It's like lynx, only on steroids. It's like the web is pure and clean again.
I like w3m well enough, but I'm still a fan of lynx for many things...
lynx doesn't make much effort to do graphical layout, all of the content
is just linearized, and presented one piece at a time. This means that
typically there will be several screens of cruft you need to page down
through to get to the main body of text, but it also means that the
"designer" can't dictate how you're going to use your screen real estate.
(Also, it doesn't leak like a sieve and crash all the time: I can open up
a lynx window showing a long document, and it will still be there a month
later after four Firefox 1.x crashes or thirty Firefox 2.x crashes. But then,
w3m is pretty much the same as far as that goes.)
Tis true, if you're looking for emacs integration, the w3m world is much
better -- I've been living without for awhile, myself (I had some random
set-up hassle I didn't feel like figuring out.)
Though I disagree with Stallman on the totality of why he insists on saying Gnu/Linux, I do understand that the kernel and OS layers are not usable without some kind of interface which, in the most basic implementations, is usually the Gnu utilities (command line tools, compilers, etc.).
In other words, you fundamentally agree with Richard Stallman, but it's necessary to state that you don't,
because all good slash kiddies know that he's a bad, baad, man.
Anyway, yeah, this story is a poster-child for the importance of distinguishing between the kernel and the whole system (which we
traditionally call "the distro").
If you actually read the article, by the way, it doesn't take long to discover that the author has some severe reading comprehension problems:
"I don't think they're equally [updated to fix word] flawed - I think Leopard is a much better system," Torvalds said. But then he added: "OS X is in some ways actually worse than Windows to program for. Their file system is complete and utter crap, which is scary."
"I think [Mac] OS X is nicer than Windows in many ways," he continued. "But neither can hold a candle to my own [Linux]. It's a race for second."
And while you would expect this kind of propaganda from the operating system's founder, does he even speak to (or for) the Linux community anymore? If you ask me, he's just another wolf howling in the night hoping someone will agree.
Here Torvalds is talking about the experience of programmers working on third-party applications, but the author seems to think he's talking about the ease-of-use for end users, which are completely different worlds.
So unless the point of this is to get the slashmob to help smarten up a dumb-bunny, there isn't any point at all in this story.
Colin Powell's speech to the UN mentioned camps in the northern part of Iraq that we cleared out within a few weeks after Baghdad fell. And Gee Whiz, they even found traces of chemical weapons at one of the camps.
Wowie zowie: the moldy remains of some shells of chemical weapons... but what happened to the mushroom clouds
the Bush regime was trying to scare us with? And if Saddam Hussein had weapons of mass destruction,
why didn't he actually use them on US troops before they made it into Bagdhad? It was all part of his scheme,
to let us destroy Iraq so that we would never discover his awesome capabilities until -- uh, until when?
Bush declared war based on his accusations that Saddam Hussein was trying
to obtain, or already had Weapons of Mass destruction.
I know this piece of fiction has been told and retold in the MSM media so many times it seems like fact, but it isn't. Google can provide you with the actual speech Bush gave. Concern that Saddam would either himself use WMD or supply them to others was ONE of the stated reasons. But not the primary one.
Yes, we all know there's a conspiracy against the Bush administration, by that damned liberal media.
(So now we're only allowed to consider one Bush speech, eh? We're supposed to forget Cheney and Rice
waving spectre's of "mushroom clouds"?).
Incidentally, can you explain what the emergency was about invading before the UN weapon inspection of Iraq
was completed? They publicly stated they needed "not weeks or years, but months", but for some reason
the Bush regime figured they'd better move before the results were in. Funny, eh?
Does anyone here believe the voting process is rigged one way or another?
In the 2004 election, there were peculiar discrepancies between the exit polls
and the reported results which correlated with the use of electronic voting machines:
Who won?.
Statistical indications like this are all you're likely to get in this game, and now you won't even get that much -- they changed the way exit poll data is handled because the 2004 situation was too embarrassing.
Remember, Bush stole the 2000 election largely because of paper ballots.
Punch cards != paper ballots.
Op scan ballots are okay, but at a minimum there needs to be random, hand-counted audits conducted in public view,
and some clear rules that state that any discrepancy should trigger a full re-count by hand. For example: if the totals for all candidates don't match the number of people who voted at a polling place, there should be a recount.
Requiring some third party to pay for audits that ought to be part of the system is ridiculous.
And yes, whether the op scan machines are nominally running "open source" software is close to irrelevant:
fancy key encryption digitial signature tricks are far too obscure to expect the general public to trust
the results -- getting the count right isn't enough, people have to believe that the count was done correctly,
and the simpler the system the better.
After 7 years, I suspect this covers the entire Perl6 development team too.
I can't tell if your comment here is supposed to be approving
perserverence or chiding stubborness, but in any case, the perl 6
development effort has achieved some notable successes over the
years (and few, if any, members of the team have been working
soley on perl 6...).
Off the top of my head:
Many Perl 6 concepts were implemented as perl 5 modules, and some
have become core features in perl 5.10
Pugs: reference implementation of perl 6 implemented in haskell
Perl 6 on Parrot (now called Rakudo
continues to progress...
Simon Cozens is impressed with the state of the Parrot
Compiler Toolkit:
"Parrot lets you implement your own languages using Perl 6 rules for the grammar and Perl 6 for the compiler."
The researchers studied a group of 26 men, 12 of whom had the A1 gene mutation for low numbers of D2 receptors.
Got that? Sample size: 26.
People just eat this shit up these days, they love biological "explanations" for human
behavior. Hey, it's not my fault, I was born this way.
Work harder in school? But if you don't have the natural talent, what's the point?
Spend more money on public
education? Oh, what the hell for? Those people will never learn.
Code written once worked on 95% of the computers. Would a better OS be superior? Yes. However, kernels, like phone companies, work best if there is only one.
Really? The Free Software Foundation doesn't have so much trouble supporting multiple kernels. You just take the source and re-compile it...
And you know, the phone system in the United States only got better after the Ma Bell breakup... long-distance rates went down,
and all of a sudden you could buy all sorts of inexpensive third-party equipment (e.g. answering machines).
Most are exceptionally lame, plus, they really missed the point of "Ceci n'est pas une pipe".
Yes, I was thinking that it would be clearer in psenudo-perl: $pipe_ref = \$pipe; $pipe_ref <> $pipe.
A side-effect of many features being "hacked-in" to perl, is that you need to think about what's going on a little more
than in some other languages, as in this case, where you take references explicitly, and you need to understand that
there's a difference between the reference and the thing referenced. The entire point of the paradox in this
painting is to make people conscious of the fact that they often conflate the two... ("the map is not the territory").
Top replies are just as easy on the reader than bottom replies. Moreso if you bottom-repliers would give up and quit it. There's nothing confusing about a bottom-to-top ordered thread, especially if you know it's in that order!
Uh, huh. If you want to understand the current message, you're supposed to back up to the beginning, and read/skim it all in order...
I repeat: this appeals to lazy writers. It's not the kind of thing a good writer demands of their audience.
Reading alternating lines of dialog, proceeding from top to bottom, is not exactly unfamiliar to English readers.
Even if it's unfamiliar to someone in email context, it doesn't take more than half a second to get it.
I like top posting. I want to see the message that is for me at the top, and not to have to scroll through a series of replies to get to where my message really begins. Plus, if it was relevant to me I've probably already seen it, although it's nice to have for context.
This style works -- to the extent that it does -- with a rapid exchange of replies among a small group of recipients who can be expected to read it immediately. Not suprisingly, this describes the corporate enviroment.
It's very, very bad if you're writing for the benefit of a larger audience, some of whom might not have been reading the entire exchange, who may be coming in the middle some time after the message was written.
It has a great failing in writing for either type of audience, in that it's often difficult to tell which part of the message the reply is referring to.... the reply-below, trim for relevence style (that people who've been using
computers since before you were born long ago figured out was the right way to do it) allows a much finer granularity of response.
My own experience is that the reply-below style is *in general* much easier on readers -- certainly it never confuses readers, if it's done right. On the other hand, the reply above style is really easy on lazy writers, and consequently leads to rapid exchanges that spiral out of control in ways that the participants are only half-way conscious of.
Dude, what we have here is a business opportunity. You need to organize corporate seminars
in email effectiveness. Spend a day putting together inane power point slides pointing
out the obvious in bulleted lists, and they'll pay you money to rant at them -- I mean, gently criticize with humble, humorous anecdotes -- their
stupid email habits.
If it comes packaged as a corporate training session, then it's got to be good.
No one listens to some computer geek jabbering in email, I mean come on.
I try it. I really do. But more often than not i get negative responses for it, because people do no longer recognize the old reply-style i learned in the early nineties. They expect my text on top of theirs and don't even look below the quoted lines for additional lines.
My experience is the opposite. No one has any trouble reading correctly formatted, reply-below-the-quote text -- though it could be
you're not trimming enough of the material if they don't even see your response down there.
I got a comment from an email-beginner once "I have no trouble following an exchange between you and , why can't the other people on the list learn how to use email?"
There is admittedly a problem with repeated exchanges between reply-below-the-quote guys talking to the lazy above-the-quote-people...
The actual point -- to the extent that there is any -- of copyright protections is not
to require that derivative works "add value" (how is the court system
supposed to judge that???). The point is to prevent someone else from
reducing the value of the copyrighted work. If you splice together
a lot of snippets from some existing movies, you may easily give
people the feeling that they don't care about seeing the originals: if
you splice together a small number of clips, that's supposed to be
covered by fair use (and an intelligent copyright holder would want
you to do that, because you may very well increase interest in the original).
In the print media world, the principles of quotation are enshrined in custom
(though not the letter of the law) and it's understood you don't need to worry if
you short quote passages of less than 50 words. The difficulty with recorded music and film is that
the money-grubbing bastards have been fighting the establishment of similar standards
for other media: e.g. rather than developing a rule-of-thumb that, say, 3 second
samples are okay, they try to milk people for "sampling rights".
Jon Stokes here in his The arstechnica article
is doing a "sing-them-to-sleep" operation, if you ask me:
For example: oh, he's so tired of all these silly internet
amateurs, just imagine what the rest of the election process is
going to be like! And: all these guys are just trying to make a
name for themselves. (They're not, for example, nervously
checking to see if there's any hope that the American Republic is intact.)
He informs us that he doesn't think there was any election fraud
here, but doesn't bother to enlighten us as to why (his reasons
are "too vague").
One of the links he points to is particularly funny:
Robert
Hansen. Here Robert Hansen is plugging through the
statistical analysis, turning up a significant correlation
on the counting technique and not on the other possible
variables, but he remains convinced that this must be wrong...
and is in the process of actively looking for "sources of
error" to fix the results.
Anyway:
if you're interested in this subject, I suggest keeping an eye
on Brad Friedman's bradblog.
(By the way: why are people assuming this would have to be Hillary's
fault? A "republican dirty trick" theory would seem a little more likely to me:
kill the momentum of the front-runner, sew dissention in the ranks, maybe steer the primary toward a more
defeatable candidate... why not?)
PostgreSQL adoption: why?
Really, I still have not heard a good reason why one should use - in 2005, let alone today - PostgreSQL over MySQL. Technology wise I've never heard a compelling reason to use PostgreSQL that applies to recent (well, no so recent) versions.
You're not looking hard enough. Postgresql reportedly scales better (to higher load, and also for multiple processors),
and it's had serious RDBMS features for much longer than mysql: I wouldn't
assume that mysql has done much more than seize bullet-points without confirming
that the features really work... and since most mysql users don't really care
about advanced RDBMS features, I doubt they're getting exercised that heavily.
The data shows that town size, voting pattern, and machine-countedness are all correlated. There are several possible explanations:
And what I'm trying to point out to you is that all you're doing is showing that the machine counts are consistent with an "urban bias" theory, you can't use them to confirm that the machine counts are not corrupt.
If you start with they hypothesis that the machine counts could be corrupt, you need another source of data to contradict that theory.
That alternate theories exist is all well and good, but not in question.
There's an important lesson here: not everything you read online is correct.
So not only do you need to RTFA, you need to think about TFA. (The horror!)
And I thought about your post, and realized that the post you were pointing to for support appears to be nonsense, and I realized that your post was also quite likely nonsense.
(Which is not to say that I think we can prove that there's no urban bias for Clinton... but I find it a little
suspicious when an effect like this is trumped up after the fact. Wouldn't it be interesting if actually had a meaningful
recount of this data -- as called for by Denis Kucinich, by the way -- and we could stop speculating?)
As a NH voter - one of those damned independents Hillary and her campaign people were complaining about before she won - and a guy who actually hates and works to remove Diebold machines from the election process here, the differences are pretty minor and seem to be easily explained. The areas using the machines differ VASTLY from those that do not, in most cases. The socioeconomic classes and lifestyles vary across the entire state, and you tend to see the machines in places you see the same types of people. I think, for some reason, Hillary fared better in the cities than did Obama.
Also, afaik all the Diebold machines here only count. All of us still scribble in a dot on a paper ballot. We have paper trails for my city, Nashua, which is one that went to Hillary over Obama.
And do you, or do you not, want to see those dots recounted?
(Ever noticed how often you see postings on the net with these kinds of claims of unverfiable creditials? "I'm a cyclist myself, but I think any bike that gets in the way of a cab deserves to be run over.")
If you take the cities from TFA (> 5,000 votes, all counted by machine)
Am I mistaken, or are you actually using data from the machine counts to confirm that the machine counts must be okay?
The "urban areas favor Clinton" hypothesis is interesting, but it strikes me that no one mentioned this fact before
the election.
Do we, or do we not, do a recount of the paper trail -- preferably by hand?
On the other hand, we place our faith in this theory trumped up after-the-fact to explain a suspect result, or on the
other hand, we gain some actual knowledge that we can use to confirm or deny the integrity of our electoral system.
After 2004, the Edison/Mitosfsky company adopted a policy of waiting for election results, and then using that data to "correct" their exit polling data. The public no longer sees raw exit polling data, and it will not be reported on CNN. What they do report "matches perfectly", because it was adjusted to do so.
Obama 'lost' to Clinton. But they each received 9 pledged delegates. Nationally, there 3515 pledged delegates at stake. Look beyond the hype at the numbers: even if fraud was detected, it shouldn't make on damn bit of difference.
Yeah, election integrity is a purely academic issue. (Of course, the same election machines are in use, with the same dubious firm in charge of them, in other states, but no doubt this is just a one-off.) So let's just let this one ride, and forget about it. And mostly likely use the same damn system next time around...
Myself, I think there are a number of interesting things about this beyond "who won". One might be "who dunnit?" If the Clinton campaign managed to arrange this, that would be a good thing to know... on the other hand, if it's a Republican Dirty Trick (to sabotage democratic momentum, sew dissention in the ranks, possibly even to promote a candidate they think is beatable), that would also be a good thing to know. And if there's no way to know who did it, that would also be an interesting thing to know.
It's important to note that in all these precincts the exit polls agreed with the actual results.
Where the hell did you get that?
In any case, after the 2004 election, the Edison/Mitofsky polling firm was so embarrassed by their
exit polls contradicting the official results, that they adopted a new policy of hanging on to the results
until the official results are in, and using those figures to "correct" the exit polls before releasing them.
(You got that? I wish I were kidding, but I am not.)
Slashdot Mirror
I like w3m well enough, but I'm still a fan of lynx for many things... lynx doesn't make much effort to do graphical layout, all of the content is just linearized, and presented one piece at a time. This means that typically there will be several screens of cruft you need to page down through to get to the main body of text, but it also means that the "designer" can't dictate how you're going to use your screen real estate.
(Also, it doesn't leak like a sieve and crash all the time: I can open up a lynx window showing a long document, and it will still be there a month later after four Firefox 1.x crashes or thirty Firefox 2.x crashes. But then, w3m is pretty much the same as far as that goes.)
Tis true, if you're looking for emacs integration, the w3m world is much better -- I've been living without for awhile, myself (I had some random set-up hassle I didn't feel like figuring out.)
In other words, you fundamentally agree with Richard Stallman, but it's necessary to state that you don't, because all good slash kiddies know that he's a bad, baad, man.
Anyway, yeah, this story is a poster-child for the importance of distinguishing between the kernel and the whole system (which we traditionally call "the distro").
If you actually read the article, by the way, it doesn't take long to discover that the author has some severe reading comprehension problems:
Here Torvalds is talking about the experience of programmers working on third-party applications, but the author seems to think he's talking about the ease-of-use for end users, which are completely different worlds.So unless the point of this is to get the slashmob to help smarten up a dumb-bunny, there isn't any point at all in this story.
Wowie zowie: the moldy remains of some shells of chemical weapons... but what happened to the mushroom clouds the Bush regime was trying to scare us with? And if Saddam Hussein had weapons of mass destruction, why didn't he actually use them on US troops before they made it into Bagdhad? It was all part of his scheme, to let us destroy Iraq so that we would never discover his awesome capabilities until -- uh, until when?
Yes, we all know there's a conspiracy against the Bush administration, by that damned liberal media. (So now we're only allowed to consider one Bush speech, eh? We're supposed to forget Cheney and Rice waving spectre's of "mushroom clouds"?).
Incidentally, can you explain what the emergency was about invading before the UN weapon inspection of Iraq was completed? They publicly stated they needed "not weeks or years, but months", but for some reason the Bush regime figured they'd better move before the results were in. Funny, eh?
In the 2004 election, there were peculiar discrepancies between the exit polls and the reported results which correlated with the use of electronic voting machines: Who won?.
Statistical indications like this are all you're likely to get in this game, and now you won't even get that much -- they changed the way exit poll data is handled because the 2004 situation was too embarrassing.
Punch cards != paper ballots.
Op scan ballots are okay, but at a minimum there needs to be random, hand-counted audits conducted in public view, and some clear rules that state that any discrepancy should trigger a full re-count by hand. For example: if the totals for all candidates don't match the number of people who voted at a polling place, there should be a recount.
Requiring some third party to pay for audits that ought to be part of the system is ridiculous.
And yes, whether the op scan machines are nominally running "open source" software is close to irrelevant: fancy key encryption digitial signature tricks are far too obscure to expect the general public to trust the results -- getting the count right isn't enough, people have to believe that the count was done correctly, and the simpler the system the better.
I can't tell if your comment here is supposed to be approving perserverence or chiding stubborness, but in any case, the perl 6 development effort has achieved some notable successes over the years (and few, if any, members of the team have been working soley on perl 6...).
Off the top of my head:
Got that? Sample size: 26.
People just eat this shit up these days, they love biological "explanations" for human behavior. Hey, it's not my fault, I was born this way. Work harder in school? But if you don't have the natural talent, what's the point? Spend more money on public education? Oh, what the hell for? Those people will never learn.
Really? The Free Software Foundation doesn't have so much trouble supporting multiple kernels. You just take the source and re-compile it...
And you know, the phone system in the United States only got better after the Ma Bell breakup... long-distance rates went down, and all of a sudden you could buy all sorts of inexpensive third-party equipment (e.g. answering machines).
Yes, I was thinking that it would be clearer in psenudo-perl: $pipe_ref = \$pipe; $pipe_ref <> $pipe.
A side-effect of many features being "hacked-in" to perl, is that you need to think about what's going on a little more than in some other languages, as in this case, where you take references explicitly, and you need to understand that there's a difference between the reference and the thing referenced. The entire point of the paradox in this painting is to make people conscious of the fact that they often conflate the two... ("the map is not the territory").
Uh, huh. If you want to understand the current message, you're supposed to back up to the beginning, and read/skim it all in order...
I repeat: this appeals to lazy writers. It's not the kind of thing a good writer demands of their audience.
Reading alternating lines of dialog, proceeding from top to bottom, is not exactly unfamiliar to English readers. Even if it's unfamiliar to someone in email context, it doesn't take more than half a second to get it.
This style works -- to the extent that it does -- with a rapid exchange of replies among a small group of recipients who can be expected to read it immediately. Not suprisingly, this describes the corporate enviroment.
It's very, very bad if you're writing for the benefit of a larger audience, some of whom might not have been reading the entire exchange, who may be coming in the middle some time after the message was written.
It has a great failing in writing for either type of audience, in that it's often difficult to tell which part of the message the reply is referring to.... the reply-below, trim for relevence style (that people who've been using computers since before you were born long ago figured out was the right way to do it) allows a much finer granularity of response.
My own experience is that the reply-below style is *in general* much easier on readers -- certainly it never confuses readers, if it's done right. On the other hand, the reply above style is really easy on lazy writers, and consequently leads to rapid exchanges that spiral out of control in ways that the participants are only half-way conscious of.
Dude, what we have here is a business opportunity. You need to organize corporate seminars in email effectiveness. Spend a day putting together inane power point slides pointing out the obvious in bulleted lists, and they'll pay you money to rant at them -- I mean, gently criticize with humble, humorous anecdotes -- their stupid email habits.
If it comes packaged as a corporate training session, then it's got to be good.
No one listens to some computer geek jabbering in email, I mean come on.
My experience is the opposite. No one has any trouble reading correctly formatted, reply-below-the-quote text -- though it could be you're not trimming enough of the material if they don't even see your response down there.
I got a comment from an email-beginner once "I have no trouble following an exchange between you and , why can't the other people on the list learn how to use email?"
There is admittedly a problem with repeated exchanges between reply-below-the-quote guys talking to the lazy above-the-quote-people...
The actual point -- to the extent that there is any -- of copyright protections is not to require that derivative works "add value" (how is the court system supposed to judge that???). The point is to prevent someone else from reducing the value of the copyrighted work. If you splice together a lot of snippets from some existing movies, you may easily give people the feeling that they don't care about seeing the originals: if you splice together a small number of clips, that's supposed to be covered by fair use (and an intelligent copyright holder would want you to do that, because you may very well increase interest in the original).
In the print media world, the principles of quotation are enshrined in custom (though not the letter of the law) and it's understood you don't need to worry if you short quote passages of less than 50 words. The difficulty with recorded music and film is that the money-grubbing bastards have been fighting the establishment of similar standards for other media: e.g. rather than developing a rule-of-thumb that, say, 3 second samples are okay, they try to milk people for "sampling rights".
For example: oh, he's so tired of all these silly internet amateurs, just imagine what the rest of the election process is going to be like! And: all these guys are just trying to make a name for themselves. (They're not, for example, nervously checking to see if there's any hope that the American Republic is intact.)
He informs us that he doesn't think there was any election fraud here, but doesn't bother to enlighten us as to why (his reasons are "too vague").
One of the links he points to is particularly funny: Robert Hansen. Here Robert Hansen is plugging through the statistical analysis, turning up a significant correlation on the counting technique and not on the other possible variables, but he remains convinced that this must be wrong... and is in the process of actively looking for "sources of error" to fix the results.
Anyway: if you're interested in this subject, I suggest keeping an eye on Brad Friedman's bradblog.
(By the way: why are people assuming this would have to be Hillary's fault? A "republican dirty trick" theory would seem a little more likely to me: kill the momentum of the front-runner, sew dissention in the ranks, maybe steer the primary toward a more defeatable candidate... why not?)
T-Ranger wrote:
You're not looking hard enough. Postgresql reportedly scales better (to higher load, and also for multiple processors), and it's had serious RDBMS features for much longer than mysql: I wouldn't assume that mysql has done much more than seize bullet-points without confirming that the features really work... and since most mysql users don't really care about advanced RDBMS features, I doubt they're getting exercised that heavily.
And what I'm trying to point out to you is that all you're doing is showing that the machine counts are consistent with an "urban bias" theory, you can't use them to confirm that the machine counts are not corrupt.
If you start with they hypothesis that the machine counts could be corrupt, you need another source of data to contradict that theory.
That alternate theories exist is all well and good, but not in question.
And I thought about your post, and realized that the post you were pointing to for support appears to be nonsense, and I realized that your post was also quite likely nonsense.
(Which is not to say that I think we can prove that there's no urban bias for Clinton... but I find it a little suspicious when an effect like this is trumped up after the fact. Wouldn't it be interesting if actually had a meaningful recount of this data -- as called for by Denis Kucinich, by the way -- and we could stop speculating?)
GenKreton wrote:
And do you, or do you not, want to see those dots recounted?
(Ever noticed how often you see postings on the net with these kinds of claims of unverfiable creditials? "I'm a cyclist myself, but I think any bike that gets in the way of a cab deserves to be run over.")
Am I mistaken, or are you actually using data from the machine counts to confirm that the machine counts must be okay?
The "urban areas favor Clinton" hypothesis is interesting, but it strikes me that no one mentioned this fact before the election.
Do we, or do we not, do a recount of the paper trail -- preferably by hand? On the other hand, we place our faith in this theory trumped up after-the-fact to explain a suspect result, or on the other hand, we gain some actual knowledge that we can use to confirm or deny the integrity of our electoral system.
Heigh-ho.
Jeremy Erwin wrote:
Yeah, election integrity is a purely academic issue. (Of course, the same election machines are in use, with the same dubious firm in charge of them, in other states, but no doubt this is just a one-off.) So let's just let this one ride, and forget about it. And mostly likely use the same damn system next time around...
Myself, I think there are a number of interesting things about this beyond "who won". One might be "who dunnit?" If the Clinton campaign managed to arrange this, that would be a good thing to know... on the other hand, if it's a Republican Dirty Trick (to sabotage democratic momentum, sew dissention in the ranks, possibly even to promote a candidate they think is beatable), that would also be a good thing to know. And if there's no way to know who did it, that would also be an interesting thing to know.
Kristoph wrote:
Where the hell did you get that?
In any case, after the 2004 election, the Edison/Mitofsky polling firm was so embarrassed by their exit polls contradicting the official results, that they adopted a new policy of hanging on to the results until the official results are in, and using those figures to "correct" the exit polls before releasing them. (You got that? I wish I were kidding, but I am not.)
Now, take note of this story by Brad Friedman: Chris Matthews: Raw EXIT POLL Data 'Indicated Significant Victory' for Obama in NH 'Was Ahead an Average of 8 Points, Even in Our Own Exit Polls': "Even the Exit Polls showed that Obama should have won, according to Chris Matthews on Hardball today. It's the first specific indication that we've seen that the raw, unadjusted Exit Poll data, which only corporate mainstream media folks, not mere mortals, are allowed to see, confirmed all of the pre-election polling which predicted an Obama win."