying on the postdoc track: work on someone else's stuff, maybe squeezing in a little bit of creativity here or there, and then move on to the next postdoc. You could do that for a decade.
The key is to start thinking like an independent researcher -- something you were supposed to do by the end of graduate school. Work on getting your own grants, start building your own network. If you focus on escape from postdoc limbo, you'll get out before you burn out.
The obvious case is a generator that writes "C" code -- after all, "C" was intended as just a metalanguage like Gnu's RTL, but for the earlier BCMP compiler. It's arguable that generated C code is more true to the spirit of the language than is hand-coded C code.
I became a code-generation convert about two years ago when I first encountered Perl Data Language -- what made PDL development feasible was a code generator by Lukka, Glazebrook, and Soeller: their generator ("PP") writes all the loops and conditionals that do vectorized processing of large arrays.
Of course, PP has its limitations -- the age old complaint of engineers is that they will eventually outgrow any code generation framework -- but it's simple enough to augment the generation language or, in extreme cases, to sidestep the parts of the functionality that you're not using.
A 747 holds several tens of thousands of gallons of fuel, but it also weighs perhaps a hundred times as much as your car. The fuel just doesn't have enough energy to get the plane up to orbital speed.
Most modern jets, in the absence of friction, could get to orbital velocity before hitting the ground, if they got a lift to 100,000 ft.
No, actually, that's not true. Jet fuel has about the same amount of energy as gasoline, and it's just not enough to supply the kinetic energy needed.
Where most people go wrong is that high speeds are pretty counterintuitive: the amount of energy goes like the square of the speed, so doubling your speed quadruples your kinetic energy. Accelerating your car to 60 mph requires about a megajoule (one snickers bar). Accelerating it to 600 mph in the absence of friction would require about 100 megajoules (ten snickers bars, or about a gallon of gasoline) -- that's about the level of kinetic energy that jets support. Accelerating your car by another factor of 10 to 6,000 mph would require 10,000 snickers bars, or about 100 gallons of gasoline -- assuming the same efficiency as your engine gets driving you around town. Rockets are considerably less efficient, so getting going that fast with a rocket engine would require more like 1,000 gallons of gasoline -- except that fuel in those quantities would affect the mass of your car, so you have to spend even more fuel to accelerate the fuel you burn at the end. All told, getting your car up to orbital speed with rockets would take something like 5 or ten thousand gallons of fuel.
The problem with incremental development of RLVs is that there's a huge leap between the size and difficulty of putting something into space for five minutes (as in the current X-prize contenders) and putting it into orbit (as in the shuttle). That will make it difficult to evolve our way into a commercial space program.
I often find myself pointing out that just getting into space isn't all that hard. Lifting yourself up 100km requires about a megajoule (that's the energy equivalent of a stick of dynamite, or about 1/12th of a gallon of gasoline (about 1/4 kg or 1/2 pound of gasoline), or a jelly doughnut, or running a hairdryer for 2 minutes) per kilogram of mass.
By contrast, orbital speed is something like 7000 meters per second, (or 16,000 miles per hour for you provincials). Getting going that fast requires an additional 24 megajoules per kilogram of mass (for a total of 25).
In short, the difference between the amount of energy you need to get into orbit and just into space is a factor of 25, for the same mass. That ratio of 25 is about equal to the difference between the latent chemical energies of broccoli and gasoline.
Except that, in the case of space travel, you better be burning something at least as energetic as gasoline to start with, or you'll never even hoist yourself up 100km.
The way we've traditionally gotten into orbit is to concentrate the kinetic energy into ever smaller bits of the vehicle: you use a huge rocket motor and tanks to get everything started moving, then ditch the empty tankage and rocket motors for the first stage -- that lets you concentrate on moving a smaller amount of stuff even faster.
Realistic reusable designs are usually not staged designs, because it's hard to recover and reuse the first stages. The problem is that you have to have incredibly lightweight tankage and engines to make everything work. But pushing stuff to lighter weight makes it more flimsy and less prone to being reusable. Darn.
The VentureStar, IIRC, ran into problems with exactly this technology -- they were using lightweight carbon fiber tanks to hold their propellant, and they couldn't make the tank light enough to boost itself into orbit.
The shuttle is NOT a reusable vehicle in any but the most technical sense of the word: it requires constant skilled redesign and intelligent (rather than scripted) maintenance, and the engines have to be overhauled after every flight.
They're not reverse-engineering the hardware, they're reverse-engineering the driver software that is used to talk to the hardware. In the absence of an authoritative interface document, the driver can serve as one. In the absence of source code for the driver, the reverse-engineered driver code can serve as one.
Actualy, no, you are entitled to play the songs to the general public. The whole point of CARP is to mitigate the restrictions inherent in monopoly control. Remember, copyright is an artificial monopoly that is ostensibly intended to encourage creativity. The problem is that, perversely, copyright discourages dissemination of valuable ideas if the owner of those ideas loses interest in exploiting them, or perversely refuses to license them. The CARP is a kludge designed to fix that problem, by forcing content providers to license their content to anyone at a reasonable rate.
(If those are all full, and you want to fill [your ballot] out in the open, that doesn't disqualify your vote.)
It certainly should. If you have a choice whether to make your ballot secret, then an evil cartel could potentially use extortion to make you reveal your ballot -- and thereby force you to vote a certain way. If your ballot must be secret to be valid, then your vote is much less susceptible to that kind of control.
If it's a large thickened section of the earth's crust "floating" on the mantle, and it moves around as a single independent unit, then it's a continent.
No, actually, then it's a tectonic plate. Geological provinces and plates are distinguished from continents, as in "California is made up of several tectonic plates that crashed into one another", or "This part of the continent is a different province than that part over there".
Off topic but I couldn't resist...
110V is more of a surprise than a hurt...
on
Solving a Wiring Mess?
·
· Score: 2, Interesting
... at least through extremities. My old high school physics teacher used to stick his finger in light sockets for laughs. Then he would pull out the Variac and let you stick your own finger in a light socket with variable AC voltage. Up to about 30 VDC you couldn't feel a thing. When you control the voltage yourself, 110VAC is no big deal at all -- just a strong tingling in the finger (in the light socket).
It sounds like one of the main buses in the box got shorted out. You really need to get the thing rewired by a licensed electrician.
If the panel is at work, it's probably illegal for you to work on it. In the immortalized words of Beowulf Schaeffer, "Its Worth Yore Life And More To Go In There!" Seriously, as an uber-l337 633K, you probably have the ability to find the place where the short occurred -- but do you really want to (A) monkey around in a box with live current (it sounds like that is your main box, not a subpanel); (B) stick your neck out by breaking the law for your employer; and (C) possibly assume liability for burning down the building? No? I didn't think so.
If you're at home, you're probably allowed to work on it -- but most places require you to get an electrical inspector to look at it once you're done. Unfortunately, doing this requires getting the local electrical code, so that you know what you're doing -- and that's a whole separate rant. The National Electrical Code is adopted into most municipalities' building codes -- but it's copyrighted by the National Fire Prevention Association. Fortunately, the Supreme Court recently decided that it's unconstitutional to attempt to copyright the law of the land. Unfortuantely, you can't get the electrical code online yet -- you probably still have to buy it.
All those little pipeline shell one-liners are inconsequential precisely because the UNIX philosophy works. Some applications (the monolithic ones) don't fit that ideal very well, but many others (all those little one-liners you use, without thinking much about it) do.
If text manipulation and piping didn't work well in UNIX, you'd know about it -- all those tasks would be a real thorn in your side. As it is, you have the right tools, so they're no big deal.
Jesus Christ, doesn't anyone look out for name collisions anymore? XForms is a GUI toolkit for X., in (slow) development since 1995 and still used in many useful apps like GeomView and Lyx.
One-time-use film cameras are more expensive to make than the cost in the store, too. But the point is that they're selling convenience. While, technically, everyone who buys a one-time camera can take it apart and reload it with film or cannibalize the
camera itself for flash parts (I've certainly done both enough times), it's just not worth the time cost for the vast majority of users.
The difference between this idea and the defunct I-opener is that geeks make up a much smaller proportion of the potential market for the camera than they did for the I-opener. Who cares if 10% of the cameras get cannibalized? They probably only have to be recycled 2-3 times for Ritz to turn a profit.
I discovered 'em recently, and bought 12 of the AAs for an old Logitech cordless optical mouse and keyboard setup that can't run off NiCds (1.3 volts) or NiMHs (1.2 volts). (4 in the keyboard, 2 in the mouse; 2x for continuous use).
I liked 'em so much I went back to Target to find more and discovered that they're not carried there anymore. Nor at Radio Schlock. Nor at a bazillion other places. I think Ray O Vac is trying to phase them out.
It's too bad -- rechargeable alkalines do something the others don't: last a long time in the drawer. It takes years for an RA to self-discharge; about 30 days for a NiCd. (The upshot is that any NiCd you don't store in a a charger is empty when you actually want it).
It doesn't matter what "causes" the crash. The OS should be essentially crashproof. That's what an OS was for, and it was why Apple got such a drubbing before OS X finally came out (twelve years later).
the undelivered sweeping promises (quality products for free; ha!)
Actually, GNU products are excellent and free-as-in-beer. They're so good they're usually invisible. When's the last time you had to work around a bug in ls, or gcc?
Dictionaries, maps, and many books have included these things for decades, perhaps centuries. Most modern maps have tiny errors that are intended to be insignificant except as copyright "markers". Dictionaries contain false entries intended to serve as markers and preserve the collection copyright. Many books (especially editions of public-domain works or collections of multiple works) contain deliberate (originally not-so-deliberate) typos that mark the particular edition.
As far as I can tell, "honeytoken" is just a nice sounding buzzword for an ages-old technique.
Slashdot Mirror
I see a lot of postdocs more or less staying on the postdoc track: work on someone else's stuff, maybe squeezing in a little bit of creativity here or there, and then move on to the next postdoc. You could do that for a decade.
The key is to start thinking like an independent researcher -- something you were supposed to do by the end of graduate school. Work on getting your own grants, start building your own network. If you focus on escape from postdoc limbo, you'll get out before you burn out.
Jeez, shouldn't post before coffee. That's "BCPL" of course, not "BCMP"...
I became a code-generation convert about two years ago when I first encountered Perl Data Language -- what made PDL development feasible was a code generator by Lukka, Glazebrook, and Soeller: their generator ("PP") writes all the loops and conditionals that do vectorized processing of large arrays.
Of course, PP has its limitations -- the age old complaint of engineers is that they will eventually outgrow any code generation framework -- but it's simple enough to augment the generation language or, in extreme cases, to sidestep the parts of the functionality that you're not using.
A 747 holds several tens of thousands of gallons of fuel, but it also weighs perhaps a hundred times as much as your car. The fuel just doesn't have enough energy to get the plane up to orbital speed.
No, actually, that's not true. Jet fuel has about
the same amount of energy as gasoline, and it's just not enough to supply the kinetic energy needed.
Where most people go wrong is that high speeds are pretty counterintuitive: the amount of energy goes like the square of the speed, so doubling your speed quadruples your kinetic energy. Accelerating your car to 60 mph requires about a megajoule (one snickers bar). Accelerating it to 600 mph in the absence of friction would require about 100 megajoules (ten snickers bars, or about a gallon of gasoline) -- that's about the level of kinetic energy that jets support. Accelerating your car by another factor of 10 to 6,000 mph would require 10,000 snickers bars, or about 100 gallons of gasoline -- assuming the same efficiency as your engine gets driving you around town. Rockets are considerably less efficient, so getting going that fast with a rocket engine would require more like 1,000 gallons of gasoline -- except that fuel in those quantities would affect the mass of your car, so you have to spend even more fuel to accelerate the fuel you burn at the end. All told, getting your car up to orbital speed with rockets would take something like 5 or ten thousand gallons of fuel.
Whup -- ya caught me! I slipped a decimal place (1kw vs. 10kw)...
The problem with incremental development of RLVs is that there's a huge
leap between the size and difficulty of putting something into space
for five minutes (as in the current X-prize contenders) and putting it
into orbit (as in the shuttle). That will make it difficult to evolve our
way into a commercial space program.
I often find myself pointing out that just getting into space isn't
all that hard. Lifting yourself up 100km requires about a megajoule
(that's the energy equivalent of a stick of dynamite, or about 1/12th
of a gallon of gasoline (about 1/4 kg or 1/2 pound of gasoline), or a
jelly doughnut, or running a hairdryer for 2 minutes) per kilogram of
mass.
By contrast, orbital speed is something like 7000 meters per second,
(or 16,000 miles per hour for you provincials). Getting going that fast
requires an additional 24 megajoules per kilogram of mass (for a total of
25).
In short, the difference between the amount of energy you need to
get into orbit and just into space is a factor of 25, for the same
mass. That ratio of 25 is about equal to the difference between the
latent chemical energies of broccoli and gasoline.
Except that, in the case of space travel, you better be burning
something at least as energetic as gasoline to start with, or you'll
never even hoist yourself up 100km.
The way we've traditionally gotten into orbit is to concentrate the
kinetic energy into ever smaller bits of the vehicle: you use a huge
rocket motor and tanks to get everything started moving, then ditch the
empty tankage and rocket motors for the first stage -- that lets you
concentrate on moving a smaller amount of stuff even faster.
Realistic reusable designs are usually not staged designs,
because it's hard to recover and reuse the first stages. The problem is
that you have to have incredibly lightweight tankage and engines to make
everything work. But pushing stuff to lighter weight makes it more
flimsy and less prone to being reusable. Darn.
The VentureStar, IIRC, ran into problems with exactly this technology --
they were using lightweight carbon fiber tanks to hold their propellant,
and they couldn't make the tank light enough to boost itself into orbit.
The shuttle is NOT a reusable vehicle in any but the most technical
sense of the word: it requires constant skilled redesign and intelligent
(rather than scripted) maintenance, and the engines have to be overhauled
after every flight.
They're not reverse-engineering the hardware, they're reverse-engineering the driver software that is used to talk to the hardware. In the absence of an authoritative interface document, the driver can serve as one. In the absence of source code for the driver, the reverse-engineered driver code can serve as one.
Actualy, no, you are entitled to play the songs to the general public. The whole point of CARP is to mitigate the restrictions inherent in monopoly control. Remember, copyright is an artificial monopoly that is ostensibly intended to encourage creativity. The problem is that, perversely, copyright discourages dissemination of valuable ideas if the owner of those ideas loses interest in exploiting them, or perversely refuses to license them. The CARP is a kludge designed to fix that problem, by forcing content providers to license their content to anyone at a reasonable rate.
Shouldn't WINE run OK under MacOS X?
It certainly should. If you have a choice whether to make your ballot secret, then an evil cartel could potentially use extortion to make you reveal your ballot -- and thereby force you to vote a certain way. If your ballot must be secret to be valid, then your vote is much less susceptible to that kind of control.
No, actually, then it's a tectonic plate. Geological provinces and plates are distinguished from continents, as in "California is made up of several tectonic plates that crashed into one another", or "This part of the continent is a different province than that part over there".
Off topic but I couldn't resist...
... at least through extremities. My old high school physics teacher used to stick his finger in light sockets for laughs. Then he would pull out the Variac and let you stick your own finger in a light socket with variable AC voltage. Up to about 30 VDC you couldn't feel a thing. When you control the voltage yourself, 110VAC is no big deal at all -- just a strong tingling in the finger (in the light socket).
It sounds like one of the main buses in the box got shorted out. You really need to get the thing rewired by a licensed electrician.
If the panel is at work, it's probably illegal for you to work on it. In the immortalized words of Beowulf Schaeffer, "Its Worth Yore Life And More To Go In There!" Seriously, as an uber-l337 633K, you probably have the ability to find the place where the short occurred -- but do you really want to (A) monkey around in a box with live current (it sounds like that is your main box, not a subpanel); (B) stick your neck out by breaking the law for your employer; and (C) possibly assume liability for burning down the building? No? I didn't think so.
If you're at home, you're probably allowed to work on it -- but most places require you to get an electrical inspector to look at it once you're done. Unfortunately, doing this requires getting the local electrical code, so that you know what you're doing -- and that's a whole separate rant. The National Electrical Code is adopted into most municipalities' building codes -- but it's copyrighted by the National Fire Prevention Association. Fortunately, the Supreme Court recently decided that it's unconstitutional to attempt to copyright the law of the land. Unfortuantely, you can't get the electrical code online yet -- you probably still have to buy it.
If text manipulation and piping didn't work well in UNIX, you'd know about it -- all those tasks would be a real thorn in your side. As it is, you have the right tools, so they're no big deal.
Now it's also "the next generation of web forms". Gag me with a buzzword.
It's not as if the original XForms were unknown, either -- it comes up second in a Google search for "Xforms". These jokers should have known better.
Feh.
The difference between this idea and the defunct I-opener is that geeks make up a much smaller proportion of the potential market for the camera than they did for the I-opener. Who cares if 10% of the cameras get cannibalized? They probably only have to be recycled 2-3 times for Ritz to turn a profit.
I discovered 'em recently, and bought 12 of the AAs for an old Logitech cordless optical mouse and keyboard setup that can't run off NiCds (1.3 volts) or NiMHs (1.2 volts). (4 in the keyboard, 2 in the mouse; 2x for continuous use).
I liked 'em so much I went back to Target to find more and discovered that they're not carried there anymore. Nor at Radio Schlock. Nor at a bazillion other places. I think Ray O Vac is trying to phase them out.
It's too bad -- rechargeable alkalines do something the others don't: last a long time in the drawer. It takes years for an RA to self-discharge; about 30 days for a NiCd.
(The upshot is that any NiCd you don't store in a a charger is empty when you actually want it).
It doesn't matter what "causes" the crash. The OS should be essentially crashproof. That's what an OS was for, and it was why Apple got such a drubbing before OS X finally came out (twelve years later).
Hmmm... if dual-booting is a pain, that forces you to make an OS choice. I know which one most slashdotters would take...
Actually, GNU products are excellent and free-as-in-beer. They're so good they're usually invisible. When's the last time you had to work around a bug in ls, or gcc?
That would definitely be from the Free Software Foundation and not the Free Seminar Foundation...
As far as I can tell, "honeytoken" is just a nice sounding buzzword for an ages-old technique.