It would presumably take the form of appointing an Attorney General who
would petiti0on to dismiss the case. Microsoft would certainly agree:),
but the other states wouldn't. This would simply remove the feds as a
party, but not otherwise affect the case.
Bush has indicated support for such a dismissal in the past, but I
expect his position would change when fully briefed. The general
free-market position is to avoid anti-trust actions, as they're
usually not justified by the economics. I agree with this, and fit
in firmly with Bork and Posner on it. However, after analysis, this
is one of the other cases.
Another possibility would be that the DOJ could push for a milder
remedy. This is possible; I'm certain there will be a remedy,
but I won't bank on the form. I believe that a split would actually
be less intrusive than any of the other possibilities, all of
which would require heavy-handed government involvement in running
the company. Were I a microsoft shareholder, I would much prefer
a split than such intrusion.
Guess what: the U.S. DOJ is *not* the only victorious plaintiff in the suit. Even if it were to dismiss, you still have a gaggle of states who would press on for remedy. He'll also probably get better economic and anti-trust advice, and see that an unfettered microsoft is *not* pro-competitive, but that's a side issue.
The question is not whether or not the suit will be dropped, but what the remedy is. While I'm at it, the folks opposing the breakup because the free-market will take care of it are right about the market--but the market will take a few (5-10?) years, during which consumers continue to lose billions of dollars to the illegal practices.
Also, the notion that Linux is tricky to configure while windcows is not seems to come from people who actually haven't configured both of them (you may have tried, but that would make you a rare exception). Installing windows from scratch is *much* harder. And when the machine ships with an OS, the user doesn't have to configure it either way.
>any of the people posting to this story seem to be implying that MS
>is just plain evil and they will do anything in their power to close
>up open-source. That is plain and simply not true.
That's right. Microsoft is a particularly complicated form of evil,
not plain evil:)
While I'm at it, it wasn't Willie Nelson, but Merle Haggard and
Willie Nelson that sang "Pancho and Lefty," on an album of the same
name (a good album, at that . ..)
Last night, in response to one of the "Do you really think that Bush
wouldn't do the same thing?" questions, I started to respond that,
"No; he wouldn't be what I thought I was voting for if he did"--and
then rememberd again that I didn't vote for him:)
And no, I didn't vote Nader--I'm an economics professor, for crying out
loud; I could never vote for Buchannan or Nader. Come to think of it,
I don't even know what color the sky is in Nader's world:)
Overall, though, the correllations that you suggest would be high, but
right now, to get the high proportion of voters that think Gore is a sore
loser, a portion of those are folks who voted for Gore. That is, there
are more people calling him a sore loser than voted for Bush.
I don't know the exact age of the machiens, but I first used one
in Jr. High in 1976.
And I have voted on butterfly ballots, though I couldn't tell you
offhand whether it was in califorina or Nevada. Anywone who was confused
by that shouldn't have been voting . . .
But they won't find out what "actually" did happen. Which standard do you use?
If you put in three teams of democrats, and three of republicans, you *will* have thre *different* counts showing Gore wond, and three *different* counts showing bush won.
We have accurate tallies for all of the counties. What we don't have is subjective tallies of what those who can't or won't fill out a ballot properly "intended."
I suppose now I'm speaking as a statistician, rather than a lawyer . . . [so many hats, so little time . ..;)]
There is no way to come up with a count from those "uncounted" ballots that shows the results were changed.
THose aren't uncounted ballots, but ballots without a valid vote. The "count" that would come from them would depend upon the biases of the "counters", as to how big of a prick on the ballot to count for their candidate, and how much larger it must be when counting for the other candidate.
The only objective standard is the one already used--any other standard leaves room for argument
THis is not legal advice, although I am an attorney. If you need legal advice, contact an attorney licensed in your own jurisdiction.
No, that's not what the Constitution requires. Here is what and
when is required:
1) Polls must close on the same day for representatives (and now
senators). Congress may set this date by law, or take the
Tuesday following the first Monday of November (which it has
set by law).
2) The Constitution requires that in some form or another, the state
legislature choose electors. All 50 states have chosen to have
elections.
3) A federal law requires that elections for these electors be held
on the same date as in 1). It does *not* require they be elected,
nore could it (this would violate the directive that the legislature
choose a method).
4) All electors must vote on the same date, which is December 18 this year.
Those are required. Addtionally, law provides that *if* electors.
have been chosen at least 6 days prior to the date in 4)
Whether or not a legislature having chosen elections on the date of
1 can change their mind after that date is not clear, but would seem
to lose the safe harbor of the statute. Also, given that the electors
*have* been chosen, it is not clear that the legislature can change its
mind at all (it's a close call).
I'm not sure, no, but my memory says that Bladerunner was the book full of missing manuscript bits--but _The Unteleported Man_ comes to mind, too. I'm dead certain that it was a heavily dystopian work, as suicided didn't seem surprising at all from one with scuch a dark view of the future . . .
1) The original "Do Androids Dream of Electric Sheep", which i believe has asterisks in a few places to show missing portions of the manuscript (due to the author's suicide--I might be thinking of _The Unteleported Man_, but I don't think so).
2) The movie "Bladerunner" as it first appeared in theaters, with the voiceover and happy ending
3) An edited, "Do Androids Dream of Electric Sheep."
4) The director's cut of _Bladerunner_, without the voiceover and with the dystopian ending.
Actually, I think the voiceover worked better in the film, but that's life:)
hawk, who's still trying to convince his wife that "They Live" and "Terminator" are dystopian science fiction, not "Rambo Movies" . . .
> So none of the guys' clothes will match in the eyes of women--great.
And this would be different than the current situation *how*?
I've resorted to the "scream test" to get dressed. I come downstairs, and if my wife doesn't scream in horror, I'm ok for the day.
I've given up. We can't agree on which pants are grey and green, and which are blue and black. And she still won't admit that the plaid shirt matched the tasteful and subdued plaid suit . . .
The day I really worried was the time I came out of the shower and found clothes laid out on hte bed. I had to check with a female friend to make sure I hadn't been committing even worse crimes against fashion than usual . . .
I've seen things that I've moderated then pop up with something different than I marked--sometimes moderating the right direction, othertimes the wrong direction.
I've been trying to open powerpoint files with staroffice for nearly three years now.
Usually what I try to open is the lecture slides that come with textbooks. It's certainly much better than it used to be, but it still loses or misplaces some of the information (but a least it no longer spits out a page-and-a-half length document that doesn't fit on the slide . ..)
Come to think of it, word documents tend to come out aline to long--the last line of one page forms ends up on the next page, and my last three installations have produced spreadsheets that can't generate charts . . .
Staroffice is a lousy product. The question isn't whether or not it's any good, but whether it's worse than ms office. [and I'm not willing to fight windows long enough to find out:) ]
I've brought both FreeBSD (3.2?) and Linux down from userland, and both are repeateable. I've never seen responses to bug reprots, so I don't know if it applies.
FreeBSD: go to image-link laden page (news collages, etc.) and middle-click on a slew of images under netscape 3. This causes many netscape windows and instances of xv to try to load. It overwhelmed the vm (but i didn't leave it until morning--it still answered pings, but eventually stopped swapping)>
Linux: debian kerenl 2.4.pre5. a) load 1.6G file into beav with only 160M of memory. Same vm oveload
b) I still don't know exactly what happened, as my fingers slipped, but I believe I selected two columns, and pasted them overlapping one of the two, causing massive memory allocation with recursion. *wham*.
All three of these were done from userland, not as root. The first two are repeatable, and I expect I could repeat the third if i knew what happened:)
That server holds pretty much *all* of the download records, and has for several years.
It may get slow, but it can saturate its link, and a mere slashdotting can't bring it down . . .
However, you will generally get *much* faster results from a regional server.
hawk
Re:"Bigger edges for Bush..."
on
eLection '04
·
· Score: 2
Yes. Expect gore to pick up a handful of votes in areas he one, and bush to pick up votes in areas he won.
Hand counting only in districts that one candidate one *will* skew the results, and is inherently inaccurate.
I hope this lets me post as me. However, I got an AC page. I'm getting that a lot recently . . .
hawk
Recounts tend to provide an additional margin for the candidate who one in that district/county/whatever--which is to be expected, if the errors are randomly distributed: if Sam has more votes than Paul, more were probably miscounted for Sam than for Paul.
I flatly don't believe the skew in Florida. Two counties show *way* to much gain for Bush, and four show even more than that for Gore.
And if this county is to be recounted by hand, what about the strongly republican counties with even bigger edges for Bush?
I have a longer piece on the statistical unbelievablity of the first recount at
I've put together a histogram of the proportion of votes that changed by county. In one county, Gore's alleged gain (past bush's) is nearly 1% of the total votes cast.
This county is 50 standard deviations out--far less likely than winning the lottery (the big tens of million ones) every week for three months straight . . .
hawk
it's at http://personal.ds.psu.edu/reh18
The big hole that remains
on
eLection '04
·
· Score: 2
is that there is no physical record to guarantee the machines are honest. I want something that we can go back and check; there's too many ways to hide things in a machine. In particualr, it should be possible to verify by eyeball before putting the card in--e.g., on the butterfly machines, if Bush is hole 3, I can make sure that chaff 3 is missing.
But physical ballots don't fully avoid the fraud: I've just put out an oped piece showing that there are two counties in which bush gained, and four in which gore gained, which are not even close to believable. It's at
http://www.personal.psu.edu/reh18
I've included the histogram as both jpeg and ps there. In a nutshell, almost everything should be within a couple of standard deviations, but bush has two counties at about 16 out, and gore has four that range from 20 to 50 . . .
hawk, wearing his statistician hat
Nevada "None of the Above"
on
eLection '04
·
· Score: 2
Yes, by law, we have "None of the Above" on the ballot for all statewide offices.
Years ago, I looked it up to see what happens. If memory serves, should none of the above win, a new election is held with all current candidates disqualified.
I don't think that could happen in a presidential election, however--the polls *must* close on the Tuesday following the first Monday in November. I assume that if none won in the presidential election, the the governor would order the legislature (the old one, not the newly elected one which wouldn't take office for several months) into a special session and that the legislature would appoint electors--very likely a delegation led by the governor.
I was quite relieved in '96 to find that choice; it's much more clear
than a general protest vote for one of the third parties.
It's more than just personal responsibility. It's an (indirect) tax on the housewifes.
Before you fly off the handle, think about it. Everyone receives a lower wage to account for the extra costs involved. The benefits, however, accrue to the dual income and single parent family. Then net result is a transfer from the housewife family to the other families.
Realize that your benefits are not without cost. Paychecks are smaller by exactly the cost of providing them (and the same applies to the so-caled "employer" section of your social security taxe: your wages are roughly 7.6% lower to accomodate this). The same "housewife tax" problem applies to tax deductions for daycare that do not also accrue to housewifes.
This is not to say that it is less efficient for employer's to provide daycare--it may well be less expensive to have the daycare there on the premises--but that this isn't the freebie it sounds like, and there's a definite subsidy from the lower income family (1 paycheck) to the higher income family (2 paychecks).
hawk, speaking as an economics professor
hmm, this could make a good test question:)
[now we see how many of my students read slashdot . . .:) ]
Slashdot Mirror
Thanks. It's always nice to be appreciated.
:),
It would presumably take the form of appointing an Attorney General who
would petiti0on to dismiss the case. Microsoft would certainly agree
but the other states wouldn't. This would simply remove the feds as a
party, but not otherwise affect the case.
Bush has indicated support for such a dismissal in the past, but I
expect his position would change when fully briefed. The general
free-market position is to avoid anti-trust actions, as they're
usually not justified by the economics. I agree with this, and fit
in firmly with Bork and Posner on it. However, after analysis, this
is one of the other cases.
Another possibility would be that the DOJ could push for a milder
remedy. This is possible; I'm certain there will be a remedy,
but I won't bank on the form. I believe that a split would actually
be less intrusive than any of the other possibilities, all of
which would require heavy-handed government involvement in running
the company. Were I a microsoft shareholder, I would much prefer
a split than such intrusion.
hawk, esq., etc.
Guess what: the U.S. DOJ is *not* the only victorious plaintiff in the suit. Even if it were to dismiss, you still have a gaggle of states who would press on for remedy. He'll also probably get better economic and anti-trust advice, and see that an unfettered microsoft is *not* pro-competitive, but that's a side issue.
The question is not whether or not the suit will be dropped, but what the remedy is. While I'm at it, the folks opposing the breakup because the free-market will take care of it are right about the market--but the market will take a few (5-10?) years, during which consumers continue to lose billions of dollars to the illegal practices.
Also, the notion that Linux is tricky to configure while windcows is not seems to come from people who actually haven't configured both of them (you may have tried, but that would make you a rare exception). Installing windows from scratch is *much* harder. And when the machine ships with an OS, the user doesn't have to configure it either way.
hawk, antitrust lawyer and economics professor
>any of the people posting to this story seem to be implying that MS
:)
.)
>is just plain evil and they will do anything in their power to close
>up open-source. That is plain and simply not true.
That's right. Microsoft is a particularly complicated form of evil,
not plain evil
While I'm at it, it wasn't Willie Nelson, but Merle Haggard and
Willie Nelson that sang "Pancho and Lefty," on an album of the same
name (a good album, at that . .
hawk
Last night, in response to one of the "Do you really think that Bush
wouldn't do the same thing?" questions, I started to respond that,
"No; he wouldn't be what I thought I was voting for if he did"--and
then rememberd again that I didn't vote for him
And no, I didn't vote Nader--I'm an economics professor, for crying out
loud; I could never vote for Buchannan or Nader. Come to think of it,
I don't even know what color the sky is in Nader's world
Overall, though, the correllations that you suggest would be high, but
right now, to get the high proportion of voters that think Gore is a sore
loser, a portion of those are folks who voted for Gore. That is, there
are more people calling him a sore loser than voted for Bush.
I don't know the exact age of the machiens, but I first used one
in Jr. High in 1976.
And I have voted on butterfly ballots, though I couldn't tell you
offhand whether it was in califorina or Nevada. Anywone who was confused
by that shouldn't have been voting . . .
hawk
But they won't find out what "actually" did happen. Which standard do you use?
If you put in three teams of democrats, and three of republicans, you *will* have thre *different* counts showing Gore wond, and three *different* counts showing bush won.
We have accurate tallies for all of the counties. What we don't have is subjective tallies of what those who can't or won't fill out a ballot properly "intended."
I suppose now I'm speaking as a statistician, rather than a lawyer . . . [so many hats, so little time . . .;)]
There is no way to come up with a count from those "uncounted" ballots that shows the results were changed.
THose aren't uncounted ballots, but ballots without a valid vote. The "count" that would come from them would depend upon the biases of the "counters", as to how big of a prick on the ballot to count for their candidate, and how much larger it must be when counting for the other candidate.
The only objective standard is the one already used--any other standard leaves room for argument
THis is not legal advice, although I am an attorney. If you need legal advice, contact an attorney licensed in your own jurisdiction.
No, that's not what the Constitution requires. Here is what and
when is required:
1) Polls must close on the same day for representatives (and now
senators). Congress may set this date by law, or take the
Tuesday following the first Monday of November (which it has
set by law).
2) The Constitution requires that in some form or another, the state
legislature choose electors. All 50 states have chosen to have
elections.
3) A federal law requires that elections for these electors be held
on the same date as in 1). It does *not* require they be elected,
nore could it (this would violate the directive that the legislature
choose a method).
4) All electors must vote on the same date, which is December 18 this year.
Those are required. Addtionally, law provides that *if* electors.
have been chosen at least 6 days prior to the date in 4)
Whether or not a legislature having chosen elections on the date of
1 can change their mind after that date is not clear, but would seem
to lose the safe harbor of the statute. Also, given that the electors
*have* been chosen, it is not clear that the legislature can change its
mind at all (it's a close call).
hawk, esq.
I'm not sure, no, but my memory says that Bladerunner was the book full of missing manuscript bits--but _The Unteleported Man_ comes to mind, too. I'm dead certain that it was a heavily dystopian work, as suicided didn't seem surprising at all from one with scuch a dark view of the future . . .
There's at least four choices that I'm aware of:
:)
1) The original "Do Androids Dream of Electric Sheep", which i believe has asterisks in a few places to show missing portions of the manuscript (due to the author's suicide--I might be thinking of _The Unteleported Man_, but I don't think so).
2) The movie "Bladerunner" as it first appeared in theaters, with the voiceover and happy ending
3) An edited, "Do Androids Dream of Electric Sheep."
4) The director's cut of _Bladerunner_, without the voiceover and with the dystopian ending.
Actually, I think the voiceover worked better in the film, but that's life
hawk, who's still trying to convince his wife that "They Live" and "Terminator" are dystopian science fiction, not "Rambo Movies" . . .
If you're going to install one of these, shouldn't it actually do an installation rather than silly slogans?
> So none of the guys' clothes will match in the eyes of women--great.
And this would be different than the current situation *how*?
I've resorted to the "scream test" to get dressed. I come downstairs, and if my wife doesn't scream in horror, I'm ok for the day.
I've given up. We can't agree on which pants are grey and green, and which are blue and black. And she still won't admit that the plaid shirt matched the tasteful and subdued plaid suit . . .
The day I really worried was the time I came out of the shower and found clothes laid out on hte bed. I had to check with a female friend to make sure I hadn't been committing even worse crimes against fashion than usual . . .
hawk
I've seen things that I've moderated then pop up with something different than I marked--sometimes moderating the right direction, othertimes the wrong direction.
hawk
Yeh, but it's abusive to run ed when you don't have a slow teletype or other printing terminal . . .
Besides, ed is for those that can't handle toggle switches and patch cords.
Damn newbies, demanding keyboards . . .
:)
hawk
I've been trying to open powerpoint files with staroffice for nearly three years now.
.)
:) ]
Usually what I try to open is the lecture slides that come with textbooks. It's certainly much better than it used to be, but it still loses or misplaces some of the information (but a least it no longer spits out a page-and-a-half length document that doesn't fit on the slide . .
Come to think of it, word documents tend to come out aline to long--the last line of one page forms ends up on the next page, and my last three installations have produced spreadsheets that can't generate charts . . .
Staroffice is a lousy product. The question isn't whether or not it's any good, but whether it's worse than ms office. [and I'm not willing to fight windows long enough to find out
hawk
I've brought both FreeBSD (3.2?) and Linux down from userland, and both are repeateable. I've never seen responses to bug reprots, so I don't know if it applies.
FreeBSD: go to image-link laden page (news collages, etc.) and middle-click on a slew of images under netscape 3. This causes many netscape windows and instances of xv to try to load. It overwhelmed the vm (but i didn't leave it until morning--it still answered pings, but eventually stopped swapping)>
Linux: debian kerenl 2.4.pre5. a) load 1.6G file into beav with only 160M of memory. Same vm oveload
b) I still don't know exactly what happened, as my fingers slipped, but I believe I selected two columns, and pasted them overlapping one of the two, causing massive memory allocation with recursion. *wham*.
All three of these were done from userland, not as root. The first two are repeatable, and I expect I could repeat the third if i knew what happened
hawk
That server holds pretty much *all* of the download records, and has for several years.
It may get slow, but it can saturate its link, and a mere slashdotting can't bring it down . . .
However, you will generally get *much* faster results from a regional server.
hawk
Yes. Expect gore to pick up a handful of votes in areas he one, and bush to pick up votes in areas he won. Hand counting only in districts that one candidate one *will* skew the results, and is inherently inaccurate. I hope this lets me post as me. However, I got an AC page. I'm getting that a lot recently . . . hawk
Recounts tend to provide an additional margin for the candidate who one in that district/county/whatever--which is to be expected, if the errors are randomly distributed: if Sam has more votes than Paul, more were probably miscounted for Sam than for Paul.
I flatly don't believe the skew in Florida. Two counties show *way* to much gain for Bush, and four show even more than that for Gore.
And if this county is to be recounted by hand, what about the strongly republican counties with even bigger edges for Bush?
I have a longer piece on the statistical unbelievablity of the first recount at
http://www.personal.psu.edu/reh18
hawk
I've put together a histogram of the proportion of votes that changed by county. In one county, Gore's alleged gain (past bush's) is nearly 1% of the total votes cast.
This county is 50 standard deviations out--far less likely than winning the lottery (the big tens of million ones) every week for three months straight . . .
hawk
it's at http://personal.ds.psu.edu/reh18
is that there is no physical record to guarantee the machines are honest. I want something that we can go back and check; there's too many ways to hide things in a machine. In particualr, it should be possible to verify by eyeball before putting the card in--e.g., on the butterfly machines, if Bush is hole 3, I can make sure that chaff 3 is missing.
But physical ballots don't fully avoid the fraud: I've just put out an oped piece showing that there are two counties in which bush gained, and four in which gore gained, which are not even close to believable. It's at
http://www.personal.psu.edu/reh18
I've included the histogram as both jpeg and ps there. In a nutshell, almost everything should be within a couple of standard deviations, but bush has two counties at about 16 out, and gore has four that range from 20 to 50 . . .
hawk, wearing his statistician hat
Yes, by law, we have "None of the Above" on the ballot for all statewide offices.
Years ago, I looked it up to see what happens. If memory serves, should none of the above win, a new election is held with all current candidates disqualified.
I don't think that could happen in a presidential election, however--the polls *must* close on the Tuesday following the first Monday in November. I assume that if none won in the presidential election, the the governor would order the legislature (the old one, not the newly elected one which wouldn't take office for several months) into a special session and that the legislature would appoint electors--very likely a delegation led by the governor.
I was quite relieved in '96 to find that choice; it's much more clear
than a general protest vote for one of the third parties.
hawk, a Nevada lawyer among his many hats . . .
Then why use 4.7 rather than 3.0???
:)
hawk, who generally uses lynx, anyway
As I type (9:45 easter), cnn's realaudio feed is down . . .
It's more than just personal responsibility. It's an (indirect) tax on the housewifes.
:)
:) ]
Before you fly off the handle, think about it. Everyone receives a lower wage to account for the extra costs involved. The benefits, however, accrue to the dual income and single parent family. Then net result is a transfer from the housewife family to the other families.
Realize that your benefits are not without cost. Paychecks are smaller by exactly the cost of providing them (and the same applies to the so-caled "employer" section of your social security taxe: your wages are roughly 7.6% lower to accomodate this). The same "housewife tax" problem applies to tax deductions for daycare that do not also accrue to housewifes.
This is not to say that it is less efficient for employer's to provide daycare--it may well be less expensive to have the daycare there on the premises--but that this isn't the freebie it sounds like, and there's a definite subsidy from the lower income family (1 paycheck) to the higher income family (2 paychecks).
hawk, speaking as an economics professor
hmm, this could make a good test question
[now we see how many of my students read slashdot . . .
>License conflict occur when you mix two different, incompatible
>licenses in the same program.
Yes, but keep in mind what "incompatible" means when speaking of the GPL:
"X is incompatible with the GPL" means that "the license of X cannot be replaced with the GPL"
It's not so much that licenses are incompatible with the GPL, but that the GPL is incompatible with just about everything except itself.
hawk