Many posts here talking about retraining time and familiarity. I have used SW for five years and PTC CREO/Wildfire/Pro-E for more than twenty. The simple fact is that while FreeCAD is a nice try and has some useful features for very simple work, it is not even in the same discussion group as "real" CAD software. A similar thing can be said regarding program for FEA and CFD analysis. The irony is that Pro-E was actually developed for Unix (OK, Solaris) and abandoned that field.
A sign of hope is some of the newer programs, like Abacus, will run in a linux environment. But there is no indication that any of the full-feature CAD and analysis combination packages are even remotely thinking about running in linux.
I manage better than most commenting here with LibreOffice replacing Word and exchanging documents with my Microsoft-laden friends. But there is a Win7 workstation under the desk to run SW and CREO that I would love to replace but cannot.
This is a very important point. And the current set of changes is not the first time there has been a cost shift because of a reduction in government support. This also occurred in the late sixties and early seventies when the DoD was forced to reduce their (rather large) support to university research programs.
Since tuition is a fraction of actual cost, a shift in the external support produces a disproportionate impact on tuition. When I started teaching at a state school 20 years ago, tuition was ~25% of cost at our institution. It is now a bit more than a third because of the reduction of external support. That means that even if there were no change in costs and no inflation, tuition would go up by 32%. Add to that inflation (and that means real inflation of the goods and services that a university uses) and you can see a serious increase in student cost.
But as they say, it is worse than that. Over the last 20 years there has been a steady stream of legislation at both the state and federal level that has introduced new tasks and concomitantly new expenses. This is over and above inflation and has little positive effect on actual instruction. I would also add clear increases in bureaucratic processes associated with accreditation. The bottom line is that student tuition and fees (don't forget fee escalation) has gone up scandalously. At the same time, I am teaching two to three times as many students (who accordingly get less of my time) than I did when I started. And it is similar for most of my colleagues.
I will be the first to agree that we should be looking at different and more efficient modalities of instruction. But we also need to be thinking clearly about all of the factors influencing tuition. --- not that thoughtfulness is a hallmark of slashdot
I use a Contour Roller Mouse and like it very much. It takes a bit of getting used to since its motion is somewhat different than a conventional mouse or trackball. This keeps my hands close to the keyboard at all times. The newer models have a number of buttons. I originally got it on the advice of an industrial ergonomist to address shoulder and neck pain from long hours at the computer (something that it has, indeed, improved significantly), but now find that I am more productive in all applications except CAD and graphics work. It is a bit expensive, but constitutes an interesting alternative.
I have been using an fi-6130 for several months now. It is quite simply the best scanner I have used. It is fast, highly reliable and very seldom misfeeds (1 per 500-800 pages in my experience). I use it for scanning archival financial records and also for technical papers. It includes a copy of Kofax Virtual ReScan, which does a great job of creating readable 1-bit monotone scans of originals with colored backgrounds. There are a number of possible target formats, and it has several automated ways of handling group separator sheets. I highly recommend it. I have seen no evidence of "marketing drone foolishness."
While it is true that fonts may be embedded, it is also true that PDF provides the mechanism for proprietary fonts to decline embedding. This is general feature of some proprietary fonts. I run into this frequently with PDF documents that contain ancient Greek or Hebrew fonts. In about a third of these documents the PDF will not render correctly because they have used proprietary fonts that I do not have on my machine. In these cases, the document is completely unreadable since the Latin font equivalent chosen in no way reflects the underlying text.
Indeed, the presumed oscillations imply that the mass of the neutrino is small, but not zero. See, for example http://focus.aps.org/story/v2/st10 for a good discussion. Getting a good experimental measure of the mass of a particle that interacts so weakly with detectors has been a very long running challenge in experimental physics.
For those interested, the paper itself can be found at http://arxiv.org/abs/0708.2251v1. The team is detecting neutrinos from Be 7 at the rate of 47 per day.
There seems to be two questions that can be asked in this regard. One is that of the parent: why should you be viewed as an expert. The CV actually has fewer than 5 refereed publications if you discount those that were published when he was a student working under the supervision of one of the other authors. This is a truly remarkably thin publication record for someone who has been a professor for this length of time. For practical purposes, there is no peer recognition of his expertise. And the other information on the CV does not, for the most part, apply to this area of expertise either. I am surprised that the court has accepted his credentials as an expert. I am not surprised that he has not been promoted past associate professor despite his time of service.
The second question would what the relationship between his testimoney in this case and his personal business interests might be. It would appear on the surface that success in RIAA cases would serve as advertising for the software he produces. The connection might be arguable, but it does seem worth asking.
A longer summary article of the recent paper whose abstract is referenced above is here. Note that this is a recent article. The Antikythera Mechanism has been discussed before on/., but this paper is recent.
The stability is greater than the earlier superheavy atoms produced. This is an indication that one may be approaching the hypothesized "island of stability". I have not seen the lifetime published yet, but was told that it was "of order milliseconds".
Two answers: 1) these results are unclassified. Classified calculations may be done for other materials and conditions which will be more directly applicable to weapons questions. And more importantly, 2) the goal of the calculations is developing an a priori understanding of the behavior of high-Z (many electrons) materials at extreme conditions. Calculating the response of a material like Mo allows comparison of the numerical result with actual experiments to see if the calculation is correctly reproducing the real physics. This may not represent exact conditions in a weapon, but it is getting the science right which is the first goal. When you establish that the science is right, you are in a position to do more things with confidence.
Note also that Blue Gene/L is used for a wide array of scientific calculations, not just things of interest for weapons physics.
So the Cell is great because there's going to be millions of them sold in PS3's so they'll be cheap. But it's only really great if a new custom variant is built. Sounds kind of contradictory.
The HPC world is substantially different from either gaming or "normal" application programming. The strong draw of the cell is that it is a production core with characteristics that are important to High Performance Computing, particularly power dissipation per flop. While conventional applications target getting the most out of a processor, HPC applications center on scalability in number of processors. This means running the largest number of processors for a given power/cooling supply, and maintaining the lowest latency in interprocessor communication. The latter is closely related to the physical layout of the processor array, which is also dependent upon cooling strategy. Hand coding, or at least hand optimization of the code, is reasonable for these applications. The resulting improvement can make possible calculations that would otherwise not be accomplished. As the number of processors increases substantially, the leading issue shifts from local execution speed to load balancing. Load balancing requires at least an initial "hand code" for a given architecture in any event.
There are several application spaces for HPC. Some, like semantic network processing do not require double precision and can be mounted on cell processors as they stand. Those which are fundamentally based on massive differential equation solution would benefit from the double precision modification. The key point here is that the double precision pipeline unit is a modification, not a different core. It is likely that IBM can make such a change at a fraction of the cost of the original core development with benefits not only to the HPC community, but also to potential workstation use.
The bottom line is than one can be easily mislead trying to think of HPC architectures and programming from the familiar standpoint of game and web server development.
I think it provides UC with some serious money and opportunity to do major research, so the geeks get attracted to it and tend to brush over any ethical concerns.
There may be a number of motivations, but it is not likely that money has been a large one...at least up to now. In the original LANL contract with UC, the maximum fee was $8.7M (about 0.4% of the LANL operating budget). Maximum because it could (and has been) reduced based on the DOE evaluation of performance relative to contract criteria. While that may sound like a lot of money, it is very small compared to the UC total research budget and certainly extremely small when viewed in the light of the controversy and difficulty of managing the Laboratory.
One of the things that the source selection board discovered in trying to solicit serious industrial interest was that the fee was too small. Even though they started with a much larger fee in the original solicitation, they had to double it in order to attract industrial participants. The fee for managing Los Alamos now stands at $79M, nine times the original fee. It is hoped that this increased fee will be recouped by improvements in efficiency.
While the University of California will be deeply involved in the new management of the Los Alamos National Laboratory, it is not strictly correct to call this a win for UC. As the DOE press release makes clear, the winner of the competition was a limited liability corporation comprised of UC, Bechtel, BWX Technologies and others. The difference is very significant in some areas. For example, LANL personnel will no longer be members of the UC staff and participants in their retirement system, but employees of the LLC. The DOE did not release details of the winning proposal yet. As they do, I believe it will become increasingly clear that there is much more to this change than just UC continuing to play the same role.
In operation, heads are supported on a very thin aerodynamic bearing film, indeed. There are still two conditions that influence the life and allowable speed of the drive. One is the fact that the heads must land at some time when the drive is off in most (although not all) designs. The other is that the aero film is much smaller than a typical particle. It is quite undesireable to have a bearing in which the slightest degree of contact or small particle immediately results in catastrophic failure. Designers of aerostatic and aerodynamic bearings of all types pay careful attention to the plain bearing properties of the materials that they use for this reason.
The article discusses spot size as resulting from the shorter wavelength. From the description in the graphic, a 50-75 nm wavelength is required. There is no available source in this range of wavelengths that does not occupy a whole lot of space, take a whole lot of power and cost a whole lot of money. This is not really ultraviolet; it is closer to soft x-ray. Some idea of the difficulty can be derived from observing what the lithography community has struggled with to get a 13 nm source for "extreme ultraviolet lithography". For that matter, the community has sharply reduced work at 157 nm (in favor of immersion 193) for lack of a workable material set for the optics. The wavelength that is apparently proposed here is quite a bit more energetic than 157 and probably nearly as difficult to produce and direct as the 13 nm. Virtually all materials absorb at these wavelengths. Moreover, the photon energy is well in excess of available semiconductor material bandgaps, implying to semiconductor laser source. Whatever may be true of the recording mechanism, there is no clear path to implementation of this kind of device.
I have been an author, reviewer and associate editor at one time or another for scientific journals. None of these three are renumerated activities. While I object to the high price of the journals, I do have a little more appreciation for the economics. It is true that the original articles are written and the review process is carried out without expense to the journal. However, the publisher does pay the cost of administering this process and also (in most cases) the cost of editing, setting the article for publication, printing and marketing. While I believe that substantial profit is still involved, it is true that there is considerable expense as well. This expense, coupled with the very small circulation of these journals is one of the reasons for the apparently high substription price. One just cannot think of archival journals with small circulations and no advertising income in the same terms as mass market magazines.
It is also the case that some professional societies derive significant income from their publications.
Having said this, I would still donate my time to review and edit for an open electronic journal. But such a journal would still have expenses, and it is unclear how they would be met.
...it seems that you misunderstand copyright. Copyright protects not an idea but an expression of an idea...
...they should be able to make a clean-room implementation and sell it, then that's fair. However, copyright protection doesn't prevent that, so it's not an argument against copyright.
Using Rowling and Harry Potter as an example is interesting. While they may not be strictly "clean room" parallels, the works which have been attacked by Rowling's publisher are nonetheless original writing. They are being attacked because they copy some part of an idea, not because they copy text. So your notion of copyright may not be objectionable, but the actual instantiation we have now may be.
You have to read the full text! While the early portions of the treaty do, indeed, enforce compliance with established international treaties, the latter portions add additional requirements. Specifically included are the life-plus-seventy copyright life and the provisions of the US DMCA. Take a look at article 17.4, item 7 which contains a detailed enumeration of requirements with regard to technical protection of copyright. This goes far beyond the current state of international agreements.
There is, indeed, a similarity between DRAM and the structure postulated by the authors. In both cases, a part of the structure stores the information and that storage would be continuous if it were not for leakage. The presence of loss in the device limites the storage time and leads to the need for refresh. Since the paper only describes a theoretical structure, it is probably a little early to be able to make any statements about the actual loss and resulting storage time for such a device.
Yes, I believe the intention is to allow such a structure to be a storage device. However, one should probably think of it more as information storage than energy storage. The entire light pulse must fit physically in the device in order to achieve the results when the refractive index is modulated. That implies that only very short pulses can be stored, since the pulse speed before modulation is a significant fraction of c.
Yes, the concept (it is only a theoretical concept, not a chip, in the paper) does store the light. When the optical pulse is completely within the postulated structure (meaning only a very short pulse can be stored), a modulation of the refractive index causes the fields associated with the pulse to be stored in the internal cavities of the crystal. Reversing the refractive index change causes the stored fields to reform a traveling wave, which exits the structure.
The way that you know that the pulse has been stored in the computer simulations is that after the first refractive index change, nothing comes out of the structure. After the second change, a pulse emerges that has the same shape as the one that was sent in.
Slashdot Mirror
Many posts here talking about retraining time and familiarity. I have used SW for five years and PTC CREO/Wildfire/Pro-E for more than twenty. The simple fact is that while FreeCAD is a nice try and has some useful features for very simple work, it is not even in the same discussion group as "real" CAD software. A similar thing can be said regarding program for FEA and CFD analysis. The irony is that Pro-E was actually developed for Unix (OK, Solaris) and abandoned that field. A sign of hope is some of the newer programs, like Abacus, will run in a linux environment. But there is no indication that any of the full-feature CAD and analysis combination packages are even remotely thinking about running in linux. I manage better than most commenting here with LibreOffice replacing Word and exchanging documents with my Microsoft-laden friends. But there is a Win7 workstation under the desk to run SW and CREO that I would love to replace but cannot.
This is a very important point. And the current set of changes is not the first time there has been a cost shift because of a reduction in government support. This also occurred in the late sixties and early seventies when the DoD was forced to reduce their (rather large) support to university research programs. Since tuition is a fraction of actual cost, a shift in the external support produces a disproportionate impact on tuition. When I started teaching at a state school 20 years ago, tuition was ~25% of cost at our institution. It is now a bit more than a third because of the reduction of external support. That means that even if there were no change in costs and no inflation, tuition would go up by 32%. Add to that inflation (and that means real inflation of the goods and services that a university uses) and you can see a serious increase in student cost. But as they say, it is worse than that. Over the last 20 years there has been a steady stream of legislation at both the state and federal level that has introduced new tasks and concomitantly new expenses. This is over and above inflation and has little positive effect on actual instruction. I would also add clear increases in bureaucratic processes associated with accreditation. The bottom line is that student tuition and fees (don't forget fee escalation) has gone up scandalously. At the same time, I am teaching two to three times as many students (who accordingly get less of my time) than I did when I started. And it is similar for most of my colleagues. I will be the first to agree that we should be looking at different and more efficient modalities of instruction. But we also need to be thinking clearly about all of the factors influencing tuition. --- not that thoughtfulness is a hallmark of slashdot
I use a Contour Roller Mouse and like it very much. It takes a bit of getting used to since its motion is somewhat different than a conventional mouse or trackball. This keeps my hands close to the keyboard at all times. The newer models have a number of buttons. I originally got it on the advice of an industrial ergonomist to address shoulder and neck pain from long hours at the computer (something that it has, indeed, improved significantly), but now find that I am more productive in all applications except CAD and graphics work. It is a bit expensive, but constitutes an interesting alternative.
I have been using an fi-6130 for several months now. It is quite simply the best scanner I have used. It is fast, highly reliable and very seldom misfeeds (1 per 500-800 pages in my experience). I use it for scanning archival financial records and also for technical papers. It includes a copy of Kofax Virtual ReScan, which does a great job of creating readable 1-bit monotone scans of originals with colored backgrounds. There are a number of possible target formats, and it has several automated ways of handling group separator sheets. I highly recommend it. I have seen no evidence of "marketing drone foolishness."
I don't work for Plexus, but I am familiar with some of their projects. From what I have seen they are a first class outfit.
While it is true that fonts may be embedded, it is also true that PDF provides the mechanism for proprietary fonts to decline embedding. This is general feature of some proprietary fonts. I run into this frequently with PDF documents that contain ancient Greek or Hebrew fonts. In about a third of these documents the PDF will not render correctly because they have used proprietary fonts that I do not have on my machine. In these cases, the document is completely unreadable since the Latin font equivalent chosen in no way reflects the underlying text.
Indeed, the presumed oscillations imply that the mass of the neutrino is small, but not zero. See, for example http://focus.aps.org/story/v2/st10 for a good discussion. Getting a good experimental measure of the mass of a particle that interacts so weakly with detectors has been a very long running challenge in experimental physics.
For those interested, the paper itself can be found at http://arxiv.org/abs/0708.2251v1. The team is detecting neutrinos from Be 7 at the rate of 47 per day.
There seems to be two questions that can be asked in this regard. One is that of the parent: why should you be viewed as an expert. The CV actually has fewer than 5 refereed publications if you discount those that were published when he was a student working under the supervision of one of the other authors. This is a truly remarkably thin publication record for someone who has been a professor for this length of time. For practical purposes, there is no peer recognition of his expertise. And the other information on the CV does not, for the most part, apply to this area of expertise either. I am surprised that the court has accepted his credentials as an expert. I am not surprised that he has not been promoted past associate professor despite his time of service.
The second question would what the relationship between his testimoney in this case and his personal business interests might be. It would appear on the surface that success in RIAA cases would serve as advertising for the software he produces. The connection might be arguable, but it does seem worth asking.
A longer summary article of the recent paper whose abstract is referenced above is here. Note that this is a recent article. The Antikythera Mechanism has been discussed before on /., but this paper is recent.
The stability is greater than the earlier superheavy atoms produced. This is an indication that one may be approaching the hypothesized "island of stability". I have not seen the lifetime published yet, but was told that it was "of order milliseconds".
Two answers: 1) these results are unclassified. Classified calculations may be done for other materials and conditions which will be more directly applicable to weapons questions. And more importantly, 2) the goal of the calculations is developing an a priori understanding of the behavior of high-Z (many electrons) materials at extreme conditions. Calculating the response of a material like Mo allows comparison of the numerical result with actual experiments to see if the calculation is correctly reproducing the real physics. This may not represent exact conditions in a weapon, but it is getting the science right which is the first goal. When you establish that the science is right, you are in a position to do more things with confidence. Note also that Blue Gene/L is used for a wide array of scientific calculations, not just things of interest for weapons physics.
The HPC world is substantially different from either gaming or "normal" application programming. The strong draw of the cell is that it is a production core with characteristics that are important to High Performance Computing, particularly power dissipation per flop. While conventional applications target getting the most out of a processor, HPC applications center on scalability in number of processors. This means running the largest number of processors for a given power/cooling supply, and maintaining the lowest latency in interprocessor communication. The latter is closely related to the physical layout of the processor array, which is also dependent upon cooling strategy. Hand coding, or at least hand optimization of the code, is reasonable for these applications. The resulting improvement can make possible calculations that would otherwise not be accomplished. As the number of processors increases substantially, the leading issue shifts from local execution speed to load balancing. Load balancing requires at least an initial "hand code" for a given architecture in any event.
There are several application spaces for HPC. Some, like semantic network processing do not require double precision and can be mounted on cell processors as they stand. Those which are fundamentally based on massive differential equation solution would benefit from the double precision modification. The key point here is that the double precision pipeline unit is a modification, not a different core. It is likely that IBM can make such a change at a fraction of the cost of the original core development with benefits not only to the HPC community, but also to potential workstation use.
The bottom line is than one can be easily mislead trying to think of HPC architectures and programming from the familiar standpoint of game and web server development.
There may be a number of motivations, but it is not likely that money has been a large one...at least up to now. In the original LANL contract with UC, the maximum fee was $8.7M (about 0.4% of the LANL operating budget). Maximum because it could (and has been) reduced based on the DOE evaluation of performance relative to contract criteria. While that may sound like a lot of money, it is very small compared to the UC total research budget and certainly extremely small when viewed in the light of the controversy and difficulty of managing the Laboratory.
One of the things that the source selection board discovered in trying to solicit serious industrial interest was that the fee was too small. Even though they started with a much larger fee in the original solicitation, they had to double it in order to attract industrial participants. The fee for managing Los Alamos now stands at $79M, nine times the original fee. It is hoped that this increased fee will be recouped by improvements in efficiency.
While the University of California will be deeply involved in the new management of the Los Alamos National Laboratory, it is not strictly correct to call this a win for UC. As the DOE press release makes clear, the winner of the competition was a limited liability corporation comprised of UC, Bechtel, BWX Technologies and others. The difference is very significant in some areas. For example, LANL personnel will no longer be members of the UC staff and participants in their retirement system, but employees of the LLC. The DOE did not release details of the winning proposal yet. As they do, I believe it will become increasingly clear that there is much more to this change than just UC continuing to play the same role.
In operation, heads are supported on a very thin aerodynamic bearing film, indeed. There are still two conditions that influence the life and allowable speed of the drive. One is the fact that the heads must land at some time when the drive is off in most (although not all) designs. The other is that the aero film is much smaller than a typical particle. It is quite undesireable to have a bearing in which the slightest degree of contact or small particle immediately results in catastrophic failure. Designers of aerostatic and aerodynamic bearings of all types pay careful attention to the plain bearing properties of the materials that they use for this reason.
The article discusses spot size as resulting from the shorter wavelength. From the description in the graphic, a 50-75 nm wavelength is required. There is no available source in this range of wavelengths that does not occupy a whole lot of space, take a whole lot of power and cost a whole lot of money. This is not really ultraviolet; it is closer to soft x-ray. Some idea of the difficulty can be derived from observing what the lithography community has struggled with to get a 13 nm source for "extreme ultraviolet lithography". For that matter, the community has sharply reduced work at 157 nm (in favor of immersion 193) for lack of a workable material set for the optics. The wavelength that is apparently proposed here is quite a bit more energetic than 157 and probably nearly as difficult to produce and direct as the 13 nm. Virtually all materials absorb at these wavelengths. Moreover, the photon energy is well in excess of available semiconductor material bandgaps, implying to semiconductor laser source. Whatever may be true of the recording mechanism, there is no clear path to implementation of this kind of device.
I have been an author, reviewer and associate editor at one time or another for scientific journals. None of these three are renumerated activities. While I object to the high price of the journals, I do have a little more appreciation for the economics. It is true that the original articles are written and the review process is carried out without expense to the journal. However, the publisher does pay the cost of administering this process and also (in most cases) the cost of editing, setting the article for publication, printing and marketing. While I believe that substantial profit is still involved, it is true that there is considerable expense as well. This expense, coupled with the very small circulation of these journals is one of the reasons for the apparently high substription price. One just cannot think of archival journals with small circulations and no advertising income in the same terms as mass market magazines. It is also the case that some professional societies derive significant income from their publications. Having said this, I would still donate my time to review and edit for an open electronic journal. But such a journal would still have expenses, and it is unclear how they would be met.
You have to read the full text! While the early portions of the treaty do, indeed, enforce compliance with established international treaties, the latter portions add additional requirements. Specifically included are the life-plus-seventy copyright life and the provisions of the US DMCA. Take a look at article 17.4, item 7 which contains a detailed enumeration of requirements with regard to technical protection of copyright. This goes far beyond the current state of international agreements.
There is, indeed, a similarity between DRAM and the structure postulated by the authors. In both cases, a part of the structure stores the information and that storage would be continuous if it were not for leakage. The presence of loss in the device limites the storage time and leads to the need for refresh. Since the paper only describes a theoretical structure, it is probably a little early to be able to make any statements about the actual loss and resulting storage time for such a device.
Yes, I believe the intention is to allow such a structure to be a storage device. However, one should probably think of it more as information storage than energy storage. The entire light pulse must fit physically in the device in order to achieve the results when the refractive index is modulated. That implies that only very short pulses can be stored, since the pulse speed before modulation is a significant fraction of c.
Yes, the concept (it is only a theoretical concept, not a chip, in the paper) does store the light. When the optical pulse is completely within the postulated structure (meaning only a very short pulse can be stored), a modulation of the refractive index causes the fields associated with the pulse to be stored in the internal cavities of the crystal. Reversing the refractive index change causes the stored fields to reform a traveling wave, which exits the structure. The way that you know that the pulse has been stored in the computer simulations is that after the first refractive index change, nothing comes out of the structure. After the second change, a pulse emerges that has the same shape as the one that was sent in.