The sad thing about this is that most professors know that this is happening. And the solution, well, a lot of people aren't going to like it. There's a principled answer (do lots of delightfully unique, practical assignments that can't just be cribbed; include a lot of 'called onto the carpet' type assessment where the students must verbally justify their essay/code/proof/whatever).
"Unique, practical assignments" are not the principled answer. The princpled answer is: whenever cheating if discovered and can be properly documented [original sources identified, etc.] the students should be expelled from the school.
Why is cheating so common? Because there are no consequences if the student is caught. The school's (university-level) I've taugh at were all too afraid of lawsuits and their reputations to do anything serious about cheating. If they would just follow their own disciplinary procedures (academic probation after a first offence, expulsion after a second), word would get around very quickly and the rate of cheating would go way down.
With a pure math BA you can basically go to any engineering, physics, biology, neuroscience, finance, econ, cs, etc masters or PhD program and do just fine.
I can only speak for physics, but I've seen people try and it doesn't work. If you want a PhD in physics, you need a strong undergraduate physics background; preferably a major but a minor with a related major might do, although you'll start way behind the curve. No one teaches new graduate students how to do basic physics--you're supposed to have learned that before you start. Graduate school is for graduate-level material; there are no remedial courses.
Again, I only know physics, but I'll be very suprised if it's any different for any of the other fields you mentioned.
What on earth were they planning on doing with such a huge stockpile of molasses?!
The were going to make run.
From pre-revolutionary times until Prohibition, Boston was a major center for rum making. Molasses was shipped up from the Carribean (cheaper to ship than raw sugar?). In Boston, they added water, allowed it to ferment, and then distilled it down to rum. The rum was then either shipped over-land or exported to Europe and Africa (completing the third leg of the old slave triangle).
The tank that ruptured was just a holding tank, but the rum was heated to keep it liquid. The people who died were burned to death. Local legend had it that for dacades after you could still get a whiff of molasses on hot August days in certain parts of north Boston, the smell coming up from the ground, a residue from the ruptured tank.
Neutrino mass has been an established fact since 1998 (courtesy work at the Super-Kamiokande).
Super-Kamiokande showed that neutrinos have mass if our models about solar neutrino production are correct. (There was a slight day/night shift observed too, but that's statistically weaker). SNO-ball provided strong supporting evidence too, by which point pretty much everyone agreed neutrino mass was right theory, but there's still the question of exactly what the solar neutrino flux is made up of.
MINOS showed the same thing (neutrion oscillations), but this time with a direct measurement of the source neutrino flux and not a model-dependent estimate of the flux. There's really no way to argue with this result, only with the exact mechanism for neutrino oscillation.
Youngsters these days, always reinventing the wheel. This whole thing about using case statements to set up labels for an iterator, it's really just a computed goto--Fortran's had 'em for at least 40 years.
C... HTML doesn't like my indentations, but you get the idea.
C... Set the initial value i = 1
100 goto (200,300,400,999), i STOP 'state variable out of range'
I'm curious why Americans are so shocked that the world preeminence we have enjoyed for a century looks like it will come an end in the next few generations (if we're lucky).
History is in fact rife with empires that rose to politcal, military and cultural dominance and then (for whatever reason) saw it slip away. The English before US. The Spanish before them. The HRE, Romans, Egyptians...
Because we're even worse at studying history than we are at science?
In my opinion (as someone who does both Unix and mainframe programming for a living) the problems you describe have a lot to do with the fact that COBOL programmers tend to be older people, who rolled into the programming field from something else, such as engineering or accounting a long time ago. They didn't study CS or IT in college, because there was not really such a field 20 years ago.
20 years? More like 40 years ago. At the universities I've worked at, CS departments split off from EE (or, occasionally, math) departments in the 60's. By 1985 (20 years ago), CS was pretty universally its own department.
"Problem: Legacy Apps You'd be surprised how many cycles the world's Sun boxes spend running decades-old FORTRAN, COBOL, C, and C++ code in monster legacy apps that work just fine and aren't getting thrown away any time soon. There aren't enough people and time in the world to re-write these suckers, plus it took person-centuries in the first place to make them correct.
Well, the Fortran programs have an easy solution---just recompile with a modern compiler designed for these CPU's. Any loop that can be automatically unrolled can be parallelized instead. Loop parallelization has been a standard Fortran optimization on parallel architectures for decades. Yes, this can be done with other languages as well, but historically it hasn't been (I expecte either due to a lack of demand, or because it's harder to accomodate language features [things like strict aliasing], or both).
Yes, fossil reserves will eventually end. But here's the fun part: we already know how to produce synthetic oil. We've known it for a long time. And not just theoretically: Germany's WW2 tank warfare was _based_ on synthetised fuel. It wasn't cheap, but it did keep the panzers rolling nevertheless.
The problem is energy, not its source. Fossil fuels are a finite supply, that goes for coal and natural gas as well as oil. Nuclear is more plentiful now, but also finite. Fusion doesn't work yet, hydro is too limited by geography. Solar works, but no one's been willing to sacrifice the surface area to make it a real commercial source (not to mention that whole night thing). Sure we know how to turn carbon into gasoline, but it takes energy to do that, and the energy has to come from somewhere. If we run out of fossil fuels, we've just put a very large dent in our available energy economy.
Tools like Matlab are great for working scientists who need to get the job done by don't have access to real programming. For those who do, the latter appears to be a much more popular choice.
I find Matlab is also a nice language for prototyping numerical codes. It has its own built-in debugger, it's interpreted so editing is fast, and its syntax for matrix and vector operations is remarkably compact. These all come in handy when you just want to play with some algorithms and see what works. On the other hand, some operations (e.g., any for-loops that it can't unroll) are egregiously slow, so I agree---once you know roughly how to proceed, you're better off scraping the Matlab code and rewriting using a compiled language.
Class III lasers are the most dangerous ones, resulting in instant damage. I believe most if not all laser pointers are class 3 devices.
Class III means the laser is an eye hazard. Class IIIa are hazardous if focused, class IIIb are focused as-is. There are also class IV lasers, which are burn hazards (I've worked with both) and are more dangerous than class III lasers. Most laser pointers are class IIIb, although I've seen lower. Note too that laser classification is based on power at the output, not at the source.
"Instant damage" is a bit over-stated, but class IIIb lasers (especially at the >100mW levels in the original article) are most definitely capable of causing retinal damage.
Don't know about right-pondians, but in the US, "cue" means to ready a performer (or, in modern times, a tape or digital clip) to begin performing and therefore "cue" is the right word and the correct spelling, not "queue".
Last time I read up on this PNG is not suitable for photo realistic images. This is what JPEG does well - does anyone know of a good alternative?
Suitable is subjective. PNG is lossless-the image you get out is exactly the same as the image you put in. JPEG is lossy, some information is lost in the process of compressing the image (thereby improving the compression ratio).
If you want photo-realistic and recovering the original image is important (e.g., medical imaging), then PNG is entirely suitable. If getting the smallest possible image is more important, then JPEG is more suitable.
Bullshit, if they want to keep their research bottled up, then don't use government grants.
For basic research (i.e., anything that won't obviously lead to a marketable product in less than 12 months), you generally don't have a choice. There is no other source of money.
I do research for a living - trust me, I've tried to find every source of grant money I can, and unless you're working in a high-profile field like breast cancer, alternate sources simply don't exist.
As far as "killing the financial base" of the scientific publication market goes: Yes, it might just do that. I don't believe that anyone guaranteed that publication market any kind of revenue stream, let alone a good one. They've had it made recently, being able to raise prices to astronomical levels. Now those prices might have to fall. It's called business, people. Get over it.
To continue your analogy, what kind of business can possibly operate if the government madates that they have to give away all their product? Any industry you do this to will either go to a 100% advertising model (not feasable here due to the small size) or go out of business. There is no other alternative. I say this both as a subcriber and an author.
Free publications [as in beer] can mean only bankrupcy and the mass extinction of journals. Remember, many of these journals (all the ones I read, certainly) are put out by non-profit scientific societies. There is no other source of income to subsidize these journals with and if they lose individual and library subscriptions they will close.
I've done grants before. There is no difference made between foreign and domestic graduate students. Research assistants (aka graduate students) get their tuition covered, so this ammounts to an indirect subsidy.
On the other hand, since many of the students will remain in the country after graduation, this isn't neccessarily a bad thing.
No offense, but thats one of the stupidest things I've heard today. Which get paid more, research scientists in the private industry whose work is generally protected by patents or research scientists in academia whose work is generally not?
This is not correct (having worked as a researcher in academia).
Both private industry and academia protect the IP generated by their research scientists. If you have a patentable idea, both will generally file the patent etc for you, for free (assuming in both cases that the IP might have enough future economic value to be worth patenting in the first place). In private industry, that's all you get (maybe a bonus if they feel generous, but it's usually not required). Academics usually get 50% of any and all future royalties.
In sort, both academics and private industry protect the researchers IP, and arguably you (as researcher) get a much better deal [exclusive of salary] in academia than in industry.
While I'm painting something of a rosy picture here, I do have a point to this rant. The US is losing *engineers* for various reasons.
I am not an engineer, I'm a scientist (specifically, a physicist) and we are losing scientists too. For about 20 years we [the science community] have been off-setting that loss by pulling in more international students, but as INS tightents the visa rules and the processing times have gone up, many internationals are opting to go elsewhere (Europe?) instead of studying in the US.
The question then becomes, do we really need these people? We've been pumping out scientists (at all levels-BS, MS, and PhD) far in excess of the avaible jobs for years which means most people end up leaving science in order to eat or support families (Wall St. was popular in the early- to mid-90's, programming jobs have been popular too for some time).
Part of the problem is the need to keep Universities well staffed with graders and research associates, without worrying if there's a reasonable chance of finding them a job when they graduate. This "shortage" may be nothing more than a well-needed adjustment to existing market conditions. God forbid, after all, that the supply of PhD scientists should every dry up so far that we have to pay them in fair portion to the value they bring to a corporation after all...
P.S. We also don't understand why quantum mechanics rules apply at very small scales, but very different rules apply at larger scales. (A photon can seem to go through two slits at once, but you won't get a baseball to do that trick, or even a really tiny speck of dust.)
You can get interference fringes with buckyballs (60 carbon atoms), so it does work with really, really, really, really tiny specks of dust! It's all a function of the deBroglie wavelength in the end.
As it happens I am a super-Copenhagen believer, that is, our function, as conscious entities, is to observe the many possible universes and 'select' the real one.
This defines consciousness, by the way.
So this means my photo-diode (which emits electrons only when it "detects" a photon) is conscious?
I believe that you have a very poor understanding of what measurement is
If you're a physicist, perhaps you can answer a question that has been puzzling me for some time when reading about the slit experiment: what exactly is 'a measurement'? Is there a scientific definition? For example, if something detects the photon, but then discards the information does it still count as a measurement (and affect the intereference pattern)?
I'm not the original poster, but I am a physicist.
Measurement in quantum mechanics is a rather badly-defined term unless you get into the full mathematics behind quantum theory. A useful rule of thumb, though, is that anything that provides information about the state of the system is a measurement. It doesn't matter what you do with the information, as long as the time of "detection" the information is there to be had.
The longer answer is that anything that forces the system into a definite eigenstate is a measuremet. This, unfortunately, tends to be a recursive definition (How do you put the system into a definite eigenstate? You make a measurement...)
It's an interesting question, do quantum wavefunctions (and their dynamical evolution) have any physical meaning or are they a mathematical fictions that happens to yield the correct answer every time?
Radioactive material, like all other toxins, requires a certain concentration to be lethal.
While many people would like this to be the case (for one thing, it would allow safety standards to be relaxed considerably) there is no clear scientific evidence either for or against this statement. The problem is distinguishing low-exposure cancers from the natural background cancer rate--the statistics simply aren't there to justify setting a non-zero threshold.
It is only a natural evolution from character based displays to bitmaps to vector graphics.
Except that vector terminals pre-date character terminals (many of the first video games like "Asteroids" used nothing but vector graphics). We've simply closed the loop.
Slashdot Mirror
"Unique, practical assignments" are not the principled answer. The princpled answer is: whenever cheating if discovered and can be properly documented [original sources identified, etc.] the students should be expelled from the school.
Why is cheating so common? Because there are no consequences if the student is caught. The school's (university-level) I've taugh at were all too afraid of lawsuits and their reputations to do anything serious about cheating. If they would just follow their own disciplinary procedures (academic probation after a first offence, expulsion after a second), word would get around very quickly and the rate of cheating would go way down.
-JS
I can only speak for physics, but I've seen people try and it doesn't work. If you want a PhD in physics, you need a strong undergraduate physics background; preferably a major but a minor with a related major might do, although you'll start way behind the curve. No one teaches new graduate students how to do basic physics--you're supposed to have learned that before you start. Graduate school is for graduate-level material; there are no remedial courses.
Again, I only know physics, but I'll be very suprised if it's any different for any of the other fields you mentioned.
-JS (PhD)
The were going to make run.
From pre-revolutionary times until Prohibition, Boston was a major center for rum making. Molasses was shipped up from the Carribean (cheaper to ship than raw sugar?). In Boston, they added water, allowed it to ferment, and then distilled it down to rum. The rum was then either shipped over-land or exported to Europe and Africa (completing the third leg of the old slave triangle).
The tank that ruptured was just a holding tank, but the rum was heated to keep it liquid. The people who died were burned to death. Local legend had it that for dacades after you could still get a whiff of molasses on hot August days in certain parts of north Boston, the smell coming up from the ground, a residue from the ruptured tank.
-JS
Super-Kamiokande showed that neutrinos have mass if our models about solar neutrino production are correct. (There was a slight day/night shift observed too, but that's statistically weaker). SNO-ball provided strong supporting evidence too, by which point pretty much everyone agreed neutrino mass was right theory, but there's still the question of exactly what the solar neutrino flux is made up of.
MINOS showed the same thing (neutrion oscillations), but this time with a direct measurement of the source neutrino flux and not a model-dependent estimate of the flux. There's really no way to argue with this result, only with the exact mechanism for neutrino oscillation.
-JS
Youngsters these days, always reinventing the wheel. This whole thing about using case statements to set up labels for an iterator, it's really just a computed goto--Fortran's had 'em for at least 40 years.
-JS
Because we're even worse at studying history than we are at science?
-JS
20 years? More like 40 years ago. At the universities I've worked at, CS departments split off from EE (or, occasionally, math) departments in the 60's. By 1985 (20 years ago), CS was pretty universally its own department.
-JS (showing my age, alas)
Well, the Fortran programs have an easy solution---just recompile with a modern compiler designed for these CPU's. Any loop that can be automatically unrolled can be parallelized instead. Loop parallelization has been a standard Fortran optimization on parallel architectures for decades. Yes, this can be done with other languages as well, but historically it hasn't been (I expecte either due to a lack of demand, or because it's harder to accomodate language features [things like strict aliasing], or both).
-JS
The problem is energy, not its source. Fossil fuels are a finite supply, that goes for coal and natural gas as well as oil. Nuclear is more plentiful now, but also finite. Fusion doesn't work yet, hydro is too limited by geography. Solar works, but no one's been willing to sacrifice the surface area to make it a real commercial source (not to mention that whole night thing). Sure we know how to turn carbon into gasoline, but it takes energy to do that, and the energy has to come from somewhere. If we run out of fossil fuels, we've just put a very large dent in our available energy economy.
-JS
I find Matlab is also a nice language for prototyping numerical codes. It has its own built-in debugger, it's interpreted so editing is fast, and its syntax for matrix and vector operations is remarkably compact. These all come in handy when you just want to play with some algorithms and see what works. On the other hand, some operations (e.g., any for-loops that it can't unroll) are egregiously slow, so I agree---once you know roughly how to proceed, you're better off scraping the Matlab code and rewriting using a compiled language.
-JS
Class III means the laser is an eye hazard. Class IIIa are hazardous if focused, class IIIb are focused as-is. There are also class IV lasers, which are burn hazards (I've worked with both) and are more dangerous than class III lasers. Most laser pointers are class IIIb, although I've seen lower. Note too that laser classification is based on power at the output, not at the source.
"Instant damage" is a bit over-stated, but class IIIb lasers (especially at the >100mW levels in the original article) are most definitely capable of causing retinal damage.
-JS
Don't know about right-pondians, but in the US, "cue" means to ready a performer (or, in modern times, a tape or digital clip) to begin performing and therefore "cue" is the right word and the correct spelling, not "queue".
-JS
Suitable is subjective. PNG is lossless-the image you get out is exactly the same as the image you put in. JPEG is lossy, some information is lost in the process of compressing the image (thereby improving the compression ratio).
If you want photo-realistic and recovering the original image is important (e.g., medical imaging), then PNG is entirely suitable. If getting the smallest possible image is more important, then JPEG is more suitable.
-JS
Unfortunately, PNG transparency is also something MS Internet Explorer can't do either.
-JS
For basic research (i.e., anything that won't obviously lead to a marketable product in less than 12 months), you generally don't have a choice. There is no other source of money. I do research for a living - trust me, I've tried to find every source of grant money I can, and unless you're working in a high-profile field like breast cancer, alternate sources simply don't exist.
-JS
To continue your analogy, what kind of business can possibly operate if the government madates that they have to give away all their product? Any industry you do this to will either go to a 100% advertising model (not feasable here due to the small size) or go out of business. There is no other alternative. I say this both as a subcriber and an author.
Free publications [as in beer] can mean only bankrupcy and the mass extinction of journals. Remember, many of these journals (all the ones I read, certainly) are put out by non-profit scientific societies. There is no other source of income to subsidize these journals with and if they lose individual and library subscriptions they will close.
-JS
I've done grants before. There is no difference made between foreign and domestic graduate students. Research assistants (aka graduate students) get their tuition covered, so this ammounts to an indirect subsidy.
On the other hand, since many of the students will remain in the country after graduation, this isn't neccessarily a bad thing.
-JS
When getting 50-100 qualified applications for a single job posting is routine, you have too many researchers.
-JS
This is not correct (having worked as a researcher in academia).
Both private industry and academia protect the IP generated by their research scientists. If you have a patentable idea, both will generally file the patent etc for you, for free (assuming in both cases that the IP might have enough future economic value to be worth patenting in the first place). In private industry, that's all you get (maybe a bonus if they feel generous, but it's usually not required). Academics usually get 50% of any and all future royalties.
In sort, both academics and private industry protect the researchers IP, and arguably you (as researcher) get a much better deal [exclusive of salary] in academia than in industry.
-JS
I am not an engineer, I'm a scientist (specifically, a physicist) and we are losing scientists too. For about 20 years we [the science community] have been off-setting that loss by pulling in more international students, but as INS tightents the visa rules and the processing times have gone up, many internationals are opting to go elsewhere (Europe?) instead of studying in the US.
The question then becomes, do we really need these people? We've been pumping out scientists (at all levels-BS, MS, and PhD) far in excess of the avaible jobs for years which means most people end up leaving science in order to eat or support families (Wall St. was popular in the early- to mid-90's, programming jobs have been popular too for some time).
Part of the problem is the need to keep Universities well staffed with graders and research associates, without worrying if there's a reasonable chance of finding them a job when they graduate. This "shortage" may be nothing more than a well-needed adjustment to existing market conditions. God forbid, after all, that the supply of PhD scientists should every dry up so far that we have to pay them in fair portion to the value they bring to a corporation after all...
-JS
You can get interference fringes with buckyballs (60 carbon atoms), so it does work with really, really, really, really tiny specks of dust! It's all a function of the deBroglie wavelength in the end.
-JS
So this means my photo-diode (which emits electrons only when it "detects" a photon) is conscious?
-JS
I'm not the original poster, but I am a physicist.
Measurement in quantum mechanics is a rather badly-defined term unless you get into the full mathematics behind quantum theory. A useful rule of thumb, though, is that anything that provides information about the state of the system is a measurement. It doesn't matter what you do with the information, as long as the time of "detection" the information is there to be had.
The longer answer is that anything that forces the system into a definite eigenstate is a measuremet. This, unfortunately, tends to be a recursive definition (How do you put the system into a definite eigenstate? You make a measurement...)
It's an interesting question, do quantum wavefunctions (and their dynamical evolution) have any physical meaning or are they a mathematical fictions that happens to yield the correct answer every time?
-JS
While many people would like this to be the case (for one thing, it would allow safety standards to be relaxed considerably) there is no clear scientific evidence either for or against this statement. The problem is distinguishing low-exposure cancers from the natural background cancer rate--the statistics simply aren't there to justify setting a non-zero threshold.
-JS
Except that vector terminals pre-date character terminals (many of the first video games like "Asteroids" used nothing but vector graphics). We've simply closed the loop.
-JS