Do you remember 1990? Everybody was doing Turbo Vision Programs using Borland Dos IDEs. Luckily at that time the funciton key assignments for DOS program where settled. Press F1, get help, press F10 and something else happes (i think it was calling the menu). AFAIR F2 was save and alt+key breought you directly in the menu. That was efficient, i can tell you. The beginning of GUIs was good. I likes OS2. clear concepcts and the programmers stuck up with the concepts. Then it go a little worse. All these "Visual Basic Morons" started to change the meaning of the ok or of the cancel button. They made seemingly unmodal guis, which where modal. Then another big thing happend: the confused, old, awkyard world of X11 on linux educated prgrammers to forget what they learned. Shaping buttons in the same way as all other buttons on the os was not longer considered a good programming practice.
Finally the web came. Everybody places buttons where he wants, they do what he thinks might be funny and the way in how to press them is different each time. Keyboard shortcuts are not available and a funciton to print does not exist (just print hte web page). Because JS and flash are used improperly, something which was executable on a 16MHz computer in 1990 now is slow on a 3GHz Pentium. You have the constant feeling that the programmer did not use the network but constantly worked against the shortcoming of packing everthing into AJAX (which most people seem not to understand).
Even the moset advanced webmail can not provide me with an user interface which is as fast and purpose-oriented as that of pine. Web Applications to me seem as if programmers would have forgotten how to program a user interface. Furthermore i can decide to read e-mail via ssh (if necessary ans ssh java web applet) and then the remote access looks exactly like the local one, and doen not just try to resemble it (and fail poorly in doing so!). In combination with a "screen session" this thing can follow you around the world exactly in the state in which you left it (I had a opened pine-session in a screen, which followed me for three weeks and 20000km). Can anybody tell me an webmail which has the same cool "remote config" feature as pine? Doe any webmail have an editor as advanced as emacs or as useful as vi, or do i really only get a stupid Test edit field in an input form where some incapable programmer decide to bind som mouse button event to something unresonable? Can I decide when to update or will my stupid provider force "new features" and "new website designs" onto me when i just got to 40% pine productivity?
No, my opinion is: Web application are good for simple things, which you do sometimes. Something like e-mail which i am doing nearly each day several times is easier to handle with a stable, locally install application completely under my control, which one should not change more often than every five years.
Sometime, too much sharing of responsibility is not good. You can save the money for the third and fourth review - at least if you tell them that they are the third and fourth. Also alwas check any design first by your pocket calculator (or emacs calc, or python command line if you wish) before inputting it to the fancy simulation. Having a paper with five lines (or 10) it is easiert to figure out which questions need to be asked.
1) Theo de Raadt behaved like a troll in this discussion. No really. If you read the thread, he is the one heating the discussion up. I would say that Rueschs original mail was typical german style. That does not make it better, since he wrote it in English; nevertheless de Raadt did his best to provoke him further. I can not understand how this can happen. The first mail was four days ago, and only two days later, the BSD driver maintainer dropped the development of the driver. Sorry guys, if i take this behaviour as a measure for you professionality i am not sure that your excellent programming skills outweight it.
a different discussion style would have been:
a) Ruesch send an e-mail, not an list-post b) The Maintainer removes the code puts the log message to the list that there are copyright issues with Ruesch - NOT that he removes the code because Ruesch attacked him. (The latter means that he would be fine with this behaviour). This would have also been the appropriate reaction to the first message, even if it was on a list c) De Raadt comments shortly that he will discuss this issue with both and agrees with the driver being removed from the repository - NOT that Ruesch is the bad guy because his code was stolen. His task would have been to cool the situation, not to drive Ruesch into arguments really making a further dialog impossible.
2) The BSD drivers author states clearly that it was intentional. "To make quick progess" in nice, but honestly, not an excuse. Also if you want to do so, do it in you own sandbox or your private repository. Even that would in a very strict interpretation violate the cleanroom implementation rules. However it would be enough not to raise such problems.
3) Theo de Raadts stream of though "we are all open source developers - thanks for slowing us down" which i see as a connotation to some of his mail is really stupid. It is HIS personal opinion that the BSD lisense is as good as the GPL. He may think whatever he likes. If i own something it is after all my decision how to use it. You may call me egoistic if i use it for personal gain. You may call me stupid if i dont make profit of it. But nobody is a bad guy because he insists that he put something under the GPL.
IMHO if they bought the code they may do whatever they want. There is no obligation for them to open the code. Fuethermore I doubt that they don't use ist. I could imagine that some of the beos ideas are very usefule in the resource-limite devices Access Inc is usually providing Software for!
> In any case, the paper in question was also published in Physical Review Letters [aip.org], Mar 9. So even by your definition...
Ok I'll consider reading it now, when i have time.
Excuse me, it is somethin ehich is called commonly a "preprint server". That defines it to be a preprint. If the Author wants to publish it, he can use his own website. This is called self-publishing and also works if you want yourself to be the editor of a book you wrote.
A few generations from now, society will be split into priests of high-tech and the masses, whose limited conception of science makes them beleive that a mobile phone of that time is magic. The priests will rule the world again and will send the masses to senseless crusades just to reduce their number to a level which is well-controllable.
If the plug can provide only a single one of the voltages it does not mean that devices will run on only one voltage. My prediction is that the device designers will quickly welcome the possibility of removing an transformer ans replacing it by an cheaper DC-DC converter, which can run on any of the three voltages.
This method is also the most efficient, sonce nobody on the host-side can guess what voltage is needed, it is best to take the one which is available easily in your design.
now, please everybody repeat after me: A link to thre preprint archive is not a link to a published paper, because a preprint is a preprint and a paper is a paper (even if it is pdf nowadays....).
I find this most annoying in the lab, when i have an oscilloscope, where i can not turn on/of measurements, switch to an advanced trigger mode or zoom into the screen without a mouse.
What what the typical application. Was it well writen or not. what was the reason for the slowdown (Memory, Network, ?). Without that 43% is just a munber as good as 42%, the answer to all performance loss questions
I had a Omnibook XE3 with an additional (500Euro!) Service contract for three years. As in so many Omnibook XE3 (according to google...), when carried on a bike in a (laptop) backpack regularly (this is speculation), the keyboard failed after two years. I call there, am redirected several times and talk to an engineer. I say that i can wait for some time, but the repair should be sone quickly, so he enters the case and we agree that the notebook is picked up after I call again. Two days later i find a notification in my postbox saying "you where not there when we tried to pick up your package". I ask if the keyboard is actually a fixed version or a new version, support says: no, it"s the same. So I ask: then it will break in two years again. Response: but the three years warranty will be over then. me: I have the feeling that this is a serial mistake and need to be corrected. He promises to talk to technicians. Nothing happens for three weeks. I call again. I am being told: your case was closed. I ignore that the problem is a serial mistake, because i already at that point made the decision to buy nothing from this company for ten years and wanted to have a working keybiard again. After being redirected several times I explain the situation. The support offered me (by himself) to send the keyboard to me and I replace it. I ask "wont this void the warranty?" answer: no, it is a "customer replacable part". When the keyboard arrives i follow the instruction and try to replace it (on an ESD protected workingplace...). At one point i find the instruction not clear and call. The supportperson: What you are trying to replace your keyboard? That voids the warranty.
Similar frustrating episodes happen to me when buying an HP scanner..... it was impossible to get the scanner driver for an englich windows in germany on CD, and i was explained, after pointing out that the website does not contain this driver, that "the support can not influence the websites contents and does not know whom to talk to". (actually they managed after three weeks to send me the wrong driver one more time on CD and support gave me the advice to unpack the package for the Mac, because it would contain the right driver for Windows; it did not).
in an simple electronics design of mine a certain circuit (SN74128) appears. This cercuit does nothing special, just an NOR gate. Actually I do not use the NOR gate as a NOR gate, i am only interested in the handy 50Ohm driver in the output to convert a logic signal to a 50Ohm line. Sound pretty standard, eh?
That was a mistake. I assume that this ceruit is available easily and everywhere, just because the supplier i previously had nearly everything. When buying my part locally i figure out that only one of the four local shops has this component, even if they all have the complete 74Series of circuits.
I though about it, and the more i thought about it i figured out that this chip is not standard at all: 1) Amateurs usually don't drive logic signals over 50 Ohm lines 2) Professional electronic engineers would probably use another driver of use anyway an higher integrated component
So I got my lesson, that probaly physicists (i am one), semi-professional in EE, who just want to build a simple circuit driving an Coaxial cable to the timing input of a device a few meters away are indeed nearly the only users of this circuit.....
It is an irony of fate that this episode happend to me just after writing the last message.
A generalized vie of it would be not to rely in an design which is non-mainstream on the availability of something which doe not get a significant market share from what you are doing.
Ok i now need a faster processor, but this is definitely cool. Why? Because it means that, as long as there is Hardware executing Java around, Dos programs can be used. Could make the transition and archiving of existing data easier.
> Yes; Understand too, how the gates work, all the details of how the chips work, (the myriad chips your program may run on,) and have
If you understand what the gate does, there is no semantical level to break it down any more. Howver, be aware that it consumes Energy.
> a good understanding of quantum mechanics, as well.
That is not so important.
> Understand all the business supply chains, as well.
Yipp, this indeed is important. Understanding why sometimes a technologically favourable solution is not the best makes you a good engineer. (E.g.: Instead of asking "Could we not transport four video streams more over our ethernet if we use an adapted divx coding with increased buffer size?", it may be better to ask: We need a camera, well tested, avaialble in thousands, weatherproof, from an reliable supplier with a standard protocol, because handling a problem in the firmware of a device which is distributed in hundreds over this large industrial facility will consume time and money.)
> Anything else, is just-- pshaw... Black boxing it.
Besides the number unaccounted for mentioned already by other, I can only say about ecology: Obviously, if you are going 100 Miles per day on a Highway, buy no hybrid car. By an light, if possible used car. If you are going to be for 1 hour per day in the traffic jam of your city, buy a hybrid (or even an electric only...).
If you are concerned about material being transported around during production, demand for duties or - the easies solution: make energy more expensive. The only reason why the normal automotive manufacuteres have more effective production chains is that they are established longer and produce a larger amount of vehicles. Otherwise they are very happy in buying part from all over the world.... I am sure that in the moment when a significant part of the nickel produced in the mine (which is NOT equivalent to the mine delivering ALL the nickel for the Prius) will be used for the Prius, a short production chain will be established.
If your nickel production facilities suck because of poor enviromental laws, request better laws and let them pay the damage.
If you just make things (engery consumption, SO2 emmission, special waste) mor expensive buy taking taxes the market will settle the problem by itself.
Why do BSoD's indicate that it is not ready? Bill Gates has shown BSoD when presenting an new OS, or is this an urban legend? It seem that they are somewhere around Windows 2000 right know.... I remember BSoD during installing W2K, so a fully compatilbe OS should show BSoDs during installing. Other companies sell proiducts in such a state....
We should have thought about that back in the early ninties. Some people were aware of it at that time; i remember that becaus i did some literature research (in 1992) for writing an article in our schools pupils magazine about genetic technology. Even at that time i found it obvious that the primary danger is not our ability to modify the gonom but to read it out. I been scared since that time about the careless discuccion in the public. It is true, we have to develop an ethical code on that. For example, insurances should not be allowed to make mandatory genetical tests; however perosns who did genetical tests should share the results with the insurances. That's very simple. If I know that I am going to have cancer and I am making an insurance, it is no risk any more. On the other hand, one should discuss that the increased "genetically inherited" risk is taken over by the state. So I think insurances should mandatory offer an plan for anybody and openly caclulate the fraction of that plan attributed to genetic disposition, which i suggest should be compensated for from either tax money or another (mandatory) insurance (macroeconolically both possibilitis are different tastes of the same thing!).
In that way, everybody could choose: * not know your risk and be insured at an fair rate * know your risk and be insured at either a higher or lower rate, in the first case the state would pay the difference
The insurances * Could use their conventional mathematics a the same fractions of uncertainity * Could try to lower the overall costs by taking preventive measures.
I am 32 now. I believe that i had the great luck of beeing born at the time - maybe a few years to late to observe the rise of microprocessors. When I was 7 years i started with simple electronics. Years from posessing a computer i read a small article about how a microprocessor works. I did not get it fully, but what i understood was not universes away from the real thing at that time. At that time i got a computer (12 years) you would at home typically buy a Commodore 64 or 128. I got one Commodore 128. This is a funny machine. One of the best features is the so called "user port". You have eith IO lines which just ouput a digital signal, freely programmable. I think that i experienced no similar thrill ever in my life than when constructin a cheap, stupid light sensor and measuring with it the 50Hz noise of the lamps in my room (Now i am a physiscist and am good at constructing data acquisition systems).
Nowadays, people have a camera built into their laptops, mobile phones, computers. Even the cheapes webcam hasa data tranfer rate 100000 higher than this light sensor. Howver, even with the most expensive webcams you may not reproduce this measurement. And that is the core of the problem. It is, without buing a microcontroller or additional hardware impossible to "just play". Instead of having LDA #1;STA $IOADDR;LDA #0;STA $IOADDR; for making a single pulse on the user port, you probably have to click-click on the activeX control (a standard defined in a thick manual) which came with the proprietary kit. This is no fun. I had more fun when i played with lego than when i got a ready made thing.
So the problem, that once computers and it attracted people who built and constructed things; sometimes proofs of concepts - you stick twi wires into you computer at take the stupid diode on the top of your monitor. It is a natural process that things are taken over by people who are more like users. Actually I experienced this one or more times with proofs of concepts which I did when i worked i a group supporting the webmaster at our university. When i sat in 1998 in a system programming course in university i was astonished how uninterested the average IT first semester students where in how to use system calls in C. To point that out: I LOVE programming languages other than C. However it is good to know what will be expensive. A simple example: to make an access over the network is ok. If you however do something semantical which requires waiting for a response, you kill performance (Everybody having connected a LED and a light sensor to the user port of his C64 knows that the light sensor does never react BEFORE the LED is turned on. Depending how cheaply you built it it might react milliseconds after that).
As an example there are three approaches to the problem if you find out that you database is too slow:
1) Speed up your hardware (chosen by people who do not undertand and are not intersete in the fact that they dont)
2) Change to a DBMS which handles if better (chosen by the majority of "dilbertized" workers, who understand that they don not understand)
3) do it better and sell it to people of point 2). Bo that by taking a cup of tea, a pen and not talking for how long it takes. Do not write a report before you understood the problem.
Nowadays, even point 3) gets more dilbertized. This is a cultural effect which beliefs in ISO9000 processes to be the best tool for everything. If they are only the future can tell.
> What you said is so true. When "programmers" do research in order to find new and better ways of doing things, or solving problems > that have not been solved, it's "science." It may not by physics or chemistry, but it's still science. I would bet that most
No, it's not. Science defines itselft no by 'solving a problem' but by 'increasing everyones knowlegde'. This happens by the means of writing an article about wgat you did. Just writing a program is not enough.
> programmers are doing a little bit of science every day. Sometimes, it's a lot.
I have no doubt that in the next years we will need an huge amount of embedded Developers. put together the funny crappy libraries which had in mind that opening the Start Menu of windows at high resolution an 32 bit color depth transfers more data than the whol payload (e.g. text Document) of the whole Apllication has.
Terribly good for saving resources (mainly: Battery!)
Slashdot Mirror
Do you remember 1990? Everybody was doing Turbo Vision Programs using Borland Dos IDEs. Luckily at that time the funciton key assignments for DOS program where settled. Press F1, get help, press F10 and something else happes (i think it was calling the menu). AFAIR F2 was save and alt+key breought you directly in the menu. That was efficient, i can tell you. The beginning of GUIs was good. I likes OS2. clear concepcts and the programmers stuck up with the concepts. Then it go a little worse. All these "Visual Basic Morons" started to change the meaning of the ok or of the cancel button. They made seemingly unmodal guis, which where modal. Then another big thing happend: the confused, old, awkyard world of X11 on linux educated prgrammers to forget what they learned. Shaping buttons in the same way as all other buttons on the os was not longer considered a good programming practice.
Finally the web came. Everybody places buttons where he wants, they do what he thinks might be funny and the way in how to press them is different each time. Keyboard shortcuts are not available and a funciton to print does not exist (just print hte web page). Because JS and flash are used improperly, something which was executable on a 16MHz computer in 1990 now is slow on a 3GHz Pentium. You have the constant feeling that the programmer did not use the network but constantly worked against the shortcoming of packing everthing into AJAX (which most people seem not to understand).
Even the moset advanced webmail can not provide me with an user interface which is as fast and purpose-oriented as that of pine. Web Applications to me seem as if programmers would have forgotten how to program a user interface. Furthermore i can decide to read e-mail via ssh (if necessary ans ssh java web applet) and then the remote access looks exactly like the local one, and doen not just try to resemble it (and fail poorly in doing so!). In combination with a "screen session" this thing can follow you around the world exactly in the state in which you left it (I had a opened pine-session in a screen, which followed me for three weeks and 20000km). Can anybody tell me an webmail which has the same cool "remote config" feature as pine? Doe any webmail have an editor as advanced as emacs or as useful as vi, or do i really only get a stupid Test edit field in an input form where some incapable programmer decide to bind som mouse button event to something unresonable? Can I decide when to update or will my stupid provider force "new features" and "new website designs" onto me when i just got to 40% pine productivity?
No, my opinion is: Web application are good for simple things, which you do sometimes. Something like e-mail which i am doing nearly each day several times is easier to handle with a stable, locally install application completely under my control, which one should not change more often than every five years.
Sometime, too much sharing of responsibility is not good. You can save the money for the third and fourth review - at least if you tell them that they are the third and fourth. Also alwas check any design first by your pocket calculator (or emacs calc, or python command line if you wish) before inputting it to the fancy simulation. Having a paper with five lines (or 10) it is easiert to figure out which questions need to be asked.
1) Theo de Raadt behaved like a troll in this discussion. No really. If you read the thread, he is the one heating the discussion up. I would say that Rueschs original mail was typical german style. That does not make it better, since he wrote it in English; nevertheless de Raadt did his best to provoke him further. I can not understand how this can happen. The first mail was four days ago, and only two days later, the BSD driver maintainer dropped the development of the driver. Sorry guys, if i take this behaviour as a measure for you professionality i am not sure that your excellent programming skills outweight it.
a different discussion style would have been:
a) Ruesch send an e-mail, not an list-post
b) The Maintainer removes the code puts the log message to the list that there are copyright issues with Ruesch - NOT that he removes the code because Ruesch attacked him. (The latter means that he would be fine with this behaviour). This would have also been the appropriate reaction to the first message, even if it was on a list
c) De Raadt comments shortly that he will discuss this issue with both and agrees with the driver being removed from the repository - NOT that Ruesch is the bad guy because his code was stolen. His task would have been to cool the situation, not to drive Ruesch into arguments really making a further dialog impossible.
2) The BSD drivers author states clearly that it was intentional. "To make quick progess" in nice, but honestly, not an excuse. Also if you want to do so, do it in you own sandbox or your private repository. Even that would in a very strict interpretation violate the cleanroom implementation rules. However it would be enough not to raise such problems.
3) Theo de Raadts stream of though "we are all open source developers - thanks for slowing us down" which i see as a connotation to some of his mail is really stupid. It is HIS personal opinion that the BSD lisense is as good as the GPL. He may think whatever he likes. If i own something it is after all my decision how to use it. You may call me egoistic if i use it for personal gain. You may call me stupid if i dont make profit of it. But nobody is a bad guy because he insists that he put something under the GPL.
IMHO if they bought the code they may do whatever they want. There is no obligation for them to open the code. Fuethermore I doubt that they don't use ist. I could imagine that some of the beos ideas are very usefule in the resource-limite devices Access Inc is usually providing Software for!
> In any case, the paper in question was also published in Physical Review Letters [aip.org], Mar 9. So even by your definition... Ok I'll consider reading it now, when i have time.
Asfar as i see dvd ram is not discussed
Excuse me, it is somethin ehich is called commonly a "preprint server". That defines it to be a preprint. If the Author wants to publish it, he can use his own website. This is called self-publishing and also works if you want yourself to be the editor of a book you wrote.
A few generations from now, society will be split into priests of high-tech and the masses, whose limited conception of science makes them beleive that a mobile phone of that time is magic. The priests will rule the world again and will send the masses to senseless crusades just to reduce their number to a level which is well-controllable.
If the plug can provide only a single one of the voltages it does not mean that devices will run on only one voltage. My prediction is that the device designers will quickly welcome the possibility of removing an transformer ans replacing it by an cheaper DC-DC converter, which can run on any of the three voltages.
This method is also the most efficient, sonce nobody on the host-side can guess what voltage is needed, it is best to take the one which is available easily in your design.
now, please everybody repeat after me: A link to thre preprint archive is not a link to a published paper, because a preprint is a preprint and a paper is a paper (even if it is pdf nowadays....).
I find this most annoying in the lab, when i have an oscilloscope, where i can not turn on/of measurements, switch to an advanced trigger mode or zoom into the screen without a mouse.
R2D2 they call him....
What what the typical application. Was it well writen or not. what was the reason for the slowdown (Memory, Network, ?). Without that 43% is just a munber as good as 42%, the answer to all performance loss questions
I had a Omnibook XE3 with an additional (500Euro!) Service contract for three years. As in so many Omnibook XE3 (according to google...), when carried on a bike in a (laptop) backpack regularly (this is speculation), the keyboard failed after two years. I call there, am redirected several times and talk to an engineer. I say that i can wait for some time, but the repair should be sone quickly, so he enters the case and we agree that the notebook is picked up after I call again. Two days later i find a notification in my postbox saying "you where not there when we tried to pick up your package". I ask if the keyboard is actually a fixed version or a new version, support says: no, it"s the same. So I ask: then it will break in two years again. Response: but the three years warranty will be over then. me: I have the feeling that this is a serial mistake and need to be corrected. He promises to talk to technicians. Nothing happens for three weeks. I call again. I am being told: your case was closed. I ignore that the problem is a serial mistake, because i already at that point made the decision to buy nothing from this company for ten years and wanted to have a working keybiard again. After being redirected several times I explain the situation. The support offered me (by himself) to send the keyboard to me and I replace it. I ask "wont this void the warranty?" answer: no, it is a "customer replacable part". When the keyboard arrives i follow the instruction and try to replace it (on an ESD protected workingplace...). At one point i find the instruction not clear and call. The supportperson: What you are trying to replace your keyboard? That voids the warranty.
Similar frustrating episodes happen to me when buying an HP scanner..... it was impossible to get the scanner driver for an englich windows in germany on CD, and i was explained, after pointing out that the website does not contain this driver, that "the support can not influence the websites contents and does not know whom to talk to". (actually they managed after three weeks to send me the wrong driver one more time on CD and support gave me the advice to unpack the package for the Mac, because it would contain the right driver for Windows; it did not).
Actually i got a lesson in that yesterday.
in an simple electronics design of mine a certain circuit (SN74128) appears. This cercuit does nothing special, just an NOR gate. Actually I do not use the NOR gate as a NOR gate, i am only interested in the handy 50Ohm driver in the output to convert a logic signal to a 50Ohm line. Sound pretty standard, eh?
That was a mistake. I assume that this ceruit is available easily and everywhere, just because the supplier i previously had nearly everything. When buying my part locally i figure out that only one of the four local shops has this component, even if they all have the complete 74Series of circuits.
I though about it, and the more i thought about it i figured out that this chip is not standard at all:
1) Amateurs usually don't drive logic signals over 50 Ohm lines
2) Professional electronic engineers would probably use another driver of use anyway an higher integrated component
So I got my lesson, that probaly physicists (i am one), semi-professional in EE, who just want to build a simple circuit driving an Coaxial cable to the timing input of a device a few meters away are indeed nearly the only users of this circuit.....
It is an irony of fate that this episode happend to me just after writing the last message.
A generalized vie of it would be not to rely in an design which is non-mainstream on the availability of something which doe not get a significant market share from what you are doing.
Ok i now need a faster processor, but this is definitely cool. Why? Because it means that, as long as there is Hardware executing Java around, Dos programs can be used. Could make the transition and archiving of existing data easier.
> Yes; Understand too, how the gates work, all the details of how the chips work, (the myriad chips your program may run on,) and have
If you understand what the gate does, there is no semantical level to break it down any more. Howver, be aware that it consumes Energy.
> a good understanding of quantum mechanics, as well.
That is not so important.
> Understand all the business supply chains, as well.
Yipp, this indeed is important. Understanding why sometimes a technologically favourable solution is not the best makes you a good engineer. (E.g.: Instead of asking "Could we not transport four video streams more over our ethernet if we use an adapted divx coding with increased buffer size?", it may be better to ask: We need a camera, well tested, avaialble in thousands, weatherproof, from an reliable supplier with a standard protocol, because handling a problem in the firmware of a device which is distributed in hundreds over this large industrial facility will consume time and money.)
> Anything else, is just-- pshaw... Black boxing it.
Yepp. More than one Comapany died in such a way.
Besides the number unaccounted for mentioned already by other, I can only say about ecology: Obviously, if you are going 100 Miles per day on a Highway, buy no hybrid car. By an light, if possible used car. If you are going to be for 1 hour per day in the traffic jam of your city, buy a hybrid (or even an electric only...).
If you are concerned about material being transported around during production, demand for duties or - the easies solution: make energy more expensive. The only reason why the normal automotive manufacuteres have more effective production chains is that they are established longer and produce a larger amount of vehicles. Otherwise they are very happy in buying part from all over the world.... I am sure that in the moment when a significant part of the nickel produced in the mine (which is NOT equivalent to the mine delivering ALL the nickel for the Prius) will be used for the Prius, a short production chain will be established.
If your nickel production facilities suck because of poor enviromental laws, request better laws and let them pay the damage.
If you just make things (engery consumption, SO2 emmission, special waste) mor expensive buy taking taxes the market will settle the problem by itself.
Why do BSoD's indicate that it is not ready? Bill Gates has shown BSoD when presenting an new OS, or is this an urban legend? It seem that they are somewhere around Windows 2000 right know.... I remember BSoD during installing W2K, so a fully compatilbe OS should show BSoDs during installing. Other companies sell proiducts in such a state....
Thanks, it's because of people like you that insurances can't afford not to force people to tests.....
We should have thought about that back in the early ninties. Some people were aware of it at that time; i remember that becaus i did some literature research (in 1992) for writing an article in our schools pupils magazine about genetic technology. Even at that time i found it obvious that the primary danger is not our ability to modify the gonom but to read it out. I been scared since that time about the careless discuccion in the public. It is true, we have to develop an ethical code on that. For example, insurances should not be allowed to make mandatory genetical tests; however perosns who did genetical tests should share the results with the insurances. That's very simple. If I know that I am going to have cancer and I am making an insurance, it is no risk any more. On the other hand, one should discuss that the increased "genetically inherited" risk is taken over by the state. So I think insurances should mandatory offer an plan for anybody and openly caclulate the fraction of that plan attributed to genetic disposition, which i suggest should be compensated for from either tax money or another (mandatory) insurance (macroeconolically both possibilitis are different tastes of the same thing!).
In that way, everybody could choose:
* not know your risk and be insured at an fair rate
* know your risk and be insured at either a higher or lower rate, in the first case the state would pay the difference
The insurances
* Could use their conventional mathematics a the same fractions of uncertainity
* Could try to lower the overall costs by taking preventive measures.
I am 32 now. I believe that i had the great luck of beeing born at the time - maybe a few years to late to observe the rise of microprocessors. When I was 7 years i started with simple electronics. Years from posessing a computer i read a small article about how a microprocessor works. I did not get it fully, but what i understood was not universes away from the real thing at that time. At that time i got a computer (12 years) you would at home typically buy a Commodore 64 or 128. I got one Commodore 128. This is a funny machine. One of the best features is the so called "user port". You have eith IO lines which just ouput a digital signal, freely programmable. I think that i experienced no similar thrill ever in my life than when constructin a cheap, stupid light sensor and measuring with it the 50Hz noise of the lamps in my room (Now i am a physiscist and am good at constructing data acquisition systems).
Nowadays, people have a camera built into their laptops, mobile phones, computers. Even the cheapes webcam hasa data tranfer rate 100000 higher than this light sensor. Howver, even with the most expensive webcams you may not reproduce this measurement. And that is the core of the problem. It is, without buing a microcontroller or additional hardware impossible to "just play". Instead of having LDA #1;STA $IOADDR;LDA #0;STA $IOADDR; for making a single pulse on the user port, you probably have to click-click on the activeX control (a standard defined in a thick manual) which came with the proprietary kit. This is no fun. I had more fun when i played with lego than when i got a ready made thing.
So the problem, that once computers and it attracted people who built and constructed things; sometimes proofs of concepts - you stick twi wires into you computer at take the stupid diode on the top of your monitor. It is a natural process that things are taken over by people who are more like users. Actually I experienced this one or more times with proofs of concepts which I did when i worked i a group supporting the webmaster at our university. When i sat in 1998 in a system programming course in university i was astonished how uninterested the average IT first semester students where in how to use system calls in C. To point that out: I LOVE programming languages other than C. However it is good to know what will be expensive. A simple example: to make an access over the network is ok. If you however do something semantical which requires waiting for a response, you kill performance (Everybody having connected a LED and a light sensor to the user port of his C64 knows that the light sensor does never react BEFORE the LED is turned on. Depending how cheaply you built it it might react milliseconds after that).
As an example there are three approaches to the problem if you find out that you database is too slow:
1) Speed up your hardware (chosen by people who do not undertand and are not intersete in the fact that they dont)
2) Change to a DBMS which handles if better (chosen by the majority of "dilbertized" workers, who understand that they don not understand)
3) do it better and sell it to people of point 2). Bo that by taking a cup of tea, a pen and not talking for how long it takes. Do not write a report before you understood the problem.
Nowadays, even point 3) gets more dilbertized. This is a cultural effect which beliefs in ISO9000 processes to be the best tool for everything. If they are only the future can tell.
> What you said is so true. When "programmers" do research in order to find new and better ways of doing things, or solving problems
> that have not been solved, it's "science." It may not by physics or chemistry, but it's still science. I would bet that most
No, it's not. Science defines itselft no by 'solving a problem' but by 'increasing everyones knowlegde'. This happens by the means of writing an article about wgat you did. Just writing a program is not enough.
> programmers are doing a little bit of science every day. Sometimes, it's a lot.
Sometimes, yes
I have no doubt that in the next years we will need an huge amount of embedded Developers. put together the funny crappy libraries which had in mind that opening the Start Menu of windows at high resolution an 32 bit color depth transfers more data than the whol payload (e.g. text Document) of the whole Apllication has. Terribly good for saving resources (mainly: Battery!)