Good lord, either I haven't had enough coffee, or you're not making any sense. Look up the word malign. It means to maliciously and falsly accuse.
Actually, no, it doesn't. There's nothing "false" necessarily or even usually implicated by the use of the word "malign." Not sure where you get that from, but it wasn't a dictionary.
I'm not saying the GOP is exploiting anything. I'm just wondering why you would make a statement like you did.
Because it is just as true as what was said in the article.
Just answer my first question before you pose one to me.
Why? The article made its statement before I made mine.
Say what? If Democrats were well known for exploiting paper ballots, why would they be protesting moving back to paper ballots?
Did you not read the article, which accused the GOP of "exploiting flaws in electronic systems"? So answer this: if the GOP is known for exploiting flaws in electronic systems, why would they be trying to move to paper ballots?
I'll grant you that a lot of RFK's numbers may be a little inflated, artificial or puffed up.
Well, no: most of them are complete bullshit with no basis in fact whatsoever.
But what about anomalies like rural voters in southern Ohio voting for Bush and against an anti-gay marriage-amendment ballot measure in the state?
What about it? I voted for Bush, and would be against such a constitutional amendment.
That doesn't make any sense at all on any criteria that I can figure out. Have you got any hypothesis to explain that?
I don't need one. I would vote that way.
That you cannot understand how someone would do that is not a basis for believing the votes have been manipulated. The only thing it may possibly be a basis for is doing further investigation: you see what you think is an anomaly, so you do more work to see if there is actual error or wrongdoing. If you find none, then... it remains at most a mere anomaly, representing nothing in particular about how the votes were counted.
And there's still the central claim that exit polls are the gold standard in every other country about how honest and fair an election is.
The central, and false, claim. Like most polls, they require the pollster to manipulate the results to get what he thinks is a representative sample of core demographics. But it's even harder to do right with exit polling than with normal polling. So Zogby does an exit poll in rural WA, and he gets a response of 150 self-identified Democrats, and 50 self-identified Republicans, all of whom voted for their party's candidate. Obviously, it would be silly to think that Kerry will win by a ratio of 3:1; so, how do they deal with that? One way is that they have pregenerated numbers representing Republican and Democratic voter ratios in the area, so if they know the area is 60 percent GOP and 40 percent Democrat, then they massage the actual results accordingly.
And there are many other problems that can be encountered, as well. How do you take absentee voters into account, since you do not touch them with exit polls? Are they more or less likely to vote for Kerry or Bush? More or less likely to vote party line? More or less likely to... vote at all?
The first, and primary, thing to look at for problems with exit polls is methodology as conducted and assumptions made. Are their democgraphic assumptions used to massage the data accurate? Did they ask good questions, from a wide variety of areas?
Once you start learning more about how exit polling is actually done, how prone to error it is, Occam's Razor will tell you (being that the most obvious answer is usually the correct one) that the error is probably in the polling itself.
But let's assume the exit polls were right, and the official count was wrong. You can't prove it, of course, by just assuming the exit polls SHOULD be right. There's a reason it's called the "official count." Unless you can prove it by showing HOW the official count was actually incorrect -- something RFK Jr. lied through his teeth to attempt -- it's meaningless supposition. Again, as with the "anomaly" you cite about people voting against the anti-gay-marriage amendment but for Bush, this is merely reason to investigate further, and not serious evidence of error or fraud by itself.
And don't get me wrong, I am not saying this just because I voted for Bush. Here in WA, our governor's race in 2004 was far closer than the Presidential race in Ohio, or even Florida in 2000. A race with over three million votes for governor was decided by about 150 votes. The final count went against the candidate I voted, and campaigned, for, Dino Rossi. Many of my GOP colleagues say that the Democrats "stole" the election. Maybe they did, but I will not consider that a reasonable possibility without evidence that actually shows it. And the evidence in WA is similar: Democratic operatives vandalizing GOP offices, missing vo
Here's the executive summary: "[RFK Jr.] claims 357,000 legal voters were denied the right to vote, or did not have their legal vote counted. He has no actual data to justify the inclusion of at least 347,000 of the 357,000, and his claim that this is mostly the fault -- let alone the intent -- of Republicans is, to be kind, specious."
Enable it in Firefox to recreate the situation I'm trying to tell you about. Now go into Internet Explorer 6 Go to PREFERENCES go to COMMENTS and all I get is whitespace below Discussion Style
You said before that you "I see it. HOWEVER... if I re-enable the University whatever option and saved it now bang."
I don't know what "now bang" means and now you say that you get nothing, whereas before you said you saw it. So I still have no idea what is happening that you actually see and experience, because you're saying conflicting things.
My point is, if you read my original post, that i made the change in firefox and then it killed IE.
And I told you that I disbelieve it, and if it really is happening, to tell me the exact steps you take to get there, and exactly what happens. You still haven't done that. You just said "it blew up" and then "if I re-enable the University whatever option and saved it now bang," neither of which tell me what actually happened, and nowhere do you say the actual steps you took.
Until you tell me the actual exact steps you take and we can attempt to narrow down whether there is actually a problem, and if so what is happening... what would you have me do? Wave a magic wand?
It is written a way that is almost equivalent of var 1 = 2;
No, it isn't. "property" != "identifier". Indeed, var c = { 1: 2} is explicitly allowed in the ECMA spec. On page 42, it says that in c = { 1: 2} that the "1" is a PropertyName which may be evaluated as an Identifier, StringLiteral, or NumericLiteral. If it only said Identifier there, you would be correct. But you are incorrect.
That object property names are parsed and interpreted differently than variable names in other browsers than Opera is the real bug here.
The real bug in this discussion is that you are asserting things that flatly contradict the specification.
lists an identifier as starting with a letter, underscore, dollar sign, or escape sequence. It canNOT start with a number.
It's not an identifier, though, it's a property.
This is probably the reason why var y = { 8388607: 1 }; fails in some browsers (IE5, for example), but var y = { "8388607": 1 }; (with the quotes) succeeds.
I'll take your word for it that it fails in IE5, but it works everywhere we've tested (including IE7, which is the only IE this ever has a chance of working with anyway).
You can't click "Preferences" at the top of this page (or any other page on the site)? Then click "Comments" to edit the comments prefs? Then click "Discussion Style" -> "Normal"? Then "Save"?
I see no reason why any of that should crash IE. I disbelieve (especially since I just tried it and it worked in IE 7.0.5346.5 Beta 2). If this really did make IE "blow up," then please tell me the version/build of IE you are using, and the exact steps you took, and I'll try it out.
Does it still show the same behavior if use a string representation of a number
Yes. I could not force a string representation (except by changing the value, such as starting it with a letter). I suspect this too is a bug, but I am not sure.
Although, that was Opera 8, and I cannot recall if I tried that part with Opera 9. So, testing again...
And yes, I see no evidence anything up to and including (2^31)-1 should be a problem, but we are a loooooong way off from that (almost literally, 128 times the number of comments we've had to this point in Slashdot's history, as we fast approach 2^24). We'll deal with it later if we have to... but at this rate, we'll hit it in... oh, a thousand years or so.:-)
We've never seen that. If you can show it to us happening, give us all the details (URL, browser, platform, etc.), then file a bug report, and we can look into it.
Do you have actual numbers to show that the "potentially thousands of concatenations" would actually add measurable overhead?
Nope. But I know it could, and I am not going to take the risk until it can be measured to NOT add overhead. This is a very difficult thing to know for sure: there are many browsers on many platforms.
We actually are going to do another round of profiling, and we may throw this into the batch of things we look at, if we have the time. Even then, though, it will be difficult to reasonably profile all common browsers on all platforms to know for sure that it won't have a deleterious effect, just to work around the bug in Opera. So, bottom line, IF we have the time to profile it, IF it turns out to not be a problem, IF we think that our profiling is representative, IF Opera has still not fixed the bug, we'll consider making the change.
But it's a whole lot easier and better for everyone if the bug gets fixed, don't you think?
Slashdot Mirror
Good lord, either I haven't had enough coffee, or you're not making any sense. Look up the word malign. It means to maliciously and falsly accuse.
Actually, no, it doesn't. There's nothing "false" necessarily or even usually implicated by the use of the word "malign." Not sure where you get that from, but it wasn't a dictionary.
I'm not saying the GOP is exploiting anything. I'm just wondering why you would make a statement like you did.
Because it is just as true as what was said in the article.
Just answer my first question before you pose one to me.
Why? The article made its statement before I made mine.
Say what? If Democrats were well known for exploiting paper ballots, why would they be protesting moving back to paper ballots?
Did you not read the article, which accused the GOP of "exploiting flaws in electronic systems"? So answer this: if the GOP is known for exploiting flaws in electronic systems, why would they be trying to move to paper ballots?
Of course, Democrats are more well-known for exploiting paper ballots.
Great, thanks!
I'll grant you that a lot of RFK's numbers may be a little inflated, artificial or puffed up.
... it remains at most a mere anomaly, representing nothing in particular about how the votes were counted.
... vote at all?
Well, no: most of them are complete bullshit with no basis in fact whatsoever.
But what about anomalies like rural voters in southern Ohio voting for Bush and against an anti-gay marriage-amendment ballot measure in the state?
What about it? I voted for Bush, and would be against such a constitutional amendment.
That doesn't make any sense at all on any criteria that I can figure out. Have you got any hypothesis to explain that?
I don't need one. I would vote that way.
That you cannot understand how someone would do that is not a basis for believing the votes have been manipulated. The only thing it may possibly be a basis for is doing further investigation: you see what you think is an anomaly, so you do more work to see if there is actual error or wrongdoing. If you find none, then
And there's still the central claim that exit polls are the gold standard in every other country about how honest and fair an election is.
The central, and false, claim. Like most polls, they require the pollster to manipulate the results to get what he thinks is a representative sample of core demographics. But it's even harder to do right with exit polling than with normal polling. So Zogby does an exit poll in rural WA, and he gets a response of 150 self-identified Democrats, and 50 self-identified Republicans, all of whom voted for their party's candidate. Obviously, it would be silly to think that Kerry will win by a ratio of 3:1; so, how do they deal with that? One way is that they have pregenerated numbers representing Republican and Democratic voter ratios in the area, so if they know the area is 60 percent GOP and 40 percent Democrat, then they massage the actual results accordingly.
And there are many other problems that can be encountered, as well. How do you take absentee voters into account, since you do not touch them with exit polls? Are they more or less likely to vote for Kerry or Bush? More or less likely to vote party line? More or less likely to
The first, and primary, thing to look at for problems with exit polls is methodology as conducted and assumptions made. Are their democgraphic assumptions used to massage the data accurate? Did they ask good questions, from a wide variety of areas?
Once you start learning more about how exit polling is actually done, how prone to error it is, Occam's Razor will tell you (being that the most obvious answer is usually the correct one) that the error is probably in the polling itself.
But let's assume the exit polls were right, and the official count was wrong. You can't prove it, of course, by just assuming the exit polls SHOULD be right. There's a reason it's called the "official count." Unless you can prove it by showing HOW the official count was actually incorrect -- something RFK Jr. lied through his teeth to attempt -- it's meaningless supposition. Again, as with the "anomaly" you cite about people voting against the anti-gay-marriage amendment but for Bush, this is merely reason to investigate further, and not serious evidence of error or fraud by itself.
And don't get me wrong, I am not saying this just because I voted for Bush. Here in WA, our governor's race in 2004 was far closer than the Presidential race in Ohio, or even Florida in 2000. A race with over three million votes for governor was decided by about 150 votes. The final count went against the candidate I voted, and campaigned, for, Dino Rossi. Many of my GOP colleagues say that the Democrats "stole" the election. Maybe they did, but I will not consider that a reasonable possibility without evidence that actually shows it. And the evidence in WA is similar: Democratic operatives vandalizing GOP offices, missing vo
Oh, also note that RFK Jr. is the same winner that blamed Bush and Barbour for Katrina, in one of the biggest bullshit articles of 2005.
RFK Jr.'s article is utter bullshit.
If you actually examine his claims, they simply do not hold up.
Here's the executive summary: "[RFK Jr.] claims 357,000 legal voters were denied the right to vote, or did not have their legal vote counted. He has no actual data to justify the inclusion of at least 347,000 of the 357,000, and his claim that this is mostly the fault -- let alone the intent -- of Republicans is, to be kind, specious."
https://slashdot.org/~pudge/journal/137466
Interesting idea, I'll look into it.
I have no idea if this would fix Opera's problems or not, though.
Enable it in Firefox to recreate the situation I'm trying to tell you about.
Now go into Internet Explorer 6
Go to PREFERENCES
go to COMMENTS
and all I get is whitespace below Discussion Style
You said before that you "I see it. HOWEVER... if I re-enable the University whatever option and saved it now bang."
I don't know what "now bang" means and now you say that you get nothing, whereas before you said you saw it. So I still have no idea what is happening that you actually see and experience, because you're saying conflicting things.
And where is the part about it "blowing up"?
My point is, if you read my original post, that i made the change in firefox and then it killed IE.
... what would you have me do? Wave a magic wand?
And I told you that I disbelieve it, and if it really is happening, to tell me the exact steps you take to get there, and exactly what happens. You still haven't done that. You just said "it blew up" and then "if I re-enable the University whatever option and saved it now bang," neither of which tell me what actually happened, and nowhere do you say the actual steps you took.
Until you tell me the actual exact steps you take and we can attempt to narrow down whether there is actually a problem, and if so what is happening
HOWEVER... if I re-enable the University whatever option and saved it now bang
Don't. Select "Normal".
Yes, there is something wrong with the code.
No, there isn't.
It is written a way that is almost equivalent of var 1 = 2;
No, it isn't. "property" != "identifier". Indeed, var c = { 1: 2} is explicitly allowed in the ECMA spec. On page 42, it says that in c = { 1: 2} that the "1" is a PropertyName which may be evaluated as an Identifier, StringLiteral, or NumericLiteral. If it only said Identifier there, you would be correct. But you are incorrect.
That object property names are parsed and interpreted differently than variable names in other browsers than Opera is the real bug here.
The real bug in this discussion is that you are asserting things that flatly contradict the specification.
lists an identifier as starting with a letter, underscore, dollar sign, or escape sequence. It canNOT start with a number.
It's not an identifier, though, it's a property.
This is probably the reason why var y = { 8388607: 1 }; fails in some browsers (IE5, for example), but var y = { "8388607": 1 }; (with the quotes) succeeds.
I'll take your word for it that it fails in IE5, but it works everywhere we've tested (including IE7, which is the only IE this ever has a chance of working with anyway).
You can't click "Preferences" at the top of this page (or any other page on the site)? Then click "Comments" to edit the comments prefs? Then click "Discussion Style" -> "Normal"? Then "Save"?
I see no reason why any of that should crash IE. I disbelieve (especially since I just tried it and it worked in IE 7.0.5346.5 Beta 2). If this really did make IE "blow up," then please tell me the version/build of IE you are using, and the exact steps you took, and I'll try it out.
Does it still show the same behavior if use a string representation of a number
...
Yes. I could not force a string representation (except by changing the value, such as starting it with a letter). I suspect this too is a bug, but I am not sure.
Although, that was Opera 8, and I cannot recall if I tried that part with Opera 9. So, testing again
Yep. Still undef for '8388608' and '16777215'.
Go to your comments prefs. Turn it off. :)
Sure. Here.
First, see this thread for more information about the Opera bug.
Second, performance problems still exist, but it is much faster now than it was "awhile ago" (for certain values of that).
Thanks for the better docs.
... but at this rate, we'll hit it in ... oh, a thousand years or so. :-)
And yes, I see no evidence anything up to and including (2^31)-1 should be a problem, but we are a loooooong way off from that (almost literally, 128 times the number of comments we've had to this point in Slashdot's history, as we fast approach 2^24). We'll deal with it later if we have to
We've never seen that. If you can show it to us happening, give us all the details (URL, browser, platform, etc.), then file a bug report, and we can look into it.
So I have to ask ... why is slashdot rolling its own Ajax library?
:-)
It is?
Why not use Dojo or Mochikit or hell, even Prototype?
We're using prototype. View source.
Yeah, was added last night.
Do you have actual numbers to show that the "potentially thousands of concatenations" would actually add measurable overhead?
Nope. But I know it could, and I am not going to take the risk until it can be measured to NOT add overhead. This is a very difficult thing to know for sure: there are many browsers on many platforms.
We actually are going to do another round of profiling, and we may throw this into the batch of things we look at, if we have the time. Even then, though, it will be difficult to reasonably profile all common browsers on all platforms to know for sure that it won't have a deleterious effect, just to work around the bug in Opera. So, bottom line, IF we have the time to profile it, IF it turns out to not be a problem, IF we think that our profiling is representative, IF Opera has still not fixed the bug, we'll consider making the change.
But it's a whole lot easier and better for everyone if the bug gets fixed, don't you think?
Here's something else to think on: D2 is only in beta. IE7 -- there's no way we'll ever get this to work on IE6 -- is also not yet released.
:)
You want our release schedule to be ahead of Microsoft's?
Silly.