Actually, I've heard things about bacteria _rejecting_ extra genes when they serve no useful purpose to the organism. The scientists add these extra genes into the genome, and in a few generations, they are gone. Organisms seem to have an amazing ability to reject extra cellular baggage. If a gene is not useful at all, it will dissapear. Natural selection is a far better optimizer than we could hope to be. I wouldn't worry about making an organism _more_ efficient.
Also, creating an organism harmful to humans would require genes that code for toxins or other harmful chemicals. The scientists, I hope, would not include such genes in any of their creations. Toxic genes would not evolve spontaneously (at least in any length of time that we care about).
A few years ago, sitting in bio class, I started thinking to myself: an organism is just a giant computer, right? The DNA is the source code, transcription into polypeptide chains is compiling (assembling, really), protein folding is linking, and then the exe's (proteins, enzymes) run on the platform of cytoplasm. This anology can be extended further quite easily. Membrane-embedded receptor proteins are input, the golgi apparatus is output, the nervous system is the central bus, etc.
So, being a programmer, I wondered how hard it would be to "write" an organism from scratch, in assembly language (amino acid sequence). We'd have to understand all the layers above it, including transcription (we understand that pretty well) and protein folding (that one will take a lot of work still). Not to mention how tertiary/quaternary structure of proteins affects their function (a veeeery hard question, as of now). I thought that writing an original organism would be out of reach for at least 50, if not several hundred years.
What it looks like is that these scientists are not using assembly language, they're using pre-build [COM/CORBA/whatever] components. In fact, biology makes it much easier than would be expected: we don't even have to understand the precise functions and interactions (component interfaces) of each component. We just throw them all in a cell and let them mingle how they like, and it works! I never considered that in my bio class musings.
However, this still makes an interesting point: until we "write" the organism in DNA base sequences (machine language) or perhaps amino acid sequence (assembly language), we cannot say that this is an "original" organism, created by man. It's more like a microscopic Frankenstein, that is, built from pieces of other organisms.
Debugging faster isn't the only reason to learn ASM. It gives you a complete perspective (at least much more complete than you would otherwise have) of what is really going on in your computer. This kind of knowledge is very much lacking in the new generation of programmers that work in fancy RAD environments. There's a reason Knuth chose to use assembly in the classic Art of Computer Programming.
Personally, I an fed up with bloated code and RAD. I haven't yet learned x86 asm, but I've played around with Z80 and 68K (on my graphing calculators) and it's pretty fun. If you have some irrational fear of assembly, please try Z80. It was easier to learn than some high-level languages that I've tried (perl, for example).
the funniest law of all...
on
Dumb Laws
·
· Score: 1
I still think the funniest law I've ever seen was when some state legislature set the value of pi to be three. Yes, they set a fundamental trancendental mathematical constant to another value. The stupidity boggles the mind.
I think it was Virginia. Or maybe Indiana... I forgot. The full text of the bill is floating around on the net somewhere, I'm just too lazy to look for it now.
I see at least one large thread on quake, which of course makes perfect sense: What else could you do with a computer to get into a zen state faster then play an immersive, exhilerating (sp?) game?
I'd just like to mention _my_ favorite computer game, SubSpace. No one has ever heard of it, so I'll describe it briefly: It's like Asteroids, but there's a huge map with hundreds of people, real people, playing at the same time. There's a full array of weapons and special items. Now, subspace is played by just about everyone with the keyboard only. The keyboard configuration is not changable, but it is so well designed that it doesn't matter. You play with your hands in one position: right hand placed over arrow keys, left hand with thumb on control (bullet firing key) and one or two fingers on tab (bomb firing key). Almost all of the special items are placed on the keyboard so that your two hands never have to move far from those positions. (They use ins, del, home, end, shift,alt...) As an added bonus, all of the "character" keys are not bound to anything, making it very easy to chat in the game.
I've entered a zen state playing subspace many times, and just like Tom said, I belive the key is in the keyboard/mouse interface. A game needs inteligently designed keyboard and mouse bindings in order for a player to enter that state. Games are bit different than vi because you aren't just trying to save time, you _need_ to react as quickly as possible. A layout where your hands (almost) never move is essential. What were the designers of TA thinking when they made the group selecting keys alt+number? That's a nasty combination to have to type for something you use so much. (They did fix it in a later patch though.) In games, it's simply essential to have the kind of interface that Tom is talking about.
(Oh, and if anyone wants to try subspace, the greatest computer game ever, try http://subspace.ds98.com/)
I see that this proposal (the original one, I mean) has been shot down pretty hard, so I'm just going to address one minor issue that my parent comment only touched on slightly: the logistics of such a licensing scheme.
As the author of this proposal said, the Internet and world wide web is a necessarily international concept. In fact, international isn't even the right word, because it really has nothing to do with nations (besides dns country codes and the like) Global or worldwide, or perhaps ageographical.... anyway, any organization to license web users would have to be of the same scope. Right now, we have nothing that fits that description. The W3C? No way. It's a relatively small group that just makes standards. There's no possible way they could set up an infrastructure to license any web user. The United Nations? It's the only thing I can think of with enough resources, but this is entirely out of their scope. They deal with nations, not people.
Perhaps the most workable proposal would be for each country to do it separetely. But then we would have conflicts like: I'm allowed to post child pornography to a site if I'm standing in a country where it is legal (there must be at least one?), but if I move a few feet to my left, I can't. This is somewhat close to what exists already, at least with regard to the export of encryption and other situations. So doing it this way would be just about useless.
If we're talking about LWE, you can get a pass for the exhibit area and a few other things for free if you register by email, web, or fax before 1/4/00. Otherwise it's $25, for anyone including students. Here's the pricing page: http://www.linuxworldexpo.com/register_info.html I hope to see you there... once I convince my parents to take me:)
Yes, I was laughing but horrified at the same time as I read that chapter. The worst part is when I try to explain the "adding the temperatures of all these stars" example to people... and they don't get it.
In my Sociology class, we are given two textbooks: one modern one from the '90's (filled with pictures) and one old one from the '70's that is almost falling apart (with just about all b&w text). The teacher tells us to be extremely careful with the old ones, because they are irreplacable (they are out of print) and they are so much better than the new ones. Indeed, most of the reading assignments come out of those older, better books.
I think we should be worried, or at least conscious, of a trend for newer textbooks to be filled with pictures and charts and colors, while the older ones are written better and are more effective for learning.
(I also remember a 20/20 special where they called the people listed as primary authors of a textbook, and the people said that they had never seen that book before!)
Does anyone know if they will allow people under 18 to be present? Several other conferences at the Javits Center did not, and I'm not sure if that (ignorant) policy is mandated by the Javits Center, or by the organizers of the specific event. I could not find any mention of it on the site either way.
BTW, I see on the LinuxWorldExpo site that they do allow people under 18, in fact, they even give them discounts!
We use bess at my school too... I found a few good ways to get around it. First, use a proxy server to surf. One of these might work: http://www3.noproxy.com/ http://www.anonymizer.com/ or an encrypted one https://lesser-magoo.lcs.mit.edu/px.html
If you need forms support (for hotmail or something) you might have problems with these. Luckily, my network admin is not quite so intelligent. To bypass bess (in navigator), simply go to the edit menu, preferences, advanced, proxies, and select direct connection! Yes, he forgot to block port 80 traffic. If you use IE or something, it's a lot eaiser to disable the control panels, so you'll need more sophisticated hacking. You're admin is probably not so stupid, so that won't work. Good luck anyway.
> MD is much more versatile than MP3 Sorry, but that makes no sense at all. MD is a physical medium that stores audio. It does it very well, but it cannot be called versatile. MP3 is a compression standard for music/on any media/. Can you email a MD song to a friend? Put it on an FTP site? Share it with napster? Store it on your hard drive? The most important feature of MP3 is that it is not tied to any medium. You can move them around just like any other kind of file. True, MP3 players do not have the capacity, battery life, construction quality, or any number of things, that MD players have. But that will change in the coming years. For now, be happy with your MD, I would be too if I had one, but when the hardware starts improving, MP3 will be the undeniable choice precisely because of versatility.
Personally, I'm waiting for a MP3 player that uses MD disks and technology for storage. Basically, a MD player that uses MP3 instead of ATRAC. This way, you get the advantages of the high MD capacity and skip-free playing, but you have 170mb or so to store MP3's. That's 148 minutes at 128kbps!
(Just as a note: What's the point of copying MP3's to a MD? That sends the audio through _two_ layers of compression. It can't sound very good that way, even with an optical out. At the very best, the quality can't be better than the original 128kps the MP3 was encoded at.)
I certaintly agree with the idea that much (important) scientific is revolutionary rather than evolutionary, but I think that idea does not apply to this specific question. We have taken the two overall domains, the microscopic and the macroscopic, and described them with amazing accuracy. The general idea of a field theory has worked well to explain the electromagnetic force and the two nuclear forces. There is a lot of work being done on a field theory of gravity. It seems that all that's left to do is (beware of exagerated metaphor) extend the field theories and meet in the middle. While it's true that some amazing new theory might explain the macro- and microscopic in one simple theory, it is probably not too likely. Instead, the current theories will ocntinue to evolve until we have a basic framework that is compatible across all four forces (which will then be turned into one).
A lot of the replies so far have been arguing about whether open source is appropriate for medical software. The problem is that everyone has a different idea of what medical software actually is. Some posters seem to think it refers to the programs that control pacemakers, ecg's, and other medical tools and electronics. Others include even programs used to transfer information among doctors. The linked page is down now (possible slashdot effect) so I have no idea what the original page was talking about. Until everyone agrees (or finds out by reading the article) on exactly what types of software are being opened up to OSS, there is not much point debating the appropriateness of it.
Concerning my personal experience, this past summer, I worked on what I would consider to "medical software." It actually dealt with testing the accuracy of various pieces of medical hardware. Although I am no longer working on the project, there are plans to sell the resulting software to hospitals and other institutions, which is clearly incompatible with open source. In fact, most medical software of this type is probably developed for profit. OSS might have applications in other areas, however. The transfer of medical information among doctors and institutions, for example, could benefit from open standards and free tools. Treatment planning systems, on the other hand, are developed (at least partially) for a profit, and OSS would probably have little application there.
On my old apple II, there were two commands in the basic language called peek and poke. peek read the value of an absolute memory location and poke set the value. I'd assume the Sinclair had similar commands. Peeking before poking was probably a good idea for error checking and the like because pokeing could really screw something up. As for the quote being in the game, it's most likely some kind of sexual reference.
This headline caught my eye because I remember playing wolf3d a while ago, and I'm always interested in classic games (which tend to be as much fun as modern ones). I wanted to visit the page, so I looked for a hyperlink to the page in the description text on the main page. The linked text wasn't the word "this" which would have made the most sense, or "Wolfenstein 3D", which would have made a little. Or even "classic game". None of these phrases that refer to the page were linked. Instead, the hyperlink was the text "Linux version". True, there is a Linux version located at the page, but there is also a Windows version, and that text was not linked to anything. I understand that this page has a prominent Linux bias. That is acceptable when, for example, a popular windows software program is released for Linux. In this case, however, the actual story has no Linux bias. There are already versions of wolf3d for Linux as well as Windows. The point of this story was that an GL version has been made, and it has been made for both platforms. Adding a Linux bias where none existed previously is misleading, unfair, and simply bad journalism.
FYI, I do not use Linux at home, but not for reasons of technological superiority. I do believe that Linux is a technologically superior OS to Widnows. However, that is no reason to make the link text only "Linux version" when the program and site is obviously OS-neutral.
a real universal remote must have x10 features
on
The Do-It-All Remote?
·
· Score: 1
A remote control cannot be called "universal" unless it can control every single electronic device in your house. The most popular and easiest way to do this is with X10, the home automation standard. You probably already know what X10 is and what it can do, so I'd just like to point out a remote that can control tv's, vcr's, etc, and it can also send out X10 commands:
http://www.x10.com/products/x10_ur47a.htm
I think there are a few similar models on that site. Look around. But remember, it's gotta have X10 or it's not really universal.
The original news post suggested using c/c++ over perl for cgi programming, supposedly because c/c++ is lower-level and thus more efficient. Other posts here have already shown that's not true, so I won't repeat those arguments. I'll just say that c/c++ is not very well suited for cgi programming. It's string handling is weak, which is one place where perl excels, and string handling is extremely important to cgi. Also, efficiency is not quite as important to cgi as it is to productivity applications or other genres.
With that in mind, let me offer another alternative: php. You've probably heard of it, but I'll list a few advantages. It's incredibly easy to learn. The syntax is very similar to c, but it's a scripting language, so it's as quick to get started with as perl. It has very well integrated database support. Note that the database modules, which is where efficiency is most important, are written in c. It's string handling is mostly as good as perl's. (I haven't used perl much, so I might be a little off on that, but it is superior to c's.) For info, check out http://www.php.net
(BTW, I'm not associated with them in any way, I just started using php recently, and I'm very impressed at how easy and powerful it is, especially in database connectivity.)
The original news post suggested using c/c++ over perl for cgi programming, supposedly because c/c++ is lower-level and thus more efficient. Other posts here have already shown that's not true, so I won't repeat those arguments. I'll just say that c/c++ is not very well suited for cgi programming. It's string handling is weak, which is one place where perl excels, and string handling is extremely important to cgi. Also, efficiency is not quite as important to cgi as it is to productivity applications or other genres.
With that in mind, let me offer another alternative: php. You've probably heard of it, but I'll list a few advantages. It's incredibly easy to learn. The syntax is very similar to c, but it's a scripting language, so it's as quick to get started with as perl. It has very well integrated database support. Note that the database modules, which is where efficiency is most important, are written in c. For info, check out http://www.php.net
(BTW, I'm not associated with them in any way, I just started using php recently, and I'm very impressed at how easy and powerful it is, especially in database connectivity.)
I'll admit Capsela has lego beat as far as OO is concerned, but it cannot compare in other aspects. First, capsela models are simply huge. To build anything remotely functional, your model would have to be a foot long. Sure, you might like buidling life size robots and things, but capsela gives you very little functionality for the size. Second, It seemed like those nice features you mentioned make it more limited and much harder to use. It may be good for rapid prototyping, but of what? Only very simple models. You could build things with legos that would take a room full of capsela to have equal functionality.
> A generation raised on television and RPG's and > techno-fairytales cannot distinguish between the > real world and the one > they desperately wish they lived in.
While I partly agree with your assesment of television as contributing to the general apathy and cluelessness of society, I take offense at your mention of RPG's. Until a few years ago, RPGs were a very small sector of the total video game market (in the US that is; they were much more popular in Asia). In my experience, RPG players were much more intellgent, on average, than players of other genres of video games, and probably more intelligent that most average people. Perhaps it has to do with the similarity of the RPG form to fantasy and SF novels, which attract many of the more intellgent, less socially accepted "nerds" and the like. Whatever the reason, I would be much more comfortable with a populace of RPG players than players of mindless fighting games and other less intellectual genres. In the last few years (starting with the release of FF7), RPGs have become much mainstream and this generalization would probably fail to hold. As for your mention of "techno-fairytales," I'm not quite sure what you're referring to.
> You can't just apply the same code across millions of them
Of course you can. The information "learned" by a neural net is contained in a big list (really one or more matrices) of numbers. Training can be perfomed once to get a set of numbers or parameters that performs a task well, and then a product can be mass produced with that specific configuration. Sometimes, the product may use a know good configuration as a starting point and allow more learning. But neural network learning certainly can be reproduced.
[I realize the original comment was meant to be more funny than correct, but I think I should point out faulty premises.]
Slashdot Mirror
Actually, I've heard things about bacteria _rejecting_ extra genes when they serve no useful purpose to the organism. The scientists add these extra genes into the genome, and in a few generations, they are gone. Organisms seem to have an amazing ability to reject extra cellular baggage. If a gene is not useful at all, it will dissapear. Natural selection is a far better optimizer than we could hope to be. I wouldn't worry about making an organism _more_ efficient.
Also, creating an organism harmful to humans would require genes that code for toxins or other harmful chemicals. The scientists, I hope, would not include such genes in any of their creations. Toxic genes would not evolve spontaneously (at least in any length of time that we care about).
A few years ago, sitting in bio class, I started thinking to myself: an organism is just a giant computer, right? The DNA is the source code, transcription into polypeptide chains is compiling (assembling, really), protein folding is linking, and then the exe's (proteins, enzymes) run on the platform of cytoplasm. This anology can be extended further quite easily. Membrane-embedded receptor proteins are input, the golgi apparatus is output, the nervous system is the central bus, etc.
So, being a programmer, I wondered how hard it would be to "write" an organism from scratch, in assembly language (amino acid sequence). We'd have to understand all the layers above it, including transcription (we understand that pretty well) and protein folding (that one will take a lot of work still). Not to mention how tertiary/quaternary structure of proteins affects their function (a veeeery hard question, as of now). I thought that writing an original organism would be out of reach for at least 50, if not several hundred years.
What it looks like is that these scientists are not using assembly language, they're using pre-build [COM/CORBA/whatever] components. In fact, biology makes it much easier than would be expected: we don't even have to understand the precise functions and interactions (component interfaces) of each component. We just throw them all in a cell and let them mingle how they like, and it works! I never considered that in my bio class musings.
However, this still makes an interesting point: until we "write" the organism in DNA base sequences (machine language) or perhaps amino acid sequence (assembly language), we cannot say that this is an "original" organism, created by man. It's more like a microscopic Frankenstein, that is, built from pieces of other organisms.
Debugging faster isn't the only reason to learn ASM. It gives you a complete perspective (at least much more complete than you would otherwise have) of what is really going on in your computer. This kind of knowledge is very much lacking in the new generation of programmers that work in fancy RAD environments. There's a reason Knuth chose to use assembly in the classic Art of Computer Programming.
Personally, I an fed up with bloated code and RAD. I haven't yet learned x86 asm, but I've played around with Z80 and 68K (on my graphing calculators) and it's pretty fun. If you have some irrational fear of assembly, please try Z80. It was easier to learn than some high-level languages that I've tried (perl, for example).
I still think the funniest law I've ever seen was when some state legislature set the value of pi to be three. Yes, they set a fundamental trancendental mathematical constant to another value. The stupidity boggles the mind.
I think it was Virginia. Or maybe Indiana... I forgot. The full text of the bill is floating around on the net somewhere, I'm just too lazy to look for it now.
I see at least one large thread on quake, which of course makes perfect sense: What else could you do with a computer to get into a zen state faster then play an immersive, exhilerating (sp?) game?
,alt...) As an added bonus, all of the "character" keys are not bound to anything, making it very easy to chat in the game.
I'd just like to mention _my_ favorite computer game, SubSpace. No one has ever heard of it, so I'll describe it briefly: It's like Asteroids, but there's a huge map with hundreds of people, real people, playing at the same time. There's a full array of weapons and special items. Now, subspace is played by just about everyone with the keyboard only. The keyboard configuration is not changable, but it is so well designed that it doesn't matter. You play with your hands in one position: right hand placed over arrow keys, left hand with thumb on control (bullet firing key) and one or two fingers on tab (bomb firing key). Almost all of the special items are placed on the keyboard so that your two hands never have to move far from those positions. (They use ins, del, home, end, shift
I've entered a zen state playing subspace many times, and just like Tom said, I belive the key is in the keyboard/mouse interface. A game needs inteligently designed keyboard and mouse bindings in order for a player to enter that state. Games are bit different than vi because you aren't just trying to save time, you _need_ to react as quickly as possible. A layout where your hands (almost) never move is essential. What were the designers of TA thinking when they made the group selecting keys alt+number? That's a nasty combination to have to type for something you use so much. (They did fix it in a later patch though.) In games, it's simply essential to have the kind of interface that Tom is talking about.
(Oh, and if anyone wants to try subspace, the greatest computer game ever, try http://subspace.ds98.com/)
I see that this proposal (the original one, I mean) has been shot down pretty hard, so I'm just going to address one minor issue that my parent comment only touched on slightly: the logistics of such a licensing scheme.
As the author of this proposal said, the Internet and world wide web is a necessarily international concept. In fact, international isn't even the right word, because it really has nothing to do with nations (besides dns country codes and the like) Global or worldwide, or perhaps ageographical.... anyway, any organization to license web users would have to be of the same scope. Right now, we have nothing that fits that description. The W3C? No way. It's a relatively small group that just makes standards. There's no possible way they could set up an infrastructure to license any web user. The United Nations? It's the only thing I can think of with enough resources, but this is entirely out of their scope. They deal with nations, not people.
Perhaps the most workable proposal would be for each country to do it separetely. But then we would have conflicts like: I'm allowed to post child pornography to a site if I'm standing in a country where it is legal (there must be at least one?), but if I move a few feet to my left, I can't. This is somewhat close to what exists already, at least with regard to the export of encryption and other situations. So doing it this way would be just about useless.
If we're talking about LWE, you can get a pass for the exhibit area and a few other things for free if you register by email, web, or fax before 1/4/00. Otherwise it's $25, for anyone including students. Here's the pricing page:l :)
http://www.linuxworldexpo.com/register_info.htm
I hope to see you there... once I convince my parents to take me
Yes, I was laughing but horrified at the same time as I read that chapter. The worst part is when I try to explain the "adding the temperatures of all these stars" example to people... and they don't get it.
In my Sociology class, we are given two textbooks: one modern one from the '90's (filled with pictures) and one old one from the '70's that is almost falling apart (with just about all b&w text). The teacher tells us to be extremely careful with the old ones, because they are irreplacable (they are out of print) and they are so much better than the new ones. Indeed, most of the reading assignments come out of those older, better books.
I think we should be worried, or at least conscious, of a trend for newer textbooks to be filled with pictures and charts and colors, while the older ones are written better and are more effective for learning.
(I also remember a 20/20 special where they called the people listed as primary authors of a textbook, and the people said that they had never seen that book before!)
Does anyone know if they will allow people under 18 to be present? Several other conferences at the Javits Center did not, and I'm not sure if that (ignorant) policy is mandated by the Javits Center, or by the organizers of the specific event. I could not find any mention of it on the site either way.
BTW, I see on the LinuxWorldExpo site that they do allow people under 18, in fact, they even give them discounts!
"they"? no. I'm sure a group of people could not sink collectively to a level this low. It must be the work of a lone loser.
We use bess at my school too... I found a few good ways to get around it.
First, use a proxy server to surf. One of these might work:
http://www3.noproxy.com/
http://www.anonymizer.com/
or an encrypted one
https://lesser-magoo.lcs.mit.edu/px.html
If you need forms support (for hotmail or something) you might have problems with these. Luckily, my network admin is not quite so intelligent. To bypass bess (in navigator), simply go to the edit menu, preferences, advanced, proxies, and select direct connection! Yes, he forgot to block port 80 traffic. If you use IE or something, it's a lot eaiser to disable the control panels, so you'll need more sophisticated hacking. You're admin is probably not so stupid, so that won't work. Good luck anyway.
> MD is much more versatile than MP3 /on any media/. Can you email a MD song to a friend? Put it on an FTP site? Share it with napster? Store it on your hard drive? The most important feature of MP3 is that it is not tied to any medium. You can move them around just like any other kind of file. True, MP3 players do not have the capacity, battery life, construction quality, or any number of things, that MD players have. But that will change in the coming years. For now, be happy with your MD, I would be too if I had one, but when the hardware starts improving, MP3 will be the undeniable choice precisely because of versatility.
Sorry, but that makes no sense at all. MD is a physical medium that stores audio. It does it very well, but it cannot be called versatile. MP3 is a compression standard for music
Personally, I'm waiting for a MP3 player that uses MD disks and technology for storage. Basically, a MD player that uses MP3 instead of ATRAC. This way, you get the advantages of the high MD capacity and skip-free playing, but you have 170mb or so to store MP3's. That's 148 minutes at 128kbps!
(Just as a note: What's the point of copying MP3's to a MD? That sends the audio through _two_ layers of compression. It can't sound very good that way, even with an optical out. At the very best, the quality can't be better than the original 128kps the MP3 was encoded at.)
I certaintly agree with the idea that much (important) scientific is revolutionary rather than evolutionary, but I think that idea does not apply to this specific question. We have taken the two overall domains, the microscopic and the macroscopic, and described them with amazing accuracy. The general idea of a field theory has worked well to explain the electromagnetic force and the two nuclear forces. There is a lot of work being done on a field theory of gravity. It seems that all that's left to do is (beware of exagerated metaphor) extend the field theories and meet in the middle. While it's true that some amazing new theory might explain the macro- and microscopic in one simple theory, it is probably not too likely. Instead, the current theories will ocntinue to evolve until we have a basic framework that is compatible across all four forces (which will then be turned into one).
I haven't seen your cycling version (VIIDCCLXXVII) but what I have seen is symbols with bars over then, which means 1000 times the value.
_
V = 5000
So, 7777 would best be written:
_
VMMDCCLXXVII
A lot of the replies so far have been arguing about whether open source is appropriate for medical software. The problem is that everyone has a different idea of what medical software actually is. Some posters seem to think it refers to the programs that control pacemakers, ecg's, and other medical tools and electronics. Others include even programs used to transfer information among doctors. The linked page is down now (possible slashdot effect) so I have no idea what the original page was talking about. Until everyone agrees (or finds out by reading the article) on exactly what types of software are being opened up to OSS, there is not much point debating the appropriateness of it.
Concerning my personal experience, this past summer, I worked on what I would consider to "medical software." It actually dealt with testing the accuracy of various pieces of medical hardware. Although I am no longer working on the project, there are plans to sell the resulting software to hospitals and other institutions, which is clearly incompatible with open source. In fact, most medical software of this type is probably developed for profit. OSS might have applications in other areas, however. The transfer of medical information among doctors and institutions, for example, could benefit from open standards and free tools. Treatment planning systems, on the other hand, are developed (at least partially) for a profit, and OSS would probably have little application there.
On my old apple II, there were two commands in the basic language called peek and poke. peek read the value of an absolute memory location and poke set the value. I'd assume the Sinclair had similar commands. Peeking before poking was probably a good idea for error checking and the like because pokeing could really screw something up. As for the quote being in the game, it's most likely some kind of sexual reference.
This headline caught my eye because I remember playing wolf3d a while ago, and I'm always interested in classic games (which tend to be as much fun as modern ones). I wanted to visit the page, so I looked for a hyperlink to the page in the description text on the main page. The linked text wasn't the word "this" which would have made the most sense, or "Wolfenstein 3D", which would have made a little. Or even "classic game". None of these phrases that refer to the page were linked. Instead, the hyperlink was the text "Linux version". True, there is a Linux version located at the page, but there is also a Windows version, and that text was not linked to anything. I understand that this page has a prominent Linux bias. That is acceptable when, for example, a popular windows software program is released for Linux. In this case, however, the actual story has no Linux bias. There are already versions of wolf3d for Linux as well as Windows. The point of this story was that an GL version has been made, and it has been made for both platforms. Adding a Linux bias where none existed previously is misleading, unfair, and simply bad journalism.
FYI, I do not use Linux at home, but not for reasons of technological superiority. I do believe that Linux is a technologically superior OS to Widnows. However, that is no reason to make the link text only "Linux version" when the program and site is obviously OS-neutral.
A remote control cannot be called "universal" unless it can control every single electronic device in your house. The most popular and easiest way to do this is with X10, the home automation standard. You probably already know what X10 is and what it can do, so I'd just like to point out a remote that can control tv's, vcr's, etc, and it can also send out X10 commands:
http://www.x10.com/products/x10_ur47a.htm
I think there are a few similar models on that site. Look around. But remember, it's gotta have X10 or it's not really universal.
arrgh.. very sorry for the double post. someone please moderate it down (this post's parent)
The original news post suggested using c/c++ over perl for cgi programming, supposedly because c/c++ is lower-level and thus more efficient. Other posts here have already shown that's not true, so I won't repeat those arguments. I'll just say that c/c++ is not very well suited for cgi programming. It's string handling is weak, which is one place where perl excels, and string handling is extremely important to cgi. Also, efficiency is not quite as important to cgi as it is to productivity applications or other genres.
With that in mind, let me offer another alternative: php. You've probably heard of it, but I'll list a few advantages. It's incredibly easy to learn. The syntax is very similar to c, but it's a scripting language, so it's as quick to get started with as perl. It has very well integrated database support. Note that the database modules, which is where efficiency is most important, are written in c. It's string handling is mostly as good as perl's. (I haven't used perl much, so I might be a little off on that, but it is superior to c's.) For info, check out http://www.php.net
(BTW, I'm not associated with them in any way, I just started using php recently, and I'm very impressed at how easy and powerful it is, especially in database connectivity.)
The original news post suggested using c/c++ over perl for cgi programming, supposedly because c/c++ is lower-level and thus more efficient. Other posts here have already shown that's not true, so I won't repeat those arguments. I'll just say that c/c++ is not very well suited for cgi programming. It's string handling is weak, which is one place where perl excels, and string handling is extremely important to cgi. Also, efficiency is not quite as important to cgi as it is to productivity applications or other genres.
With that in mind, let me offer another alternative: php. You've probably heard of it, but I'll list a few advantages. It's incredibly easy to learn. The syntax is very similar to c, but it's a scripting language, so it's as quick to get started with as perl. It has very well integrated database support. Note that the database modules, which is where efficiency is most important, are written in c. For info, check out http://www.php.net
(BTW, I'm not associated with them in any way, I just started using php recently, and I'm very impressed at how easy and powerful it is, especially in database connectivity.)
I'll admit Capsela has lego beat as far as OO is concerned, but it cannot compare in other aspects. First, capsela models are simply huge. To build anything remotely functional, your model would have to be a foot long. Sure, you might like buidling life size robots and things, but capsela gives you very little functionality for the size. Second, It seemed like those nice features you mentioned make it more limited and much harder to use. It may be good for rapid prototyping, but of what? Only very simple models. You could build things with legos that would take a room full of capsela to have equal functionality.
> A generation raised on television and RPG's and
> techno-fairytales cannot distinguish between the
> real world and the one
> they desperately wish they lived in.
While I partly agree with your assesment of television as contributing to the general apathy and cluelessness of society, I take offense at your mention of RPG's. Until a few years ago, RPGs were a very small sector of the total video game market (in the US that is; they were much more popular in Asia). In my experience, RPG players were much more intellgent, on average, than players of other genres of video games, and probably more intelligent that most average people. Perhaps it has to do with the similarity of the RPG form to fantasy and SF novels, which attract many of the more intellgent, less socially accepted "nerds" and the like. Whatever the reason, I would be much more comfortable with a populace of RPG players than players of mindless fighting games and other less intellectual genres. In the last few years (starting with the release of FF7), RPGs have become much mainstream and this generalization would probably fail to hold. As for your mention of "techno-fairytales," I'm not quite sure what you're referring to.
Yes :)
I noticed the same thing a few weeks ago:
http://slashdot.org/co mments.pl?sid=99/09/23/0036227&cid=508
[And insert it at the beginning of my post too.]
> You can't just apply the same code across millions of them
Of course you can. The information "learned" by a neural net is contained in a big list (really one or more matrices) of numbers. Training can be perfomed once to get a set of numbers or parameters that performs a task well, and then a product can be mass produced with that specific configuration. Sometimes, the product may use a know good configuration as a starting point and allow more learning. But neural network learning certainly can be reproduced.
[I realize the original comment was meant to be more funny than correct, but I think I should point out faulty premises.]