Various German universities also offer study programs in English with the possibility of learning German on the side. Arguably the most systematic and broadest choice of options is at Jacobs University (http://m.jacobs-university.de/). (Disclaimer: I teach there.) These programs are typically not free (ours are not), but they still provide very good value as compared to out-of-state tuition or tuition at private universities in the US.
It should be mentioned that American students tend to struggle more than others, especially in the theoretical science and engineering disciplines, even though enthusiasm and motivation is high. Still, for students prepared to work hard, a very good alternative to some of the top-ranked US universities. And studying in Europe will provide a perspective on the world unlike any one would get at home...
If Google gets this onto enough devices, then little by little they will dictate the de-facto standard for OOXML, i.e., how it's "really" supposed to work and render beyond the ECMA paperwork, and Microsoft will have to make sure that their programs will be Google compatible. Google might just have enough market share to fully commoditize word processors and spread sheets, something that OO/LO started but did not get beyond the tipping point.
I second the previous poster. There are strange over-generalizations and omissions in Moore's statements that make me suspect that money rather than conviction is driving him.
Here a list, maybe other Slashdotters have more current information on these points (it's basically what I remember from observing this issue from the distance over more than 20 years or so).
No mention of geothermal energy, although it apparently has very significant potential (I have seen estimates in the range of 100% of the current global electricity consumption).
No discussion of the potential to intelligently schedule energy usage to best match the predicted supply curve from wind/solar and other non-constant sources.
The impact of the Chernobyl desaster on millions of people and the economy of many countries is inadequately described by mentioning just the 56 people whos "deaths could be directly attributed to the accident."
One of the big problems with reprocessing spent nuclear fuel is that the volume of nuclear waste is increased sharply (I remember reading the factor of 10 some time ago in connection with current plants in England and France). The specific radioactivity is obviously decreased in this process, but the waste still needs to be stored safely and is contaminated with all sorts of aggresive chemicals used in reprocessing that make handling tricky. I am not saying that this cannot be done, but stating that "it will be possible to use that energy and to greatly reduce the amount of waste that needs treatment and disposal" sounds more like an industry advert.
There are a bunch of alternative reactor designs which have the following properties: No chain reaction possible even when all control is lost, use of low grade uranium an/or thorium (the latter being more ubiquitous and less toxic than uranium) so that the fuel is not weapon grade, relatively cold nuclear waste which can be be permanently locked away after a short cool-down period and doesn't give economic incentive for reprosessing (much less of a mess, and no plutonium cycle with its unfortunate non-peaceful dual use). Some designs have prototype implementations (one example is the Thorium High Temperature Reactor design; a commercial version was built in Germany, but quickly dismantled because of engineering screw-ups - the thing kept breaking down and was leaking small amounts of radioactivity - local opposition and apparently uncertain economic prospects due to its breaking down all the time, but the basic design was, as far as I know, validated). A lot of this is not well tested, none without drawbacks (an uncontrolled THTR accident cannot melt the core, but would apparently cause the concrete shield to chemically decompose on a time scale of weeks or so, so there is no magical universal safety). I believe that anybody seriously discussing "green" nuclear energy with a long-term perspective must take such developments into consideration, in particular because of the obvious advantage for nonproliferation (e.g., if Iran had currently access to well-developed economically sound THTR technology, there would be absolutely no ambiguity with regards to their nuclear ambition).
The denial of vulnerability to terrorist attack is very naive. Even if the reactor core was safe from a direct hit (which is not the case with many currently existing reactors, but this is not the point here), there are plenty of vulnerable points in the supply and waste processing chain that could wreck at the very least economic havoc in densly populated areas (and most regions on this planet are, or will be shortly, in this category).
That said, I personally believe that there is a good case for nuclear power as a long term energy source, but let's not get fooled by the global warming scares from an ex-Greenpeacer-turned-industry-consultant into accepting short term profit-maximizing all-out nuclear solutions.
The professor tells the students: "It's a good idea to check your paper for plagiarism. Even careful scholars sometimes go overboard with their sources, and an automated check is extremely effective and quick. I recommend using favorite-cheatchecker, but of course the choice is yours. I will routinely cheat-check the papers that you submit, and expect that they are free of even minor infringements. Any infringement will be treated to the fullest extent consistent with the code of academic integrity, with a minimum penalty of something-very-nasty."
Collecting tariffs on software imports would be the most stupid thing to do. Billions of dollars each year flow into the US economy just because of the global Microsoft Tax. This only works for as long as other countries continue to play nicely, and don't get funny ideas like trying to get their cut of Microsoft's tax by themselves levying tax on software imports (and imports of other "intellectual" property like movies that have zero material value but nonethelesse generate huge flows of cash into American pockets).
In short, the US is arguably the worlds greatest profiteer from trading software, and would be damn stupid to give this up to just "protect" the lifelyhoods of a bunch of hackers. Nobody's going to shake up a system that's probably the only reason for keeping the US trade deficit from causing instant economic implosion.
In fact, the algorithm as a computational method goes back to Jacobi 1804-1851, and is essentially an iterative solver for large systems of linear equations. <p> Of course, it's still a significant contribution to see the application of the Jacobi method to ranking web pages, and I assume that they have done some clever and many more dirty tricks to get more realistic results, weed out duplicate pages, etc., which may or may not be part or the patent. <p> In any case, the basic page rank algorithm is quite intuitive to anyone who has worked with iterative numerical methods, and in fact a very nice illustration of the power of such methods.
To judge whether proprietary software is good or bad, I think it helps to look beyond the big bullies like Microsoft, who few people would dispute is more of a globally successful extortion scheme than a legitimate business, and to consider some smaller markets where proprietary software is used. So I'll offer my view on some example I am familiar with.
Although I try to use free software whenever I can, there is one product that I am deeply locked into (for almost a decade now), and this is the Mathematica computer algebra system. Basically, I don't think one can argue against the proprietary nature of this program on pragmatic grounds, the usual arguments of why "Open Source" is better just don't seem to apply (apart from pains with the current licensing policies, see below). So I think that the moral dimension of proprietary software cannot be avoided here, and everybody may judge for themselves. Here are the facts:
Mathematica has originally been conceived by a single person (Steven Wolfram) who left academics to set up a company which then developed the product to market.
The company, Wolfram Research, does little else than further develop and support the product.
Mathematica is a product of high quality, this is true for the underlying design as well as for the implementations. Of course there are always some bugs, but for a product of this complexity it has surprisingly few.
The release schedules are reasonable: most updates take several years and compatibility between versions has always been very good (I have seen quite a few major version changes). The company does not seem to confuse regular users with beta-testers - new major releases tend to be stable.
There is one major competitor, Maple, which is also a proprietary spin-off from academics (the University of Waterloo in this case). So the market place appears reasonably healthy. At the same time, there is no free alternative that comes even close to what either of these programs offers.
Needless to say, after using a product of this type more or less happily for many years, you'll have acquired a good collections of codes and expertise, so you would not want to change system without very good reason.
Now to the more ugly bits:
The price of the program is so out of range (about $1000 and up for a single user license) that even many academic departments do not want to pay for it. Currently I have to access it over a sometimes slow cross campus X connection from Central Computing.
Even the student price (around $150, I don't legally qualify) is too much to make purchase of Mathematica a reasonable class requirement. So Wolfram Research has effectively priced themselves out of the mass-purchase educational market (although still affordable to dedicated student), and one has no choice but to use the somewhat inferior, but cheaper and still proprietary Maple in-class. I doubt that this is a wise business decision in the long run, but this is something the company has to answer to themselves. I know even of departments who did high-profile pilot deployments of Mathematica to be currently backing out of it in undergraduate education because of the high licensing costs involved.
Mathematica has a rather effective copy protection scheme. In fact, due to the otherwise good quality of the software, the license server is by far the most significant single point of failure. You may blame it on the support staff, but even so it is a major pain to deal with and requires constant staff attention for absolutely no benefit. For example, major Linux kernel upgrades will make Mathematica think it has been copied onto a different machine and refuses to start up, sorting this out requires non-trivial paperwork with Wolfram.
Generally I feel uneasy about having centuries of scientific knowledge incorporated into an expert system, and then marketed as proprietary. Moreover, it raises questions about having a closed box in the loop of scientific discovery (part of the Mathematica code is actually itself written in the Mathematica language, and open for inspection), as it is increasingly used to derive of proof mathematical knowledge. It seems to me preposterous for small companies to occupy such a crucial link in a worldwide human and not-for-direct-profit human endeavor.
So while I continue to feel uneasy about this, I will also continue to use Mathematica for the foreseeable future, and dream that academic institutions (who probably pay most of Wolfram Research's revenues anyway) could pool enough money to buy out the company can convert it into an international center for Computational Mathematics.
I've heard people scream bloody murder at me for years for simply trying to sell various little odds and ends i've made, rather than just declare it public domain and give it out for free.
This is almost certainly not due to people's intention to deny you compensation for what you have done. Either they did not think your contribution was big enough to warrant any non-trivial charge, or, more likely, if what your program does is important to them, they didn't trust you enough to invest the time to learn using your product, and to lock in their data formats to your product, because they felt that you could not guarantee sufficient maintenance to meet their potentially evolving needs.
I believe that a small developer has for precisely this reason only a good chance to actually market a nontrivial product if it comes with a free license, i.e. a license that would enable the user to maintain, or have someone maintain the product for him should the need arise.
An unrelated comment: what seems to be often forgotten in the discussion is the following: FSF's rhetoric does not go against programmers working on proprietary code for a company that wants the code to be secret because they think that this will give their business a competitive advantage. Here the user of the software has a perfectly acceptable interest in non-distribution of the software, and is willing to pay the programmer for such a proprietary solution.
My guess is that most programmers acutally generate their income on work of this type, and relatively few work on software that is indented to be mass-marketed. The companies that mass-market are just a lot more visible, because they have lots of their man-power in public relations, accounting and market research, and rather less in actually developing their core products.
Although Python is elegant, it was not designed as a beginers language. Almost no languages have been designed in this way, with Basic (and Pascal) being the ones I can think of. But I would not advocate the use of these languages.
When I was at high school in Germany during the eighties, we were using a language called ELAN, which was specifically designed as a teaching language. ELAN is probably the most human readable computer language I have seen to this day. It was designed to encourage structured and abstract programming techniques.
Yet, the language is very powerful and easily exstensible. In fact, it was used to implement all higher level features of an operating system called EUMEL which was running in the early eighties as a true multi-user multi-tasking operating system on a machine with a Z80 processor, 256KB of RAM, 10MB of hard-disk (of which 4 were addressable through the operating system) and 6(!) ASCII terminals.
It seems that interest in this language has almost ceased, but here is still some web presence (follow the links for a detailed description of the language) at
http://os.inf.tu-dresden.de/L4/l3elan.html
I would be happy if the ELAN compiler was ported to Linux (yes, it is a compiled language, but it has also a default "compile and run" feature which produces almost instantaneous output for small programs). IMO it would make a superior teaching language, and is also quite capable as an implementation language for serious work provided one does not need direct access to low level system features.
But what's the benefit for science and humanity?
on
Cassini visits Earth
·
· Score: 1
Most of the people who have posted comments so far seem to be missing the point what the opposition to the Cassini probe is really about.
First of all, it is true that the worst case scenario that is painted by even the most fervent opponents of the mission is rather mild compared to other dangers that are facing us all the time. Still, I wonder why people feel the need to ridiculously downplay the dangers of plutonium. Who cares if it is really ``the most deadly substance'' - the stuff is nasty enough to cause a big headache for any organization charged with handling it responsibly.
However, the real issue is what are the benefits for us (as human beings in general) that justify putting money in the billion dollar range? Since very few people have a true interest in the advance of astronomy as a basic science, the real reason for putting so much effort in these missions are the technological spin-offs that may result. If you see it from this point of view, the Cassini mission becomes really scary!
One ``dream'' application is the further militarization of space. Plutonium battery powered spacecraft could be smaller, less vulnerable, and less restricted in their orbits than solar or conventional fuel powered ones. Should anyone want this? - Definitely not!
The development of civilian plutonium-based space missions helps to support a nuclear industry based on the plutonium cycle. The only people who really benefit from the plutonium cycle is the military (and I don't really think of the United States so much as they have enough nuclear bombs to blow us all up anyway). For civilian use it is a complete folly (economically as well as environmentally: The by far the worst radioactive polluters are reprocessing plants (plutonium extraction!), not from nuclear power stations).
In short, it leads us away from responsible use of nuclear power, which I believe is possible.
One the other hand, the rejection of a solar power source for Cassini meant what could have been a major push for solar power technology did not happen.
I am not making this statement as a blind advocate of solar power. But I do notice that a choice was made to spend a lot of money, officially on a basic science mission, in such a way that indirect benefits go to the military, and not to sustainable civilian uses.
Slashdot Mirror
Various German universities also offer study programs in English with the possibility of learning German on the side. Arguably the most systematic and broadest choice of options is at Jacobs University (http://m.jacobs-university.de/). (Disclaimer: I teach there.) These programs are typically not free (ours are not), but they still provide very good value as compared to out-of-state tuition or tuition at private universities in the US.
It should be mentioned that American students tend to struggle more than others, especially in the theoretical science and engineering disciplines, even though enthusiasm and motivation is high. Still, for students prepared to work hard, a very good alternative to some of the top-ranked US universities. And studying in Europe will provide a perspective on the world unlike any one would get at home...
If Google gets this onto enough devices, then little by little they will dictate the de-facto standard for OOXML, i.e., how it's "really" supposed to work and render beyond the ECMA paperwork, and Microsoft will have to make sure that their programs will be Google compatible. Google might just have enough market share to fully commoditize word processors and spread sheets, something that OO/LO started but did not get beyond the tipping point.
Here a list, maybe other Slashdotters have more current information on these points (it's basically what I remember from observing this issue from the distance over more than 20 years or so).
That said, I personally believe that there is a good case for nuclear power as a long term energy source, but let's not get fooled by the global warming scares from an ex-Greenpeacer-turned-industry-consultant into accepting short term profit-maximizing all-out nuclear solutions.
There is a very simple solution to all this.
The professor tells the students: "It's a good idea to check your paper for plagiarism. Even careful scholars sometimes go overboard with their sources, and an automated check is extremely effective and quick. I recommend using favorite-cheatchecker, but of course the choice is yours.
I will routinely cheat-check the papers that you submit, and expect that they are free of even minor infringements. Any infringement will be treated to the fullest extent consistent with the code of academic integrity, with a minimum penalty of something-very-nasty."
You get the idea...
Collecting tariffs on software imports would be the most stupid thing to do. Billions of dollars each year flow into the US economy just because of the global Microsoft Tax. This only works for as long as other countries continue to play nicely, and don't get funny ideas like trying to get their cut of Microsoft's tax by themselves levying tax on software imports (and imports of other "intellectual" property like movies that have zero material value but nonethelesse generate huge flows of cash into American pockets).
In short, the US is arguably the worlds greatest profiteer from trading software, and would be damn stupid to give this up to just "protect" the lifelyhoods of a bunch of hackers. Nobody's going to shake up a system that's probably the only reason for keeping the US trade deficit from causing instant economic implosion.
In fact, the algorithm as a computational method goes back to Jacobi 1804-1851, and is essentially an iterative solver for large systems of linear equations.
<p>
Of course, it's still a significant contribution to see the application of the Jacobi method to ranking web pages, and I assume that they have done some clever and many more dirty tricks to get more realistic results, weed out duplicate pages, etc., which may or may not be part or the patent.
<p>
In any case, the basic page rank algorithm is quite intuitive to anyone who has worked with iterative numerical methods, and in fact a very nice illustration of the power of such methods.
Although I try to use free software whenever I can, there is one product that I am deeply locked into (for almost a decade now), and this is the Mathematica computer algebra system. Basically, I don't think one can argue against the proprietary nature of this program on pragmatic grounds, the usual arguments of why "Open Source" is better just don't seem to apply (apart from pains with the current licensing policies, see below). So I think that the moral dimension of proprietary software cannot be avoided here, and everybody may judge for themselves. Here are the facts:
- Mathematica has originally been conceived by a single person (Steven Wolfram) who left academics to set up a company which then developed the product to market.
- The company, Wolfram Research, does little else than further develop and support the product.
- Mathematica is a product of high quality, this is true for the underlying design as well as for the implementations. Of course there are always some bugs, but for a product of this complexity it has surprisingly few.
- The release schedules are reasonable: most updates take several years and compatibility between versions has always been very good (I have seen quite a few major version changes). The company does not seem to confuse regular users with beta-testers - new major releases tend to be stable.
- There is one major competitor, Maple, which is also a proprietary spin-off from academics (the University of Waterloo in this case). So the market place appears reasonably healthy. At the same time, there is no free alternative that comes even close to what either of these programs offers.
- Needless to say, after using a product of this type more or less happily for many years, you'll have acquired a good collections of codes and expertise, so you would not want to change system without very good reason.
Now to the more ugly bits:- The price of the program is so out of range (about $1000 and up for a single user license) that even many academic departments do not want to pay for it. Currently I have to access it over a sometimes slow cross campus X connection from Central Computing.
- Even the student price (around $150, I don't legally qualify) is too much to make purchase of Mathematica a reasonable class requirement. So Wolfram Research has effectively priced themselves out of the mass-purchase educational market (although still affordable to dedicated student), and one has no choice but to use the somewhat inferior, but cheaper and still proprietary Maple in-class. I doubt that this is a wise business decision in the long run, but this is something the company has to answer to themselves. I know even of departments who did high-profile pilot deployments of Mathematica to be currently backing out of it in undergraduate education because of the high licensing costs involved.
- Mathematica has a rather effective copy protection scheme. In fact, due to the otherwise good quality of the software, the license server is by far the most significant single point of failure. You may blame it on the support staff, but even so it is a major pain to deal with and requires constant staff attention for absolutely no benefit. For example, major Linux kernel upgrades will make Mathematica think it has been copied onto a different machine and refuses to start up, sorting this out requires non-trivial paperwork with Wolfram.
- Generally I feel uneasy about having centuries of scientific knowledge incorporated into an expert system, and then marketed as proprietary. Moreover, it raises questions about having a closed box in the loop of scientific discovery (part of the Mathematica code is actually itself written in the Mathematica language, and open for inspection), as it is increasingly used to derive of proof mathematical knowledge. It seems to me preposterous for small companies to occupy such a crucial link in a worldwide human and not-for-direct-profit human endeavor.
So while I continue to feel uneasy about this, I will also continue to use Mathematica for the foreseeable future, and dream that academic institutions (who probably pay most of Wolfram Research's revenues anyway) could pool enough money to buy out the company can convert it into an international center for Computational Mathematics.This is almost certainly not due to people's intention to deny you compensation for what you have done. Either they did not think your contribution was big enough to warrant any non-trivial charge, or, more likely, if what your program does is important to them, they didn't trust you enough to invest the time to learn using your product, and to lock in their data formats to your product, because they felt that you could not guarantee sufficient maintenance to meet their potentially evolving needs.
I believe that a small developer has for precisely this reason only a good chance to actually market a nontrivial product if it comes with a free license, i.e. a license that would enable the user to maintain, or have someone maintain the product for him should the need arise.
An unrelated comment: what seems to be often forgotten in the discussion is the following: FSF's rhetoric does not go against programmers working on proprietary code for a company that wants the code to be secret because they think that this will give their business a competitive advantage. Here the user of the software has a perfectly acceptable interest in non-distribution of the software, and is willing to pay the programmer for such a proprietary solution.
My guess is that most programmers acutally generate their income on work of this type, and relatively few work on software that is indented to be mass-marketed. The companies that mass-market are just a lot more visible, because they have lots of their man-power in public relations, accounting and market research, and rather less in actually developing their core products.
When I was at high school in Germany during the eighties, we were using a language called ELAN, which was specifically designed as a teaching language. ELAN is probably the most human readable computer language I have seen to this day. It was designed to encourage structured and abstract programming techniques.
Yet, the language is very powerful and easily exstensible. In fact, it was used to implement all higher level features of an operating system called EUMEL which was running in the early eighties as a true multi-user multi-tasking operating system on a machine with a Z80 processor, 256KB of RAM, 10MB of hard-disk (of which 4 were addressable through the operating system) and 6(!) ASCII terminals.
It seems that interest in this language has almost ceased, but here is still some web presence (follow the links for a detailed description of the language) at
http://os.inf.tu-dresden.de/L4/l3elan.html
I would be happy if the ELAN compiler was ported to Linux (yes, it is a compiled language, but it has also a default "compile and run" feature which produces almost instantaneous output for small programs). IMO it would make a superior teaching language, and is also quite capable as an implementation language for serious work provided one does not need direct access to low level system features.
First of all, it is true that the worst case scenario that is painted by even the most fervent opponents of the mission is rather mild compared to other dangers that are facing us all the time. Still, I wonder why people feel the need to ridiculously downplay the dangers of plutonium. Who cares if it is really ``the most deadly substance'' - the stuff is nasty enough to cause a big headache for any organization charged with handling it responsibly.
However, the real issue is what are the benefits for us (as human beings in general) that justify putting money in the billion dollar range? Since very few people have a true interest in the advance of astronomy as a basic science, the real reason for putting so much effort in these missions are the technological spin-offs that may result. If you see it from this point of view, the Cassini mission becomes really scary!
- One ``dream'' application is the further militarization of space. Plutonium battery powered spacecraft could be smaller, less vulnerable, and less restricted in their orbits than solar or conventional fuel powered ones. Should anyone want this? - Definitely not!
- The development of civilian plutonium-based space missions helps to support a nuclear industry based on the plutonium cycle. The only people who really benefit from the plutonium cycle is the military (and I don't really think of the United States so much as they have enough nuclear bombs to blow us all up anyway). For civilian use it is a complete folly (economically as well as environmentally: The by far the worst radioactive polluters are reprocessing plants (plutonium extraction!), not from nuclear power stations).
- One the other hand, the rejection of a solar power source for Cassini meant what could have been a major push for solar power technology did not happen.
I am not making this statement as a blind advocate of solar power. But I do notice that a choice was made to spend a lot of money, officially on a basic science mission, in such a way that indirect benefits go to the military, and not to sustainable civilian uses.In short, it leads us away from responsible use of nuclear power, which I believe is possible.
Marcel (oliver@member.ams.nospam.org)