Well, if you are trying to use emacs with the Control key in the wrong place, you will certainly have issues. That's why almost all emacs users with lame keyboards remap the Caps Lock key to Control, as God meant it to be. This greatly cuts down on repetitive stress, and additionally eliminates the nuisance of typing something in ALL CAPS. Of course, with emacs fixing something in caps does not involve deleting it, as with lesser editors and word processors, but simply typing M-c to capitalise it or M-l to lowercase it.
Well, I don't believe any other editor is as extensible as emacs--not even Eclipse. Lisp (in this case, elisp) is an ideal language for interactive development, and thus extending emacs is very straightforward. There's an easy learning curve as one goes from setting a few variables to writing some functions to eventually creating one's own editing modes (I'm in the process of writing one to manage Blosxom blog entries).
Emacs also runs the same way, with the same keybindings, on every platform: Unix, Windows, Mac OS--anywhere it's been ported. It even runs both in GUI and console modes. This latter is very convenient when SSHed into a remote host. The fact that the keybindings are identical everywhere (and that one can customise them to one's taste) means that one really doesn't care so much about the host OS; emacs can make Windows feel as though it's actually a real OS.
Now, you raised the issues of keybindings: certainly, the defaults are not those nowadays used. But remember that emacs has been around longer than any other OS's user guidelines, and that its target audience is cross-platform. We're quite happy to use C-s to search forward and C-r to search backwards, and we like the fact that we have more ways to cut & copy than any other editor--because cutting & copying is so common that there really should be some customised ways to do so. We can cut the current region (like a selection); we can cut to the end of the line; we can cut the current rectangle (useful when dealing with tabular data); we can add this cut or copy to the current 'clipboard'; we actually have multiple 'clipboards,' known as the kill ring; we can cut or copy the next paragraph or word, or the previous paragraph or word; and so on.
Emacs is highly specialised for editing text of all kinds: source code; documents; email; news articles; data entry forms, whatever. It's really nice not to have to relearn different ways of doing each of these tasks.
Ah, but Eclipse is unfortunately written in Java, a language which lacks the hacker nature. It's also GUI-only, which means it cannot be run across an SSH connexion or in low-resource situations. Finally, writing a web browser, mail client or news reader for Eclipse would be nowhere near as easy as using w3m and gnus for emacs. Why does this matter? Because it's nice to be able to use the same keystrokes for reading & editing text everywhere; those two tasks are most of what most of us use computers for, and having the exact same interface everywhere rocks.
Except of course that it's almost certain that it would be a heavily anti-Christian course, and that due to the First Amendment the State may neither promote nor disparage any religion--and this is a State-run school...
And thanks to the Electoral College, if you don't vote the same way the majority of the people in your state did, then your vote is rendered irrelevant in the federal election. You don't win, and thus, the ultimate outcome is just the same as if you had not voted at all; you could've voted for Daffy Duck or the Unabomber, and either way, it'd be the same as voting for a candidate that lost - mainstream or not.
That's not much different from saying that in a popular-vote system one's vote doesn't count if one's side loses. Remember, in the US it's the states and the people which vote for the president.
IMHO the best way to handle the electoral votes would be for the state legislatures to assign two votes, and give the other votes to the winner of the state popular vote. But that's just me. Each state should be (and I think still is) allowed to select electors however it likes.
M-x shell is nice, but it doesn't provide a real terminal to the client apps. M-x term does, but it necessitates using funny key combos to talk to normal emacs stuff (C-c before any regular emacs command IIRC). Both are highly useful, for different purposes.
That doesn't make much sense in the context of this project. if the goal is to help people - why put this software ideology and zealotry ahead of the wants or needs of users?
Perhaps they want the third-worlders to be able to hack their own OS. They're not fools; many are quite intelligent and might really enjoy being able to write their own software.
Pity the article didn't mention any members of the blosxom family. This family is distinguished by using the filesystem as its article store, rather than an SQL database; also it's known for the extensive use of plugins to provide features like comments, SQL databases, calendars and so forth.
I started using blosxom for Octopodial Chrome several years ago and have been very happy with it. Besides the original perl blosxom, there's PyBlosxom (a port to Python) and my own pre-alpha Lisp Blosxom. This last is a port to Common Lisp; it doesn't work yet, but someday I hope that it'll be pretty nice.
Well, a lot of it is culture--you absorb this stuff over time.
I'm not sure what you mean by asking 'how does Unix boot?'; on a PC the BIOS boots Linux or FreeBSD essentially the same way that it boots Windows: searches the known drives for a boot loader and runs the first one it finds. Of course, the Unix boot loaders typically offer more features than the Windows boot loader.
Likewise with most of the other questions, actually. Perhaps if I grokked what you're asking better I would have a better idea of how to answer.
I realise that the odds aren't really that astronomical--didn't you see that part?
And as for how difficult it is for an app to roll its own protocols on top of IP, that's kinda my point. Why should HTTP need to roll its own verification protocol on top of TCP/IPv6, and SMTP its own, and FTP its own, and SSH its own, and so forth, when TCP/IPv6 could provide the same?
The justification I saw was that 'everyone's doing it already.' Well, right now we have three or four levels of error-checking--Ethernet, IP, TCP and possibly application-level--and errors still get through, despite the fact that four different checksum algorithms must all fail. Does it make any sense at all to remove two of those, thereby increasing the number of undetected errors exponentially?
If anything, we need to introduce more error-checking as part of the underlying comms protocols. Another byte per packet, along with a bit more CPU, really isn't that high a price to pay, esp. now that more and more important data are transmitted than previously.
The reason that we manufacture stuff in one country and ship it to another is that the first country is better at manufacturing whichever good is under discussion. Generally this is because of cheaper labour, because First World labour is insanely expensive--and not terribly productive either. However, it can also be that the first country has ready access to raw materials or somesuch.
The key, of course, is for the First World to try to be more competitive. Simple steps like reducing taxes at all levels would go a long way in this direction, without affecting quality of life. Reducing regulatory overhead would also do a good deal. Of course, both of those approaches also have downsides: in the first case, reduced State services (but First World states often provide too much anyway, so maybe this is a good thing); in the second, increased risk of the kind of troubles addressed by regulation. We would need to weigh the options: would lower prices and better competitiveness be worth the decreased services and increased risk? That's a judgement call.
In the meantime, though, First World labour is migrating to areas where it is efficient despite its mind-boggling expense. Those areas are often things like the IP realm, where even at $200/hour and hundreds of hours the costs can be recouped fairly quickly. E.g. pharmaceuticals (unbelievably expensive to develop), software, music, patents, entertainment and so on.
Which isn't to say that all of these IP realms are necessarily good. For my own part, I think that pharmaceutical patents are a good thing: give a company incentive to develop a drug, and after some number of years give everyone the rights to the drug. Yeah, for those intermediate years some people who cannot afford the drug won't get it--but without the drug's development, no-one would get it.
For software, I think on the balance copyright is good but patents are bad. Copyright is what gives things like the GPL its teeth--without copyright, the GPL would be the BSD license, and vendors would just provide binary-only software. And software is expensive to develop; developers should have the right to choose to contribute their source to the commons (the moral choice) or to keep it proprietary (selfish, but understandable--a man has a powerful need to eat). But software patents serve no discernable need: by seeing the software, one knows what it does as well as by reading the patent app (that is, a software patent provides none of the insight into design that a traditional patent offers), and the patents serve no purpose in fostering creativity.
For music, I think that copyright could stand to be relaxed a great deal. I can see a reason for artists to get 1-6 years of copyright protection, but after that let others make use of that material. Music is very much a fashion thing, and the music of six years ago isn't necessarily that popular. Maybe provide a few (somewhat expensive, perhaps taxed) renewals of the term. Live music should be encouraged, and recorded, engineered music not so much so. Musicians will also more typically be driven to make music, even in their free time; they lack the need for encouragement that a soul-less corporation has.
For books, I can see a 6-12 year term, possibly with a renewal or two--the current practically-until-the-end-of-time term is absurd. Much of our arts are founded upon what has gone before, and the current regime is getting ridiculous. The publishing industry serves a valuable purpose (that of a garbage filter), and we need a structure in which it can be rewarded, but books don't need to be copyrighted for decades.
I can see a fairly long term for film, about as long as books--although there are some arguments for slightly shorter or slightly longer terms. Films are extremely expensive (big films cost about what new drugs cost, IIRC: in the hundreds of millions), and they may require a fair bit of time to make a profit.
IP is not bad; the problem is not with IP but with the lack of consideration given to the subject. Moreover, the IP producers perpetually petition for extensions of their monopolies while consumers remain relatively silent.
I've always thought that eliminating checksums was a major mistake. Even today I sometimes get corrupted files, despite the several layers of checksums involved (Ethernet, IP & TCP level, IIRC)--one would think that the odds are astronomically against it, but it happens (probably because the average computer user sends and receives an astronomical number of bits every year). IIRC, the assumption is that each app would roll its own checksumming. Well, each app could roll its own connection protocol over IP if it wanted--but it makes sense to provide a standard TCP. Likewise, it makes sense for apps to be able to treat their pipe as error-free because the underlying protocol handles checksumming in a standard way.
It's been aeons since I've played with any of this, so perhaps my objections have been addressed; I'd love to hear so.
They weren't atheists--they were Christians. Turkey (along with much of the Moslem world) sits on what used to be the heart of Christendom: the Byzantine (or Eastern Roman) Empire. I don't know if Turkey was 50% Christian fifty years ago, but 70-80 years ago it had a large Christian population. Those people have been exterminated or left, due to persecution. Not that they were entirely blameless--the Greeks started a war to take back all of Anatolia in managed to lose the little bits they already had.
A smuggler isn't a pirate--he's a smuggler. Pirates attack opther ships, board them and steal from them (generally also killing passengers & crew). And we need to revoke the treaty which forbids commercial ships from carrying heavy weapons. A rocket into the pirate boat would do a lot better good than a little smoke-alarm noise.
Re #2: No argument there, the information that comes with a well-documented UNIX is the best way to achieve wizard or guru knowledge levels. Not quite so well suited for getting as far novice, though: A lot of the man pages - at least back in the day - were written by experts who assumed the reader was close to expert, or at least was a C coding system hacker. Like it or not, not all of us were. C coder? Yes. Sysadmin? Eventually? Kernel hacker? Nope. Library hacker? Only at gunpoint. Shell hacker? Oh, yes, please anytime.
I disagree, actually. They weren't written for experts, but they were written for folks willing to spend a fair amount of time learning. I was 13 when I started using Unix in the lab at ODU, and I didn't have much of a problem: I just read the man pages, and read 'em again (that was back when more couldn't back up a page, and so if I wanted to read the previous page I had to quit and start all over again), and read 'em yet again until I began to figure 'em out.
I'm a Unix admin by trade now, so something must've stuck...
And of course ultra-long-term planning has worked wonders for the Japanese economy...
Not that more long-term planning wouldn't help us. E.g. Social Security.
I remember in college reading up on von Tirpitz and discovering that the Imperial German Navy had plans for ship construction up to the 1970s. Think about that for a second.
The folks who stick with Lisp pretty quickly learn how to handle things such that the parens are No Big Deal. And what they buy is a pretty big deal: macros. Not C/C++ can't-use-em-for-anything macros, but real honest-to-goodness macros which let one create one's own syntactic structures. Very cool stuff.
Well, a lot of us are consistent in our view: we decry the murder involved in stem cell research, and we decry the murder involved in 'throwing away' unwanted children. There are even some who will adopt the unwanted, have them implanted, give them birth and raise them as their own.
Actually, there's a religious argument that an embryo does not have a soul (essentially, it boils down to the fact that it can split into identical twins, but that a soul cannot be split, and thus that an early-term embryo cannot have a soul), but the scientific evidence is quite clear: from the moment of conception there is a new member of homo sapiens. Kinda amusing that this is the one case where religion might be less didactic than science.
You know, I actually enjoyed The Avengers. But then, I'm a big fan of Uma Thurman. Still, I thought it was aperfectly decent bit of over-the-top action/spy/goofy fluff.
Slashdot Mirror
Well, if you are trying to use emacs with the Control key in the wrong place, you will certainly have issues. That's why almost all emacs users with lame keyboards remap the Caps Lock key to Control, as God meant it to be. This greatly cuts down on repetitive stress, and additionally eliminates the nuisance of typing something in ALL CAPS. Of course, with emacs fixing something in caps does not involve deleting it, as with lesser editors and word processors, but simply typing M-c to capitalise it or M-l to lowercase it.
Emacs also runs the same way, with the same keybindings, on every platform: Unix, Windows, Mac OS--anywhere it's been ported. It even runs both in GUI and console modes. This latter is very convenient when SSHed into a remote host. The fact that the keybindings are identical everywhere (and that one can customise them to one's taste) means that one really doesn't care so much about the host OS; emacs can make Windows feel as though it's actually a real OS.
Now, you raised the issues of keybindings: certainly, the defaults are not those nowadays used. But remember that emacs has been around longer than any other OS's user guidelines, and that its target audience is cross-platform. We're quite happy to use C-s to search forward and C-r to search backwards, and we like the fact that we have more ways to cut & copy than any other editor--because cutting & copying is so common that there really should be some customised ways to do so. We can cut the current region (like a selection); we can cut to the end of the line; we can cut the current rectangle (useful when dealing with tabular data); we can add this cut or copy to the current 'clipboard'; we actually have multiple 'clipboards,' known as the kill ring; we can cut or copy the next paragraph or word, or the previous paragraph or word; and so on.
Emacs is highly specialised for editing text of all kinds: source code; documents; email; news articles; data entry forms, whatever. It's really nice not to have to relearn different ways of doing each of these tasks.
Ah, but Eclipse is unfortunately written in Java, a language which lacks the hacker nature. It's also GUI-only, which means it cannot be run across an SSH connexion or in low-resource situations. Finally, writing a web browser, mail client or news reader for Eclipse would be nowhere near as easy as using w3m and gnus for emacs. Why does this matter? Because it's nice to be able to use the same keystrokes for reading & editing text everywhere; those two tasks are most of what most of us use computers for, and having the exact same interface everywhere rocks.
Except of course that it's almost certain that it would be a heavily anti-Christian course, and that due to the First Amendment the State may neither promote nor disparage any religion--and this is a State-run school...
Funny, then, that you picked on tobacco earlier in your post. The Frogs smoke like chimneys...
Not that smoking is good for one in general. A nice pipe sure is tasty, though!
Oh, get a grip--why on earth would they fund someone who's competing with them? That makes no sense whatsoever.
Don't forget Space:1889, the game of Victorian space exploration. Man did I love that one...
Except of course they're not in jail because of this. The US isn't nearly so eeevil as some think. We're imperfect, of course.
That's not much different from saying that in a popular-vote system one's vote doesn't count if one's side loses. Remember, in the US it's the states and the people which vote for the president.
IMHO the best way to handle the electoral votes would be for the state legislatures to assign two votes, and give the other votes to the winner of the state popular vote. But that's just me. Each state should be (and I think still is) allowed to select electors however it likes.
M-x shell is nice, but it doesn't provide a real terminal to the client apps. M-x term does, but it necessitates using funny key combos to talk to normal emacs stuff (C-c before any regular emacs command IIRC). Both are highly useful, for different purposes.
Perhaps they want the third-worlders to be able to hack their own OS. They're not fools; many are quite intelligent and might really enjoy being able to write their own software.
I started using blosxom for Octopodial Chrome several years ago and have been very happy with it. Besides the original perl blosxom, there's PyBlosxom (a port to Python) and my own pre-alpha Lisp Blosxom. This last is a port to Common Lisp; it doesn't work yet, but someday I hope that it'll be pretty nice.
I'm not sure what you mean by asking 'how does Unix boot?'; on a PC the BIOS boots Linux or FreeBSD essentially the same way that it boots Windows: searches the known drives for a boot loader and runs the first one it finds. Of course, the Unix boot loaders typically offer more features than the Windows boot loader.
Likewise with most of the other questions, actually. Perhaps if I grokked what you're asking better I would have a better idea of how to answer.
And as for how difficult it is for an app to roll its own protocols on top of IP, that's kinda my point. Why should HTTP need to roll its own verification protocol on top of TCP/IPv6, and SMTP its own, and FTP its own, and SSH its own, and so forth, when TCP/IPv6 could provide the same?
The justification I saw was that 'everyone's doing it already.' Well, right now we have three or four levels of error-checking--Ethernet, IP, TCP and possibly application-level--and errors still get through, despite the fact that four different checksum algorithms must all fail. Does it make any sense at all to remove two of those, thereby increasing the number of undetected errors exponentially?
If anything, we need to introduce more error-checking as part of the underlying comms protocols. Another byte per packet, along with a bit more CPU, really isn't that high a price to pay, esp. now that more and more important data are transmitted than previously.
The key, of course, is for the First World to try to be more competitive. Simple steps like reducing taxes at all levels would go a long way in this direction, without affecting quality of life. Reducing regulatory overhead would also do a good deal. Of course, both of those approaches also have downsides: in the first case, reduced State services (but First World states often provide too much anyway, so maybe this is a good thing); in the second, increased risk of the kind of troubles addressed by regulation. We would need to weigh the options: would lower prices and better competitiveness be worth the decreased services and increased risk? That's a judgement call.
In the meantime, though, First World labour is migrating to areas where it is efficient despite its mind-boggling expense. Those areas are often things like the IP realm, where even at $200/hour and hundreds of hours the costs can be recouped fairly quickly. E.g. pharmaceuticals (unbelievably expensive to develop), software, music, patents, entertainment and so on.
Which isn't to say that all of these IP realms are necessarily good. For my own part, I think that pharmaceutical patents are a good thing: give a company incentive to develop a drug, and after some number of years give everyone the rights to the drug. Yeah, for those intermediate years some people who cannot afford the drug won't get it--but without the drug's development, no-one would get it.
For software, I think on the balance copyright is good but patents are bad. Copyright is what gives things like the GPL its teeth--without copyright, the GPL would be the BSD license, and vendors would just provide binary-only software. And software is expensive to develop; developers should have the right to choose to contribute their source to the commons (the moral choice) or to keep it proprietary (selfish, but understandable--a man has a powerful need to eat). But software patents serve no discernable need: by seeing the software, one knows what it does as well as by reading the patent app (that is, a software patent provides none of the insight into design that a traditional patent offers), and the patents serve no purpose in fostering creativity.
For music, I think that copyright could stand to be relaxed a great deal. I can see a reason for artists to get 1-6 years of copyright protection, but after that let others make use of that material. Music is very much a fashion thing, and the music of six years ago isn't necessarily that popular. Maybe provide a few (somewhat expensive, perhaps taxed) renewals of the term. Live music should be encouraged, and recorded, engineered music not so much so. Musicians will also more typically be driven to make music, even in their free time; they lack the need for encouragement that a soul-less corporation has.
For books, I can see a 6-12 year term, possibly with a renewal or two--the current practically-until-the-end-of-time term is absurd. Much of our arts are founded upon what has gone before, and the current regime is getting ridiculous. The publishing industry serves a valuable purpose (that of a garbage filter), and we need a structure in which it can be rewarded, but books don't need to be copyrighted for decades.
I can see a fairly long term for film, about as long as books--although there are some arguments for slightly shorter or slightly longer terms. Films are extremely expensive (big films cost about what new drugs cost, IIRC: in the hundreds of millions), and they may require a fair bit of time to make a profit.
IP is not bad; the problem is not with IP but with the lack of consideration given to the subject. Moreover, the IP producers perpetually petition for extensions of their monopolies while consumers remain relatively silent.
It's been aeons since I've played with any of this, so perhaps my objections have been addressed; I'd love to hear so.
They weren't atheists--they were Christians. Turkey (along with much of the Moslem world) sits on what used to be the heart of Christendom: the Byzantine (or Eastern Roman) Empire. I don't know if Turkey was 50% Christian fifty years ago, but 70-80 years ago it had a large Christian population. Those people have been exterminated or left, due to persecution. Not that they were entirely blameless--the Greeks started a war to take back all of Anatolia in managed to lose the little bits they already had.
A smuggler isn't a pirate--he's a smuggler. Pirates attack opther ships, board them and steal from them (generally also killing passengers & crew). And we need to revoke the treaty which forbids commercial ships from carrying heavy weapons. A rocket into the pirate boat would do a lot better good than a little smoke-alarm noise.
I disagree, actually. They weren't written for experts, but they were written for folks willing to spend a fair amount of time learning. I was 13 when I started using Unix in the lab at ODU, and I didn't have much of a problem: I just read the man pages, and read 'em again (that was back when more couldn't back up a page, and so if I wanted to read the previous page I had to quit and start all over again), and read 'em yet again until I began to figure 'em out.
I'm a Unix admin by trade now, so something must've stuck...
Not that more long-term planning wouldn't help us. E.g. Social Security.
I remember in college reading up on von Tirpitz and discovering that the Imperial German Navy had plans for ship construction up to the 1970s. Think about that for a second.
Wonderfalls was great--I've been loving it from Netflix. I can't imagine why it wasn't given a chance. Sigh.
And in fact can add just about any paradigm you like onto the base language. It's pretty cool.
The folks who stick with Lisp pretty quickly learn how to handle things such that the parens are No Big Deal. And what they buy is a pretty big deal: macros. Not C/C++ can't-use-em-for-anything macros, but real honest-to-goodness macros which let one create one's own syntactic structures. Very cool stuff.
Actually, there's a religious argument that an embryo does not have a soul (essentially, it boils down to the fact that it can split into identical twins, but that a soul cannot be split, and thus that an early-term embryo cannot have a soul), but the scientific evidence is quite clear: from the moment of conception there is a new member of homo sapiens. Kinda amusing that this is the one case where religion might be less didactic than science.
You know, I actually enjoyed The Avengers. But then, I'm a big fan of Uma Thurman. Still, I thought it was aperfectly decent bit of over-the-top action/spy/goofy fluff.