The question remains: what better way is there to evaluate code in an automated fashion?
The other method you mention, commonly known as cyclomatic complexity, is also a valid metric, and we use it where I work as well as SLOC. However, there remains no real good ways to easily estimate the the quality and progress on code without a tremendous abmount of work. These metrics allow compile time statistics to be generated that give ways to track progress and quality (in conjuction with tools like checkstyle and junit/cppunit).
Its easy to complain about the metric - my point is that management needs to justify expenses and people that provide money need to get reports on progress. We, as programmers, have to come up with ways to provide that. SLOC is one such way. That is why it's useful.
Counting lines of code *is* important, just not for estimating how "powerful" a program is (what the hell does that mean, anyway?)
It's a good metric to track on enterprise development to get a handle on time usage of a development team, as well as progress towards whatever source estimates you originally set for a particular method/class/program.
There are a number of build-system integrated SLOC estimators like JavaNCSS and SLOCCount that help with tracking it, and will also provide other useful style-indicative metrics, like average number of lines per function/method.
Yeah, I did. =) Never used 5. I already replied to myself saying this, but since you figured it out, I thought I'd reply and let you know "You're right!".
Just a couple of things. Drop the "broomstick" thing. That has nothing to do with hydrogen any more than any other high pressure gas. Trust me. The broomstick trick is a classic in high pressure (600 or the old 1200 lb) steam plants in the Navy. I know because I worked in one.
Second, while I think you do in fact have your facts straight, I think your dad was right to be more worried about hydrogen mostly because petroleum products on the whole are very difficult to ignite. Gasoline comes off of the top of the column though, and it very high octane compared to other petroleum products, so it would be more dangerous from a flammability standpoint than other petroleum products (for example, those present in your dad's workplace).
You can put out matches in standing deisal fuel.
As for the issues of mixing ratios, I don't know much, but I gather that letting an amount of gasoline stand in an enclosed space will fill the space with gasoline vapors (gas has pretty high volatility, right?) which are directly ignitable. This results in those studies that find that driving with a full tank of gas is actually safer in case of accident than driving with a tank almost empty, because the empty tank has more room for the gasoline vapor/air mixture.
If URLs on your site are old (i.e. 404s) and are only indexed in Google, and yet you find MSN crawling them, only to find that their index is updated with those results shortly thereafter, well, that qualifies as something more than "'ms sucks' rhetoric". "Who cares?" might be a more appropriate retort.
Bloggers are just people. So are reporters. Just because some dude said it in a blog doesn't make it unreliable, any more than a journalist saying it makes it reliable.
In my post above, I meant to be speaking about 3 when I was talking about 5.
Winamp 5, well, I have no clue, because I've never used it. I was thinking "Oh, Winamp 5? That's that new release that was horrible!" I was remembering my experience with 3.
Sorry for the mis-post - kind of ruined my whole point.
You're kidding right? Winamp 5 was a steaming turd compared to 3. After Justin Frankel left the team, they lost their edge and "rewrote" the amazing Winamp 3 to do a bunch of crap that didn't matter, mainly adding a bunch of kruft that made it unstable and slow. I know a bunch of people that rolled back versions to 3 after trying 5 for a while.
I don't know about everyone else, but despite my love of eye candy, Winamp 5 just wasn't worth it (and took ages to get out of beta). I won't even get into how they evolved their business model to start selling their software. Its OK if you start off and are selling, and move to giving some stuff away free, but going the other way (free --> selling) doesn't typically bode as well.
On the other hand, this is a sad day, since Nullsoft did have the Gnullsoft arm that were the original creators of some stuff that ended up being very successful: Gnutella and WASTE. Even though both were pulled from the site hours after release, they still made their mark. Good job, guys. Maybe the original team will get back together and form GNN (GNN's Not Nullsoft) and make more things like Gnutella and WASTE - lord knows they have the talent.
First off, neither is stealing. Its copyright infringment. Big difference.
Second, the vast majority of people who download music in violation of copyright don't do it to make money. If they downloaded music and used it to press CDs and sell them, then we'd be getting somewhere close to what these guys are doing.
As it is, the comparison is not really valid. Enjoying someone else's work as-is for free is quite a bit different (on a philosophical and legal level, at least) than 1) taking someone's work and 2) modifying it and 3) selling it as your own 4) without permission 5) for your own personal monetary gain.
Yeah, just like "Western Union", "Internet Explorer", "Lotus Notes", "Word", "Excel", "Access", "Windows" and lets not forget things that are similar, too, like "Lindows".
So, yeah, while I agree with your sentiment, deal with it. Don't take your anger out on "Open Source" of all things, try one of the others. "Windows" is always a good starting point. =P
Huh? AMD was shipping over half the new desktop CPUs last I checked (wasn't this a/. story a few weeks ago?)...this isn't like saying "...help Linux crack Microsoft's dominance..." or something like that.
AMD may be an underdog, but they're competing quite well, and may still be shipping over half the new desktops.
I was very excited about Ubuntu, and I installed it on my new-distro-of-the-month laptop, a LAC Linux made machine that originally came with Debian. Anyway, Ubuntu is good looking and installs well. The big problem is that it kept hanging when I would log off X, and various other apps were crashing etc.
Not sure what the deal was, but the Gentoo install I had on it earler worked great, and I installed Mepis right after, and it worked wonderfully. I'm sold on Mepis and would recommnend it in lieu of Ubuntu, although they still use XFree86 - I assume they'll be switching next release.
Sony vs. Universal did exactly what you describe: they ruled that recording programs for the purpose of timeshifting was fair-use. Call this an "entire species" if you like, but that's what the decision said.
People are not entitled to fair use - you are correct. However, that in no way means that fair use is "merely" a defense: it is not. It is a gap in the applicability of copyright (and DMCA) law. In other words, a specific provision under which the resitrction does not apply. If you were looking for something that is "merely" a defense, you should look at patent law's "prior art" or something more along those lines. But "fair use" and "prior art" are different in their natures.
Lastly, the broadcast flag will not fall under anything *but* copyright law, or an extension thereof, like the DMCA.
You bring up one very good point thought that I rarely see here: even though its entirely LEGAL for you to make copies as a paying user of [digital cable, satellite TV, satellite radio, whatever], there is no law saying that the broadcasters cannot make it difficult or [relatively] impossible for you to do so. I think this is a legal loophole for providers that needs to be closed to protect consumer's rights. (As you pointed out, fair use is not a right, but I think it should be.) As it stands now, we are in a situation where we are legally allowed to copy something, but the providers are also legally allowed to take every measure to stop us from doing so, including outlawing devices that would permit such an action. Circumvention of those restriction on the device would then fall under a legal exclusion, but you have to ask yourself at some point if we're being honest with ourselves...you basically are legally allowing something and then making it so hard to accomplish that only a very small portion of the population can benefit from that legal provision. This is a sort of legislative dishonestly - you're saying one thing while allowing something completely different to actually occur. We need to decide what we really want and then put laws into place that [protect/prohibit] those actions.
If you read the Sony vs. Universal decision, there is a lot of commentary by the judge "schooling" people in copyright law, its true purpose, and how it needs to be revamped when new technologies emerge. We're approaching that time, and I'm not so sure I'm going to like the outcome.
Its actually just as tough as you paint it to be: in our military (the US military), you do have to refuse to carry out ilegal orders. You also have to carry out legal orders. So the moment someone issues an order whose legality is suspect, you need to make a very impoartant decsion, because you can get hammered if you don't make the right judgement.
This applies at all levels in the chain of command, and it keeps soldiers/sailors/marines/pilots thinking. I never found it to be too much of an issue, but I can imagine scenarios where it could be.
Whether grandparent knows it or not, he's right, even in the face of all your arguments.
You keep asking "why?" - well, I'll tell you. Because copyright was not designed with the benefit of the artist as its primary concern. The REAL purpose of copyright is to motivate the artist enough to release his work, and encourage him to keep producing other artistic works.
It is NOT the primary interest of copyright to make people millionaires so they *don't* have to continue working, it is the exact opposite. Believe it or not, copyright is primarily intended as a benefit to the general public, and a motivator to encourage the copyright holder to release his work for the public good.
For reference on th above, check the US Constitution, the 1908 (I think) copyright convention documents, and the "Betamax" case in the early 80s.
For an in depth article relating to this and how it was mis-used in the "bnetd" case, check my blog, where I cite those sources in a bit more informative way.
Wow! I'm the same age as you, and went through the *exact* same phases, although I also downloaded movies in college. Now I don't pirate anything, although I couldn't stomach paying so much for software, so I moved to Linux completely.
I regualrly buy CDs and rip them to Ogg for my two (!) portable players, a Neuros and a Rio Karma.
I buy lots of DVDs and games (that support Linux) and have learned that buying is power, because I can choose who I give my money to. I can't remember the last time I downloaded anything illegally. It was really an issue of income/budet when I pirated things.
I've been subscribing to Netflix for over a year, and they just reduced their prices to 17.99 - no return dates, three movies at a time.
Competition is good. Comparing quality, wait time and convenience, I don't think I'll be downloading.
Then again, movies and music are *very* different animals. If you think about it, the type of entertainment they are is completely different - movies are more akin to games and music. And getting people to stop pirating games, well, that's gonna be tough.
He wasn't joking...he's a backer of the Republican party, and, acting in that capacity, said he was committed to helping Ohio deliver its votes to Bush. Maybe poor judgement, but its not a joke. He didn't say "I'm going to rig the election in Ohio using my machines to illegally give Bush more votes", he's talking about party support. Everybody supports someone, it just a question of whether you voicing your biases. He did, which proabably was in poor judgement, but not poor taste.
Supporting someone in their campaign is different than rigging an election. I think what Diebold's Chairman said cleanly falls into "party support", not "illegal tampering".
Grandparent (despite the -1, Flamebait mod) makes a good point: exit polls agree with what the voting machines/systems counted. If Kerry had won, I doubt there would be so many people with conspiracy theories.
You are correct that it leads to the choice "Either the exit polls are wrong or the actual voting is being miscounted."
However, saying things like "We may never know" is a bit dramatic. Although I am not an expert in exit polling methods, I did major in OR, of which statistics is a very large part (specifically, the statistics necessary to model complex processes accurately).
In general, building an accurate model is the hard part, because it is is usually build on top of a small sample used to gauge the behavior of a real system (in this case, votes in a state). But the difficulty of actually going out and gathering a representative sample is really quite high.
In the case of elections, two major factors alone make it hard: time and location. If you gather exit polls from just a handful of counties, it certainly won't be representative (look at the county chart of Ohio or Florida and you'll see what I mean). So unless you are taking 100 exit poll samples in each county, you're probably going to get burnt.
Second: time. In past elections it has been shown that swing voters tend to be in higher concentration later on, and that they tend to vote against the incumbant. Of course, this doesn't hold in all counties equally, but it would lead you to believe that a sample of 100 taken in the morning and at night (say, 10 hours later) would have materially different compositions in "battleground" areas.
If you combine the above variables, you can see that it would practically be quite difficult to gather a good, random sample on which to base the models. Combine that with a very close election, and the usual "few percent (~5) error" suddenly decides everything.
Of course, this is the principle of Occam's Razor: I find it much more plausible that the modelling was done incorrectly (because it is so difficult and the race so close) than there is a vast Republican conspiracy to rig voting machines across the country to dump Democratic votes and replace them with Republican votes. Indeed, I think many of the problems remain from 2000: poorer counties that have a majority democratic vote use older punch-style machines, which have a failure rate approaching 1/3. This would of course sway the vote, minimizing the impact of such a demographic. However, I do not attribute this to malice; I think it is a problem to be fixed however.
Even the federal government doesn't have enough coordination over the voting process in each state and county to to make all the voting methods consistent. I doubt there's a ubiquitous underground, well-coordinated, Republican subversion of all the voting machines in the battleground states so as to fudge the election. Too many people would know, and you'd certainly have some headlines on the cover of the National Enquirer or Star. Of course, I don't read those, so maybe there have been!
That's where I am...I don't even surf from work's machines...they'll see me spending time on/.!
When I do surf from work (not much, but now happens to be a time), I just ssh into my home linux box, fire up firefox with X11 forwarding (cygnus X on my work computer) and surf.
Funny you should bring this up. I came from a Christian (Protestant) family, grew up and atheist, and have recently taken a great interest in Zen Buddhism.
One of the concepts in Zen is that everything is constantly being reborn, in every instant. Essentially that not only was the whole created last Thursday, but also in the last minute, second, and now.
This is actually quite liberating. There really is no age, per se, but just a state of things right now. And that state could be creating that way in any moment, and, in fact, it is.
It challenges our common conceptions of time, while making you realize that each moment is really new, and that you are not the same person you were a minute ago. This doesn't mean the world is manufactured, just that there is no inertia. Rather, it is an series of states, each with no "memory" of another. (Then again, what is memory if not a state?)
Gosh, I hope this makes sense to people that haven't been doing Zen reading. I'm afraid I'm not very good at explaining the idea, and why it was an epiphany for me.
Actually, punctuation only goes inside quotes if the punctuation is in the quote (amazing, huh?) If you're quoting a fragment, then the punctuation goes outside the quotes.
I'm not a fan of being pedantic, but if you're going to be, at least be right.
Gentoo is not necessarily good because of the product, but in large part because of the process. When you finish doing a stage whatever (especially 1) install, you end up learning an awful lot about Linux that someone that drops in a SuSe/RedHat/Fedora Core/whatever disk doesn't know. Most experienced Linux users will see that a user that understands whats going on under the hood will fare better than one who gives you a thousand yard stare when you mention the/etc/inittab file.
I think the benefits of compiling from source on everything are varied at best, and only sometimes outweighed by the time necessary to do it. That said, in some cases it is a good thing - if used correctly, the USE flags are nifty and let you compile without support for features you don't need. This can be quite useful, and provide a modest speed up in some cases.
Ricers aside, Gentoo provides a superb package management system in the spirit of apt/yum, and is also source based. It boosts users with moderate knowledge level to a better understanding of the architecture of a Linux system, and this can lead to some absurd enthusiasm about the distro for the younger/more impressioanable types, but I take it much the same way I take any fanboy mentality: you'll see the upsides and the downsides as time goes on. I happen to think Gentoo is great on the whole, so I use it.
Its just as childish for the folks annoyed by the Gentoo zealots to turn around be be anti-Gentoo zealots, creating webpages and ranting on about how horrible a community it is. Stop by the forums and you'll see its a responsive, well informed group, the majority of whom are quite reasonable.
Yes, but KDE isn't dealing with only network transparency. It also deals with compression transparency, format transparency and protocol transparency. This includes handling of tar.gz files without a compression program, kamera://, and, as the original post pointed out, other formats like info://.
Basically its an extensible system that allows for any protocol to be used. No one else has that yet.
3) Low yield on top-end products (moving to a smaller process == $$$)
So I suppose you could simplify and say that is just "research", but its also simply the nature of the market. There simply is no market like that in software - its very monopoly-ish (no hardware vendor has 90% of the market, even in processors), and the product lifecycle is much longer, with relatively little advancement between generations, unlike the hardware generational gap, which is large (Moore's law).
Slashdot Mirror
I understand what you're saying.
The question remains: what better way is there to evaluate code in an automated fashion?
The other method you mention, commonly known as cyclomatic complexity, is also a valid metric, and we use it where I work as well as SLOC. However, there remains no real good ways to easily estimate the the quality and progress on code without a tremendous abmount of work. These metrics allow compile time statistics to be generated that give ways to track progress and quality (in conjuction with tools like checkstyle and junit/cppunit).
Its easy to complain about the metric - my point is that management needs to justify expenses and people that provide money need to get reports on progress. We, as programmers, have to come up with ways to provide that. SLOC is one such way. That is why it's useful.
Counting lines of code *is* important, just not for estimating how "powerful" a program is (what the hell does that mean, anyway?)
It's a good metric to track on enterprise development to get a handle on time usage of a development team, as well as progress towards whatever source estimates you originally set for a particular method/class/program.
There are a number of build-system integrated SLOC estimators like JavaNCSS and SLOCCount that help with tracking it, and will also provide other useful style-indicative metrics, like average number of lines per function/method.
Yeah, I did. =) Never used 5. I already replied to myself saying this, but since you figured it out, I thought I'd reply and let you know "You're right!".
Gosh Rei, you're really replying like mad.
Just a couple of things. Drop the "broomstick" thing. That has nothing to do with hydrogen any more than any other high pressure gas. Trust me. The broomstick trick is a classic in high pressure (600 or the old 1200 lb) steam plants in the Navy. I know because I worked in one.
Second, while I think you do in fact have your facts straight, I think your dad was right to be more worried about hydrogen mostly because petroleum products on the whole are very difficult to ignite. Gasoline comes off of the top of the column though, and it very high octane compared to other petroleum products, so it would be more dangerous from a flammability standpoint than other petroleum products (for example, those present in your dad's workplace).
You can put out matches in standing deisal fuel.
As for the issues of mixing ratios, I don't know much, but I gather that letting an amount of gasoline stand in an enclosed space will fill the space with gasoline vapors (gas has pretty high volatility, right?) which are directly ignitable. This results in those studies that find that driving with a full tank of gas is actually safer in case of accident than driving with a tank almost empty, because the empty tank has more room for the gasoline vapor/air mixture.
Hmm, did you actually *read* the article?
If URLs on your site are old (i.e. 404s) and are only indexed in Google, and yet you find MSN crawling them, only to find that their index is updated with those results shortly thereafter, well, that qualifies as something more than "'ms sucks' rhetoric". "Who cares?" might be a more appropriate retort.
Bloggers are just people. So are reporters. Just because some dude said it in a blog doesn't make it unreliable, any more than a journalist saying it makes it reliable.
I'm retarded.
In my post above, I meant to be speaking about 3 when I was talking about 5.
Winamp 5, well, I have no clue, because I've never used it. I was thinking "Oh, Winamp 5? That's that new release that was horrible!" I was remembering my experience with 3.
Sorry for the mis-post - kind of ruined my whole point.
You're kidding right? Winamp 5 was a steaming turd compared to 3. After Justin Frankel left the team, they lost their edge and "rewrote" the amazing Winamp 3 to do a bunch of crap that didn't matter, mainly adding a bunch of kruft that made it unstable and slow. I know a bunch of people that rolled back versions to 3 after trying 5 for a while.
I don't know about everyone else, but despite my love of eye candy, Winamp 5 just wasn't worth it (and took ages to get out of beta). I won't even get into how they evolved their business model to start selling their software. Its OK if you start off and are selling, and move to giving some stuff away free, but going the other way (free --> selling) doesn't typically bode as well.
On the other hand, this is a sad day, since Nullsoft did have the Gnullsoft arm that were the original creators of some stuff that ended up being very successful: Gnutella and WASTE. Even though both were pulled from the site hours after release, they still made their mark. Good job, guys. Maybe the original team will get back together and form GNN (GNN's Not Nullsoft) and make more things like Gnutella and WASTE - lord knows they have the talent.
Well, its not really a valid observation.
First off, neither is stealing. Its copyright infringment. Big difference.
Second, the vast majority of people who download music in violation of copyright don't do it to make money. If they downloaded music and used it to press CDs and sell them, then we'd be getting somewhere close to what these guys are doing.
As it is, the comparison is not really valid. Enjoying someone else's work as-is for free is quite a bit different (on a philosophical and legal level, at least) than
1) taking someone's work and
2) modifying it and
3) selling it as your own
4) without permission
5) for your own personal monetary gain.
Yeah, just like "Western Union", "Internet Explorer", "Lotus Notes", "Word", "Excel", "Access", "Windows" and lets not forget things that are similar, too, like "Lindows".
So, yeah, while I agree with your sentiment, deal with it. Don't take your anger out on "Open Source" of all things, try one of the others. "Windows" is always a good starting point. =P
Huh? AMD was shipping over half the new desktop CPUs last I checked (wasn't this a /. story a few weeks ago?)...this isn't like saying "...help Linux crack Microsoft's dominance..." or something like that.
AMD may be an underdog, but they're competing quite well, and may still be shipping over half the new desktops.
I was very excited about Ubuntu, and I installed it on my new-distro-of-the-month laptop, a LAC Linux made machine that originally came with Debian. Anyway, Ubuntu is good looking and installs well. The big problem is that it kept hanging when I would log off X, and various other apps were crashing etc.
Not sure what the deal was, but the Gentoo install I had on it earler worked great, and I installed Mepis right after, and it worked wonderfully. I'm sold on Mepis and would recommnend it in lieu of Ubuntu, although they still use XFree86 - I assume they'll be switching next release.
Your ideas are good, if a bit malformed.
Sony vs. Universal did exactly what you describe: they ruled that recording programs for the purpose of timeshifting was fair-use. Call this an "entire species" if you like, but that's what the decision said.
People are not entitled to fair use - you are correct. However, that in no way means that fair use is "merely" a defense: it is not. It is a gap in the applicability of copyright (and DMCA) law. In other words, a specific provision under which the resitrction does not apply. If you were looking for something that is "merely" a defense, you should look at patent law's "prior art" or something more along those lines. But "fair use" and "prior art" are different in their natures.
Lastly, the broadcast flag will not fall under anything *but* copyright law, or an extension thereof, like the DMCA.
You bring up one very good point thought that I rarely see here: even though its entirely LEGAL for you to make copies as a paying user of [digital cable, satellite TV, satellite radio, whatever], there is no law saying that the broadcasters cannot make it difficult or [relatively] impossible for you to do so. I think this is a legal loophole for providers that needs to be closed to protect consumer's rights. (As you pointed out, fair use is not a right, but I think it should be.) As it stands now, we are in a situation where we are legally allowed to copy something, but the providers are also legally allowed to take every measure to stop us from doing so, including outlawing devices that would permit such an action. Circumvention of those restriction on the device would then fall under a legal exclusion, but you have to ask yourself at some point if we're being honest with ourselves...you basically are legally allowing something and then making it so hard to accomplish that only a very small portion of the population can benefit from that legal provision. This is a sort of legislative dishonestly - you're saying one thing while allowing something completely different to actually occur. We need to decide what we really want and then put laws into place that [protect/prohibit] those actions.
If you read the Sony vs. Universal decision, there is a lot of commentary by the judge "schooling" people in copyright law, its true purpose, and how it needs to be revamped when new technologies emerge. We're approaching that time, and I'm not so sure I'm going to like the outcome.
Its actually just as tough as you paint it to be: in our military (the US military), you do have to refuse to carry out ilegal orders. You also have to carry out legal orders. So the moment someone issues an order whose legality is suspect, you need to make a very impoartant decsion, because you can get hammered if you don't make the right judgement.
This applies at all levels in the chain of command, and it keeps soldiers/sailors/marines/pilots thinking. I never found it to be too much of an issue, but I can imagine scenarios where it could be.
Whether grandparent knows it or not, he's right, even in the face of all your arguments.
You keep asking "why?" - well, I'll tell you. Because copyright was not designed with the benefit of the artist as its primary concern. The REAL purpose of copyright is to motivate the artist enough to release his work, and encourage him to keep producing other artistic works.
It is NOT the primary interest of copyright to make people millionaires so they *don't* have to continue working, it is the exact opposite. Believe it or not, copyright is primarily intended as a benefit to the general public, and a motivator to encourage the copyright holder to release his work for the public good.
For reference on th above, check the US Constitution, the 1908 (I think) copyright convention documents, and the "Betamax" case in the early 80s.
For an in depth article relating to this and how it was mis-used in the "bnetd" case, check my blog, where I cite those sources in a bit more informative way.
Wow! I'm the same age as you, and went through the *exact* same phases, although I also downloaded movies in college. Now I don't pirate anything, although I couldn't stomach paying so much for software, so I moved to Linux completely.
I regualrly buy CDs and rip them to Ogg for my two (!) portable players, a Neuros and a Rio Karma.
I buy lots of DVDs and games (that support Linux) and have learned that buying is power, because I can choose who I give my money to. I can't remember the last time I downloaded anything illegally. It was really an issue of income/budet when I pirated things.
I've been subscribing to Netflix for over a year, and they just reduced their prices to 17.99 - no return dates, three movies at a time.
Competition is good. Comparing quality, wait time and convenience, I don't think I'll be downloading.
Then again, movies and music are *very* different animals. If you think about it, the type of entertainment they are is completely different - movies are more akin to games and music. And getting people to stop pirating games, well, that's gonna be tough.
He wasn't joking...he's a backer of the Republican party, and, acting in that capacity, said he was committed to helping Ohio deliver its votes to Bush. Maybe poor judgement, but its not a joke. He didn't say "I'm going to rig the election in Ohio using my machines to illegally give Bush more votes", he's talking about party support. Everybody supports someone, it just a question of whether you voicing your biases. He did, which proabably was in poor judgement, but not poor taste.
Supporting someone in their campaign is different than rigging an election. I think what Diebold's Chairman said cleanly falls into "party support", not "illegal tampering".
Grandparent (despite the -1, Flamebait mod) makes a good point: exit polls agree with what the voting machines/systems counted. If Kerry had won, I doubt there would be so many people with conspiracy theories.
You are correct that it leads to the choice "Either the exit polls are wrong or the actual voting is being miscounted."
However, saying things like "We may never know" is a bit dramatic. Although I am not an expert in exit polling methods, I did major in OR, of which statistics is a very large part (specifically, the statistics necessary to model complex processes accurately).
In general, building an accurate model is the hard part, because it is is usually build on top of a small sample used to gauge the behavior of a real system (in this case, votes in a state). But the difficulty of actually going out and gathering a representative sample is really quite high.
In the case of elections, two major factors alone make it hard: time and location. If you gather exit polls from just a handful of counties, it certainly won't be representative (look at the county chart of Ohio or Florida and you'll see what I mean). So unless you are taking 100 exit poll samples in each county, you're probably going to get burnt.
Second: time. In past elections it has been shown that swing voters tend to be in higher concentration later on, and that they tend to vote against the incumbant. Of course, this doesn't hold in all counties equally, but it would lead you to believe that a sample of 100 taken in the morning and at night (say, 10 hours later) would have materially different compositions in "battleground" areas.
If you combine the above variables, you can see that it would practically be quite difficult to gather a good, random sample on which to base the models. Combine that with a very close election, and the usual "few percent (~5) error" suddenly decides everything.
Of course, this is the principle of Occam's Razor: I find it much more plausible that the modelling was done incorrectly (because it is so difficult and the race so close) than there is a vast Republican conspiracy to rig voting machines across the country to dump Democratic votes and replace them with Republican votes. Indeed, I think many of the problems remain from 2000: poorer counties that have a majority democratic vote use older punch-style machines, which have a failure rate approaching 1/3. This would of course sway the vote, minimizing the impact of such a demographic. However, I do not attribute this to malice; I think it is a problem to be fixed however.
Even the federal government doesn't have enough coordination over the voting process in each state and county to to make all the voting methods consistent. I doubt there's a ubiquitous underground, well-coordinated, Republican subversion of all the voting machines in the battleground states so as to fudge the election. Too many people would know, and you'd certainly have some headlines on the cover of the National Enquirer or Star. Of course, I don't read those, so maybe there have been!
That's where I am...I don't even surf from work's machines...they'll see me spending time on /.!
When I do surf from work (not much, but now happens to be a time), I just ssh into my home linux box, fire up firefox with X11 forwarding (cygnus X on my work computer) and surf.
Funny you should bring this up. I came from a Christian (Protestant) family, grew up and atheist, and have recently taken a great interest in Zen Buddhism.
One of the concepts in Zen is that everything is constantly being reborn, in every instant. Essentially that not only was the whole created last Thursday, but also in the last minute, second, and now.
This is actually quite liberating. There really is no age, per se, but just a state of things right now. And that state could be creating that way in any moment, and, in fact, it is.
It challenges our common conceptions of time, while making you realize that each moment is really new, and that you are not the same person you were a minute ago. This doesn't mean the world is manufactured, just that there is no inertia. Rather, it is an series of states, each with no "memory" of another. (Then again, what is memory if not a state?)
Gosh, I hope this makes sense to people that haven't been doing Zen reading. I'm afraid I'm not very good at explaining the idea, and why it was an epiphany for me.
Actually, punctuation only goes inside quotes if the punctuation is in the quote (amazing, huh?) If you're quoting a fragment, then the punctuation goes outside the quotes.
I'm not a fan of being pedantic, but if you're going to be, at least be right.
...I am a Gentoo user and fan.
/etc/inittab file.
Gentoo is not necessarily good because of the product, but in large part because of the process. When you finish doing a stage whatever (especially 1) install, you end up learning an awful lot about Linux that someone that drops in a SuSe/RedHat/Fedora Core/whatever disk doesn't know. Most experienced Linux users will see that a user that understands whats going on under the hood will fare better than one who gives you a thousand yard stare when you mention the
I think the benefits of compiling from source on everything are varied at best, and only sometimes outweighed by the time necessary to do it. That said, in some cases it is a good thing - if used correctly, the USE flags are nifty and let you compile without support for features you don't need. This can be quite useful, and provide a modest speed up in some cases.
Ricers aside, Gentoo provides a superb package management system in the spirit of apt/yum, and is also source based. It boosts users with moderate knowledge level to a better understanding of the architecture of a Linux system, and this can lead to some absurd enthusiasm about the distro for the younger/more impressioanable types, but I take it much the same way I take any fanboy mentality: you'll see the upsides and the downsides as time goes on. I happen to think Gentoo is great on the whole, so I use it.
Its just as childish for the folks annoyed by the Gentoo zealots to turn around be be anti-Gentoo zealots, creating webpages and ranting on about how horrible a community it is. Stop by the forums and you'll see its a responsive, well informed group, the majority of whom are quite reasonable.
Yes, but KDE isn't dealing with only network transparency. It also deals with compression transparency, format transparency and protocol transparency. This includes handling of tar.gz files without a compression program, kamera://, and, as the original post pointed out, other formats like info://.
Basically its an extensible system that allows for any protocol to be used. No one else has that yet.
And it doesn't play Ogg Vorbis yet? You've got to be kidding me.
That's the only thing I've been waiting for to pick one of these up...
The low hardware profit margin is because of:
1) Intense competition
2) Incredibly fast product cycle
3) Low yield on top-end products (moving to a smaller process == $$$)
So I suppose you could simplify and say that is just "research", but its also simply the nature of the market. There simply is no market like that in software - its very monopoly-ish (no hardware vendor has 90% of the market, even in processors), and the product lifecycle is much longer, with relatively little advancement between generations, unlike the hardware generational gap, which is large (Moore's law).