Better standards and techniques are being adopted, but as cars will still break down so will software.
The parts of a car or any mechanical system suffer from wear and tear, which is an unpredictable process that can cause the system to fail, so it's understandable that engineers cannot predict exactly when a given mechanical system will fail.
Bridges and other structures do not suffer from wear and tear, and so if one collapses, its usually because of an intrinsic flaw in the design, or in the way the bridge was built, or in the way the bridge was being used.
Computer code does not suffer from wear and tear, and code can be written to enforce proper use, so there is no reason for code to break except for an intrinsic design or implementation flaw.
It is company suicide to build a perfect car, because new ones won't replace them and the cash flow stops. This is the same for software. People got to have an incentive to upgrade. Managers make those deadlines so tight for a reason: so you wouldn't make it perfect.
I think I've heard that a company that made titanium wrenches went broke because all of their customers bought one wrench and never needed to buy another one. Seriously, though, this just about summarizes the Achilles heel of the software industry. It is impossible to build a car that never breaks down because when moving parts rub against each other, they wear each other down, and eventually they will wear each other down so much that one or the other or both will fail. On the other hand, it is theoretically possible to write code that will never fail. (For example, there is no reason for properly written "Hello, World!" code to fail, ever) It is just that as the complexity of functionality grows, the complexity of the code to enforce proper functionality grows even faster, and people are lazy and prefer to focus on functionality rather than correctness.
Some of the cool things on display at the Science Museum in London that would probably appeal to the Slashdot crowd are an implementation of Babbage's Difference Engine and a fully mechanical printer for it, and a prototype of Danny Hillis' Clock of the Long Now.
Strictly speaking, light is a subset of electromagnetic radiation. Thus all light is electromagnetic radiation but, as a general rule, not all electromagnetic radiation is considered light.
Actually, there are lots of tunable lasers, and the diode lasers used in communications are intrisically tunable because the refractive index of the semiconductors used to make diode lasers depends on temperature. Thus to tune the emission wavelength over the laser diode's range, you only have to make it run a little hotter or a little cooler.
I think the bigger problem it that it's difficult and expensive to build electronics to demodulate extremely high frequency signals, so you'd only rather have a few of those expensive boxes feed a bunch of cheap boxes instead of having to give everyone the expensive box.
To first order, bending fiber does not affect the refractive index. The main problem with bending fiber is that you lose confinement because, from a ray perspective, the light comes in at too steep an angle and you no longer get total internal reflection and you start experiencing significant signal loss. If you go into more detail, bending fiber probably does cause stress-induced birefringence, which does change the refractive index, but this effect is probably small compared with the significant attenuation due to simple geometrical optics considerations.
Some areas of physics are better represented on arXiv.org than others. My interest is in optics, and not many optics articles appear as preprints on arXiv, so I'm better off keeping current through the journals than through arXiv. Also, materials in journals have at least gone through some peer-review, which helps filter out noise.
I believe that the fact that the scientific community catches its own frauds like cold fusion, or the Schon controversy at Bell Labs mentioned a couple of times on Slashdot within the past few months, or this controversy, reaffirms the validity of the scientific method. If they aren't familiar with a particular subject, scientists seek the opinion of someone who is familiar with the subject and use that person's judgement to determine whether or not the results or theories in question are valid. Contrast this to critical theory, which did not seek a physicist's opinion of Sokal's article and had to have Sokal to tell them the article was nonsense.
And if IBM acquires Eolas, that's a battle that I don't think Microsoft could win. IBM generates more patents every year than any other company--more than Lucent/Bell Labs, and certainly more than Microsoft.
There are an infinite number of ways to express software design A. There is only one way to build the cotton gin, as the patent describes it.
That is not true. If that were how patents on machines were enforced, thise patents would become basically unenforcable. You can always insert a pair of gears or a belt drive with a 1:1 ratio without altering the functionality of the machine. That's why the patents cover functionality and not the implementation.
Most military electronics are shielded against pretty much any kind of electromagnetic pulse short of a nuclear blast. And if you're that close to a nuclear blast, you have more things to worry about than the EMP frying your electronics.
Now civilian electronics are something to worry about. However, I imagine that even civilian avionics are shielded from EMP. Most aircraft are designed to be able to withstand a lightning strike, which is an electromagnetic pulse, so I imagine their electronics would be sufficiently protected. And since electromagnetic pulses are a transient interference source, communications equipment shouldn't be too badly affected.
When I was in high school, I was busy programming my TI-85 to double check all of my 1 variable integral calculus, but I didn't have any idea what the difference between a deep and shallow copy of data was. If I had access to college level materials, I could have had a substantial head start, all in my spare time, all due to my personal enthusiasm for the material.
I think you were limited more by the TI-85 than by your access to college-level course material. AFAIK, the programming language used in TI calculators doesn't have local variables, i.e. all variables are global variables, and variables are copied by value, i.e. all copies are shallow copies. After all, since everything is global, if you want to change the original, just change it--no need to introduce deep copies.
Most of this material is, to be blunt, pirated. (I'm speaking as an instructional tech guy here: I have to deal with these issues.) Faculty will happy scan entire books worth of art, digitize huge tracts of books and in one notable case last year, actually *making multiple photocopies of an entire textbook.*
The fact that professors will photocopy or digitize entire books rather than require students to purchase them should be a red flag to publishers that some textbooks are too damn expensive. At least in engineering, I've seen textbooks that are less than 200 pages long and cost over $100! One of my textbooks this semester is 600 pages and costs $200 new. Is it any surprise that some people are unwilling to buy or to require others to buy those books?
There's no reason that this same system can't be adopted for web publication of coursepacks. Copyright clearance need not be time-consuming or painful. The trouble, of course, is that whereas the students were willing to pay for their coursepacks (even with the added premium of the royalties), no one is going to pay for stuff on the web. Unless we make the current students pay higher tuition to subsidize web publication of their coursepacks or get the government to subsidize the effort, the publishers won't want to adopt the licensing scheme to this new use.
But copyright clearance does become very difficult for publicly-available course materials like OpenCourseWare, since, by virtue of the materials being public, there is no limitation on the number of copies that can be made from a single digital master. Obviously the copyright holders will want to be compensated for each copy, but you can't tell from the access logs whether a single student is accessing the material from different computers or each access is from a different individual. Also, you can't tell how many copies of the material were printed out. Without protection of fair use, projects like OpenCourseWare will be killed by copyright issues.
Because Microsoft is planning to, and has a good chance of, bludgeoning PDF to death with their overwhelming market share in productivity software, not with the technical merits of their product. If that's not anti-competitive abuse of monopoly power, I don't know what is.
Until the light wavefronts can be accurately manipulated, no 3D display will ever be able to approach this level of realism. I don't see it happening within my lifetime.
It's already possible to manipulate wavefronts very accurately, the problem is calculating the wavefront quickly enough to render at video rates.
That's really cool that you had the opportunity to work on holovideo with Prof. Benton. Even though I'm currently in graduate school somewhere else, it's a secret hope in the back of my mind to do my doctoral studies at MIT on holovideo.
As for the folks complaining about the image quality of holovideo, they need to be a little more forward-thinking. The image quality will improve as the efficiency and speed of computing the holographic elements improves. The first raster displays probably looked pretty crummy, too, but they've obviously improved a lot over the years.
Both times, the story poster focused on the stereoscopic display being developed at NYU instead of the much more ambitious (and much cooler, IMHO) holographic display being developed at MIT.
Just because something looks 3D doesn't mean it's a hologram, i.e. the stereoscopic display is not a holographic display.
Each system has its advantages and disadvantages. The stereoscopic display has the advantage of requiring less computation and viewer selectivity. It has the disadvantage of the complication of viewer tracking, and it requires that a separate image be rendered for each viewer. The holographic display has the advantage of being a true three-dimensional image--you can move your head to see the object at different angles without re-rendering, and a single rendered image can be viewed by multiple viewers. The disadvantage is that rendering a holographic is very computation-intensive, and most of the information rendered in a holographic image is not seen by the viewer.
The terms are sufficiently non-derogatory that Stanford uses them in their recruiting material...at least they did 6 years ago when I applied there. And Stanford being the place it is (but not quite Berkeley), they wouldn't use those terms unless there was no risk of offending someone.
Personally, I like the terms and use them all the time, even though I declined the offer of admission from Stanford. I think they're very descriptive.
I tried one on High St. (I was studying at Univ) and the experience was sufficiently traumatizing that I didn't explore those any further...:-p
My fondest memories of Oxford are Blackwell's (damn that's a big bookstore), the Eagle and Child (hoisting a few pints at C.S. Lewis's favorite hangout), and the fact that everything there was so old (I'm still impressed that Univ is three times as old as the U.S.... now that's history).
Likewise, I've been getting into trance/progressive house, but some of the pioneering albums from the early 90s (e.g. Renaissance the Mix Collection 1-3) are out of print, and it costs upwards of $70 to buy a three disc set on eBay. Even some recent albums (e.g. Renaissance Progression [2000]) are out of print and hard to find.
Both are strategically-named to create confusion among American visitors, for whom "student union" describes a building which houses cafeterias and restaurants, convenience stores, meeting rooms, and other student facilities.
When I spent a summer studying in Oxford, I got so lost looking for OUSU, and I don't think I ever saw the Oxford Union.
Does it have: TiBook battery life? TiBook operating temperature?
Sure, just underclock it from 2 GHz to 667 MHz and, voila, you've tripled your battery life and run at 1/3rd the operating temperature. At that clock speed, it might run about the same speed as a TiBook, too.
Dude, some writer for CNet needs to feed his kids, so he resorts to the time-tested journalistic technique of sensationalism to make a story where there is none. VPR Matrix decided they wanted their computers to look kinda cool--not much of the story there. But VPR Matrix asks the Porsche design firm to do it, which allows the author to drop a recognizable name and make this a story.
Slashdot Mirror
I think I've heard that a company that made titanium wrenches went broke because all of their customers bought one wrench and never needed to buy another one. Seriously, though, this just about summarizes the Achilles heel of the software industry. It is impossible to build a car that never breaks down because when moving parts rub against each other, they wear each other down, and eventually they will wear each other down so much that one or the other or both will fail. On the other hand, it is theoretically possible to write code that will never fail. (For example, there is no reason for properly written "Hello, World!" code to fail, ever) It is just that as the complexity of functionality grows, the complexity of the code to enforce proper functionality grows even faster, and people are lazy and prefer to focus on functionality rather than correctness.
Some of the cool things on display at the Science Museum in London that would probably appeal to the Slashdot crowd are an implementation of Babbage's Difference Engine and a fully mechanical printer for it, and a prototype of Danny Hillis' Clock of the Long Now.
More heat is a bad thing. Dissipating that much heat in the atmosphere makes everyone a crispy critter, even if none of the pieces hit the surface.
Strictly speaking, light is a subset of electromagnetic radiation. Thus all light is electromagnetic radiation but, as a general rule, not all electromagnetic radiation is considered light.
Actually, there are lots of tunable lasers, and the diode lasers used in communications are intrisically tunable because the refractive index of the semiconductors used to make diode lasers depends on temperature. Thus to tune the emission wavelength over the laser diode's range, you only have to make it run a little hotter or a little cooler.
I think the bigger problem it that it's difficult and expensive to build electronics to demodulate extremely high frequency signals, so you'd only rather have a few of those expensive boxes feed a bunch of cheap boxes instead of having to give everyone the expensive box.
To first order, bending fiber does not affect the refractive index. The main problem with bending fiber is that you lose confinement because, from a ray perspective, the light comes in at too steep an angle and you no longer get total internal reflection and you start experiencing significant signal loss. If you go into more detail, bending fiber probably does cause stress-induced birefringence, which does change the refractive index, but this effect is probably small compared with the significant attenuation due to simple geometrical optics considerations.
Some areas of physics are better represented on arXiv.org than others. My interest is in optics, and not many optics articles appear as preprints on arXiv, so I'm better off keeping current through the journals than through arXiv. Also, materials in journals have at least gone through some peer-review, which helps filter out noise.
I believe that the fact that the scientific community catches its own frauds like cold fusion, or the Schon controversy at Bell Labs mentioned a couple of times on Slashdot within the past few months, or this controversy, reaffirms the validity of the scientific method. If they aren't familiar with a particular subject, scientists seek the opinion of someone who is familiar with the subject and use that person's judgement to determine whether or not the results or theories in question are valid. Contrast this to critical theory, which did not seek a physicist's opinion of Sokal's article and had to have Sokal to tell them the article was nonsense.
And if IBM acquires Eolas, that's a battle that I don't think Microsoft could win. IBM generates more patents every year than any other company--more than Lucent/Bell Labs, and certainly more than Microsoft.
Most military electronics are shielded against pretty much any kind of electromagnetic pulse short of a nuclear blast. And if you're that close to a nuclear blast, you have more things to worry about than the EMP frying your electronics.
Now civilian electronics are something to worry about. However, I imagine that even civilian avionics are shielded from EMP. Most aircraft are designed to be able to withstand a lightning strike, which is an electromagnetic pulse, so I imagine their electronics would be sufficiently protected. And since electromagnetic pulses are a transient interference source, communications equipment shouldn't be too badly affected.
Because Microsoft is planning to, and has a good chance of, bludgeoning PDF to death with their overwhelming market share in productivity software, not with the technical merits of their product. If that's not anti-competitive abuse of monopoly power, I don't know what is.
That's really cool that you had the opportunity to work on holovideo with Prof. Benton. Even though I'm currently in graduate school somewhere else, it's a secret hope in the back of my mind to do my doctoral studies at MIT on holovideo.
As for the folks complaining about the image quality of holovideo, they need to be a little more forward-thinking. The image quality will improve as the efficiency and speed of computing the holographic elements improves. The first raster displays probably looked pretty crummy, too, but they've obviously improved a lot over the years.
The terms are sufficiently non-derogatory that Stanford uses them in their recruiting material...at least they did 6 years ago when I applied there. And Stanford being the place it is (but not quite Berkeley), they wouldn't use those terms unless there was no risk of offending someone.
Personally, I like the terms and use them all the time, even though I declined the offer of admission from Stanford. I think they're very descriptive.
I tried one on High St. (I was studying at Univ) and the experience was sufficiently traumatizing that I didn't explore those any further... :-p
... now that's history).
My fondest memories of Oxford are Blackwell's (damn that's a big bookstore), the Eagle and Child (hoisting a few pints at C.S. Lewis's favorite hangout), and the fact that everything there was so old (I'm still impressed that Univ is three times as old as the U.S.
Likewise, I've been getting into trance/progressive house, but some of the pioneering albums from the early 90s (e.g. Renaissance the Mix Collection 1-3) are out of print, and it costs upwards of $70 to buy a three disc set on eBay. Even some recent albums (e.g. Renaissance Progression [2000]) are out of print and hard to find.
Both are strategically-named to create confusion among American visitors, for whom "student union" describes a building which houses cafeterias and restaurants, convenience stores, meeting rooms, and other student facilities.
When I spent a summer studying in Oxford, I got so lost looking for OUSU, and I don't think I ever saw the Oxford Union.
*ducks*
Dude, some writer for CNet needs to feed his kids, so he resorts to the time-tested journalistic technique of sensationalism to make a story where there is none. VPR Matrix decided they wanted their computers to look kinda cool--not much of the story there. But VPR Matrix asks the Porsche design firm to do it, which allows the author to drop a recognizable name and make this a story.