I had skill, but no opening. I tried to learn C++, but found it too abstract and rather difficult.
You have to learn C, and do it by the K&R book. After that you would be able to make conscious choice, to keep using C for everything, study C++, or to learn some other language, but without knowledge of C the programmer is deaf and blind in any kind of language choice or understanding. And, seaprate issue, I have no idea, what kind of monster will a programmer turn into if he will learn C++ without prior knowledge of C, but I don't want to use anything he will write.
As I see it, EULA is only valid if it is presented at the moment before transaction is performed -- Microsoft may think that running self-extracting executable is a transaction that makes the text of specification available. However then ff transaction can be performed without the user doing anything that may indicate that he accepts the license (uncompressing the file without running it), then only copyright applies, with all fair use provisions untouched.
Of course, there are other issues that may invalidate the EULA even if it was accepted, and may invalidate the claims about trade secrets (you can't call something a secret and distribute it to everyone undiscriminately, so if something was taken from Microsoft, it definitely wasn't a secret).
Now, don't get silly. The basis of the government's case (at least part of it) is that, in their opinion, a web browser cannot be part of an OS. Microsoft disagrees.
There is (or at least was) a sepatate market for web browsers, different from market for operating systems -- Netscape, Opera and others competed in it, and none of them at any degree suffered from the lack of "integration" that Microsoft "innovated" later. OTOH, at no point in Unix history there was a market for basic-functionality text editor or telnet client for Unix, and even though that, say, Unix tar competed with GNU tar, none of two versions actually managed to hurt another one.
Hell, all the linux zealots are a bunch of hypocrites anyway - all y'all have Win98 on your machines for games. So basically, your ethics can be bought for a video game. Ha.
I don't have Windows anywhere -- five desktop boxes at home, desktop at work and a laptop, all run Linux and FreeBSD (one of desktops is a powermac 8100/80 -- it also runs MacOS as a bootloader for mkLinux). And, yes, I have bought UT, and I am absolutely sure, some Epic bean-counter counted it as a sold Windows version, even though I run it on my Linux box.
Unicode is one of "semi-proprietary" standards -- documents aren't available for free (be it ones from ISO or Unicode), however there is no legal barriers for making an implementation -- just the size of the table makes a job of creating fonts unreasonably huge. OTOH, the tables necessary for determining, what the characters are, are available for free.
The problem however is different -- people already use their own charsets, and those charsets were designed to reflect the structure of language, or just to be most convenient for their language, sometimes made quite different from part of Unicode that is supposed to be used for the same language. If instead of trying to _convert_ everything to Unicode, people adopted a reasonable (iso 2022 isn't reasonable) way to label, which charset and which language are where in their strings, the implementation would be able to use all known charsets, and programs that aren't concerned with operation thats depend on them can just ignore the whole thing and treat text as a sequence of bytes until charset-specific procedures are called to process/display/compare/convert/input/... the text where "real" size and mapping of characters will emerge -- and those procedures can be language-dependent, replaceable and expandable if they will implement an easy mechanism of mapping charset/language names to sets of procedures. Unicode could be used as one of possible charsets, and UTF-8 could be used as one of possible encodings in such a system, however it won't be "the" thing, that everyone is supposed to support and be aware about. At most some programs would have to know how label delimiters look.
It can be a very easy solution for the real problem, however it requires an agreement on how charsets/languages should be labeled (their "real" names should be used to make the thing expandable, however how those things should be separated from "normal" text remains a question).
Well, mebbe not. I am waiting to hear more comments from non-English slashdotters on this subject, the comments so far reflect a definite world view -- the English world.
Non-English slashdotters that at the same time use iso8859-1, most likely see Unicode (or UTF-8) as a good thing because first 256 characters of Unicode are the same as iso8859-1, and they don't give a damn about everything else, while non-English slashdotters that use other local encodings/charsets (like me, whose native language is Russian, with koi8-r as the charset used in unixlike systems) see Unicode as a monstrosity, forced on them by a bunch of dumbasses at Unicode Consortium, ISO and software vendors thart benefit from every incompatibility that can force people to upgrade.
If charset/language labeling was standardized, everyone would be able to use their own charset, and all software that is not directly involved in text editing/displaying would be able to continue working as it was before, however by STUPID decision made by "standard bodies" the priority is given to sticking "should support Unicode/UTF-8" into every standard in the place of "should pass the data as a stream of bytes regardless of the actual size of character, encoding and their possible meaning, except special characters involved in protocol" that would actually accomplish something.
In the Middle Ages in societies with limited literacy, high cost of information distribution and overwhelming politically-backed power of religion science was in the situation, similar to what commercial software is now -- consisted of "secrets" (like modern secret algorithms), obscure terminology (like modern secret data formats) pieces of religious texts and ancient books, quoted everywhere, distorted and taken out of context (like modern botched and mutilated standards and marketing buzzwords), religious "laws" severely restricted what can be done (like modern patent and copyright laws), and were vague enough to support arbitrary witch hunts against anyone (like modern frivolous lawsuits). This allowed "scientists" to produce huge amount of "work" that has little or no relation to reality, and people who practiced activities that required scientific knowledge either did their job poorly because of lack of one (engineering, like modern AI), used huge amount of resources for nothing (architecture, like modern bloatware) or turned their work into complete quackery (medicine, like most of areas of software design now).
Improvements in the information dissemination technology and weakening of the grip of religion over society caused science to become more open, and openness with wide peer review allowed scientific methods (that were known long ago yet were ignored in the absence of openness -- this is why it was called "renaissance" and was percieved by people as return to earlier, better ways of science, engineering and art) to weed out the bullshit and keep the true achievements, hidden under piles and piles of "mental masturbation". The use of this open science in various kinds of trades/businesses improved, so the job of scientist switched from "secret lab" of what amounted of skilled craftsman or "library" of religious philosopher to education institutions (that also sponsored research and books publishing) and "service" work for large engineering projects. Secrecy remained only in military-related projects, and those never amounted to much before being opened or reproduced.
As the result the progress in science went on with reasonable pace, results became openly accessible and reliable (as scientists had to make them possible to reproduce), quackery was reduced to fringes of society (some modern societies have large fringes, however this is a different problem).
When software design, the area that has properties of both theoretical science (practiced by scientists) and engineering (handled mostly by businesses) appeared, more appropriate science-like model of development was very soon replaced by business-driven model of secrets, commonly used for engineering, and it ended up at the same place where science was many centuries ago. And the same processes that opened science will open the software design/development -- the key ingredients such as widely accessible education in the area, means of dissemination of information and large amount of freely-distributed information are there.
I was using the term "ethics" in the sense of "Professional Ethics". As in: ethics n - the principles of conduct governing an individual or a group.
For me no specific "ethical code" can override the general ethics -- if some rules of professional conduct (or "GNU ethics" if such thing ever existed) contradict with general ethics that I accept, following the "specific" rule is unethical, no matter what. While I may be forced to do an unethical thing, it doesn't make it right.
Also I don't see why should I quit a job just because I don't want to follow all the bosses' words -- I am as much part of that work as he is, and if he doesn't quit his job because I disagree with him, why should I do that? There were a lot of situations in my life when managers were wrong, and they acknowledged that my actions were better for the benefit of the company/customers/whatever-thing-they-consider-imp ortant afterward.
I don't think there are clear ethical guidelines in many of these situations. If the boss has stated flatly "NO OPEN SOURCE TO BE USED", then it's clearly unethical.
Why? Ignoring bosses' orders can be ethical in a lot of cases. Company's policy has very little to do with ethics, and it's still a moral choice of a person to follow or to reject it. The key is the responsibility -- if a person can be responsible for his action, does not want to push that responsibility to the boss and can defend his decision, he can do whatever he thinks is better.
The fact of the matter is that it is an INTRINSIC property of open source that is must prove itself above and beyond that of commercial software simply because there is simply no legal recourse for companies who use it
All commercial software is released under licenses that disclaim all responsibility of the manufacturer -- sometimes with the exception of defective media shipped.
You want an example of the opposite? Look at Russia during the communist days. When the iron curtain was lifted, we saw that there wasn't much to fear. Everything in Russia sucked, technology and standard of living were basically below USA. Why?
Economy in Russia started to suck after it was dismantled in the hope of becoming "more free" -- I have lived there and moved to US in 1993, so I can compare the life before and after "perestroika" in Russia, and in US now. The comparison from the point of view of more or less "normal" person is definitely not in the favor of what I see in US now -- it's better than what is in Russia now and definitely better than what american propaganda lets you to see, however I felt better in 1989 in Russia than I do in 2000 in US. It's true that there was little freedom in Russia compared to what I would want, however in US there is even less if I discount as nonexistent the "freedoms" that are taken away by inefficiency of legal system, lobbying and lack of any mechanism that restricts a power of businesses over their employees and consumers. I am probably supposed to be considered "successful" in american society since I have decent job and make enough money for life much more comfortable than one that I care to actually have, however I feel neither "free" nor "happy", and this is what counts in the end.
The file is an ATTACHMENT. In order for it to run, the user has to doubleclick it. It would be like sending a unix user a perl script that had rm -rf ~/* in it.
File with.pl extension and content-type "application/octet-stream" is never executed by any decent mailreader -- mailreader even will warn you that there is no viewer defined for this content-type. If someone had "application/x-perl" in.mailcap pointing to "/bin/perl %s" (mailreaders never write files with executable permissions), AND it was used in the mail header, then and only then it will run, however no sane person will do such a thing and no system comes with this kind of configuration.
Every idea can be turned into absurd when taken to extreme and beyond. In this example the extreme opposite to yours one will be to ban male masturbation and condoms as it kills "unborn babies" in the form of sperm.
Every idea can be turned into absurd when taken to extreme and beyond. In this example the extreme opposite to yours one will be to ban male masturbation and condoms as it kills "unborn babies" in the form of sperm.
While it would be expensive, it can be done slowly, and the value to every local library in the US would be immense.
In 1990 or 1991 when I lived in Gomel (USSR, then Belarus), all I had were some PC-AT (286) boxes and two 2400 bps modems. I have decided that it sucks that in the whole city (that had about half a million people at the time -- comparable to San Francisco) there is no public-accessible network, and that it will be a good idea to join Fidonet (that was the only thing available without insane fees) as a "point", open a BBS on it and maintain it. I did that, then "point" became a "node", some other people joined, and soon it turned into Net 2:452 of Fidonet. With cheap and primitive equipment, flaky phone lines, DOS software (that I had to rewrite to adapt to conditions where it was sometimes impossible to complete mail packet download between having the carrier lost and redialing for half an hour to some overloaded node), it supported reasonable speed of netmail (Fidonet equivalent of email, with store-and-forward routing) and echomail (newsgroups very similar to early days of Usenet), people started using "freq" (file request, an equivalent of FTP transfer, but as opposed to FTP and HTTP, it was queued as a "special" mail message, so it could be delayed until long distance discount hours, recovered from failure, etc.) to get things on remote servers, send files as "attachments" (as opposed to MIME attacments those were kept in original format, weren't routed and were processed separately from message to preserve precious bandwidth), use search-engine-like "robots", etc. -- all that using few 2400bps modems. The amount of information that can be found by Fidonet wasn't, of course, comparable to LoC, but it made a huge difference in the lives of a bunch of programmers that operated and used that network, including myself. Forgive me, Borland/Inprise/whatever, I had Borland C++ 3.1 available for download -- my excuse is that you certainly didn't lose any profits from people whose monthly salary was equivalent to $30 or less, and the nearest store with your products was way beyond their reach.
It was done by people with no money at the time when Internet (or any network) access really was hard to obtain, and it still worked.
Now almost a decade passed. Neither in Gomel nor anywhere in US it's necessary to deal with 2400bps modems or 286. Even poorest of the poor, if they want to access the Internet, can buy some used 386 box with 4M of RAM ( => capable of running DOS or Linux) with at least 9600bps modem -- I know, I did that in 1994 after arriving here, and I had literally no money then -- it certainly didn't become harder now except that p166's replaced 386dx-25 as the most likely throwavay item. Even the worst school in US with the dumbest technician in the world can find some mac and subscribe to some cheap-ass dialup. And the poorest library still can find 2-3 computers and modems. Heck, a library can put one Linux box and all other boxes can be 286 with NCSA telnet. Someone who wants to get "enterntainment", will be very disappointed by all mentioned solutions, but a person who needs to get some particular book, especially if LoC will make it available in plain text, will find those ways of accessing it through the Internet extremely useful.
So, it's not some "rich elite" that will be able to access digitized books, it's precisely the same set of people, whom libraries are supposed to serve. Library's primary function is not to serve as a place for conversations, waiting, "looking smart" or visiting a public bathroom -- it's to allow people to find and read books, and accessibility through the Internet helps to perform it.
Nope. They are just expanding to places where they weren't present or successful before (like cancer). X-box is a renamed PC, and Pocket PC is a renamed Palm PC that uses their Windows CE, that happened to be an application of Windows design and GUI ideas to completely different idea and market.
Proxiweb is a real browser -- it requests pages through proxy that re-formats them into a format, readable by the client running on the Palm, so even though it doesn't parse HTML on the Palm, result is the same. I use it for things like searching for products specs while at the store (which of cards is supported under Linux?, what chipset is in this thing?, etc.), maps from mapblast.com (using coordinates of the nearest retransmitter taken from Metricom modem), etc.
Seeing how the Palm does great Wireless networking (802.11b) -- (yeah, right) and how great the Web capabilities are, I think one offers more than the other.
Wireless Ethernet works only for few hundreds feet -- it means that you are in the same building with your network, and you need such a speed only to so something time/resource-consuming anyway, so for things like that small laptop/notebook will be a better choice, (I have vaio n505vx with wireless ethernet). However if you need to do something remotely where "remote" is another building (or city), it won't be of any use -- you need a cellular or cellular-like radiomodem. All versions of Palm support those things pretty well (VII has builtin wireless networking, but I prefer III{"",x,c} with faster Metricom modem) -- and when I need to use a notebook at some remote location I can just connect the same Metricom modem to it.
Another problem with wireless networking is power that it requires. Those things eat a lot, and unless separate power supply is used for them, it has a good chance to drain the battery very fast. PalmVII and everything that involves external radiomodem use separate battery for wireless networking, but I am not sure that WinCE/Palm PC/Pocket PC/whatever have enough space inside for a separate battery that powers PCMCIA/CardBus cards -- so the only real alternative to keeping wireless networking disabled is having an external power supply attached. Oops, that completely negates the point of having a PDA with wireless networking, and it's easier just to carry a notebook that has big enough battery to support radio Ethernet.
gcc can be built as crosscompiler on a platform that has a bunch of unrelated to gcc compilers, build itself (and libraries that it uses) for another platform and then used on another platform to build itself. Truly paranoid people can build it on many different platforms and compare the result of first self-built version -- if it's not identical, some versions are infected. The only case when it won't work is when all compilers involved are infected, and infection of all of them affects gcc in the same way -- something that I find hard to believe.
Slashdot Mirror
I had skill, but no opening. I tried to learn C++, but found it too abstract and rather difficult.
You have to learn C, and do it by the K&R book. After that you would be able to make conscious choice, to keep using C for everything, study C++, or to learn some other language, but without knowledge of C the programmer is deaf and blind in any kind of language choice or understanding. And, seaprate issue, I have no idea, what kind of monster will a programmer turn into if he will learn C++ without prior knowledge of C, but I don't want to use anything he will write.
As I see it, EULA is only valid if it is presented at the moment before transaction is performed -- Microsoft may think that running self-extracting executable is a transaction that makes the text of specification available. However then ff transaction can be performed without the user doing anything that may indicate that he accepts the license (uncompressing the file without running it), then only copyright applies, with all fair use provisions untouched.
Of course, there are other issues that may invalidate the EULA even if it was accepted, and may invalidate the claims about trade secrets (you can't call something a secret and distribute it to everyone undiscriminately, so if something was taken from Microsoft, it definitely wasn't a secret).
Now, don't get silly. The basis of the government's case (at least part of it) is that, in their opinion, a web browser cannot be part of an OS. Microsoft disagrees.
There is (or at least was) a sepatate market for web browsers, different from market for operating systems -- Netscape, Opera and others competed in it, and none of them at any degree suffered from the lack of "integration" that Microsoft "innovated" later. OTOH, at no point in Unix history there was a market for basic-functionality text editor or telnet client for Unix, and even though that, say, Unix tar competed with GNU tar, none of two versions actually managed to hurt another one.
Actually, if these were standardized on hardware, then the API would never evolve. No cubic environment mapping for you!
This is why decent standards include standardized way to extend them.
Hell, all the linux zealots are a bunch of hypocrites anyway - all y'all have Win98 on your machines for games. So basically, your ethics can be bought for a video game. Ha.
I don't have Windows anywhere -- five desktop boxes at home, desktop at work and a laptop, all run Linux and FreeBSD (one of desktops is a powermac 8100/80 -- it also runs MacOS as a bootloader for mkLinux). And, yes, I have bought UT, and I am absolutely sure, some Epic bean-counter counted it as a sold Windows version, even though I run it on my Linux box.
Unicode is one of "semi-proprietary" standards -- documents aren't available for free (be it ones from ISO or Unicode), however there is no legal barriers for making an implementation -- just the size of the table makes a job of creating fonts unreasonably huge. OTOH, the tables necessary for determining, what the characters are, are available for free.
The problem however is different -- people already use their own charsets, and those charsets were designed to reflect the structure of language, or just to be most convenient for their language, sometimes made quite different from part of Unicode that is supposed to be used for the same language. If instead of trying to _convert_ everything to Unicode, people adopted a reasonable (iso 2022 isn't reasonable) way to label, which charset and which language are where in their strings, the implementation would be able to use all known charsets, and programs that aren't concerned with operation thats depend on them can just ignore the whole thing and treat text as a sequence of bytes until charset-specific procedures are called to process/display/compare/convert/input/... the text where "real" size and mapping of characters will emerge -- and those procedures can be language-dependent, replaceable and expandable if they will implement an easy mechanism of mapping charset/language names to sets of procedures. Unicode could be used as one of possible charsets, and UTF-8 could be used as one of possible encodings in such a system, however it won't be "the" thing, that everyone is supposed to support and be aware about. At most some programs would have to know how label delimiters look.
It can be a very easy solution for the real problem, however it requires an agreement on how charsets/languages should be labeled (their "real" names should be used to make the thing expandable, however how those things should be separated from "normal" text remains a question).
Well, mebbe not. I am waiting to hear more comments from non-English slashdotters on this subject, the comments so far reflect a definite world view -- the English world.
Non-English slashdotters that at the same time use iso8859-1, most likely see Unicode (or UTF-8) as a good thing because first 256 characters of Unicode are the same as iso8859-1, and they don't give a damn about everything else, while non-English slashdotters that use other local encodings/charsets (like me, whose native language is Russian, with koi8-r as the charset used in unixlike systems) see Unicode as a monstrosity, forced on them by a bunch of dumbasses at Unicode Consortium, ISO and software vendors thart benefit from every incompatibility that can force people to upgrade.
If charset/language labeling was standardized, everyone would be able to use their own charset, and all software that is not directly involved in text editing/displaying would be able to continue working as it was before, however by STUPID decision made by "standard bodies" the priority is given to sticking "should support Unicode/UTF-8" into every standard in the place of "should pass the data as a stream of bytes regardless of the actual size of character, encoding and their possible meaning, except special characters involved in protocol" that would actually accomplish something.
In the Middle Ages in societies with limited literacy, high cost of information distribution and overwhelming politically-backed power of religion science was in the situation, similar to what commercial software is now -- consisted of "secrets" (like modern secret algorithms), obscure terminology (like modern secret data formats) pieces of religious texts and ancient books, quoted everywhere, distorted and taken out of context (like modern botched and mutilated standards and marketing buzzwords), religious "laws" severely restricted what can be done (like modern patent and copyright laws), and were vague enough to support arbitrary witch hunts against anyone (like modern frivolous lawsuits). This allowed "scientists" to produce huge amount of "work" that has little or no relation to reality, and people who practiced activities that required scientific knowledge either did their job poorly because of lack of one (engineering, like modern AI), used huge amount of resources for nothing (architecture, like modern bloatware) or turned their work into complete quackery (medicine, like most of areas of software design now).
Improvements in the information dissemination technology and weakening of the grip of religion over society caused science to become more open, and openness with wide peer review allowed scientific methods (that were known long ago yet were ignored in the absence of openness -- this is why it was called "renaissance" and was percieved by people as return to earlier, better ways of science, engineering and art) to weed out the bullshit and keep the true achievements, hidden under piles and piles of "mental masturbation". The use of this open science in various kinds of trades/businesses improved, so the job of scientist switched from "secret lab" of what amounted of skilled craftsman or "library" of religious philosopher to education institutions (that also sponsored research and books publishing) and "service" work for large engineering projects. Secrecy remained only in military-related projects, and those never amounted to much before being opened or reproduced.
As the result the progress in science went on with reasonable pace, results became openly accessible and reliable (as scientists had to make them possible to reproduce), quackery was reduced to fringes of society (some modern societies have large fringes, however this is a different problem).
When software design, the area that has properties of both theoretical science (practiced by scientists) and engineering (handled mostly by businesses) appeared, more appropriate science-like model of development was very soon replaced by business-driven model of secrets, commonly used for engineering, and it ended up at the same place where science was many centuries ago. And the same processes that opened science will open the software design/development -- the key ingredients such as widely accessible education in the area, means of dissemination of information and large amount of freely-distributed information are there.
IEEE and ACM have documented codes of ethics.
Most of engineers aren't members of either, so they are at most remotely relevant.
I was using the term "ethics" in the sense of "Professional Ethics". As in: ethics n - the principles of conduct governing an individual or a group.
For me no specific "ethical code" can override the general ethics -- if some rules of professional conduct (or "GNU ethics" if such thing ever existed) contradict with general ethics that I accept, following the "specific" rule is unethical, no matter what. While I may be forced to do an unethical thing, it doesn't make it right.
Also I don't see why should I quit a job just because I don't want to follow all the bosses' words -- I am as much part of that work as he is, and if he doesn't quit his job because I disagree with him, why should I do that? There were a lot of situations in my life when managers were wrong, and they acknowledged that my actions were better for the benefit of the company/customers/whatever-thing-they-consider-imp ortant afterward.
I don't think there are clear ethical guidelines in many of these situations. If the boss has stated flatly "NO OPEN SOURCE TO BE USED", then it's clearly unethical.
Why? Ignoring bosses' orders can be ethical in a lot of cases. Company's policy has very little to do with ethics, and it's still a moral choice of a person to follow or to reject it. The key is the responsibility -- if a person can be responsible for his action, does not want to push that responsibility to the boss and can defend his decision, he can do whatever he thinks is better.
The fact of the matter is that it is an INTRINSIC property of open source that is must prove itself above and beyond that of commercial software simply because there is simply no legal recourse for companies who use it
All commercial software is released under licenses that disclaim all responsibility of the manufacturer -- sometimes with the exception of defective media shipped.
You want an example of the opposite? Look at Russia during the communist days. When the iron curtain was lifted, we saw that there wasn't much to fear. Everything in Russia sucked, technology and standard of living were basically below USA. Why?
Economy in Russia started to suck after it was dismantled in the hope of becoming "more free" -- I have lived there and moved to US in 1993, so I can compare the life before and after "perestroika" in Russia, and in US now. The comparison from the point of view of more or less "normal" person is definitely not in the favor of what I see in US now -- it's better than what is in Russia now and definitely better than what american propaganda lets you to see, however I felt better in 1989 in Russia than I do in 2000 in US. It's true that there was little freedom in Russia compared to what I would want, however in US there is even less if I discount as nonexistent the "freedoms" that are taken away by inefficiency of legal system, lobbying and lack of any mechanism that restricts a power of businesses over their employees and consumers. I am probably supposed to be considered "successful" in american society since I have decent job and make enough money for life much more comfortable than one that I care to actually have, however I feel neither "free" nor "happy", and this is what counts in the end.
You could do the same damn thing as a bash script
Unix mailreadres can't execute scripts from attachments.
The file is an ATTACHMENT. In order for it to run, the user has to doubleclick it. It would be like sending a unix user a perl script that had rm -rf ~/* in it.
File with .pl extension and content-type "application/octet-stream" is never executed by any decent mailreader -- mailreader even will warn you that there is no viewer defined for this content-type. If someone had "application/x-perl" in .mailcap pointing to "/bin/perl %s" (mailreaders never write files with executable permissions), AND it was used in the mail header, then and only then it will run, however no sane person will do such a thing and no system comes with this kind of configuration.
It was stupid of them to even think about trying to get away with that service without consulting anyone in the industry first
OTOH, ZicoKnows always consults someone "in the industry" before posting here.
Every idea can be turned into absurd when taken to extreme and beyond. In this example the extreme opposite to yours one will be to ban male masturbation and condoms as it kills "unborn babies" in the form of sperm.
Every idea can be turned into absurd when taken to extreme and beyond. In this example the extreme opposite to yours one will be to ban male masturbation and condoms as it kills "unborn babies" in the form of sperm.
Even better would be a TLD of ".", not "dot", just ".".
There already is -- this is the dot _after_ TLD (traditionally omitted, but is still there).
"screw it! go to H-T-T-P-colon-shash-slash-three-five-two-zero-zero -six-one-four-eight-zero!"
While it would be expensive, it can be done slowly, and the value to every local library in the US would be immense.
In 1990 or 1991 when I lived in Gomel (USSR, then Belarus), all I had were some PC-AT (286) boxes and two 2400 bps modems. I have decided that it sucks that in the whole city (that had about half a million people at the time -- comparable to San Francisco) there is no public-accessible network, and that it will be a good idea to join Fidonet (that was the only thing available without insane fees) as a "point", open a BBS on it and maintain it. I did that, then "point" became a "node", some other people joined, and soon it turned into Net 2:452 of Fidonet. With cheap and primitive equipment, flaky phone lines, DOS software (that I had to rewrite to adapt to conditions where it was sometimes impossible to complete mail packet download between having the carrier lost and redialing for half an hour to some overloaded node), it supported reasonable speed of netmail (Fidonet equivalent of email, with store-and-forward routing) and echomail (newsgroups very similar to early days of Usenet), people started using "freq" (file request, an equivalent of FTP transfer, but as opposed to FTP and HTTP, it was queued as a "special" mail message, so it could be delayed until long distance discount hours, recovered from failure, etc.) to get things on remote servers, send files as "attachments" (as opposed to MIME attacments those were kept in original format, weren't routed and were processed separately from message to preserve precious bandwidth), use search-engine-like "robots", etc. -- all that using few 2400bps modems. The amount of information that can be found by Fidonet wasn't, of course, comparable to LoC, but it made a huge difference in the lives of a bunch of programmers that operated and used that network, including myself. Forgive me, Borland/Inprise/whatever, I had Borland C++ 3.1 available for download -- my excuse is that you certainly didn't lose any profits from people whose monthly salary was equivalent to $30 or less, and the nearest store with your products was way beyond their reach.
It was done by people with no money at the time when Internet (or any network) access really was hard to obtain, and it still worked.
Now almost a decade passed. Neither in Gomel nor anywhere in US it's necessary to deal with 2400bps modems or 286. Even poorest of the poor, if they want to access the Internet, can buy some used 386 box with 4M of RAM ( => capable of running DOS or Linux) with at least 9600bps modem -- I know, I did that in 1994 after arriving here, and I had literally no money then -- it certainly didn't become harder now except that p166's replaced 386dx-25 as the most likely throwavay item. Even the worst school in US with the dumbest technician in the world can find some mac and subscribe to some cheap-ass dialup. And the poorest library still can find 2-3 computers and modems. Heck, a library can put one Linux box and all other boxes can be 286 with NCSA telnet. Someone who wants to get "enterntainment", will be very disappointed by all mentioned solutions, but a person who needs to get some particular book, especially if LoC will make it available in plain text, will find those ways of accessing it through the Internet extremely useful.
So, it's not some "rich elite" that will be able to access digitized books, it's precisely the same set of people, whom libraries are supposed to serve. Library's primary function is not to serve as a place for conversations, waiting, "looking smart" or visiting a public bathroom -- it's to allow people to find and read books, and accessibility through the Internet helps to perform it.
Nope. They are just expanding to places where they weren't present or successful before (like cancer). X-box is a renamed PC, and Pocket PC is a renamed Palm PC that uses their Windows CE, that happened to be an application of Windows design and GUI ideas to completely different idea and market.
Proxiweb is a real browser -- it requests pages through proxy that re-formats them into a format, readable by the client running on the Palm, so even though it doesn't parse HTML on the Palm, result is the same. I use it for things like searching for products specs while at the store (which of cards is supported under Linux?, what chipset is in this thing?, etc.), maps from mapblast.com (using coordinates of the nearest retransmitter taken from Metricom modem), etc.
Seeing how the Palm does great Wireless networking (802.11b) -- (yeah, right) and how great the Web capabilities are, I think one offers more than the other.
Wireless Ethernet works only for few hundreds feet -- it means that you are in the same building with your network, and you need such a speed only to so something time/resource-consuming anyway, so for things like that small laptop/notebook will be a better choice, (I have vaio n505vx with wireless ethernet). However if you need to do something remotely where "remote" is another building (or city), it won't be of any use -- you need a cellular or cellular-like radiomodem. All versions of Palm support those things pretty well (VII has builtin wireless networking, but I prefer III{"",x,c} with faster Metricom modem) -- and when I need to use a notebook at some remote location I can just connect the same Metricom modem to it.
Another problem with wireless networking is power that it requires. Those things eat a lot, and unless separate power supply is used for them, it has a good chance to drain the battery very fast. PalmVII and everything that involves external radiomodem use separate battery for wireless networking, but I am not sure that WinCE/Palm PC/Pocket PC/whatever have enough space inside for a separate battery that powers PCMCIA/CardBus cards -- so the only real alternative to keeping wireless networking disabled is having an external power supply attached. Oops, that completely negates the point of having a PDA with wireless networking, and it's easier just to carry a notebook that has big enough battery to support radio Ethernet.
gcc can be built as crosscompiler on a platform that has a bunch of unrelated to gcc compilers, build itself (and libraries that it uses) for another platform and then used on another platform to build itself. Truly paranoid people can build it on many different platforms and compare the result of first self-built version -- if it's not identical, some versions are infected. The only case when it won't work is when all compilers involved are infected, and infection of all of them affects gcc in the same way -- something that I find hard to believe.