raw notes on war crimes
![spam]@world.std.com Blair P. Houghton's profile photo](https://lh3.googleusercontent.com/a/default-user=s40-c)
spam]@world.std.com Blair P. Houghton
exciting play-by-play of the test pattern.
hey! it's her! from, uh...
why is this a white house matter?
commercial break: oboy. american beauty clones. it wasn't all that great either. shades of ordinary people.
i hate it when the homily lacks panache.
i love it when jed bartlet spreads panache on the homily.
then cj clouds up and rains all over it.
cliffy! that table's kind of big, but i bet they're playing footsie in their minds.
does she keep a diary?
commercial break: i'm not buying any of this crap.
i like this guy better than danny concannon. he's less of a yutz.
the temp doesn't rate a pull on the focus ring.
hey! it's major dad! and he's been promoted! and lost his name tag.
donna shoots. donna scores. donna shakes the rain off her umbrella and folds it away. there was no diary cliff.
commercial break: frasier stopped being interesting about 110 episodes ago...
i use my pennies. some newspaper machines take pennies. they're handy to keep you from getting more pennies in change.
commercial break: mamet's bad enough, but some critic said "more twists than a bag of pretzels"...oy. ving rhames is a pretty big guy. howie long is freaking *huge*. hey! it's monument valley! later tonight, gary busey *is* robert blake.
ominous music and toby's coming. four dead in the west wing mess.
is anyone in this group wearing a red shirt?
wa = week-ahead
we don't need lincoln on the penny. we could put lincoln on the receipt each of us gets when we grow up and sell out.
poor leo. a thing like that can turn your whole world-view around.
sorkin doesn't understand the 2d amendment.
"it's not easy being my vice president"...dick cheney hears that in his head every time he looks at his boss.
hoynes gets his seat at the table. bartlet needs him. he's going to get his way more and more.
this guy's a republican and he's got proof he slept with a girl. he's not going to take a threat of revelation as a threat.
--Blair
"That diary's posted by now."
spam]@world.std.com Blair P. Houghton
DaveMoore
P. Houghton) wrote:
>(Reposted for speakers of languages that include carriage returns.)
>
>exciting play-by-play of the test pattern.
>
>hey! it's her! from, uh...
The Volkswagon Sychronicity ad, the one where the couple
drives down a rainy street, and everything happens in time
to the music they have in the tape player.
I used to play that ad over and over again, just to hear
the music.
>
>sorkin doesn't understand the 2d amendment.
No kidding. He tried, he really did, to treat the other side
with respect and courtesy, but he simply doesn't understand
why it's there.
--
Dave Moore == djm...@uh.edu == I speak for me.
In the wrong hands, sanity is a dangerous weapon.
Scott Stevenson
P. Houghton) wrote:
>cliffy! that table's kind of big, but i bet they're playing footsie in their minds.
Every time I hear the name "Cliff", I think of the IHOP guy. IMHO,
this Steve Young doppelganger needs to be in an accident.
>
>does she keep a diary?
>
>commercial break: i'm not buying any of this crap.
>
>i like this guy better than danny concannon. he's less of a yutz.
>
>the temp doesn't rate a pull on the focus ring.
>
>hey! it's major dad! and he's been promoted! and lost his name tag.
>
>donna shoots. donna scores. donna shakes the rain off her umbrella and folds it away. there was no diary cliff.
If Cliff takes his hat off, his head will still be in it.
>
>commercial break: frasier stopped being interesting about 110 episodes ago...
>
>i use my pennies. some newspaper machines take pennies. they're handy to keep you from getting more pennies in change.
:-) Best reason to dump cents--they cost more than $.01 each to
make and distribute.
If you're scoring at home, 14 billion cents weigh more than 77
million pounds. Lay them edge to edge and they'd stretch 2/3 of the
way to the moon. We make that many every year.
>
>commercial break: mamet's bad enough, but some critic said "more twists than a bag of pretzels"...oy.
That person is next year's winner in the Bulwer-Lytton contest...
>
>ominous music and toby's coming. four dead in the west wing mess.
"Next week, on a very special 'Blossom'..."
>
>is anyone in this group wearing a red shirt?
Abbey: "Dammit, Jed, I'm a surgeon, not a miracle worker"
>
>wa = week-ahead
>
>we don't need lincoln on the penny. we could put lincoln on the receipt each of us gets when we grow up and sell out.
>
>poor leo. a thing like that can turn your whole world-view around.
Best line of the night--"All war is a crime".
>
>sorkin doesn't understand the 2d amendment.
>
>"it's not easy being my vice president"...dick cheney hears that in his head every time he looks at his boss.
>
>hoynes gets his seat at the table. bartlet needs him. he's going to get his way more and more.
>
>this guy's a republican and he's got proof he slept with a girl. he's not going to take a threat of revelation as a threat.
We're not all virgins, you know--although mine may be growing back.
spam]@world.std.com Blair P. Houghton
>On Thu, 8 Nov 2001 06:16:13 GMT, blair[no spam]@world.std.com (Blair
>P. Houghton) wrote:
>>exciting play-by-play of the test pattern.
It occurs to me that there was play-by-play before all the
pool bets were in. You're not s'posed to do that.
>>hey! it's her! from, uh...
>
>The Volkswagon Sychronicity ad, the one where the couple
>drives down a rainy street, and everything happens in time
>to the music they have in the tape player.
No, I've seen this one in something else. There's a
similar woman in a Woody Allen movie, but I don't think
it's her. And she's got a sort of Mercedes Ruel thing
going on, but Mercedes has her own league and this lady
wasn't in it. But I've seen her in major speaking parts
before and it's gonna bug me...
>>sorkin doesn't understand the 2d amendment.
>
>No kidding. He tried, he really did, to treat the other side
>with respect and courtesy, but he simply doesn't understand
>why it's there.
No. I mean when Bartlet talks about the "civilian Militia".
The 2d Amendment as consistently interpreted by the Supreme
Court doesn't in any way guarantee things like concealed
carry. Morton Grove, IL, still has a perfectly legal
handgun ban in place; the Supreme Court didn't even bother
to hear the inevitable challenge to it. When Bartlet
dismissed the Amendment as "silly" he deflated his position
by throwing up his hands. The Amendment isn't nearly as
silly as is the NRA's abusive misreading of it. Bartlet
had plenty of room to decry the injustice of it all. It
indicated that either Sorkin or maybe just Bartlet doesn't
understand that nothing we saw in that episode was
protected by that Amendment.
In so doing, he gave the "other side" way more respect than
it could ever deserve.
--Blair
"And no, our Nation isn't a
Democracy, either..."
Phred Jackson
>>
>> The Volkswagon Sychronicity ad, the one where the couple
>> drives down a rainy street, and everything happens in time
>> to the music they have in the tape player.
You get different commercials than we do, I was watching an LA feed, BUT, I
was watching 24 and couldn't figure where I had seen Nina [Sarah
Clarke/Lively]. The only thing I could find is that she was on Ed but when
you mentioned the VW commercial, I recall that she is the One... I should
post this on the 24 forum, one of the posters went to school w/ her.
€Phred
David Moore
(that would be me) wrote:
>On Thu, 8 Nov 2001 06:16:13 GMT, blair[no spam]@world.std.com (Blair
>P. Houghton) wrote:
>
>>(Reposted for speakers of languages that include carriage returns.)
>>
>>exciting play-by-play of the test pattern.
>>
>>hey! it's her! from, uh...
>The Volkswagon Sychronicity ad, the one where the couple
>drives down a rainy street, and everything happens in time
>to the music they have in the tape player.
Er, no. I saw this actress somewhere else, had trouble
figuring out who she was, and somehow connected her
to your moment of almost-recognition.
But I just went through my tape, twice, and she's not
there. So who were you talking about, and where did I
see...her? I caught glimpses of the following L&O, was
it there?
In any event, complete and total brain fart. My apologies.
spam]@world.std.com Blair P. Houghton
>On Thu, 08 Nov 2001 00:45:51 -0600, DaveMoore <djm...@uh.edu>
>(that would be me) wrote:
>>On Thu, 8 Nov 2001 06:16:13 GMT, blair[no spam]@world.std.com (Blair
>>P. Houghton) wrote:
>>>
>>>hey! it's her! from, uh...
>
>there. So who were you talking about, and where did I
>see...her? I caught glimpses of the following L&O, was
>it there?
The one I mean is in the beginning, the reporter with the
sultry voice who asks CJ the first question.
--Blair
"IMDB needs a 'hey! it's her! from, uh...'
option on the search box..."
Semmens
<blair[no spam]@world.std.com (Blair P. Houghton)> wrote in message
news:GMH49...@world.std.com...
> David Moore <djm...@uh.edu> wrote:
> >On Thu, 08 Nov 2001 00:45:51 -0600, DaveMoore <djm...@uh.edu>
> >(that would be me) wrote:
> >>On Thu, 8 Nov 2001 06:16:13 GMT, blair[no spam]@world.std.com (Blair
> >>P. Houghton) wrote:
> >>>
> >>>hey! it's her! from, uh...
> >
> >But I just went through my tape, twice, and she's not
> >there. So who were you talking about, and where did I
> >see...her? I caught glimpses of the following L&O, was
> >it there?
>
> The one I mean is in the beginning, the reporter with the
> sultry voice who asks CJ the first question.
>
>
with Jeannie (and Jorja Fox's character, Maggie Doyle) over at Doc Magoo's
to discuss whether Jeannie has a case; Jeannie believes she's been fired
because she tested HIV positive.
Laura
someday, God willing, I will forget all this trivia
Stephen Fuld
>
> i use my pennies. some newspaper machines take pennies.
> they're handy to keep you from getting more pennies in
> change.
Having pennies costs us a fairly large amount of money. The mint spends far
more money creating pennies than any other coin. And as Sam said, most of
them end up not circulating. We should simply round cash transfers to the
nearest nickle (instead of the nearest penny as we do today - for example,
how do you pay the 6.5 cents tax on a dollar purchase?) and do away with the
penny. Furthermore, if we did that, it would make room in all the nations
cash drawers for the dollar coin, which could help its success. That, in
turn, would save us even more money as the cost of the coin, amortized over
its life, is more than the number of bills that have to be printed for the
equivalent life.
>
> poor leo. a thing like that can turn your whole world-view
> around.
>
Yup. I thought perhaps the General (?) should have apologized to Leo before
leaving. It was a powerfull moment.
> sorkin doesn't understand the 2d amendment.
Nor the concealed carry stuff. He had Hoynes wimp out and not give the
reasonable pro argument. It is that it makes the bad guys fearful that
anyone could be carrying and thus more cautious all the time rather than
only when they see a gun. Much the same argument for unmarked police cars.
I am not sure I buy the argument, but it is a reaasonable one.
>
> "it's not easy being my vice president"...dick cheney hears
> that in his head every time he looks at his boss.
I think Cheney loves his position. Lots of influence/power granted him by a
boss that obviously trusts him, yet little of the onerous responsibilities,
press conferences, photo ops, etc. required of the President.
>
> hoynes gets his seat at the table. bartlet needs him.
> he's going to get his way more and more.
Maybe that means that we will see him more and more. I hope so.
>
> this guy's a republican and he's got proof he slept with a
> girl. he's not going to take a threat of revelation as a
> threat.
Cheap shot.
--
- Stephen Fuld
e-mail address disguised to prevent spam
Chris Crandall
: Nor the concealed carry stuff. He had Hoynes wimp out and not give the
: reasonable pro argument. It is that it makes the bad guys fearful that
: anyone could be carrying and thus more cautious all the time rather than
: only when they see a gun. Much the same argument for unmarked police cars.
: I am not sure I buy the argument, but it is a reaasonable one.
Yes, that's the argument. The research shows it doesn't work that way,
though.
Bruce Mills
Really? To which research do you refer? Care to provide some
citations?
Lott in "More Guns, Less Crime" seems to indicate otherwise.
Lynn
>
> i like this guy better than danny concannon. he's less of
> a yutz.
He's seems more passionate, and less "career" oriented. He does it for the
love of the sport.
> i use my pennies. some newspaper machines take pennies.
> they're handy to keep you from getting more pennies in
> change.
I found a tub in the back of the closet with about $7 worth of of the
worthless things in it. I put a handfull at a time in my wallet. $X.31?
Oh, would you like some pennies? 1..2..3... Got rid of 'em. Watch the
government stop minting them and the value skyrocket.
> hoynes gets his seat at the table. bartlet needs him.
> he's going to get his way more and more.
Let's hope it includes a bit more screen time, too.
--
Lynn
on new compooter with no speel chek
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Life if a search for the truth; and there is no truth
-Chinese Proverb
======================================================
*West Wing: http://www.lynnsland.com/Tww.html *
*WingNuts: http://www.lynnsland.com/WingNuts.html *
******************************************************
Chris Crandall
: Really? To which research do you refer? Care to provide some
: citations?
: Lott in "More Guns, Less Crime" seems to indicate otherwise.
Well, let's agree that Lott (and Lott & Mustard)'s work is
"controversial". You might mention that Lott is part of the "Law and
Economics" movement, a very ideological right-wing movement which seeks to
redefine most every legal issue in terms of economic value.
You might also mention that there is widespread criticism of Lott's work.
Here's an online taste:
http://www.asahi-net.or.jp/~zj5j-gttl/teret.htm
THere's plenty more out there. You'll find that most of the research
cited on web sites back's up the authors' point of view that existed prior
to doing the research--regardless of which side of the debate they take.
Priscilla H Ballou
: > hoynes gets his seat at the table. bartlet needs him.
: > he's going to get his way more and more.
: Let's hope it includes a bit more screen time, too.
Tim Matheson has aged well, hasn't he?
Priscilla
--
"I love cats because I enjoy my home; and little by little, they become
its visible soul." -- Jean Cocteau
DaveMoore
wrote:
>Bruce Mills (aki...@sprint.ca) wrote:
>: Really? To which research do you refer? Care to provide some
>: citations?
>
>: Lott in "More Guns, Less Crime" seems to indicate otherwise.
>
>
>Well, let's agree that Lott (and Lott & Mustard)'s work is
>"controversial". You might mention that Lott is part of the "Law and
>Economics" movement, a very ideological right-wing movement which seeks to
>redefine most every legal issue in terms of economic value.
Can we agree that there is no gun control research which is
not controversial? Kellerman, for instance, is far from
widely accepted.
I'll also note that in MGLC, Lott used the largest database
ever compiled -- every county in every state of the Union.
He has made his dataset available to anyone for the asking,
so folks who disagree with his methodology are free to try
their own. (I have not heard of anyone doing this and coming
up with radically different results, but I could easily have
missed a study; if you know of one please pass it on.)
I suspect that part of the problem here is that the data
collection techniques that enabled Lott to do such a wide-
ranging study are so new that we simply do not have enough
data over time to know for sure.
david
>
> Well, let's agree that Lott (and Lott & Mustard)'s work is
> "controversial". You might mention that Lott is part of the "Law
> and Economics" movement, a very ideological right-wing movement
> which seeks to redefine most every legal issue in terms of economic
> value.
We might also mention that Lott is, let's say "controversial", even
within the Law and Economics movement. He's certainly no Posner.
David Marc Nieporent
>: Really? To which research do you refer? Care to provide some
>: citations?
>: Lott in "More Guns, Less Crime" seems to indicate otherwise.
>Well, let's agree that Lott (and Lott & Mustard)'s work is
>"controversial".
Of course. That's not to say it's right or wrong. It just means that some
people disagree with it. (Actually, what the word typically means is that
the journalist telling the story disagrees with it, but can't come out and
say so explicitly because he needs to appear to be objective. In theory,
it's a value-neutral word; in practice, it means "extreme.")
> You might mention that Lott is part of the "Law and
>Economics" movement, a very ideological right-wing movement which seeks to
>redefine most every legal issue in terms of economic value.
I don't know how something can be "very ideological." That's like being
very dead. You are, or you're not. Of course, I suspect if you agreed
with this movement, you'd call it "principled" rather than "ideological";
the latter is just a pejorative for "a philosophy I don't like."
As for being "right-wing," given the connotations of that phrase, it's not
particularly so, no. It's on a different axis altogether.
---------------------------------------------
David M. Nieporent niep...@alumni.princeton.edu
spam]@world.std.com Blair P. Houghton
>On Thu, 8 Nov 2001 06:07:17 GMT, blair[no spam]@world.std.com (Blair
>P. Houghton) wrote:
>>we don't need lincoln on the penny. we could put lincoln on the receipt each of us gets when we grow up and sell out.
>
>already has Washington on the quarter.
Lincoln is on the five dollar bill, which is more likely to
survive than the one dollar bill or the penny, but
Jefferson is on the two, which gets less play than the
Sacajawea dollar, so let Jefferson stay on the nickel and
Lincoln on the fin.
>>sorkin doesn't understand the 2d amendment.
>
>Try to explain it yourself in a way that doesn't generate an argument
>one way or the other.
I did. About 5 years ago. Then I decided the reason I was
still getting arguments was I didn't understand it fully,
so I went digging and found out all about the Supreme Court's
history on the Amendment, and the cases, and what it all
actually means.
And that's when I got really pissed off at the "conventional
wisdom" and how thoroughly wrong it is and how that is
husbanded by the NRA. They don't want the truth to come
out because they enjoy the political climate that their
brand of the facts profits them.
Now when I post the truth I still get arguments. But I
know the reason this time is people don't want to hear
shattering truths. So I don't argue any more, because the
truth sticks, and eventually they'll be in the middle of
the argument and they'll have to choose between the
truth and political inertia. Most people not posing for
the cameras (or the net) will choose truth.
--Blair
"Individual, but not absolute."
Bruce Mills
>
> On 8 Nov 2001 22:25:29 GMT, cran...@lark.cc.ku.edu (Chris Crandall)
> wrote:
>
> >Bruce Mills (aki...@sprint.ca) wrote:
> >: Really? To which research do you refer? Care to provide some
> >: citations?
> >
> >: Lott in "More Guns, Less Crime" seems to indicate otherwise.
> >
> >
> >Well, let's agree that Lott (and Lott & Mustard)'s work is
> >"controversial". You might mention that Lott is part of the "Law and
> >Economics" movement, a very ideological right-wing movement which seeks to
> >redefine most every legal issue in terms of economic value.
> Can we agree that there is no gun control research which is
> not controversial? Kellerman, for instance, is far from
> widely accepted.
Kellerman has been widely repudiated.
Jay R. Ashworth
> Having pennies costs us a fairly large amount of money. The mint spends far
> more money creating pennies than any other coin. And as Sam said, most of
> them end up not circulating. We should simply round cash transfers to the
> nearest nickle (instead of the nearest penny as we do today - for example,
> how do you pay the 6.5 cents tax on a dollar purchase?) and do away with the
> penny. Furthermore, if we did that, it would make room in all the nations
> cash drawers for the dollar coin, which could help its success. That, in
> turn, would save us even more money as the cost of the coin, amortized over
> its life, is more than the number of bills that have to be printed for the
> equivalent life.
You know, that *sounds* sane, but it suggests that you aren't privy to
the parallel conversations that surrounded allowing digits other than 0
and 1 in the center of an Area Code, and which still accompany the idea
of lengthening the size of an North American phone number.
The idea of rounding to the nearest nickle is a *REALLY BIG DEAL*.
-=*!REALLY!*=- big.
I can't *begin* to explain how trivial it's not.
All they have to do is sharply cut back on how many pennies they mint.
Cheers,
-- jra
--
Jay R. Ashworth j...@baylink.com
Member of the Technical Staff Baylink
The Suncoast Freenet The Things I Think
Tampa Bay, Florida http://baylink.pitas.com +1 727 804 5015
Fanfic: it's enough to make you loose your mind.
-- me
Jay R. Ashworth
Cite?
Jay R. Ashworth
> We're not all virgins, you know--although mine may be growing back.
It does.
Cheers,
-- jr 'cause, y'know, you're embarassed, you don't know what to do...' a
Jay R. Ashworth
> Of course. That's not to say it's right or wrong. It just means that some
> people disagree with it. (Actually, what the word typically means is that
> the journalist telling the story disagrees with it, but can't come out and
> say so explicitly because he needs to appear to be objective. In theory,
> it's a value-neutral word; in practice, it means "extreme.")
> I don't know how something can be "very ideological." That's like being
> very dead. You are, or you're not. Of course, I suspect if you agreed
> with this movement, you'd call it "principled" rather than "ideological";
> the latter is just a pejorative for "a philosophy I don't like."
*Ghod* I love this newsfroup.
Cheers,
-- jra
Lynn
> :
> : Let's hope it includes a bit more screen time, too.
"Priscilla H Ballou" <p...@world.std.com> wrote...
>
> Tim Matheson has aged well, hasn't he?
You can say that again.
Getteur
>I can't *begin* to explain how trivial it's not.
>
>All they have to do is sharply cut back on how many pennies they mint.
>
>Cheers,
>-- jra
And have the chaos that would result from the shortage of pennies.
Gary
Stephen Fuld
news:slrn9uo1h...@dorothy.msas.net...
> C.J. stepped down from the podium, whereupon Stephen Fuld said:
> > Having pennies costs us a fairly large amount of money. The mint
spends far
> > more money creating pennies than any other coin. And as Sam said, most
of
> > them end up not circulating. We should simply round cash transfers to
the
> > nearest nickle (instead of the nearest penny as we do today - for
example,
> > how do you pay the 6.5 cents tax on a dollar purchase?) and do away
with the
> > penny. Furthermore, if we did that, it would make room in all the
nations
> > cash drawers for the dollar coin, which could help its success. That,
in
> > turn, would save us even more money as the cost of the coin, amortized
over
> > its life, is more than the number of bills that have to be printed for
the
> > equivalent life.
>
> You know, that *sounds* sane, but it suggests that you aren't privy to
> the parallel conversations that surrounded allowing digits other than 0
> and 1 in the center of an Area Code, and which still accompany the idea
> of lengthening the size of an North American phone number.
Actually, I do understand those issues, and the similar problems that will
occur when we run out of nine digit Social Security numbers. However, I
think you misunderstood what I proposed. I suggested "nickle rounding" for
retail, currency transactions only.
Thus, when you get to the register, if the total ends in 1 or 2 cents, you
round down as it it ended in zero. It it ends in 3, 4, 6 or 7 you "pretend"
it ends in 5 and if it ends in 8 or 9, you round up to the next zero. Now,
at the end of the day, the register contents and the record will probably be
off a few cents (it will be correct on average, but some will be high, some
low), but most such systems have some tolerance built in. Many stores will
"let you get away with" being short a penny on a transaction. Their systems
must have that tolerance built in.
For things like checks, credit card payments, etc. no change would be made.
Over time, if it made sense to do so, the systems underlying these could be
changed, but that is not necessary.
Note that we already have an example of a system that uses rounding in that
Federal Income Tax data can be rounded to the nearest dollar. There are
error tolerances built into the Social Security earning tracking system (to
avoid spending hundreds of dollars collecting tens of dollars). There are
other examples.
BTW, "nickle rounding" can certainly be phased in to allow for system
changes.
>
> The idea of rounding to the nearest nickle is a *REALLY BIG DEAL*.
>
> -=*!REALLY!*=- big.
>
> I can't *begin* to explain how trivial it's not.
see above.
>
> All they have to do is sharply cut back on how many pennies they mint.
How does that help? If I still "need" a penny because a transaction
requires me to pay say $1.01, I need that penny or I will get four penny's
in change. You need to eliminate the requirement before you eliminate the
penny.
Mary Rosh
I had him for a PhD level empirical methods class when he taught at
the Wharton School at the University of Pennsylvania back in the early
1990s, well before he gained national attention, and I have to say
that he was the best professor that I ever had. You wouldn't know
that he was a "right-wing" ideologue from the class. He argued both
sides of different issues. He tore apart empirical work whether you
thought that it might be right-wing or left-wing. At least at Wharton
for graduate school or Stanford for undergraduate, Lott taught me more
about analysis than any other professor that I had and I was not
alone. There were a group of us students who would try to take any
class that he taught. Lott finally had to tell us that it was best
for us to try and take classes from other professors more to be
exposed to other ways of teaching graduate material.
Jay R. Ashworth
> > You know, that *sounds* sane, but it suggests that you aren't privy to
> > the parallel conversations that surrounded allowing digits other than 0
> > and 1 in the center of an Area Code, and which still accompany the idea
> > of lengthening the size of an North American phone number.
>
> Actually, I do understand those issues, and the similar problems that will
> occur when we run out of nine digit Social Security numbers. However, I
> think you misunderstood what I proposed. I suggested "nickle rounding" for
> retail, currency transactions only.
>
> Thus, when you get to the register, if the total ends in 1 or 2 cents, you
> round down as it it ended in zero. It it ends in 3, 4, 6 or 7 you "pretend"
> it ends in 5 and if it ends in 8 or 9, you round up to the next zero. Now,
> at the end of the day, the register contents and the record will probably be
> off a few cents (it will be correct on average, but some will be high, some
> low), but most such systems have some tolerance built in. Many stores will
> "let you get away with" being short a penny on a transaction. Their systems
> must have that tolerance built in.
No system *I've* ever worked with had such tolerance. I've been off
cash twice in my life; every place I worked let you get away with it
maybe a couple times, less than a buck.
Such rounding would put you off by 3 or 4 dollars *per drawer* in the
average day. Look (I'll try to find it, but can't guarantee it) for a
bit of research which proves that in a "random" collection of numbers
from a given domain, lower first digits predominate.
> For things like checks, credit card payments, etc. no change would be made.
> Over time, if it made sense to do so, the systems underlying these could be
> changed, but that is not necessary.
>
> Note that we already have an example of a system that uses rounding in that
> Federal Income Tax data can be rounded to the nearest dollar. There are
> error tolerances built into the Social Security earning tracking system (to
> avoid spending hundreds of dollars collecting tens of dollars). There are
> other examples.
None of which actually pertain to source transactions, but you've
already clarified.
> BTW, "nickle rounding" can certainly be phased in to allow for system
> changes.
Or 'nickel rounding', even.
> > The idea of rounding to the nearest nickle is a *REALLY BIG DEAL*.
> > -=*!REALLY!*=- big.
> > I can't *begin* to explain how trivial it's not.
>
> see above.
You've never managed cashiers?
> > All they have to do is sharply cut back on how many pennies they mint.
>
> How does that help? If I still "need" a penny because a transaction
> requires me to pay say $1.01, I need that penny or I will get four penny's
> in change. You need to eliminate the requirement before you eliminate the
> penny.
Naw; they'll come out of the woodwork. I think.
Jay R. Ashworth
> "Priscilla H Ballou" <p...@world.std.com> wrote...
> > Tim Matheson has aged well, hasn't he?
>
> You can say that again.
Tim Matheson has aged well, hasn't he?
Cheers,
spam]@world.std.com Blair P. Houghton
>C.J. stepped down from the podium, whereupon David Marc Nieporent said:
>> Of course. That's not to say it's right or wrong. It just means that some
>> people disagree with it. (Actually, what the word typically means is that
>> the journalist telling the story disagrees with it, but can't come out and
>> say so explicitly because he needs to appear to be objective. In theory,
>> it's a value-neutral word; in practice, it means "extreme.")
>[ ... ]
>> I don't know how something can be "very ideological." That's like being
>> very dead. You are, or you're not. Of course, I suspect if you agreed
>> with this movement, you'd call it "principled" rather than "ideological";
>> the latter is just a pejorative for "a philosophy I don't like."
>
>*Ghod* I love this newsfroup.
I hope you mean for the entertainment value.
http://dictionary.cambridge.org/define.asp?key=ideology*1+0
"Ideological" means of or pertaining to ideas. Being ideological
is having an attitude[*] and using it when forming opinions.
Being an ideologue, however, is like being a cleric on the
opinion.
Maybe that's what Dave's confusing.
Think of it as the difference between being political and
being a politician.
And yes, you can be a bit ideological or a lot ideological,
depending on how vociferous or intractable you are about
your attitudes. Calling a decision or a person "ideological"
doesn't necessarily confer an insult. You have to expose
your own ideology as contrasting his beforehand, and then
not call your own ideology an ideology. And then say it
with a sneer. And a wad of mud lofted over the dais.
* - the taxonomy of situational belief:
more [mo-ray]: a central belief applied to all situations
(thou shalt not kill)
attitude: a belief applied to a set of situations
(war is bad for children and other living things)
opinion: a specific decision in a given situation
(bomb the taliban)
Note that your mores, attitudes, and opinions don't have to
respect each other. If they do, always, you're probably
playing the character for political reasons, and guaranteed
to get a lot of "what if?" arguments.
--Blair
"See? High school sociology
wasn't a waste of a semester."
david
> C.J. stepped down from the podium, whereupon Stephen Fuld said:
>> Actually, I do understand those issues, and the similar problems that will
>> occur when we run out of nine digit Social Security numbers. However, I
>> think you misunderstood what I proposed. I suggested "nickle rounding" for
>> retail, currency transactions only.
>>
>> Thus, when you get to the register, if the total ends in 1 or 2 cents, you
>> round down as it it ended in zero. It it ends in 3, 4, 6 or 7 you "pretend"
>> it ends in 5 and if it ends in 8 or 9, you round up to the next zero. Now,
>> at the end of the day, the register contents and the record will probably be
>> off a few cents (it will be correct on average, but some will be high, some
>> low), but most such systems have some tolerance built in. Many stores will
>> "let you get away with" being short a penny on a transaction. Their systems
>> must have that tolerance built in.
>
> No system *I've* ever worked with had such tolerance. I've been off
> cash twice in my life; every place I worked let you get away with it
> maybe a couple times, less than a buck.
Actually, cash registers work like that now. With sales tax, most
transactions come to some percentage of a penny. We used to mint
things like half-pennies to handle this, but haven't done so in quite
some time. Nobody seems to care.
> Such rounding would put you off by 3 or 4 dollars *per drawer* in the
> average day.
This is a larger problem than most anti-penny folks realize. Ideally,
the register should handle the rounding itself, which means the
drawers would remain accurate. But there are a lot of registers out
there.
The main concept which keeps the penny alive is the idea that it
represents "one unit" of currency. But it doesn't. It really
represents 1/100 of a unit, and there's no inherent reason why our
smallest coin should be 1/100 rather than 1/200 (as it used to be)
or 1/20 (as proposed).
>> For things like checks, credit card payments, etc. no change would be made.
>> Over time, if it made sense to do so, the systems underlying these could be
>> changed, but that is not necessary.
>>
>> Note that we already have an example of a system that uses rounding in that
>> Federal Income Tax data can be rounded to the nearest dollar. There are
>> error tolerances built into the Social Security earning tracking system (to
>> avoid spending hundreds of dollars collecting tens of dollars). There are
>> other examples.
>
> None of which actually pertain to source transactions, but you've
> already clarified.
A better example might be that of the stock market. Stocks and bonds
are generally traded in units of 1/8 of a cent, but are rounded to the
nearest cent when converted into cash.
Stephen Fuld
Now that you mention this, I vaguely remember such a bias. I would
appreciate your effort in looking for it and posting a cite. Of course, it
may not apply in this situation as prices tend to end in fives, eights or
nines and presumably there is some reasonably stable distribution of number
of items per purchase, and the tax rate is fixed for any given area. That
would be an interesting research problem.
But bacK to the problem, remember, over time, the registers would be
reporgrammed, so the "off by a little" would be a temporary problem.
>
> > For things like checks, credit card payments, etc. no change would be
made.
> > Over time, if it made sense to do so, the systems underlying these
could be
> > changed, but that is not necessary.
> >
> > Note that we already have an example of a system that uses rounding in
that
> > Federal Income Tax data can be rounded to the nearest dollar. There
are
> > error tolerances built into the Social Security earning tracking system
(to
> > avoid spending hundreds of dollars collecting tens of dollars). There
are
> > other examples.
>
> None of which actually pertain to source transactions, but you've
> already clarified.
>
> > BTW, "nickle rounding" can certainly be phased in to allow for system
> > changes.
>
> Or 'nickel rounding', even.
Yeah, but nickels cost five cents wheras you can get nickles for less :-)
>
> > > The idea of rounding to the nearest nickle is a *REALLY BIG DEAL*.
> > > -=*!REALLY!*=- big.
> > > I can't *begin* to explain how trivial it's not.
> >
> > see above.
>
> You've never managed cashiers?
>
No. Tell me about what the issues are.
> > > All they have to do is sharply cut back on how many pennies they mint.
> >
> > How does that help? If I still "need" a penny because a transaction
> > requires me to pay say $1.01, I need that penny or I will get four
penny's
> > in change. You need to eliminate the requirement before you eliminate
the
> > penny.
>
> Naw; they'll come out of the woodwork. I think.
I think you're wrong but besides, even if they did, when the "woodwork"
supply ran out, what would happen then? But that is a testable hypothesisI
suspect it won't actually be tested though.
Getteur
I've dealt with numbers all my life (a career accountant) and never heard of
such a thing.
Gary
Getteur
>> average day.
>
I'd like for this person to explain, in mathematical terms, just how this
would--or could--happen.>A better example might be that of the stock market.
Stocks and bonds
>are generally traded in units of 1/8 of a cent, but are rounded to the
>nearest cent when converted into cash.
>
No lately. Apparently you haven't been keeping up with the market. All
setting is now done in dollars and *cents* down to the penny. Or should I say
hundreth of a dollar. ;-)
Gary
spam]@world.std.com Blair P. Houghton
>A better example might be that of the stock market. Stocks and bonds
>are generally traded in units of 1/8 of a cent, but are rounded to the
>nearest cent when converted into cash.
Stock markets decimalized earlier this year. Both the NYSE
and NASDAQ got rid of fractions. Some commodity markets
still use them. And even when dealing in fractions (of a
dollar, not of a cent) the numbers of units were huge, so
the penny rounding was on the order of a few
parts-per-million of an average transaction. And most
trading tools will allow you to go out two extra places past
the penny when making limit orders.
The biggest expense of dumping the penny would be reprogramming
every cash register in America to handle penniless computations.
Many of them just won't get it and will have to be replaced.
--Blair
david
> david <da...@example.com> wrote:
>>A better example might be that of the stock market. Stocks and bonds
>>are generally traded in units of 1/8 of a cent, but are rounded to the
>>nearest cent when converted into cash.
>
> Stock markets decimalized earlier this year. Both the NYSE
> and NASDAQ got rid of fractions.
I did know that, which makes me wonder why I didn't use past tense
(ironically, I work in investment banking and deal with these numbers
constantly).
> Some commodity markets
> still use them. And even when dealing in fractions (of a
> dollar, not of a cent) the numbers of units were huge, so
> the penny rounding was on the order of a few
> parts-per-million of an average transaction. And most
> trading tools will allow you to go out two extra places past
> the penny when making limit orders.
Good point. So the sales tax thing is probably a better example.
What I was trying to get across is that there's two obstacles to
eliminating the penny: one is the real costs of doing so (like the
cash register issue), and the other is the mental obstacle of losing
the unit value.
Jay R. Ashworth
> The biggest expense of dumping the penny would be reprogramming
> every cash register in America to handle penniless computations.
> Many of them just won't get it and will have to be replaced.
Precisely.
Stephen Fuld
news:GMMBE...@world.std.com...
Assuming the cashier can handle the rounding - I know, I know :-( with a
posted card that says what to do with each possible last digit, and the
afor-mentioned tolerances, you wouldn't have to replace (or even reporgram)
any. (Well, not absolutely true. There are probably some that would have to
be reporgrammed rather quickly. Such as the ones connected to automatic
change dispensers.) Of course, over time, they will get replaced anyway, so
the problem, will go away anyway.
And remember, this can be phased in. Stores can be encouraged to use the
rounding manually at first, then to reporgram their systems. Over the
course of several years, the need for pennies will decrease until the point
when we can stop minting them.
Scott Stevenson
It really shouldn't be a mental obstacle, since the cent isn't the
monetary unit of the US--the dollar is.
I'm not even sure that the cash register programming would be that
big a deal, since rounding would only be done on the final total of
cash transactions. The only place that I've worked that had a cash
register, the actual program could be updated at the central office
and then downloaded when needed (which was how sales were programmed
into the register--it called in the numbers at night, and any new
prices were downloaded. The same thing could be done with the rounding
function).
For places that don't have that sort of system (mom&pop stores, etc)
it could probably be done with a software/firmware upgrade. Given
enough lead time (say a year), I don't think it would be that big a
problem.
(Pleeeeze don't ask about why someone with 12 years of schooling
can't remember that totals ending with 1, 2, 6, or 7 should round
down, and 3, 4, 8, and 9 should round up).
take care,
Scott
who, while in a terribly pendantic mode, wishes to remind everyone
that the British use "pennies", and 'Murcans use "cents"
spam]@world.std.com Blair P. Houghton
> Scott
> who, while in a terribly pendantic mode, wishes to remind everyone
>that the British use "pennies", and 'Murcans use "cents"
Isn't pennies just short for pence.
--Blair
"Short."
Maggie
Scott
> who, while in a terribly pendantic mode, wishes to remind
everyone
> that the British use "pennies", and 'Murcans use "cents"
Er, I don't get this point. I have 3 American pennies. a nickel,
and a quarter in my pocket for a total of 33 cents.
I've been enjoying this penny candy thread, even if inflation
means there isn't any more penny candy.
Maggie
david
> On 11 Nov 2001 17:03:01 GMT, david <da...@example.com> wrote:
>
>>
>>Good point. So the sales tax thing is probably a better example.
>>What I was trying to get across is that there's two obstacles to
>>eliminating the penny: one is the real costs of doing so (like the
>>cash register issue), and the other is the mental obstacle of losing
>>the unit value.
>
> It really shouldn't be a mental obstacle, since the cent isn't the
> monetary unit of the US--the dollar is.
That point was made just two posts ago.
> I'm not even sure that the cash register programming would be that
> big a deal, since rounding would only be done on the final total of
> cash transactions. The only place that I've worked that had a cash
> register, the actual program could be updated at the central office
> and then downloaded when needed (which was how sales were programmed
> into the register--it called in the numbers at night, and any new
> prices were downloaded. The same thing could be done with the rounding
> function).
I seriously doubt the "actual program" was updated at the central
office. It's much more likely that the data (prices, receipts, etc.)
was transferred. Downloading prices is one thing, modern cash
registers are *designed* to do that. Adding a rounding function would
take re-programming.
> For places that don't have that sort of system (mom&pop stores, etc)
> it could probably be done with a software/firmware upgrade. Given
> enough lead time (say a year), I don't think it would be that big a
> problem.
I don't claim incredibly wide knowledge of all the cash registers out
there, but I doubt most of them are set up for easy firmware upgrades.
Although I'm curious how many electronic registers already have the
capability built in. Given the fact that this penny thing has been
kicking around for years, it seems like a natural (and cheap) feature
to have added years ago.
> (Pleeeeze don't ask about why someone with 12 years of schooling
> can't remember that totals ending with 1, 2, 6, or 7 should round
> down, and 3, 4, 8, and 9 should round up).
That's not the issue. The issue is that unless the register does it
automatically, the drawer counts will *always* be off. Also, having
the register display the real total will cut down on complaints.
Tim Lambert
> C.J. stepped down from the podium, whereupon Stephen Fuld said:
> > Thus, when you get to the register, if the total ends in 1 or 2 cents, you
> > round down as it it ended in zero. It it ends in 3, 4, 6 or 7 you "pretend"
> > it ends in 5 and if it ends in 8 or 9, you round up to the next zero. Now,
> > at the end of the day, the register contents and the record will probably be
> > off a few cents (it will be correct on average, but some will be high, some
> > low), but most such systems have some tolerance built in. Many stores will
> > "let you get away with" being short a penny on a transaction. Their systems
> > must have that tolerance built in.
>
> No system *I've* ever worked with had such tolerance. I've been off
> cash twice in my life; every place I worked let you get away with it
> maybe a couple times, less than a buck.
>
> Such rounding would put you off by 3 or 4 dollars *per drawer* in the
> average day. Look (I'll try to find it, but can't guarantee it) for a
> bit of research which proves that in a "random" collection of numbers
> from a given domain, lower first digits predominate.
Zipf's law, but that applies to *first* digits, not last digits.
Rounding up is just as likely as rounding down, so the accumulated
error follows the statistics of a random walk. If you have n
transactions, the amount you will be off by will be roughly the square
root of n. To be out by even as much as one dollar would require
100*100= 10,000 transactions. In an eight hour day, that would mean a
transaction every 3 seconds.
Here in Australia, we got rid of a one and two cent coins several
years ago, and nobody misses them.
Tim
Chris Crandall
: >Economics" movement, a very ideological right-wing movement which seeks to
: >redefine most every legal issue in terms of economic value.
David Marc Nieporent (niep...@alumni.princeton.edu) wrote:
: I don't know how something can be "very ideological." That's like being
: very dead. You are, or you're not.
No. Some people are very ideological in their decision making, e.g.,
Scalia & Thomas, and others are somewhat ideological, e.g., Kennedy &
O'Connor. There are degrees of "ideological".
: Of course, I suspect if you agreed
: with this movement, you'd call it "principled" rather than "ideological";
: the latter is just a pejorative for "a philosophy I don't like."
I disagree. I am often quite ideological. And correct when I do it, too.
And I like my thinking. Ideology is no perforative--it means a consistent
or coherent set of thoughts, values, and beliefs that one is committed to.
: As for being "right-wing," given the connotations of that phrase, it's not
: particularly so, no. It's on a different axis altogether.
Oh, well, you might convince one or two folk with that, but law and
economics movement has a home among the right-wing, and no place at all
among the left-wing.
Chris Crandall
: BTW, "nickle rounding" can certainly be phased in to allow for system
: changes.
Works in Australia. I didn't even notice it.
spam]@world.std.com Blair P. Houghton
>I don't claim incredibly wide knowledge of all the cash registers out
>there, but I doubt most of them are set up for easy firmware upgrades.
Think "Y2K Bug" for the scope and pissantry of the
programming involved. It's not rocket science, but the
sheer numbers and variety of machines, some of which need
major code changes to make them round in other than pennies,
will make it a big deal.
The difference between this and Y2K is that this would
actually be guaranteed to affect your pocket if it weren't
fixed. Y2K was a magnified shadow of a pocket-sized
straw-man for the pols to knock down.
--Blair
"Ooh! Hit it again!"
spam]@world.std.com Blair P. Houghton
>"Jay R. Ashworth" <j...@dorothy.msas.net> writes:
>> Such rounding would put you off by 3 or 4 dollars *per drawer* in the
>> average day. Look (I'll try to find it, but can't guarantee it) for a
>> bit of research which proves that in a "random" collection of numbers
>> from a given domain, lower first digits predominate.
>
>Zipf's law, but that applies to *first* digits, not last digits.
I looked it up at the newly resurrected Mathworld.wolfram.com.
Zipf is actually a similar law for word frequency. The law
for leading digits is Benford's Law.
It comes from the fact that numbers are actually logarithmically
distributed, and we don't count "leading" zeroes. If you look
at a logarithmic grid, the 1-2 step is about 30% (0.30103...)
of the total distance from 1 to 10.
(see eq. 3; it's really just log(this integer)-log(previous integer))
http://mathworld.wolfram.com/BenfordsLaw.html
Numbers are logarithmically distributed because there are a
lot of things with a range of 0-(a few) and fewer things
with a range of 0-(a lot) and very few things with a range
of (something bigger than 1)-(a lot).
It's just one of those things that mathematicians call
"elegant" but really work out to "creepy" and "twilight
zone" for most people.
>Rounding up is just as likely as rounding down, so the accumulated
>error follows the statistics of a random walk. If you have n
>transactions, the amount you will be off by will be roughly the square
>root of n. To be out by even as much as one dollar would require
>100*100= 10,000 transactions. In an eight hour day, that would mean a
>transaction every 3 seconds.
That's an average. Once a month you'd be a couple of
dollars out. Now multiply the rounding error by 5. How
many minimum-wage checkers could stand the pressure of not
knowing if they're $10 out?
>Here in Australia, we got rid of a one and two cent coins several
>years ago, and nobody misses them.
Oh yeah? They're clogging up our drains!
--Blair
"Thruppence?"
Tim Lambert
> Tim Lambert <lam...@cse.unsw.EDU.AU> wrote:
> >"Jay R. Ashworth" <j...@dorothy.msas.net> writes:
> >> Such rounding would put you off by 3 or 4 dollars *per drawer* in the
> >> average day. Look (I'll try to find it, but can't guarantee it) for a
> >> bit of research which proves that in a "random" collection of numbers
> >> from a given domain, lower first digits predominate.
> >
> >Zipf's law, but that applies to *first* digits, not last digits.
>
> I looked it up at the newly resurrected Mathworld.wolfram.com.
> Zipf is actually a similar law for word frequency. The law
> for leading digits is Benford's Law.
Quite right, my mistake. Did you read Eric's commentary on why it was
shutdown? CRC is evil.
> >Rounding up is just as likely as rounding down, so the accumulated
> >error follows the statistics of a random walk. If you have n
> >transactions, the amount you will be off by will be roughly the square
> >root of n. To be out by even as much as one dollar would require
> >100*100= 10,000 transactions. In an eight hour day, that would mean a
> >transaction every 3 seconds.
>
> That's an average. Once a month you'd be a couple of
> dollars out.
Nope. Let's work it out. If they do a transaction every 30 seconds
for eight hours, that's roughly 1,000 transactions. The variance is
of one transcation is ((-2)^2 + (-1)^2 + 0^2 + 1^2 + 2^2)/5 = 2.
For 1,000 transactions the variance is 2*1,000= 2,000 and the standard
deviation is the square root of that, about 45c. Hence, 95% of the
time they're going to less than two standard deviations out, that is
less than a dollar. About once a month, the amount would be as high
as a dollar. And remember that thay're just as likely to be ahead as
behind.
> Now multiply the rounding error by 5.
Huh? Why?
> >Here in Australia, we got rid of a one and two cent coins several
> >years ago, and nobody misses them.
>
> Oh yeah? They're clogging up our drains!
They are? Which way do they rotate when they go down down the drain?
Stephen Fuld
news:GMq5q...@world.std.com...
Two comments:
"Nickle rounding" is not similar as Y2K. There is no fixed deadline. The
code changes are probably more localized and easily tested. There is a
manual workaround. No changes to file formats are required (e.g. expanding
files to provide for four digit years). So that isn't a valid comparison.
That Y2K didn't cause any major problems is a tribute to an awful lot of
work by a lot of people. I have talked to people who were involved in doing
the testing and changes. Believe me, had they not fixed the programs things
would have been a lot worse. You are right that many systems did not
require changing. But they all had to be checked. It was a massive job,
and one done very well.
Paul Brinkley
<j...@dorothy.msas.net> wrote:
>C.J. stepped down from the podium, whereupon Lynn said:
>> "Priscilla H Ballou" <p...@world.std.com> wrote...
>> > Tim Matheson has aged well, hasn't he?
>>
>> You can say that again.
>
>Tim Matheson has aged well, hasn't he?
Yep, that's why I hang around here, the comedy....
spam]@world.std.com Blair P. Houghton
>blair[no spam]@world.std.com (Blair P. Houghton) writes:
>> I looked it up at the newly resurrected Mathworld.wolfram.com.
>
>shutdown? CRC is evil.
I've been waiting for this to come back ever since the day
it was took down. CRC is evil with a self-serving cherry
on top. They're all that's wrong about copyright law. I'm
all for protecting artists, but to hell with corporations
that victimize naive artists and treat the consumer like
we should be glad they did.
>> >transactions, the amount you will be off by will be roughly the square
>> >root of n. To be out by even as much as one dollar would require
>> >100*100= 10,000 transactions. In an eight hour day, that would mean a
>> >transaction every 3 seconds.
>>
>> That's an average. Once a month you'd be a couple of
>> dollars out.
>
>Nope. Let's work it out. If they do a transaction every 30 seconds
Changing the rules? It was once every 3 seconds a second ago.
>for eight hours, that's roughly 1,000 transactions. The variance is
>of one transcation is ((-2)^2 + (-1)^2 + 0^2 + 1^2 + 2^2)/5 = 2.
>For 1,000 transactions the variance is 2*1,000= 2,000 and the standard
>deviation is the square root of that, about 45c. Hence, 95% of the
>time they're going to less than two standard deviations out, that is
>less than a dollar. About once a month, the amount would be as high
>as a dollar. And remember that thay're just as likely to be ahead as
>behind.
But at 10,000 transactions it's sqrt(20,000)=141. That
puts the 50% confidence level (the median, .675*sigma)
at 95 cents.
>> Now multiply the rounding error by 5.
>
>Huh? Why?
I thought you were working in pennies rather than nickels
when you came up with 10k transactions to be a dollar out.
But if I'd done the math I'd have seen that the standard
deviation for half-penny rounding is sqrt(10000*(.25+.25)/3) = 40 cents.
Or 13 cents if you only do 1000 transactions a day.
>> >Here in Australia, we got rid of a one and two cent coins several
>> >years ago, and nobody misses them.
>>
>> Oh yeah? They're clogging up our drains!
>
>They are? Which way do they rotate when they go down down the drain?
They're spolling clockwise.
--Blair
"And fucking hard to get out
of the trap..."
spam]@world.std.com Blair P. Houghton
><blair[no spam]@world.std.com (Blair P. Houghton)> wrote in message
>news:GMq5q...@world.std.com...
>> programming involved. It's not rocket science, but the
>> sheer numbers and variety of machines, some of which need
>> major code changes to make them round in other than pennies,
>> will make it a big deal.
>>
>> The difference between this and Y2K is that this would
>> actually be guaranteed to affect your pocket if it weren't
>> fixed. Y2K was a magnified shadow of a pocket-sized
>> straw-man for the pols to knock down.
>
Note the word "would". I'm being subjunctive. And if
they made the change, there would be a deadline, since
it would affect tax collection.
>code changes are probably more localized and easily tested.
Every calculation in the machine is liable to be a non-integer
scaling, with a rounding error applied. Only a few will be,
but like Y2K you don't know until you check.
>There is a
>manual workaround. No changes to file formats are required (e.g. expanding
>files to provide for four digit years). So that isn't a valid comparison.
It's a rough comparison. Nobody's offered me my billing
rate to do the real requirements analysis. The point was
the expansiveness and the piddling fixes, not the algorithm.
>That Y2K didn't cause any major problems is a tribute to an awful lot of
>work by a lot of people. I have talked to people who were involved in doing
>the testing and changes. Believe me, had they not fixed the programs things
>would have been a lot worse. You are right that many systems did not
>require changing. But they all had to be checked. It was a massive job,
>and one done very well.
A few major systems needed the change critically, a bunch
more needed it to avoid inconvenience. Most would have
been fine if they had crashed on 1/1/2000 and rebooted a
few minutes later. The draconian measures taken were
orders of magnitude more trouble than the problem. I
personally had to sign a statement--under penalty of
perjury--that a certain piece of test equipment, which
happened to be a computer used for time measurement, would
not crash due to the Y2K bug, because there was no chance
anyone would be running it at midnight on New Year's Eve.
It didn't matter that it measured time in microseconds, not
years. The delegees heard "computer" and "time" and
there was no way out except to accept responsibility on
pain of jail, or decommission the machine for several
months to have its code validated. It was a classic case
of the bureaucracy failing to have a sense of perspective.
--Blair
"Which might be a tautology."
Stephen Fuld
> Stephen Fuld <s.fuld.pl...@att.net> wrote:
> ><blair[no spam]@world.std.com (Blair P. Houghton)> wrote in message
> >news:GMq5q...@world.std.com...
> >> Think "Y2K Bug" for the scope and pissantry of the
> >> programming involved. It's not rocket science, but the
> >> sheer numbers and variety of machines, some of which need
> >> major code changes to make them round in other than pennies,
> >> will make it a big deal.
> >>
> >> The difference between this and Y2K is that this would
> >> actually be guaranteed to affect your pocket if it weren't
> >> fixed. Y2K was a magnified shadow of a pocket-sized
> >> straw-man for the pols to knock down.
> >
> >"Nickle rounding" is not similar as Y2K. There is no fixed deadline. The
>
> Note the word "would". I'm being subjunctive. And if
> they made the change, there would be a deadline, since
> it would affect tax collection.
If you mean sales tax, probably not. For the individual transaction, the tax
"rates" are actually "tables" saying that for items from X cents to Y cents,
the tax is Z cents. The tables would continue to work perfectly well. If
you mean the aggregate tax owed by the store to the government, remember the
rounding is fair, so with very high probability, the amount owed will be
very close to the amount collected and the probability of overage is exactly
the same as the probability of underage. I don't see a problem.
>
> >code changes are probably more localized and easily tested.
>
> Every calculation in the machine is liable to be a non-integer
> scaling, with a rounding error applied. Only a few will be,
> but like Y2K you don't know until you check.
>
I must be missing something. The tax is on the total, not the individual
items. The total would be computed exactly as before, the tax would be
computed exactly as before, but at the very end, the rounding function would
be applied. I agree, everything has to be checked, but the checking should
be very much easier than finding all the places dates are used in a multi
thousand or million line program.
> >There is a
> >manual workaround. No changes to file formats are required (e.g.
expanding
> >files to provide for four digit years). So that isn't a valid
comparison.
>
> It's a rough comparison. Nobody's offered me my billing
> rate to do the real requirements analysis. The point was
> the expansiveness and the piddling fixes, not the algorithm.
I agree there is a lot of work. However, my point was that it is different
from Y2K for a vriety of reasons and that the expense is one time, but the
savings from not having to deal with pennies continues forever.
>
> >That Y2K didn't cause any major problems is a tribute to an awful lot of
> >work by a lot of people. I have talked to people who were involved in
doing
> >the testing and changes. Believe me, had they not fixed the programs
things
> >would have been a lot worse. You are right that many systems did not
> >require changing. But they all had to be checked. It was a massive job,
> >and one done very well.
>
> A few major systems needed the change critically, a bunch
> more needed it to avoid inconvenience. Most would have
> been fine if they had crashed on 1/1/2000 and rebooted a
> few minutes later. The draconian measures taken were
> orders of magnitude more trouble than the problem. I
> personally had to sign a statement--under penalty of
> perjury--that a certain piece of test equipment, which
> happened to be a computer used for time measurement, would
> not crash due to the Y2K bug, because there was no chance
> anyone would be running it at midnight on New Year's Eve.
> It didn't matter that it measured time in microseconds, not
> years. The delegees heard "computer" and "time" and
> there was no way out except to accept responsibility on
> pain of jail, or decommission the machine for several
> months to have its code validated. It was a classic case
> of the bureaucracy failing to have a sense of perspective.
I agree with the lunacy of your experience. Many people, especially those
who talked to lawyers went way overboard. But the trial lawyers generally
support your end of the political spectrum not mine :-)
Chris Free
> >"Jay R. Ashworth" <j...@dorothy.msas.net> writes:
> >> Such rounding would put you off by 3 or 4 dollars *per drawer* in the
> >> average day. Look (I'll try to find it, but can't guarantee it) for a
> >> bit of research which proves that in a "random" collection of numbers
> >> from a given domain, lower first digits predominate.
> >
> >Zipf's law, but that applies to *first* digits, not last digits.
>
> I looked it up at the newly resurrected Mathworld.wolfram.com.
> Zipf is actually a similar law for word frequency. The law
> for leading digits is Benford's Law.
just a quick, minor, thought.
george kingsley zipf's law was
generalized by benoit mandelbrot --
in his work about information theory --
that was formulated by claude shannon --
who helped in the formulation of a way to
shoot down german planes during WWII --
by giving fodder to norbert wiener at MIT --
where john sheehan had developed RDX --
that also helped win the war against germany.
the zipf and mandelbrot connection is explained
in the book "the quark and the jaguar", written
by murray gell-mann -- who was friends with
richard feynman -- who helped create the
atomic bomb that helped win the war against japan.
without all of which we might be minting
pfennig and yen rather than lincoln pennies
and folksy firearms.
onward to the next "the west wing" and
sorkin's new melody, "gone quiet."
spam]@world.std.com Blair P. Houghton
><blair[no spam]@world.std.com (Blair P. Houghton)> wrote in message
>news:GMs9x...@world.std.com...
>> Stephen Fuld <s.fuld.pl...@att.net> wrote:
>
>rounding is fair, so with very high probability, the amount owed will be
>very close to the amount collected and the probability of overage is exactly
>the same as the probability of underage. I don't see a problem.
Tax collectors aren't known to trust people innately.
They send revenooers around to calibrate the cash
registers. Ever notice the sticker on the side?
Gas pumps too. It's an industry, contracting to the
government for standards and accountancy.
>> >code changes are probably more localized and easily tested.
>>
>> Every calculation in the machine is liable to be a non-integer
>> scaling, with a rounding error applied. Only a few will be,
>> but like Y2K you don't know until you check.
>>
>
>I must be missing something. The tax is on the total, not the individual
>items.
What state do you live in again? Check a grocery receipt.
Each taxable item will be marked with a star or a T.
Sometimes the taxes are different. Liquor, butts, etc.
And it's not just the taxes. Any item sold by weight will
need scaling and rounding to fix the price.
Once you find a few exceptions you realize you need to vet
almost every requirement in the box.
The look that fact puts on the faces of clients is when I
know I can buy a new car.
>I agree, everything has to be checked, but the checking should
>be very much easier than finding all the places dates are used in a multi
>thousand or million line program.
You won't find a cash register using a million lines of code,
but ten thousand would be kind of small.
>> It's a rough comparison. Nobody's offered me my billing
>> rate to do the real requirements analysis. The point was
>> the expansiveness and the piddling fixes, not the algorithm.
>
>I agree there is a lot of work. However, my point was that it is different
>from Y2K for a vriety of reasons and that the expense is one time, but the
>savings from not having to deal with pennies continues forever.
Well, yeah. I'm just saying there's going to be a tedious
and lucrative one-time charge for it. And then we get to lose
the nickel. And the dime. And by then cash will go away and
we'll start using pennies again, they'll just be electronic
ones. Maybe we'll round to the hundredth of a penny.
--Blair
"Depends on who's selling the upgrade."
Jay R. Ashworth
> You won't find a cash register using a million lines of code,
> but ten thousand would be kind of small.
Naw. I could do a standard retail cash register with multiple taxing
and preset/PLU keys in not much more than 2 or 3 kloc.
Well, ok, do I have to count the 4gl interpreter and the libraries?
:-)
Cheers,
- jra
spam]@world.std.com Blair P. Houghton
>C.J. stepped down from the podium, whereupon (Blair P. Houghton) said:
>> You won't find a cash register using a million lines of code,
>> but ten thousand would be kind of small.
>
>Naw. I could do a standard retail cash register with multiple taxing
>and preset/PLU keys in not much more than 2 or 3 kloc.
>
>Well, ok, do I have to count the 4gl interpreter and the libraries?
>:-)
Not if they're separately validated. That's actually a major
feature of off-the-shelf software. Spend once on the testing,
include many times in projects. It's about the only thing
really that keeps software from costing a nickel more than the
blank media it's delivered on. But someone, somewhere, would
have to validate it, at least once, to generalized (i.e.,
all-encompassing and therefore rigorously huge) standards.
--Blair
"And you thought politics was the
boring part of life."
Chris Free
throw in an ncr 2126 -- i hear they were y2k fun.
i also hear estimating cots integration costs isn't
np-complete -- no matter how good the estimates ;)
Jay R. Ashworth
> Jay R. Ashworth <j...@baylink.com> wrote:
> >C.J. stepped down from the podium, whereupon (Blair P. Houghton) said:
> >> You won't find a cash register using a million lines of code,
> >> but ten thousand would be kind of small.
> >
> >Naw. I could do a standard retail cash register with multiple taxing
> >and preset/PLU keys in not much more than 2 or 3 kloc.
> >
> >Well, ok, do I have to count the 4gl interpreter and the libraries?
> >:-)
>
> Not if they're separately validated. That's actually a major
> feature of off-the-shelf software. Spend once on the testing,
> include many times in projects.
*Bingo*. Thank you; I'd been trying to capsulize that for some time.
> It's about the only thing
> really that keeps software from costing a nickel more than the
> blank media it's delivered on. But someone, somewhere, would
> have to validate it, at least once, to generalized (i.e.,
> all-encompassing and therefore rigorously huge) standards.
Python has a regression test suite.
> --Blair
> "And you thought politics was the
> boring part of life."
Naw; coding's fun -- if you're in the right frame of mind.
Cheers,
-- jra
--
Jay R. Ashworth
> i also hear estimating cots integration costs isn't
> np-complete -- no matter how good the estimates ;)
No, it's NP-annoying.
Cheers,
-- jra
--
Stephen Fuld
news:nfII7.8116$uS1.86...@newssvr15.news.prodigy.com...
snip
>
> just a quick, minor, thought.
>
> george kingsley zipf's law was
> generalized by benoit mandelbrot --
> in his work about information theory --
> that was formulated by claude shannon --
> who helped in the formulation of a way to
> shoot down german planes during WWII --
> by giving fodder to norbert wiener at MIT --
> where john sheehan had developed RDX --
> that also helped win the war against germany.
>
> the zipf and mandelbrot connection is explained
> in the book "the quark and the jaguar", written
> by murray gell-mann -- who was friends with
> richard feynman -- who helped create the
> atomic bomb that helped win the war against japan.
>
> without all of which we might be minting
> pfennig and yen rather than lincoln pennies
> and folksy firearms.
Thank you James Burke! :-)
Getteur
>and lucrative one-time charge for it. And then we get to lose
>the nickel. And the dime. And by then cash will go away and
>we'll start using pennies again, they'll just be electronic
>ones. Maybe we'll round to the hundredth of a penny.
>
Perhaps that's what should be done, bypass the rounding to 5 cents and
eliminate both pennies and nickels in one fell swoop and round all cash
transactions to the next decimal.
Gary
spam]@world.std.com Blair P. Houghton
Lower all prices by a factor of 10.
--Blair
"Two bits! Thanks, mister!"