Let's have a look at what is done in its country of origin (Denmark, for the uninitiated): we say "lego brick" or "lego piece" (I'm not sure which translation is most accurate), and "lego bricks" or "lego pieces".
You might say "some lego" (as in a collection of lego pieces), just as you might say "some pot" (as in a collection of some--most often weight--unit of pot)--it's uncountable.
"Lego", when said on its own, refers to the company that produces those wonderful toys.
"Legos"? That's like saying "Linuxes" (or "Linucies", or "Linuces", or "Linuxen", or...) when referring to multiple identical installations (and not referring to what it really means: multiple distros; ex: "I don't care which of the linucies you like, be it Mandrake, or SuSE, or Debian GNU/, or...").
its kinda anying ot see all those headlines: "license held up in court". A license must be like "and you allow me to eat your children" to be not upheld in court. A owner of the copyright can create what ever license he wants and it will allways be valid in court!!!
Sure, it's about as groundshaking as "`true' didn't segfault", which only happens if the program says something LIKE { char *p = 0; *p = 0; } (for interesting implementations of LIKE).
But, and here's the difference, this is (as far as I know) the first time a `hippie underground' license has been tested in court. That's like running your own kernel for the first time: even if it just does {cli; a: jmp a;}, it is interesting.
But I do feel that the artists (...) have a right to be compensated.
Oh absolutely. I'm all for compensating the artists. However, very little of the retail price for a CD is spent on compensating the artist, and a lot of it is spent rewarding the RIAA for tyranizing the population. That, I think, is wrong. I'd gladly donate a dollar here and there for the songs I love, if I'm confident that the payment system is reasonably secure, and that three nines or more of the money goes to the donatee.
Don't try to justify your theft (and yes, it is theft) by saying that I'm an idiot for paying for my music. Don't worry, it's not theft, so there's nothing to justify, and you're not an idiot for paying for your music, you're an idiot for paying money to be oppressed by the RIAA.
Anyways, let me explain: If I copied your rhetoric, you would not lose your ability to use it. If I copied all your CDs, you would not lose the raw materials, or the ability to listen to them. If I copied your car (don't ask), you wouldn't lose your car.
It is NOT theft (and I'm sick and tired of people calling it that), it's copying. We as a society (I'll only speak for "my"self, and that's Denmark), had at a time come to the conclusion that giving up our right to copy information (since we were unable to, by virtue of not owning printing presses and CD toaster) would be a good trade, since more works would be produced. Now, the assumptions are wrong: we are able to copy information, almost effortlessly.
I'm just saying that this whole notion that music should be free simply because it can easily be pirated anyway is stupid. So, if we could copy food limitlessly, we shouldn't let the hungry people have it, because the food-producers need to be compensated?
Yes, it is a valid analogy: copyright creates an artificial limit on the copying on music, the above example is an artificial limitation on the copying of food.
Maybe an economical argument can convince you: once a piece of music is on the internet, supply is infinite. In a free market, that would imply that the prices ought to be zero(*). However, copyright law and the RIAA artificially (try to) limit the supply, thereby artificially keeping the price up.
(*) Note, though, that even if the price is zero, that doesn't prevent people from donating money; surprise: some people do that.
It's actually very simple to set someone up a Linux desktop account, show them their email and browser, and just let them use it.
I haven't tried that, but I'm willing to take your word for it--it's not like it's difficult to use, it's just different. And if the users have a sysadmin, that ought to make it workable for them. So... yeah, perhaps that is what I should do to my mom...
RMS says that he choose Unix because he knew that hardware would be much different by the time his OS was complete, and Unix was the most portable OS of the time.
That too! Perhaps that's just a function of building the right amount and kind of layers of abstraction over the hardware? I'm not sure, but as you point out, the designs converge with "some kind of unix" as the limit. Is that because of the technical superiority of unix, or the economical superiority of unix? (go read "worse is better" and "unix-haters handbook" if you haven't already) I would think that it's actually because of the economic superiority; sure, unix is also a great system in many technical ways (sure as hell it beats windows), but by virtue of being built out of small idependent components, almost all system properties and system-wide intergration efforts are belong to the kernel;
Then again, limiting monolithicity to the kernel is much better than unlimited monolithicity. To paraphrase Churchill: "Unix is the worst OS ever, except for all the other OSes";)
[...] systems like Windows, which promote the idea that end-users can administrate computers.
Right on target! Windows (falsely) promotes the idea that end-users (aka. the Joe Sixpaxen of the world) can admin computers.
There are two ways around this: one is to alleviate most needs for administration--i.e. "Just Work"--and the other is to create a high enough barrier to entry that only reasonably competent people will run the system in the first place. Let's call them the "OS X way" and the "GNU/Linux way".
sidetrack {
I'm all for making GNU/Linux accessible to the end-user. In fact, I think GNOME is doing great on this aspect. Unfortunately, it does so by "dumbing down" the interface. I could probably live with that, if I had the option of editing ~/.gnome/settings/input/keyboard/keybindings/deskt opbindings/obscure-setting.conf to bind "C-t a M-x ESC F4 # q" to "maximize window, then move it to the next workspace if that workspace is empty; otherwise, blink the screen and run the command 'uptime >> ~/uptimelog'". However, gnome consistently removes those options, much to the frustration of power users.
It's not like I crave for customizability, I'm willing to accept reasonable defaults for keybindings (case in point: I don't even remap my vi keys even though I use the dvorak keyboard layout, which is y/hjkl/jcvp/). However, for a feature I really want, "" is not a reasonable default--I need to be able to change that. Gnome won't let me. (and it's not only keybindings, although that's what sprung to mind right now--I'm sure you can find other examples if you use or dislike gnome enough).
However, gnome is not the OS (...)
sidetrack {
I define "OS" the Stallmanian way, meaning that GNU/Linux is an OS and Linux is a kernel. If you define an OS the Tanenbaumian way (I name it that way because it was in OSes:D&I I first encountered the other definition), deal with it for now. I'd like to discuss that, but it's a different issue.
Off stage: GET ON WITH IT!!!
}
(...) which is what saves it: due to it not being the OS, it can't get in the way of me doing the things where "sorry, we won't let you do that" would reeeeally hurt (managing (grub|xorg|modules|ld.so).conf, setting the MTU, iptables,... basically, all the down-and-dirty low-level hacking).
That is, GNOME doesn't prevent you from doing any sysadminly things, it only prevents you from doing some userly things. Even if I didn't have an alternative, I'd probably be able to accept it. Luckily, I do have an alternative (if you really care--why would you?--I use fluxbox).
Windows (all of them?) also dumbs itself down. Sure, it may make easy tasks easy (wget -qO - http://www.joelonsoftware.com/printerFriendly/uibo ok/fog0000000249.html | grep excel | grep lists--okay, it's for office, not windows, but you get the idea), but it also makes hard tasks impossible (I recall hearing that setting the MTU is impossible, and that the default of 1500 is really asinine, due to it being larger than 1492).
Disclaimer: Modifying the registry can cause serious problems that may require you to reinstall your operating system. We cannot guarantee that problems resulting from modifications to the registry can be solved. Use the information provided at your own risk.
Now that made me smile.
}
Anyways, I guess I can't prove my point about windows, so I'll resort to vigorous gesticulation (handwaving in particular): I feel antagonized when
"Don't do that, then" is not helpful advice to the people who are suffering from this attack.
That's not what parent said. What he said was (depending on exactly what you mean by your analogy) "don't let him do that then" or "have the third party that lets that guy over there do that stop letting him do that then". Okay, so if it involves the third party, the attackee doesn't have much they can do to defend themselves except asking the third party to do the Right Thing, but the doctor is equally out of options.
is it really necessary to sacrafice teh productivity of a standard keyboard in order to gain a convenient, compact form factor?
Uhmm.... yeah, right. Like the standard keyboard is worth more than a dime-a-dozen.
I'm serious: I treated myself with a Kinesis Ergo Elan (http://www.kinesis-ergo.com/elan.htm), and I'm really glad I did--it's a really pleasant keyboard to work with. I only noticed when I first used a standard keyboard after adapting to using the Ergo Elan--not only did it feel uncomfortable, it felt painful. I'm serious--and it wasn't just because I had adjusted to something different (if it were, I should have felt pain when switching to Ergo Elan, something I did not). Also, one thing that completely confounds me is why modern keyboards still have stagger--it's just being mindlessly carried along from the 1890's up until today (sure, it was reasonable due to mechanical constraints on typewriters; newsflash: computer keyboards isn't a new invention).
(End of rant about the sad state keyboard design is in these days)
Actually, I think the AlphaGrip might be better than a (physical) keyboard layout designed for typewriters. I'm open to the possibility that I might be wrong (in general I dislike small keyboards, but this one might have been done well).
I hope you don't argue that the usefulness of the standard keyboard layout is high just because it's a standard; if that was true, Windows 95 was once a good OS, and Bush was a good president (at election time, he was the "standard" choice).
Also, the keyboard permutation matters. I'll replay this anecdote (I'm not playing any role, and I can't remember who participates in it, and I'm not even sure I got the phrases exactly right, or that I've go the gender information right, but... oh well, someone once said something to the effect of...):
Alice: "Excuse me. If you don't mind me asking, are you using that alternative keyboard layout?"
Bob:: "Yes--it's called Dvorak. How could you tell?"
Alice: "Your fingers aren't moving."
I like the promise that the virtual keyboards have It's not a bad idea (at all), but it's amazing that they haven't shuffled the keys around (okay, so I was almost done ranting about keyboard design): it must be almost trivial to do--all there is to it is basically detect surface contact in a different polygon.
Give a man a fish, he'll eat for a day. Try to teach a man to fish, and he'll complain that "I DON'T CARE!! You take take of catching fish! I mean, you're the sysadmin!"
Now I can boot a non-free and a non-free OS (I will tell you which is which: windows is the non-free one, and OS X is the non-free one).
Of course, if it means moving users away from (maybe even out of) the Microsoft Monoculture (tm), substituting one non-free OS for another one is just a sideways step, not a step away from freedom. Plus, even if people still use Losedose, using OS X on the side may make a few of them competent around a "real" unix system; that is to say, some of them might learn to use the terminal and cd, cp, ls, mv, seq, wget,.... I'm not getting my hopes up, though: when I ran a terminal on my neighbour's mac, she said "What's that?"--which really shows that Apple has done well (at least on the usability and "just works" part).
Erm... I lost track of what I meant to say. Oh yeah, I remember: [unethusiastically]Wow, this matters a lot to me[/unenthusiastically]
Who'd have the stress of a brain surgeon if you could have the same lifestyle cooking burgers?
Obviously people who think brain surgery is intrinsically rewarding!
It's like saying "Who'd have the stress of being a Gentoo user if you could have the same lifestyle using Debian?" I believe gentoo users waste a lot of resources on compiling things, but obviously most gentoo users disagree--so instead of saying "gee, using gentoo is really stupid!" I say "gee, I'd really like to understand--why do you use gentoo?"
Perhaps you could ask some brain surgeons that have tried flipping burgers (so they seen both sides of the deal) if they'd rather flip burgers if the salaries were the same. I predict that most of them would say "hell no!".
As much as one might agree with RMS and prefer the term "free software" to "open source" (I do), one must admit that ESR got his marketing skills worked out (at least for advocating free software).
One important point that he makes, that you apparently miss (and I'm convinced it's true): you want to emphasise risk, not cost.
(which doesn't ship with mplayer or mp3 support, fer christsake)
Define `ship'. mplayer is in contrib, so it's definitely in the repositories--you just have to set up apt properly.
Oh, and yeah, there's no mp3 support. totem-(gstreamer|xine), vlc, mpd, playsound, xmms, and rythmbox do not exist, despite me believing I have played mp3s with every one of them on my debian "out-of-the-post-installation-box" box. I admit I have no first-hand experience with amaroK, juk, noatun or zinf, but I guess they play mp3-files just fine.
As [Score:-1,] Has Been Said, it might just be because you're not using the non-US repositories. I assume they're there so that debian can avoid breaking the law.
So this means that you're either (1) blaming yourself for not editing your sources.list; or (2) blaming debian for not breaking the law.
Do you also blame microsoft for not breaking the l... oh, never mind;)
Hmm... I guess I came off as rather intending to start a (holy|distro|flame) war. Not intended. I'm just trying to say that you should do your research before you judge any distro. If you need help with configuring debian, I'm sure there's plenty of help available at http://www.linuxquestions.org/.
Good question. The theoretical answer is that one can't. I'll replicate the proof I prepared for my Computability and Logic exam.
I assume you know what turing machines, hoare triples are, and what recursive and recursively enumerable means.
Assumption: there exists an encoding of turing machines, such that the language of all turing machines is recursively enumerable (this is fairly easy to show, but I'll skip it).
The problem is to show that there doesn't exist any turing machine which will take any hoare triple (pre, code, post), and show that under the precondition, running the code enters a state satisfying the postcondition.
I want to show that the `semi-negation' of the problem--that is, proving programs to be incorrect--is recursively enumerable. Consider the hoare triple (true, anything, false). That is, the program is incorrect iff it halts. So any machine solving the incorrectness problem must--among others--solve the halting problem. But we know the halting problem to be recursively enumerable but not recursive. Hence incorrectness is recursively enumerable, but not recusive (this is actually inspired by Dijkstra: "by testing a program, you can only prove the presence of bugs, not the absence").
But then the complement--that is, ('the code part is a turing machine' AND `the hoare triple is correct')--can not be recursively enumerable.
That is, assuming turing machines are as powerful as general-purpose computers, there is no program which will verify correctness of *any* hoare triple.
However, there might be programs that catch most of the typical mistakes, which would of course be very valuable.
Also, there's the whole question of counting lines (as has been pointed out is the not-a-dupe-at-the-time thread).
Also, a very interesting question is whether or not the human brain is more than turing complete--that is, can humans every prove any program to be correct? Nourishing substances for activities in the frontal lobes.
Of coruse, then the new question becomes how to manage PKI in a system which is understandable and easy for the entire population. (I'm sure a system can be made though.)
Aunt Tillie (to her nephew Mervin): "But it won't install the smiley icons if I click no in the pop-up thingie... security something, it says".
Heck, even I don't read all the "do you want to trust this signature?" popups that firefox shoves into my face. It'd take a lot of education to make that work at what I'd call an accptable level. Also, some people actively resist any attempt to educate them, so... I highly doubt it. But then again, I haven't successfully predicted any social revolutions (mostly because I don't go around predicting social revolutions).
Anyways... I better go look at some cryptography (I haven't had that course yet...)
Spot on! Another question: How can you trust the net card (there was post not long ago about IPMI, and the potential for hiding complete remote control backdoors in network interfaces).
Then again, how can you trust humans to count perfectly?
What's wrong with paper Nothing! Absolutely nothing. As much as my mom suggests I should write my masters or Ph.D thesis on "on-line voting", I relly think this is one of the areas where correctness is just too important to ever trust networked computers (which, afaict, is a strict requirement). Also, there's a lot of long-haired issues: how do we make sure each voter can vote at most once? How do we make selling votes difficult (My best guess: through social ways, i.e. education)? How do we make sure each vote is anonymous, yet at the same time make sure each voter can vote at most once? These are just hard problems in my mind.
In effect, since it's the foundation of democracy (for those nations that are still at least somewhat democratic, i.e. excluding the USA), we need to treat this as safety-critical systems. That includes the provability (and proof!) of correctness, one-time pads for encryption and all that rigamarole.
No. Please. That's the one area where a two-mile bike ride plus waiting in line to put a cross on a piece of paper really pays off well. Stick with the primitive technology here, due to the advantages of being primitive.
Todays features for tracking changes in documents fail terrible when multiple persons are involved, and their changes overlap.
I guess all the users of cvs and svn would agree completely with that statement.
Of course, it (for most practical purposes) requires a readily hackable textual representation of the data in question, but at least AbiWord has that (xml-based), and I suspect (I haven't researched this) that OOo does xml as well.
Of course, most people won't be able to grok that, so your point is still valid in some specific cases. However, saying "generally, three-way merge is broken" is just flat out wrong, and everyone with a minimum of clue knows this.
That's why I give a big thumbs up to the poster using CVS with LaTeX--I think that's a really great way for more than one person to work on the same document (only topped--maybe--by using svn and LaTeX).
Plus, as a minor benefit, LaTeX looks better than >90% of what's not typeset in SomeKindOfTeX, and a helluvalot better than 99.9% of what comes out of Microsoft Office.
Being open source though, it probably started out a little behind the closed source, because it likely did not have a paid and well-organized development group working on it.
That's probably the most absurd argument I've ever heard. Firstly, the assumptions are wrong: people also get paid to work on free software, and some of them work in well-organized groups.
Secondly, what's the definition of `well-organized'? Working to implement a fixed feature set to a fixed deadline (which can be roughly equivalent to being told "run 100 meters in two seconds")? The feature list being driven by whatever hype-words the marketing folks heard from your company's competitor, and the ship date being "before the christmas sales"? Of course, the team is being well-organized by a manager who hasn't read The Mythical Man-Month, so as the schedule slips, he keeps adding programmers--does that help? ("if a project is late, adding programmers will make it even more late", due to quadratic communication overhead).
On the other hand, with free software developed with the Bazaar model, you have one very key benefit: the coders are (mostly) self-selected. Why do the work on the code? Doesn't matter (here), but most do it because they get some kind of intrinsic reward, and are thus better motivated. Also, you have parallelizable (and hopefully -lized) debugging, for which there's linear communication overhead (every J. Random Hacker talks to the project lead dev, or a small O(1) core).
Oh, by the way, I think Mac OS classic was written by "paid, well-organized" people, yet it did not support preemptive scheduling or memory protection. Nourishing substances for activities in the (pre)frontal lobes.
(note: I don't love hating Mac OS classic--I bash all OSes which doesn't have memory protection and preemptive scheduling, and I don't love hating Apple either--In fact I think that Mac OS X may be a very good choice for mom, pop, my gf and aunt tillie; I won't use it, because not all of the software is free, afaik, but in a strictly technical sense it's great for joe sixpack).
Slashdot Mirror
dig through pages and pages of Google results.
/. ramblings about "Windows is a toy OS" -- "No it isn't" -- "Yes it is!"
Yes! Because it's so much more productive to dig through pages and pages of
Why? WHY??? Now I can't get my mind off cx.goatse!
Let's have a look at what is done in its country of origin (Denmark, for the uninitiated): we say "lego brick" or "lego piece" (I'm not sure which translation is most accurate), and "lego bricks" or "lego pieces".
...) when referring to multiple identical installations (and not referring to what it really means: multiple distros; ex: "I don't care which of the linucies you like, be it Mandrake, or SuSE, or Debian GNU/, or...").
You might say "some lego" (as in a collection of lego pieces), just as you might say "some pot" (as in a collection of some--most often weight--unit of pot)--it's uncountable.
"Lego", when said on its own, refers to the company that produces those wonderful toys.
"Legos"? That's like saying "Linuxes" (or "Linucies", or "Linuces", or "Linuxen", or
Now, all I've got to do is think of something interesting to say in my first first post...
We humans are just really good at reconciling incompatible ideas and actions in our heads.
Yeah. It's called doublethink.
what I don't get is why they don't just use babelfish.
its kinda anying ot see all those headlines: "license held up in court". A license must be like "and you allow me to eat your children" to be not upheld in court. A owner of the copyright can create what ever license he wants and it will allways be valid in court!!!
Sure, it's about as groundshaking as "`true' didn't segfault", which only happens if the program says something LIKE { char *p = 0; *p = 0; } (for interesting implementations of LIKE).
But, and here's the difference, this is (as far as I know) the first time a `hippie underground' license has been tested in court. That's like running your own kernel for the first time: even if it just does {cli; a: jmp a;}, it is interesting.
Sorry for biting...
But I do feel that the artists (...) have a right to be compensated.
Oh absolutely. I'm all for compensating the artists. However, very little of the retail price for a CD is spent on compensating the artist, and a lot of it is spent rewarding the RIAA for tyranizing the population. That, I think, is wrong. I'd gladly donate a dollar here and there for the songs I love, if I'm confident that the payment system is reasonably secure, and that three nines or more of the money goes to the donatee.
Don't try to justify your theft (and yes, it is theft) by saying that I'm an idiot for paying for my music.
Don't worry, it's not theft, so there's nothing to justify, and you're not an idiot for paying for your music, you're an idiot for paying money to be oppressed by the RIAA.
Anyways, let me explain: If I copied your rhetoric, you would not lose your ability to use it. If I copied all your CDs, you would not lose the raw materials, or the ability to listen to them. If I copied your car (don't ask), you wouldn't lose your car.
It is NOT theft (and I'm sick and tired of people calling it that), it's copying. We as a society (I'll only speak for "my"self, and that's Denmark), had at a time come to the conclusion that giving up our right to copy information (since we were unable to, by virtue of not owning printing presses and CD toaster) would be a good trade, since more works would be produced. Now, the assumptions are wrong: we are able to copy information, almost effortlessly.
I'm just saying that this whole notion that music should be free simply because it can easily be pirated anyway is stupid.
So, if we could copy food limitlessly, we shouldn't let the hungry people have it, because the food-producers need to be compensated?
Yes, it is a valid analogy: copyright creates an artificial limit on the copying on music, the above example is an artificial limitation on the copying of food.
Maybe an economical argument can convince you: once a piece of music is on the internet, supply is infinite. In a free market, that would imply that the prices ought to be zero(*). However, copyright law and the RIAA artificially (try to) limit the supply, thereby artificially keeping the price up.
(*) Note, though, that even if the price is zero, that doesn't prevent people from donating money; surprise: some people do that.
The short version: I agree!
;)
The longer version:
It's actually very simple to set someone up a Linux desktop account, show them their email and browser, and just let them use it.
I haven't tried that, but I'm willing to take your word for it--it's not like it's difficult to use, it's just different. And if the users have a sysadmin, that ought to make it workable for them. So... yeah, perhaps that is what I should do to my mom...
RMS says that he choose Unix because he knew that hardware would be much different by the time his OS was complete, and Unix was the most portable OS of the time.
That too! Perhaps that's just a function of building the right amount and kind of layers of abstraction over the hardware? I'm not sure, but as you point out, the designs converge with "some kind of unix" as the limit. Is that because of the technical superiority of unix, or the economical superiority of unix? (go read "worse is better" and "unix-haters handbook" if you haven't already) I would think that it's actually because of the economic superiority; sure, unix is also a great system in many technical ways (sure as hell it beats windows), but by virtue of being built out of small idependent components, almost all system properties and system-wide intergration efforts are belong to the kernel;
Then again, limiting monolithicity to the kernel is much better than unlimited monolithicity. To paraphrase Churchill: "Unix is the worst OS ever, except for all the other OSes"
Toy Story? The creators are totally unoriginal--they named all the characters after Debian releases!
Right on target! Windows (falsely) promotes the idea that end-users (aka. the Joe Sixpaxen of the world) can admin computers.
There are two ways around this: one is to alleviate most needs for administration--i.e. "Just Work"--and the other is to create a high enough barrier to entry that only reasonably competent people will run the system in the first place. Let's call them the "OS X way" and the "GNU/Linux way".
sidetrack {
This sentence no verb.
"Don't do that, then" is not helpful advice to the people who are suffering from this attack.
That's not what parent said. What he said was (depending on exactly what you mean by your analogy) "don't let him do that then" or "have the third party that lets that guy over there do that stop letting him do that then". Okay, so if it involves the third party, the attackee doesn't have much they can do to defend themselves except asking the third party to do the Right Thing, but the doctor is equally out of options.
is it really necessary to sacrafice teh productivity of a standard keyboard in order to gain a convenient, compact form factor?
Uhmm.... yeah, right. Like the standard keyboard is worth more than a dime-a-dozen.
I'm serious: I treated myself with a Kinesis Ergo Elan (http://www.kinesis-ergo.com/elan.htm), and I'm really glad I did--it's a really pleasant keyboard to work with. I only noticed when I first used a standard keyboard after adapting to using the Ergo Elan--not only did it feel uncomfortable, it felt painful. I'm serious--and it wasn't just because I had adjusted to something different (if it were, I should have felt pain when switching to Ergo Elan, something I did not). Also, one thing that completely confounds me is why modern keyboards still have stagger--it's just being mindlessly carried along from the 1890's up until today (sure, it was reasonable due to mechanical constraints on typewriters; newsflash: computer keyboards isn't a new invention).
(End of rant about the sad state keyboard design is in these days)
Actually, I think the AlphaGrip might be better than a (physical) keyboard layout designed for typewriters. I'm open to the possibility that I might be wrong (in general I dislike small keyboards, but this one might have been done well).
I hope you don't argue that the usefulness of the standard keyboard layout is high just because it's a standard; if that was true, Windows 95 was once a good OS, and Bush was a good president (at election time, he was the "standard" choice).
Also, the keyboard permutation matters. I'll replay this anecdote (I'm not playing any role, and I can't remember who participates in it, and I'm not even sure I got the phrases exactly right, or that I've go the gender information right, but... oh well, someone once said something to the effect of...):
Alice: "Excuse me. If you don't mind me asking, are you using that alternative keyboard layout?"
Bob:: "Yes--it's called Dvorak. How could you tell?"
Alice: "Your fingers aren't moving."
I like the promise that the virtual keyboards have
It's not a bad idea (at all), but it's amazing that they haven't shuffled the keys around (okay, so I was almost done ranting about keyboard design): it must be almost trivial to do--all there is to it is basically detect surface contact in a different polygon.
Give a man a fish, he'll eat for a day. Try to teach a man to fish, and he'll complain that "I DON'T CARE!! You take take of catching fish! I mean, you're the sysadmin!"
Now I can boot a non-free and a non-free OS (I will tell you which is which: windows is the non-free one, and OS X is the non-free one).
.... I'm not getting my hopes up, though: when I ran a terminal on my neighbour's mac, she said "What's that?"--which really shows that Apple has done well (at least on the usability and "just works" part).
Of course, if it means moving users away from (maybe even out of) the Microsoft Monoculture (tm), substituting one non-free OS for another one is just a sideways step, not a step away from freedom. Plus, even if people still use Losedose, using OS X on the side may make a few of them competent around a "real" unix system; that is to say, some of them might learn to use the terminal and cd, cp, ls, mv, seq, wget,
Erm... I lost track of what I meant to say. Oh yeah, I remember: [unethusiastically]Wow, this matters a lot to me[/unenthusiastically]
Who'd have the stress of a brain surgeon if you could have the same lifestyle cooking burgers?
Obviously people who think brain surgery is intrinsically rewarding!
It's like saying "Who'd have the stress of being a Gentoo user if you could have the same lifestyle using Debian?" I believe gentoo users waste a lot of resources on compiling things, but obviously most gentoo users disagree--so instead of saying "gee, using gentoo is really stupid!" I say "gee, I'd really like to understand--why do you use gentoo?"
Perhaps you could ask some brain surgeons that have tried flipping burgers (so they seen both sides of the deal) if they'd rather flip burgers if the salaries were the same. I predict that most of them would say "hell no!".
Funny, I thought of proper robes for teaching the gospel of the flying spaghetti monster (http://www.venganza.org/)...
See also "Meme hacking for fun and profit" by ESR (http://technetcast.ddj.com/tnc_play_stream.html?s tream_id=318). It's a bit dated, but hasn't lost any truth or relevance.
As much as one might agree with RMS and prefer the term "free software" to "open source" (I do), one must admit that ESR got his marketing skills worked out (at least for advocating free software).
One important point that he makes, that you apparently miss (and I'm convinced it's true): you want to emphasise risk, not cost.
(which doesn't ship with mplayer or mp3 support, fer christsake)
;)
Define `ship'. mplayer is in contrib, so it's definitely in the repositories--you just have to set up apt properly.
Oh, and yeah, there's no mp3 support. totem-(gstreamer|xine), vlc, mpd, playsound, xmms, and rythmbox do not exist, despite me believing I have played mp3s with every one of them on my debian "out-of-the-post-installation-box" box. I admit I have no first-hand experience with amaroK, juk, noatun or zinf, but I guess they play mp3-files just fine.
As [Score:-1,] Has Been Said, it might just be because you're not using the non-US repositories. I assume they're there so that debian can avoid breaking the law.
So this means that you're either (1) blaming yourself for not editing your sources.list; or (2) blaming debian for not breaking the law.
Do you also blame microsoft for not breaking the l... oh, never mind
Hmm... I guess I came off as rather intending to start a (holy|distro|flame) war. Not intended. I'm just trying to say that you should do your research before you judge any distro. If you need help with configuring debian, I'm sure there's plenty of help available at http://www.linuxquestions.org/.
How can one ever count the defects/bugs per line?
Good question. The theoretical answer is that one can't. I'll replicate the proof I prepared for my Computability and Logic exam.
I assume you know what turing machines, hoare triples are, and what recursive and recursively enumerable means.
Assumption: there exists an encoding of turing machines, such that the language of all turing machines is recursively enumerable (this is fairly easy to show, but I'll skip it).
The problem is to show that there doesn't exist any turing machine which will take any hoare triple (pre, code, post), and show that under the precondition, running the code enters a state satisfying the postcondition.
I want to show that the `semi-negation' of the problem--that is, proving programs to be incorrect--is recursively enumerable. Consider the hoare triple (true, anything, false). That is, the program is incorrect iff it halts. So any machine solving the incorrectness problem must--among others--solve the halting problem. But we know the halting problem to be recursively enumerable but not recursive. Hence incorrectness is recursively enumerable, but not recusive (this is actually inspired by Dijkstra: "by testing a program, you can only prove the presence of bugs, not the absence").
But then the complement--that is, ('the code part is a turing machine' AND `the hoare triple is correct')--can not be recursively enumerable.
That is, assuming turing machines are as powerful as general-purpose computers, there is no program which will verify correctness of *any* hoare triple.
However, there might be programs that catch most of the typical mistakes, which would of course be very valuable.
Also, there's the whole question of counting lines (as has been pointed out is the not-a-dupe-at-the-time thread).
Also, a very interesting question is whether or not the human brain is more than turing complete--that is, can humans every prove any program to be correct? Nourishing substances for activities in the frontal lobes.
Of coruse, then the new question becomes how to manage PKI in a system which is understandable and easy for the entire population. (I'm sure a system can be made though.)
Aunt Tillie (to her nephew Mervin): "But it won't install the smiley icons if I click no in the pop-up thingie... security something, it says".
Heck, even I don't read all the "do you want to trust this signature?" popups that firefox shoves into my face. It'd take a lot of education to make that work at what I'd call an accptable level. Also, some people actively resist any attempt to educate them, so... I highly doubt it. But then again, I haven't successfully predicted any social revolutions (mostly because I don't go around predicting social revolutions).
Anyways... I better go look at some cryptography (I haven't had that course yet...)
(Mandatory reference) Reflections on trusting trust: http://www.acm.org/classics/sep95/
Spot on! Another question: How can you trust the net card (there was post not long ago about IPMI, and the potential for hiding complete remote control backdoors in network interfaces).
Then again, how can you trust humans to count perfectly?
What's wrong with paper
Nothing! Absolutely nothing. As much as my mom suggests I should write my masters or Ph.D thesis on "on-line voting", I relly think this is one of the areas where correctness is just too important to ever trust networked computers (which, afaict, is a strict requirement). Also, there's a lot of long-haired issues: how do we make sure each voter can vote at most once? How do we make selling votes difficult (My best guess: through social ways, i.e. education)? How do we make sure each vote is anonymous, yet at the same time make sure each voter can vote at most once? These are just hard problems in my mind.
In effect, since it's the foundation of democracy (for those nations that are still at least somewhat democratic, i.e. excluding the USA), we need to treat this as safety-critical systems. That includes the provability (and proof!) of correctness, one-time pads for encryption and all that rigamarole.
No. Please. That's the one area where a two-mile bike ride plus waiting in line to put a cross on a piece of paper really pays off well. Stick with the primitive technology here, due to the advantages of being primitive.
Todays features for tracking changes in documents fail terrible when multiple persons are involved, and their changes overlap.
I guess all the users of cvs and svn would agree completely with that statement.
Of course, it (for most practical purposes) requires a readily hackable textual representation of the data in question, but at least AbiWord has that (xml-based), and I suspect (I haven't researched this) that OOo does xml as well.
Of course, most people won't be able to grok that, so your point is still valid in some specific cases. However, saying "generally, three-way merge is broken" is just flat out wrong, and everyone with a minimum of clue knows this.
That's why I give a big thumbs up to the poster using CVS with LaTeX--I think that's a really great way for more than one person to work on the same document (only topped--maybe--by using svn and LaTeX).
Plus, as a minor benefit, LaTeX looks better than >90% of what's not typeset in SomeKindOfTeX, and a helluvalot better than 99.9% of what comes out of Microsoft Office.
Being open source though, it probably started out a little behind the closed source, because it likely did not have a paid and well-organized development group working on it.
That's probably the most absurd argument I've ever heard. Firstly, the assumptions are wrong: people also get paid to work on free software, and some of them work in well-organized groups.
Secondly, what's the definition of `well-organized'? Working to implement a fixed feature set to a fixed deadline (which can be roughly equivalent to being told "run 100 meters in two seconds")? The feature list being driven by whatever hype-words the marketing folks heard from your company's competitor, and the ship date being "before the christmas sales"? Of course, the team is being well-organized by a manager who hasn't read The Mythical Man-Month, so as the schedule slips, he keeps adding programmers--does that help? ("if a project is late, adding programmers will make it even more late", due to quadratic communication overhead).
On the other hand, with free software developed with the Bazaar model, you have one very key benefit: the coders are (mostly) self-selected. Why do the work on the code? Doesn't matter (here), but most do it because they get some kind of intrinsic reward, and are thus better motivated. Also, you have parallelizable (and hopefully -lized) debugging, for which there's linear communication overhead (every J. Random Hacker talks to the project lead dev, or a small O(1) core).
Oh, by the way, I think Mac OS classic was written by "paid, well-organized" people, yet it did not support preemptive scheduling or memory protection. Nourishing substances for activities in the (pre)frontal lobes.
(note: I don't love hating Mac OS classic--I bash all OSes which doesn't have memory protection and preemptive scheduling, and I don't love hating Apple either--In fact I think that Mac OS X may be a very good choice for mom, pop, my gf and aunt tillie; I won't use it, because not all of the software is free, afaik, but in a strictly technical sense it's great for joe sixpack).