She won despite her age, not because she took drugs or anything. I think she deserves her medal. The only scandal here are the documents, not her competing.
Speaking narrowly to the issue of rule-making and rule-enforcement in general, and ignoring he question of the truth
of the specific allegations in this specific case:
Any rule not applied fairly is a risk to equal competition.
Just because you don't know whether a rule introduces a bias on the outcome does not mean that it
doesn't.
For example, let's suppose some country (any country) did have an athlete participating in an
event contrary to some rule. It doesn't matter if it's age or drugs or taste in
music. If some number of countries select from their entire population and others select from
only the people in the approved group, then whether or not any given country was able to show its
most competitive face is purely a question of whether the excluded group contains the most competitive
person.
Let's suppose the games are closed to anyone who likes hip hop music, for example.
Why might it matter if some hypothetical Foozania were to field a swimmer who secretly
likes hip hop music when the other countries voluntarily held back?
Absent Michael Phelps (we all know from US airtime allotments that there are not really any other
swimmers of note in the US), who would be voluntarily withheld because of his professed like of
hip hop, the Foozanian swimmer's scores might seem very good. By your reasoning, which seems to
amount to absence of competition, he deserves his medal fair and square, right? But if the absence
of competition can be caused by uneven application of rules, that's where the problem comes.
But beyond this, there is also a human rights question: Are there sports in which people
are pressured at a younger and younger age to get into the sport, before they are ready to
make a free choice? Are there sports in which the toll the sport takes on the athlete is
damaging before a certain age? These are complex questions of ethics that it seems fair for
the Olympic committee to at least consider, so you can understand why there might be such rules.
And once there are such rules, my examples above hopefully show why they must be applied fairly in order for
the Olympics to mean anything at all.
Imagine a world where current higher education materials are available to ALL OF HUMANITY instead of a select few rich enough to go to college and pay these "rich people only please" prices.
College prices are out of control, but the cost of books is not the problem.
Books even at these prices are still among the most cost-effective aspect of education of any kind,
including higher education. The value per dollar you get from a book even at such inflated prices
is probably well worth it.
Eliminating the economic motivation for people to write books is not going to end well for any of us.
With software, the alleged answer (yes, I'm paraphrasing with a slight cynical bias injected) says
that although the software is initially free, it's full of bugs
and people with money will gladly pay you to maintain it for them.
Even if you, unlike me, think that paradigm is approaching some kind of programmer Utopia,
I don't think with books there is a similar argument. People with lots of money are not waiting to
pay writers to maintain the books they give away for free. Nor are there large publishing houses
looking to hire people to fix typos...
Natural selection isn't creating some noble super-race. It favors the strong, but also the violent and the crafty. It looks only at outcome; it doesn't moralize about tactics. And its measure of outcome seems, by modern theory, limited narrowly to "has offspring ready to play the game anew". That's a possible theory of "good", but not the only possible theory. It seems just a little limiting, in fact. Which is why society tries to circumvent it through conscious thought and group policy, for better or worse.
Following on my prior remark about how selection is selecting the physically strong, not other attributes,
I wanted to add one thing: If right now we just let the weak in the gene pool die, and we saved only the
people who were physically strong, I doubt that would save us from Climate Change. Nature might recover from that on its own, but it's likely that its way of doing it will be to shed most of the species on Earth, including us, and start back with bacteria. It's nearly certain
that the reason Nature has the cataclysms it does is that it's subject to the Hill Climbing Problem in its broad brush attempt to survive. That's robust in a sense. We are, in the end, a fragile species and perhaps there is a robust one that Nature might eventually come up with if the Heat Death of the Universe didn't loom so tangibly close (at least, when weighed against the speed of average case convergence for the algorithm Nature uses for coming up with good species).
Now, it's theoretically possible that Climate Change will stop suddenly of its own accord
and that either brain power miraculously won't matter or
adaptive physical strength will be all that's really called on to live in a world of altered air and
water. So it's possible brains won't be what we need. But those who do have brainpower right now are
not bullish on these options, and I think it's not because they're worried about their jobs or ego. I think
they're worried for their kids and grandkids. So we'd better pay some heed to preserving their ability to
see not just the calamity of the moment but the potential calamities of the future in time to do something
other than the narrowly selfish to the needs of the business quarter.
Mankind right now, through the actions of scientists, not football players, has the self-awareness to recognize that an adaptive algorithm inserted just about now might improve matters for both Nature and ourselves, and brainpower, not musclepower, is what's going to give us our only real chance.
It's the only tool man has ever had that allowed it to rise above the other animals, and we dare not
at this point suggest that it is of no consequence.
There may come a day when indeed we can't save everyone. But when that day comes, what we'll probably need
is brains, not muscle. And if that day comes, it will be the rich and the politicians, if that's not redundant, making the choices. So let's hope we keep the world safe for all now to avoid having it safe only for the Dr. Strangelove contingent later.
Obligatory Relevance Comment: What has this to do with medical records being sold? Because they are not
being sold to people who are optimizing the outcome of humanity. They are being sold to those who are
optimizing their portfolio, a portfolio that is not taxed in proportion to the dissonance between its own size and the good meta-health world. In meta, they are playing out the same survival game in business as we're playing out in the world, with Capitalism substituted for Nature. But they're not equipped to survive the storm if Mankind is wiped out through their follies. So all of these things are interconnected, and Climate Change is, I allege, not as off topic here as one might be tempted to presuppose.
However, catering to the sickly weakens our gene pool.
If our society were based on blugeoning each other with clubs, this would be a relevant
argument. We'd need the specific quality of physical strength and resilience to survive.
But the fact is that people who are "sickly" (to use your word) can make
important contributions to society exactly because the aspects of the person
that is required to make
those contributions is often unrelated to the health issue they may confront.
Look at Stephen Hawking for example. Nothing wrong with his brain, so as long as the
essential aspects of his body are functionally maintained,
he can continue to make his contribution. And even when
reasoning in some sort of cold/mercenary way, the cost of maintaining
such a person may be much less than the cost of losing such a person's potential contribution.
Besides, natural selection is intensely focused on the high order bit--whether people
survive to breeding age at all. It's not very concerned with selecting for good writers, philosphers, mathematicians,
teachers, etc. Nor does it appear to care a whole lot about diseases that come up after
breeding age. So the argument about the gene pool being affected by caring for the so-called "sickly"
seems bogus given that a lot of people who we care for are older than breeding age and do not,
at that point, contribute
to the gene pool.
Natural selection isn't creating some noble super-race. It favors the strong, but also the violent
and the crafty. It looks only at outcome; it doesn't moralize about tactics. And its measure of outcome
seems, by modern theory, limited narrowly to "has offspring ready to play the game anew". That's a possible
theory of "good", but not the only possible theory. It seems just a little limiting, in fact.
Which is why society tries to circumvent it through conscious thought and group policy, for better or worse.
This sounds like the latest incarnation of the dream of television executives who in the early '90s talked about the "information superhighway," before it was clear that the Internet was going to fill that role. What they envisioned was "interactive TV," i.e. buying stuff with your remote.
The last piece of the puzzle, completing the cycle of virtualization, will be Wii integration, so that we
can look back fondly on the days when we were able to go to the store to buy it,
pull it off the shelf, and carry it home.
The multiplayer version will allow our visiting friends to be the person at the counter, who talks us into buying.
Watch for extra points to be racked up quickly if they get you to buy lots. Perhaps even a commission, paid in wee dollars.
just because you don't know what a patch did doesn't make it evidence of tampering
Although I am not an unconditional fan of open source. I think it has its uses and its non uses. But this is one
definite use. It might be the case that as a matter of law right now, what you say is true--it's not evidence.
But I actually don't see how we'd be hurt by having a law that makes it into evidence by making it, ipso facto,
a crime to apply an undocumented patch at all.
If you're going to have voting machines at all, it seems to me they ought to have open source programs
running on them. (I personally don't care if the programs are free in cost, though I imagine they could be.
I only care that they are inspectable by anyone
and that it is fair use to copy them to other machines for the purpose of verifying their correctness and aherence
to advertised and required spec.)
I don't see that anyone other than someone doing something illegal is served by keeping the source secret.
The data, of course, might be protected. But the program should be auditable by anyone.
And in such a world, I don't see any reason whatsoever that it shouldn't be a crime for there to be any
software applied where its entire content and purpose were not carefully documented. Then it wouldn't require any
proof other than that it happened in order to detect wrongdoing; and it could be further a crime to phony up a faulty
description of what the patch does, so that then it could be audited for validity by anyone who wants to.
The weak link in all this seems to be the hardware. It's quite hard to look at a box and know what's going on inside it.
It requires specialized skill. It likely wouldn't be hard to make a machine with a dummy system for show that had the
right program on it and another shadow machine tucked away that had the wrong program, with just some subtle wire somewhere leading to one doing the real voting. Call it "democracy theater" if you like, borrowing on the "security
theater" moniker used a lot in other venues these days.
Such a charade could be hard to notice. Which is why I prefer paper balloting.
It means the entire process is exposed.
Employing workers abroad often involves flying management and workers back and forth,
even for non-manufacturing jobs like the ones near and dear to most Slashdot readers: software.
I wonder if the higher prices of oil will result in
better teleconferencing or if there is a cross-over point at which such coordinated
management at a distance will be deemed to impractical.
If it were at some point deemed impractical,
I wonder whether that would tend to result in jobs staying here or just locating entirely abroad,
since shipping end user software is really cost-free and so the "manufacturing" could be anywhere.
While it may be unfair to target fat people (or smokers or drinkers or what-have-you), isn't it equally unfair to make healthy people pay a lot of extra money to support the unhealthy lifestyles of their neighbors?
... unless the reason those people are overweight is that they are stuck at a desk job making business succeed while the people who are out playing tennis are not.
Yes, some people gain weight because they are loafing, which I think is what is at the heart of this--a belief that it's as simple as couch potatoes or self-discipline. But some people may be working overtime and not have time for a healthy meal, and the success or failure of businesses may depend on those people.
And at this point some people may be innocent victims of genetic predisposition after generations of this. Shall we tax people for bad DNA?
To reduce the whole discussion to "those people cost me money" is naive and self-serving unless there is the data to back up the claim that these are all independent variables, rather than a complex web of things. At the heart of all bad social campaigns is an attempt to depersonalize the data and assume that the Other Side is all of one kind and all trivially characterizable by some stereotype.
Those who have struggled with diet know it is immensely complicated, and in fact one thing that can work against their succeeding is a feeling of lack of self worth, which certainly a tax is going to only make worse.
Also, the other very very subtle thing is that one thing that may make the US fatter is that it's free (in the sense of rights). The more freedoms you give up, the more you can help certain problems. But not all problems. Freedoms turn out to be intrinsically important sometimes.
There was a Garfield comic I read once where he was lazing around and noting that you could be sure Power Steering wasn't invented by an exercise freak.
p.s. In my browser (IE7), the new Slashdot page format for editing posts has me typing this text inside a 1.5 inch by one 1.5 inch square, non-stretching box, so apologies for any typos but I can't be bothered to do serious editing in this tiny thing. I can barely see what I'm writing.
Creation of new domains is like extortion. For example, Disney will have to pay for disney.fun, disney.kids,
(And for end users, it's a fraud dilemma since there will be potentially a zillion names that look authentic, with no way to know
what's really the thing.)
Having open season on TLD's is computationally equivalent to eliminating TLD's and just saying there's only.COM.
It just seems a bad idea.
What's next? TLD regexps? Will ICANN be selling "(^|.*[.])DISNEY($|[.].*)" for a bundled fee?
AS far as credentials go, I think the university system is important. However, science doesn't require that people -have- Phds
Actually, I didn't mean credentials have to be of the university kind.
In suggesting exploring credentials, I didn't mean to be restrictive at all. Slashdot's karma points
and personal-relationships (friends and friends of friends) are not category-specific, so wouldn't do much
for science, but are examples of credentials at least. In Blade Runner, Roy Batty visits his creator using
his chess expertise as a credential. I meant it in the most general sense, not to presuppose any particular
mechanism, only to say that ultimately everyone uses some mechanism and the only question is whether there
is institutional help in figuring out what you need to test and what not. After all, the number of testable questions in the world exceeds the time you have to
test them all, so even if the answer to P or NP was P, at some point you'd have to just trust again, because there isn't even enough time in the world to run even a polynomial-time query on every factoid you end up trusting in a single day. So the question isn't whether you'll trust, it's only whether when you do trust, you "trust well" or "trust poorly"...
The beauty of good science is that you DON'T have to trust someone else's eyes. You can trust your own.
If you go back and re-read my post, you'll see I didn't say what you think you're responding to.
I didn't say science wasn't testable, just the opposite. I suggested societal experimentation with
coming up with decent credentials, precisely based on the notion that there is some hope of objective
truth at that level.
And, in part, my remarks about consistency-checking is exactly what you just said as well.
By not trusting someone else's eyes, you're saying to check what they say for consistency.
The place where you probably went astray is when you thought I was talking about doubting science.
I was talking about doubting one's senses. That is the fundamental leap of faith. You know what science tells
you because you believe your senses. But how do you know your senses tell you anything useful, and that life
is not one big VR game, like in the Matrix. You don't, actually. You simply accept it as the rules of the game because there seems no alternative. This is what you would have learned from my Dark Star reference.
But relating it back to Wikipedia, my point was simply by analogy, Wikipedia has a hard time knowing what
is true because it only knows what it is told. You can make credentials for it, and then it will know what credentialed people think, but then that's only as good as the credentials and the will of the people with them to get to the truth. Even if the Wikipedia standard were one of truth (and I'd like that better, frankly, imperfect as it might be) rather than a standard of documentability, it would still be wrong sometimes. But not because of some sinister reason--just because there is no objective standard to actually measure truth and to know when it was right and wrong.
So my post wasn't off-topic, it was just marked that way, presumably by someone with credentials. Credentials get you only so far toward the truth. You can credential the credentialers, as happens here with metamoderation. But at some point there's always a leap of faith and you just shrug and say "good enough" and get on with life, even knowing it's imperfect.
As to judging tin whiskers, my point was not that there wasn't truth to be had, it was that for all its faults, the one virtue of Wikipedia's approach is that at least when it's wrong, it's supposed to contain a paper trail that leads you to original source material to learn why someone thought the wrong thing. That's not perfect either. But it is at least a defensible theory. And for more than that, there's the free market and other models.
I took away my own karma bonus on this to save someone the trouble of moderating me down if they continue to erroneously think I'm off topic. I just wanted to respond here and say I don't really have a disagreement with anything you said. I am a strong proponent of the scientific method. I just think you mis-read my intent.
I used to buy into Wikipedia's stated ethos until I realized that any one person can... hijack articles to push and protect their point of view and once that happens you can forget about the "Five Pillars" and objectivity.
Well, trust is not a binary thing. You can trust someone for one thing and not another. And even if you do
trust, you can have trust at a wide variety of different levels.
I'm not a big fan of Wikipedia in some ways either, not so much because it lies, but because it doesn't want
the truth. If I know something true, and I'm the only person in the world, it doesn't want it. But if I
know something false, and I write it up, then it's referenceable, and it becomes closer to something Wikipedia
does want. I can understand both of those at some level, but I think Wikipedia should care a lot more than it
does about creating new mechanisms to let in real truth (perhaps creating a mechanism by which individual
knowledge can be vetted) and keep out falsehoods (perhaps creating mechanisms for peer review of referenced
documents). The fact that it doesn't is, of course, why other competitors have come up.
I guess on that point, you have to score one for the marketplace for at least creating the idea and allowing
competition to crank out alternatives.
But as to what to trust in Wikipedia, their strength is that the things they say are supposed to be
things that can be backed up by reference. Where you see a strong claim and no reference, find a way to
flag that fact and maybe the person who put it in will add a reference. Where you see a reference, follow
the chain back to the original source. That source may ultimately be believable or not, of course.
In some sense, by its choice of paradigm, Wikipedia is just a complicated, statically-enumerated set
of search engine results. It gets you started, but it isn't the whole of the thing you want.
If I knew more about phenomenology, I'd probably say that's just the nature of the Universe, and that
Wikipedia can no more escape it than anyone can escape the Three Laws of Thermodynamics.
That is, no one ever really knows anything about the Universe other than what they're told, and what they can work out
in terms of internal consistency checks on what they're told. But all I know of that is what I've
seen mentioned in
Dark Star. So I'll let you do your own research there.
Whether to direct you to Wikipedia or the movie though, to study more of phenomenology... I dunno, that's a hard choice. Probably I'd say just see the movie. It's worth more than the 6.5 stars IMDB gives it.
One could just imagine what Bomb #20 might have to say on the matter of Wikipedia...
Kirk: Mister Saavik, punch up the data charts of Reliant's command console.
No, that's MISS Saavik, as in female
It's not an error.
I don't know if it's done commonly any more, but mister has
sometimes been used in military settings as a unisex term.
We used to do that sometimes in the ROTC program I was in (a small number of years
before that movie was made).
The military sometimes likes people to be interchangeable (which is part of the reason for
wearing uniforms and normalizing haircuts), and differences in such forms of address
tend to emphasize
individuality, so they're inclined to suppress it for reasons of suppressing a sense of identity.
That might seem a negative, but by deemphasizing individual differences, so the theory goes,
everyone starts at the same place and has the same advancement opportunities. There was a theory
then, and I don't know if it's still adhered to (there's presumably been additional statistics
taking since then about whether it worked, which maybe someone knows about and can share)
that the use of the female honorific would draw attention
to the person's gender rather than to their role, and so it was thought better to emphasize role.
The use of "mister" itself in the military is really a role thing.
It's not used to address someone's marital status or gender,
but rather as a way to refer to someone who has a rank but
without reference to their rank. So instead of saying "Corporal Jones" you say "Mr. Jones",
to deemphasize rank. If in doing so, you had to bring in gender, the thought was that this would
waste some of the value of moving to a rank-neutral term. You'd only be trading a gender-neutral title of rank for a rank-neutral term of gender. By making the term genderless, you increase its
utility. Or so the theory went.
In the days where I was in that program, women were just finally being allowed
a foothold as equals and my sense at the time (I was quite
an advocate of that equality) was that it tended to produce the desired outcome, causing people
to be treated more as equals. I can't say as I know for sure
what the emotional impact was, but my guess
knowing the people involved was that they wouldn't have been in the program in the first place if
they were worried about something like that. They were deliberately pushing gender stereotypes
just to be allowed into those roles, and a few non-standard word usages weren't going to deter them.
Of course, one of the things that Star Trek has always done is mirror society's progress (but on time delay), so the fact that the term fell out of use in later movies (and maybe in the real
world military, too) may well have been a mark
of the fact that once the gender barrier was broken, the fact of the gender reference in someone's
name was not as big a deal as it had been until then. In that era (1970's), it was
a big deal
that we had a female brigade commander, just for example. I'd like to think that nowadays, that's
just the ordinary course of business--though some of the recent political debate about Hillary as a possible Commander in Chief makes me think maybe public acceptance on this has a ways to go yet.
Still, Star Trek and its many good role models give hope.
The conversation is there, recorded, ongoing, and a more careful consideration than just casual water cooler talk. Reputable peer reviewed journals are just another form of subjectively valued moderation, fame, and reputation.
The key, though, is not just to have people be able to use such alternate media,
but to have it be legit as a reference/credential for advancement
in schools. Right now, anyone has a voice, but it doesn't mean
citeseer accepts them and it doesn't mean professorships are granted based on them.
In some ways I think the open question almost isn't "will schools accept this?" but rather
"will traditional schools be able to retain their credibility if they continue to lock out these
other media?" Schools in the US continue to get more expensive at a rate that far outpaces the rest of
the society's recent cost of living increases. As it becomes more and more expensive, the question
becomes whether alternate education isn't just as good on a benefit-per-dollar basis, and whether
companies won't start saying "Forget traditional colleges. Just show me you know what you're doing."
It may not have happened yet, but at some point it will. And the failure to accept alternate peer
review mechanisms as credible may hasten the process, since it will create easy-to-see showcases of
alternate knowledge sources and alternate ways of contributing to research and development that is
untapped by conventional schools. I don't see how they can ultimately afford not to consider
such mechanisms.
Without peer reviewed journals
then there would be no accountability in science.
Well, in fairness, although I agree that peer review is quite important, what is an open question is
whether peer review needs to happen in advance of publication, or if there isn't some way for
peer review to happen after-the-fact. This would accomplish three things: (1) it would allow faster
access to raw ideas in the marketplace [at the sometimes expense of reliability], (2) it would allow
more opportunity for someone who thought something was valuable to make themselves known, since in at least
some cases the reason something isn't accepted to a journal may simply be lack of room or lack of interest
by a particular target community, and (3) it would allow dynamic assignment and unassignment of
support over a long period of time. Sometimes the sense of what's important at the moment is wrong among
peers, and this would assure that such things could be later corrected.
I'm not especially advocating after-the-fact peer review. I'm just saying it's not obvious that it can't
occur, and it does seem like there are at least some benefits (even along with what I am sure are also
new risks that I've not detailed, but that almost certainly involve the need for society to come to grips
with the notion that publication does not imply community acceptance, something that may never have been
true even in the past but that may be doubly uncertain in a world where there was no barrier to publication).
Then again, the web is available as a form of unfettered publication path to anyone who wants it, so nothing is stopping individuals on a case-by-case basis from publishing without peer review. The sad thing is that there's no way to take a finding that is already out on the web and promote it to journal status in terms of peer acceptance. If anything, I suspect there's some hidden desire, as probably happens in the regular
publishing industry too, to punish those who go outside the system for not having followed the rules. And
that kind of thing seems a less good side of the peer review process.
The fact is, metered bandwidth is good for our own freedom because it gives us a greater argument for demanding a hands-off approach to regulating protocols. If you pay for the bandwidth itself, rather than just a simple monthly access fee, it's easier to argue that it's your bandwidth now and the ISP needs to piss off if they think they'll tell you how to use it, the law notwithstanding.
Great point, succinctly expressed. I totally agree. This is related to a point I was going to make: It annoys me that one cannot resell bandwidth.
The notion that one person having 3 people in the house can grab bandwidth for all of those people at one
price, but three people living separately have to buy 3 separate services seems unfair. In practice, it
means that lots of people cheat and get away with it, while the people who don't cheat are charged a premium (or, more specifically: several premiums)
for operating to the letter of the rule. For quite a while, I've been pushing the need for Universal Business Access, and have only just recently written it up, but it relates to that.
Email has a similar kind of issue, where not paying is more expensive than paying, since we all pay for spam due to
email being free, and the cost of that spam is certainly way higher than the cost that email itself would be--except to the spammers.
Where I think it is dangerous to pick and choose improvements to the floating point standard is that the exact properties of this standard are also the basis for axioms of computer arithmetic
Can you say what axioms you think are violated by what they did? I see what they did as subsetting, not as changing
the rules. I'd like to understand better otherwise.
In general, floating point from the outset doesn't even guarantee x+1-1=x, so it's hard to say that someone trying to make subsets of floating point is ill-advised in doing so. Floating point takes risks and for you to say:
I think it's much better to know about and rely on existing code that works
seems a hair's breadth from a religious rather than a computational statement.
I don't personally acknowledge the concept legitimacy of the unadorned word "works". I prefer to use the terminology
"works for " and to judge "works" in a specific context. The problem with using it in general is that
it presupposes a one-mode-fits-all paradigm, and I don't think that's a good assumption.
I'd rather equip people to think, to use their brains, and that means equipping them to resist phrases like
"look, we've worked this out and it works so you don't have to think about it."
One of the most empowering things that I ever learned about floating point was when I saw someone
with a great knowledge of floating point say "I don't understand floating point." I suddenly became conscious
that one could "not know" for reasons that are wise, not just reasons that are uneducated. I have come to understand
that a little healthy fear and respect of what is really nothing more than a calculated gamble goes a long way.
I do not understand floating point. I also doubt many do, and especially I doubt that many people who claim to understand
it do. I'm sure a few do, but not as many as are intended to be users of it. That's not to say I don't think people
should study it--it's to say I think they really need to know a lot to really appreciate its subtleties. And it's also
to say that I think teaching them "it works" is not the path to that understanding.
I don't really disagree with your premise of wanting algebra to work out. I just feel like I may disagree with what
you may be saying follows from that. But feel free to set me straight.
I don't disagree with this kind of statement you made though:
There is often a serious difference in performance and accuracy between a hand rolled method, and a properly debugged and optimized implementation such as can be found on netlib.
Absolutely I think people should be educated in this kind of subtlety, even if they are only going to use such algorithms, so they will be good shoppers.
... Excel doesn't adhere to IEEE 754. Why Microsoft feels that they know more about numerics than William Kahan I cannot fathom.
I appreciate seeing this cross-reference--thanks for posting it.
However, I do disagree with your conclusion. The regions of non-conformance that Excel cites are issues that
are commonly discussed and sometimes actively disputed, and in any case that often require very skilled
understanding to navigate. Although I've not been in on any internal discussions at Microsoft about how they
organize their product line and who they intend to sell to, I feel confident in saying that Excel's target
customer is not typifically a physics student. This is a product for business, and the mathematics of
binary-coded decimal (BCD) seems more appropriate to
its intended end, even if they don't actually use BCD in practice, probably for efficiency or convenience.
But the banking community is not wrong in preferring to do things in variance from IEEE's floating point standard
either.
Note that Microsoft has not assigned an incompatible meaning to the situations on which they don't conform (although
it might not be unreasonable if they did), it has merely said that it doesn't
support some cases. I actually think that even the brief passage you cite offers credible justifications for their
decisions. (I'm frankly surprised that the things they do are classified as non-conformances. There are other standards where the decision to simply not implement part of a standard and to signal an error in that case is considered a subset and therefore to be strictly conforming. Maybe there is some subtle way they deviate, or maybe they're just being careful in classifying what they did as non-conforming when it's really a judgment call so they aren't accused of hiding the issue.)
But back to the point of this thread, there are two take-home points:
(1) Excel does not seem to me an appropriate first choice as a tool for doing math or science, and especially
as a tool for learning about that. A program in those areas should pick one or more
tools that accommodate the notation, computational power, library base, and other factors required to support
serious computations.
(2) Students of math and science should be taught not that there is one particular way that numerical computation
is done, but rather should be taught why fixed point / BCD is different than floating point / IEEE 754 so that they can make informed decisions about which tools
suffice in which circumstances. For that matter, I'd also like them to understand the difference between numeric computation and symbolic computation, and why each of those has its place. School ought not be about dispensing dogma. It ought to be about
preparing people to be good decision makers as circumstances change.
As to the language not being important, I don't know that that's entirely
true. Each language offers a choice of types and some sets of choices are
more instructive than others. I personally think Lisp or Scheme are good
teaching
candidates because they offer arbitrary precision integers
and rational numbers
in addition to basic floating point number types
so that it's easy to see side-by-side
the trade-offs being made between correctness on one hand and space/speed
on the other hand that go along with choices in this regard.
I'm curious which free software has eliminated all competition in its niche? People still pay for web servers, C compilers, and operating systems. Perhaps web browsers. Anything else significant?
<tongue-in-cheek>
Nope, you've pretty much enumerated the entire space of things anyone would want to make out
of software and certainly that they might have wanted to receive money for.
Moreover, now that you've prodded me to think on it further, I realize there's no chance that anyone would ever have wanted to make software of some other kind than you've enumerated and yet has
then declined to make such software because of the likelihood that an
immediate free-software knockoff made by a
free software "sniper" might undercut their investment.
So, upon reflection, I stand corrected. Sorry for any confusion.
</tongue-in-cheek>
By the way, the point of my original post was about Google, and my mention of free software was
only incidental. I've replied here in fun so as not to ignore your post, but I plan not to carry
this further and have declined my otherwise-automatic karma bonus on this post as a self-imposed penalty for drifting even this far off topic.
I voluntarily discounted my karma bonus to make this reply, which is really just a courtesy
response and won't advance the debate.
Free software hasn't killed the proprietary software business.
You're right that the effect is not total. However, some free software projects have killed some
proprietary software efforts. It's a subjective matter to decide how significant that effect is--I happen to rate the negatives more highly than some other people.
Anyway, my intent was not to open the whole free software debate, it was merely to allude to it,
so I'll agree to disagree with you on this point rather than re-do the entire debate here. I wasn't intending to troll about free software, only to make an allusion about free software to those who, like me, believe that effect to be significant. If you're not one of them, and you think I'm nuts for thinking as I do, I guess I can live with that for today. We'll debate the issue of free software another time in another forum in a more serious way, I'm quite sure.
I really think you are jumping the gun here.
Yes and no. After all, I did say "If Google becomes the standard of mail," so that's of course a hypothetical.
I didn't say it had happened. I was hypothesizing about an effect that could happen
by applying observations about a trend in another area. I hardly think it's jumping the gun
to speculate. All speculation jumps the gun. That's the point of speculation, to be ahead of
the gun.
but also no one can afford not to use the free thing because the cost of the luxury of buying an alternative brand will be exposed by the market as superfluous if passed along to end users.
Yes. That is a feature, not a bug.
Either the alternative brand has some value, which end users will pay for or its value is not worth anything and the end users are not willing to pay for it.
The mistake you are making is that you value competition for its own sake.
And the mistake you are making is that you think that things that competition
does not value are not of value. Competition is not sentient and functions
only when proper assignments of dollar value have been created. It is
possible to badly assign dollar values to things and in so doing to create
situations in which competition will kill things of value as, for example,
it might be argued is occurring with things like global warming, where
a competitive market is seeking various earth-unfriendly situations because
the dollar value of having a world that continues to function correctly has
not been appropriately assigned. So please don't imply that the God
of Competition is all-knowing about what is good and right and proper in the
world without adequately investigating the complete assignment of human value
to dollar value.
I'll take a neutral point: Let's say we believe that there is intrinsic
human decency value to health care, and yet there are companies in the world
who don't pay it. What we will find (and are finding) is that the market
will seek supply from those who are in the market at the least cost. In
some cases, we can suppose, there's the theoretical possibility that these
remote locations to which jobs and businesses are moving offer equivalent
health care and other standard of living to what the US offers, and perhaps
in those cases what it's saying is that the US simply charges more than it
needs to in order to provide these things. But certainly it is the case that
in some cases, one reason the cost is lower in those other places is that there
is not health care (or not at the same level) or there is not environmental
regulation (or not at the same level). Competition will squeeze that out as
a luxury because the person buying the sneakers or the iPod or the toaster
or whatever doesn't think they're buying someone health care or buying the
world the right to survive climate change--they just selfishly want cheaper
products without regard to any of that.
And so if I own a company where I would like to offer better health care to
my employees, I may not be able to afford to because my competition does not
and I cannot compete without offering the lowest of what anyone else does or
that cost of my "indulging" my employees with good health care will show
through to my product. But that is not to say that that health care
had no value. The problem is that in a layered product world, value occurs
at different levels and value at one level is not always tolerated at another,
so that platitudes like your (repeated for emphasis)
Either the alternative brand has some value, which end users will pay for or its value is not worth anything and the end users are not willing to pay for it.
do not hold up. The fact is that if I have a company where my
employees really will be "happier" using a commercial piece of
software internally, but if it won't crank out a better end-product,
the competitive marketplace rates the value of the employee happiness
at zero. And so you're quite right that the market will crank that out.
But what you're wrong about is that there is no loss in the process.
Because in the end, taken to its extreme, competition is capable of cranking
out all fun and happiness from all the world in the name of serving
people better. But if no one is having any fun or happiness, to what end
is that?
partnering with Google offers DreamHost a way to offload many of its trouble tickets, reducing the support overhead. Is Google starting to make web hosts less necessary?
One of the things I don't like about free software is that it basically pays for itself off the
profits of an unrelated industry, eliminating competition in an otherwise viable industry because
someone can afford to offer the service for free as a loss leader to other business.
A thing that is especially troublesome is that not only does it basically make it so that no one can afford to
be in the business area (software development for money) competing with the free thing
(software given away for nothing), but also no one can afford not to use the free
thing because the cost of the luxury of buying an alternative brand will be exposed by the market
as superfluous if passed along to end users.
It seems to me that if this becomes a trend, it will be the effective continuation of that paradigm
shift by Google into another area, and that the logical continuation of this, by analogy, would
be that not only can no one afford to compete with Google and other agencies giving away free mail
but no one will be able to afford not to use Google's mail.
That would be sad if it turns out that there are reasons why using Google's mail is not a good
idea... such as, for example, concerns about privacy.
If Google becomes the standard of mail, the problem is that it can afford to add incidental
services in parity with any nuisance it causes, making it impossible for would-be competitors to
match on a value-point by value-point basis even if they find a way that should theoretically be
able to compete.
Just like me, my employer pays me to have produced the software and then I see no further renumeration. From my POV, the software is free.
Well, if I understand your situation correctly, I suspect the reason your employer can pay you is that the software is not free, and so that's why they have money to pay you.
My guess is that you'd notice it if your employer had no money coming in from that software.
From your "POV", that could look like unemployment.
So if you want something in apache, you either do it yourself or pay someone to produce it. Or wait to see if it gets done for someone else.
The software is still free, but the time is paid by creating the software from someone's requirements or from personal need.
If that's all you (and probably others) aspire to in terms of design, that explains why
I don't like this paradigm and you don't see the problem. To me, anything that involves adding
a widget or module to Apache is not program design, it's program maintenance.
To have programming reduced to this is like telling an would-be orchestral composer
there's plenty of work available writing advertising jingles.
Slashdot Mirror
Speaking narrowly to the issue of rule-making and rule-enforcement in general, and ignoring he question of the truth of the specific allegations in this specific case:
Any rule not applied fairly is a risk to equal competition. Just because you don't know whether a rule introduces a bias on the outcome does not mean that it doesn't.
For example, let's suppose some country (any country) did have an athlete participating in an event contrary to some rule. It doesn't matter if it's age or drugs or taste in music. If some number of countries select from their entire population and others select from only the people in the approved group, then whether or not any given country was able to show its most competitive face is purely a question of whether the excluded group contains the most competitive person.
Let's suppose the games are closed to anyone who likes hip hop music, for example. Why might it matter if some hypothetical Foozania were to field a swimmer who secretly likes hip hop music when the other countries voluntarily held back? Absent Michael Phelps (we all know from US airtime allotments that there are not really any other swimmers of note in the US), who would be voluntarily withheld because of his professed like of hip hop, the Foozanian swimmer's scores might seem very good. By your reasoning, which seems to amount to absence of competition, he deserves his medal fair and square, right? But if the absence of competition can be caused by uneven application of rules, that's where the problem comes.
But beyond this, there is also a human rights question: Are there sports in which people are pressured at a younger and younger age to get into the sport, before they are ready to make a free choice? Are there sports in which the toll the sport takes on the athlete is damaging before a certain age? These are complex questions of ethics that it seems fair for the Olympic committee to at least consider, so you can understand why there might be such rules. And once there are such rules, my examples above hopefully show why they must be applied fairly in order for the Olympics to mean anything at all.
College prices are out of control, but the cost of books is not the problem. Books even at these prices are still among the most cost-effective aspect of education of any kind, including higher education. The value per dollar you get from a book even at such inflated prices is probably well worth it.
Eliminating the economic motivation for people to write books is not going to end well for any of us.
With software, the alleged answer (yes, I'm paraphrasing with a slight cynical bias injected) says that although the software is initially free, it's full of bugs and people with money will gladly pay you to maintain it for them. Even if you, unlike me, think that paradigm is approaching some kind of programmer Utopia, I don't think with books there is a similar argument. People with lots of money are not waiting to pay writers to maintain the books they give away for free. Nor are there large publishing houses looking to hire people to fix typos...
Following on my prior remark about how selection is selecting the physically strong, not other attributes, I wanted to add one thing: If right now we just let the weak in the gene pool die, and we saved only the people who were physically strong, I doubt that would save us from Climate Change. Nature might recover from that on its own, but it's likely that its way of doing it will be to shed most of the species on Earth, including us, and start back with bacteria. It's nearly certain that the reason Nature has the cataclysms it does is that it's subject to the Hill Climbing Problem in its broad brush attempt to survive. That's robust in a sense. We are, in the end, a fragile species and perhaps there is a robust one that Nature might eventually come up with if the Heat Death of the Universe didn't loom so tangibly close (at least, when weighed against the speed of average case convergence for the algorithm Nature uses for coming up with good species).
Now, it's theoretically possible that Climate Change will stop suddenly of its own accord and that either brain power miraculously won't matter or adaptive physical strength will be all that's really called on to live in a world of altered air and water. So it's possible brains won't be what we need. But those who do have brainpower right now are not bullish on these options, and I think it's not because they're worried about their jobs or ego. I think they're worried for their kids and grandkids. So we'd better pay some heed to preserving their ability to see not just the calamity of the moment but the potential calamities of the future in time to do something other than the narrowly selfish to the needs of the business quarter.
Mankind right now, through the actions of scientists, not football players, has the self-awareness to recognize that an adaptive algorithm inserted just about now might improve matters for both Nature and ourselves, and brainpower, not musclepower, is what's going to give us our only real chance. It's the only tool man has ever had that allowed it to rise above the other animals, and we dare not at this point suggest that it is of no consequence. There may come a day when indeed we can't save everyone. But when that day comes, what we'll probably need is brains, not muscle. And if that day comes, it will be the rich and the politicians, if that's not redundant, making the choices. So let's hope we keep the world safe for all now to avoid having it safe only for the Dr. Strangelove contingent later.
Obligatory Relevance Comment: What has this to do with medical records being sold? Because they are not being sold to people who are optimizing the outcome of humanity. They are being sold to those who are optimizing their portfolio, a portfolio that is not taxed in proportion to the dissonance between its own size and the good meta-health world. In meta, they are playing out the same survival game in business as we're playing out in the world, with Capitalism substituted for Nature. But they're not equipped to survive the storm if Mankind is wiped out through their follies. So all of these things are interconnected, and Climate Change is, I allege, not as off topic here as one might be tempted to presuppose.
If our society were based on blugeoning each other with clubs, this would be a relevant argument. We'd need the specific quality of physical strength and resilience to survive. But the fact is that people who are "sickly" (to use your word) can make important contributions to society exactly because the aspects of the person that is required to make those contributions is often unrelated to the health issue they may confront. Look at Stephen Hawking for example. Nothing wrong with his brain, so as long as the essential aspects of his body are functionally maintained, he can continue to make his contribution. And even when reasoning in some sort of cold/mercenary way, the cost of maintaining such a person may be much less than the cost of losing such a person's potential contribution.
Besides, natural selection is intensely focused on the high order bit--whether people survive to breeding age at all. It's not very concerned with selecting for good writers, philosphers, mathematicians, teachers, etc. Nor does it appear to care a whole lot about diseases that come up after breeding age. So the argument about the gene pool being affected by caring for the so-called "sickly" seems bogus given that a lot of people who we care for are older than breeding age and do not, at that point, contribute to the gene pool.
Natural selection isn't creating some noble super-race. It favors the strong, but also the violent and the crafty. It looks only at outcome; it doesn't moralize about tactics. And its measure of outcome seems, by modern theory, limited narrowly to "has offspring ready to play the game anew". That's a possible theory of "good", but not the only possible theory. It seems just a little limiting, in fact. Which is why society tries to circumvent it through conscious thought and group policy, for better or worse.
Finally an iron-clad reason to keep the Republicans from aborting Mars missions...
At least until we find actual life, when I guess they'll stop caring and start suggesting that such life invest in its own individual retirement plan.
The last piece of the puzzle, completing the cycle of virtualization, will be Wii integration, so that we can look back fondly on the days when we were able to go to the store to buy it, pull it off the shelf, and carry it home.
The multiplayer version will allow our visiting friends to be the person at the counter, who talks us into buying. Watch for extra points to be racked up quickly if they get you to buy lots. Perhaps even a commission, paid in wee dollars.
Although I am not an unconditional fan of open source. I think it has its uses and its non uses. But this is one definite use. It might be the case that as a matter of law right now, what you say is true--it's not evidence. But I actually don't see how we'd be hurt by having a law that makes it into evidence by making it, ipso facto, a crime to apply an undocumented patch at all.
If you're going to have voting machines at all, it seems to me they ought to have open source programs running on them. (I personally don't care if the programs are free in cost, though I imagine they could be. I only care that they are inspectable by anyone and that it is fair use to copy them to other machines for the purpose of verifying their correctness and aherence to advertised and required spec.) I don't see that anyone other than someone doing something illegal is served by keeping the source secret. The data, of course, might be protected. But the program should be auditable by anyone.
And in such a world, I don't see any reason whatsoever that it shouldn't be a crime for there to be any software applied where its entire content and purpose were not carefully documented. Then it wouldn't require any proof other than that it happened in order to detect wrongdoing; and it could be further a crime to phony up a faulty description of what the patch does, so that then it could be audited for validity by anyone who wants to.
The weak link in all this seems to be the hardware. It's quite hard to look at a box and know what's going on inside it. It requires specialized skill. It likely wouldn't be hard to make a machine with a dummy system for show that had the right program on it and another shadow machine tucked away that had the wrong program, with just some subtle wire somewhere leading to one doing the real voting. Call it "democracy theater" if you like, borrowing on the "security theater" moniker used a lot in other venues these days. Such a charade could be hard to notice. Which is why I prefer paper balloting. It means the entire process is exposed.
Employing workers abroad often involves flying management and workers back and forth, even for non-manufacturing jobs like the ones near and dear to most Slashdot readers: software.
I wonder if the higher prices of oil will result in better teleconferencing or if there is a cross-over point at which such coordinated management at a distance will be deemed to impractical.
If it were at some point deemed impractical, I wonder whether that would tend to result in jobs staying here or just locating entirely abroad, since shipping end user software is really cost-free and so the "manufacturing" could be anywhere.
... unless the reason those people are overweight is that they are stuck at a desk job making business succeed while the people who are out playing tennis are not.
Yes, some people gain weight because they are loafing, which I think is what is at the heart of this--a belief that it's as simple as couch potatoes or self-discipline. But some people may be working overtime and not have time for a healthy meal, and the success or failure of businesses may depend on those people.
And at this point some people may be innocent victims of genetic predisposition after generations of this. Shall we tax people for bad DNA?
To reduce the whole discussion to "those people cost me money" is naive and self-serving unless there is the data to back up the claim that these are all independent variables, rather than a complex web of things. At the heart of all bad social campaigns is an attempt to depersonalize the data and assume that the Other Side is all of one kind and all trivially characterizable by some stereotype.
Those who have struggled with diet know it is immensely complicated, and in fact one thing that can work against their succeeding is a feeling of lack of self worth, which certainly a tax is going to only make worse.
Also, the other very very subtle thing is that one thing that may make the US fatter is that it's free (in the sense of rights). The more freedoms you give up, the more you can help certain problems. But not all problems. Freedoms turn out to be intrinsically important sometimes.
There was a Garfield comic I read once where he was lazing around and noting that you could be sure Power Steering wasn't invented by an exercise freak.
p.s. In my browser (IE7), the new Slashdot page format for editing posts has me typing this text inside a 1.5 inch by one 1.5 inch square, non-stretching box, so apologies for any typos but I can't be bothered to do serious editing in this tiny thing. I can barely see what I'm writing.
(And for end users, it's a fraud dilemma since there will be potentially a zillion names that look authentic, with no way to know what's really the thing.)
Having open season on TLD's is computationally equivalent to eliminating TLD's and just saying there's only .COM.
It just seems a bad idea.
What's next? TLD regexps? Will ICANN be selling "(^|.*[.])DISNEY($|[.].*)" for a bundled fee?
Actually, I didn't mean credentials have to be of the university kind. In suggesting exploring credentials, I didn't mean to be restrictive at all. Slashdot's karma points and personal-relationships (friends and friends of friends) are not category-specific, so wouldn't do much for science, but are examples of credentials at least. In Blade Runner, Roy Batty visits his creator using his chess expertise as a credential. I meant it in the most general sense, not to presuppose any particular mechanism, only to say that ultimately everyone uses some mechanism and the only question is whether there is institutional help in figuring out what you need to test and what not. After all, the number of testable questions in the world exceeds the time you have to test them all, so even if the answer to P or NP was P, at some point you'd have to just trust again, because there isn't even enough time in the world to run even a polynomial-time query on every factoid you end up trusting in a single day. So the question isn't whether you'll trust, it's only whether when you do trust, you "trust well" or "trust poorly"...
If you go back and re-read my post, you'll see I didn't say what you think you're responding to. I didn't say science wasn't testable, just the opposite. I suggested societal experimentation with coming up with decent credentials, precisely based on the notion that there is some hope of objective truth at that level.
And, in part, my remarks about consistency-checking is exactly what you just said as well. By not trusting someone else's eyes, you're saying to check what they say for consistency.
The place where you probably went astray is when you thought I was talking about doubting science. I was talking about doubting one's senses. That is the fundamental leap of faith. You know what science tells you because you believe your senses. But how do you know your senses tell you anything useful, and that life is not one big VR game, like in the Matrix. You don't, actually. You simply accept it as the rules of the game because there seems no alternative. This is what you would have learned from my Dark Star reference.
But relating it back to Wikipedia, my point was simply by analogy, Wikipedia has a hard time knowing what is true because it only knows what it is told. You can make credentials for it, and then it will know what credentialed people think, but then that's only as good as the credentials and the will of the people with them to get to the truth. Even if the Wikipedia standard were one of truth (and I'd like that better, frankly, imperfect as it might be) rather than a standard of documentability, it would still be wrong sometimes. But not because of some sinister reason--just because there is no objective standard to actually measure truth and to know when it was right and wrong.
So my post wasn't off-topic, it was just marked that way, presumably by someone with credentials. Credentials get you only so far toward the truth. You can credential the credentialers, as happens here with metamoderation. But at some point there's always a leap of faith and you just shrug and say "good enough" and get on with life, even knowing it's imperfect.
As to judging tin whiskers, my point was not that there wasn't truth to be had, it was that for all its faults, the one virtue of Wikipedia's approach is that at least when it's wrong, it's supposed to contain a paper trail that leads you to original source material to learn why someone thought the wrong thing. That's not perfect either. But it is at least a defensible theory. And for more than that, there's the free market and other models.
I took away my own karma bonus on this to save someone the trouble of moderating me down if they continue to erroneously think I'm off topic. I just wanted to respond here and say I don't really have a disagreement with anything you said. I am a strong proponent of the scientific method. I just think you mis-read my intent.
Well, trust is not a binary thing. You can trust someone for one thing and not another. And even if you do trust, you can have trust at a wide variety of different levels.
I'm not a big fan of Wikipedia in some ways either, not so much because it lies, but because it doesn't want the truth. If I know something true, and I'm the only person in the world, it doesn't want it. But if I know something false, and I write it up, then it's referenceable, and it becomes closer to something Wikipedia does want. I can understand both of those at some level, but I think Wikipedia should care a lot more than it does about creating new mechanisms to let in real truth (perhaps creating a mechanism by which individual knowledge can be vetted) and keep out falsehoods (perhaps creating mechanisms for peer review of referenced documents). The fact that it doesn't is, of course, why other competitors have come up. I guess on that point, you have to score one for the marketplace for at least creating the idea and allowing competition to crank out alternatives.
But as to what to trust in Wikipedia, their strength is that the things they say are supposed to be things that can be backed up by reference. Where you see a strong claim and no reference, find a way to flag that fact and maybe the person who put it in will add a reference. Where you see a reference, follow the chain back to the original source. That source may ultimately be believable or not, of course. In some sense, by its choice of paradigm, Wikipedia is just a complicated, statically-enumerated set of search engine results. It gets you started, but it isn't the whole of the thing you want.
If I knew more about phenomenology, I'd probably say that's just the nature of the Universe, and that Wikipedia can no more escape it than anyone can escape the Three Laws of Thermodynamics. That is, no one ever really knows anything about the Universe other than what they're told, and what they can work out in terms of internal consistency checks on what they're told. But all I know of that is what I've seen mentioned in Dark Star. So I'll let you do your own research there. Whether to direct you to Wikipedia or the movie though, to study more of phenomenology... I dunno, that's a hard choice. Probably I'd say just see the movie. It's worth more than the 6.5 stars IMDB gives it. One could just imagine what Bomb #20 might have to say on the matter of Wikipedia...
It's not an error. I don't know if it's done commonly any more, but mister has sometimes been used in military settings as a unisex term. We used to do that sometimes in the ROTC program I was in (a small number of years before that movie was made).
The military sometimes likes people to be interchangeable (which is part of the reason for wearing uniforms and normalizing haircuts), and differences in such forms of address tend to emphasize individuality, so they're inclined to suppress it for reasons of suppressing a sense of identity. That might seem a negative, but by deemphasizing individual differences, so the theory goes, everyone starts at the same place and has the same advancement opportunities. There was a theory then, and I don't know if it's still adhered to (there's presumably been additional statistics taking since then about whether it worked, which maybe someone knows about and can share) that the use of the female honorific would draw attention to the person's gender rather than to their role, and so it was thought better to emphasize role.
The use of "mister" itself in the military is really a role thing. It's not used to address someone's marital status or gender, but rather as a way to refer to someone who has a rank but without reference to their rank. So instead of saying "Corporal Jones" you say "Mr. Jones", to deemphasize rank. If in doing so, you had to bring in gender, the thought was that this would waste some of the value of moving to a rank-neutral term. You'd only be trading a gender-neutral title of rank for a rank-neutral term of gender. By making the term genderless, you increase its utility. Or so the theory went.
In the days where I was in that program, women were just finally being allowed a foothold as equals and my sense at the time (I was quite an advocate of that equality) was that it tended to produce the desired outcome, causing people to be treated more as equals. I can't say as I know for sure what the emotional impact was, but my guess knowing the people involved was that they wouldn't have been in the program in the first place if they were worried about something like that. They were deliberately pushing gender stereotypes just to be allowed into those roles, and a few non-standard word usages weren't going to deter them.
Of course, one of the things that Star Trek has always done is mirror society's progress (but on time delay), so the fact that the term fell out of use in later movies (and maybe in the real world military, too) may well have been a mark of the fact that once the gender barrier was broken, the fact of the gender reference in someone's name was not as big a deal as it had been until then. In that era (1970's), it was a big deal that we had a female brigade commander, just for example. I'd like to think that nowadays, that's just the ordinary course of business--though some of the recent political debate about Hillary as a possible Commander in Chief makes me think maybe public acceptance on this has a ways to go yet. Still, Star Trek and its many good role models give hope.
The key, though, is not just to have people be able to use such alternate media, but to have it be legit as a reference/credential for advancement in schools. Right now, anyone has a voice, but it doesn't mean citeseer accepts them and it doesn't mean professorships are granted based on them.
In some ways I think the open question almost isn't "will schools accept this?" but rather "will traditional schools be able to retain their credibility if they continue to lock out these other media?" Schools in the US continue to get more expensive at a rate that far outpaces the rest of the society's recent cost of living increases. As it becomes more and more expensive, the question becomes whether alternate education isn't just as good on a benefit-per-dollar basis, and whether companies won't start saying "Forget traditional colleges. Just show me you know what you're doing." It may not have happened yet, but at some point it will. And the failure to accept alternate peer review mechanisms as credible may hasten the process, since it will create easy-to-see showcases of alternate knowledge sources and alternate ways of contributing to research and development that is untapped by conventional schools. I don't see how they can ultimately afford not to consider such mechanisms.
Well, in fairness, although I agree that peer review is quite important, what is an open question is whether peer review needs to happen in advance of publication, or if there isn't some way for peer review to happen after-the-fact. This would accomplish three things: (1) it would allow faster access to raw ideas in the marketplace [at the sometimes expense of reliability], (2) it would allow more opportunity for someone who thought something was valuable to make themselves known, since in at least some cases the reason something isn't accepted to a journal may simply be lack of room or lack of interest by a particular target community, and (3) it would allow dynamic assignment and unassignment of support over a long period of time. Sometimes the sense of what's important at the moment is wrong among peers, and this would assure that such things could be later corrected.
I'm not especially advocating after-the-fact peer review. I'm just saying it's not obvious that it can't occur, and it does seem like there are at least some benefits (even along with what I am sure are also new risks that I've not detailed, but that almost certainly involve the need for society to come to grips with the notion that publication does not imply community acceptance, something that may never have been true even in the past but that may be doubly uncertain in a world where there was no barrier to publication).
Then again, the web is available as a form of unfettered publication path to anyone who wants it, so nothing is stopping individuals on a case-by-case basis from publishing without peer review. The sad thing is that there's no way to take a finding that is already out on the web and promote it to journal status in terms of peer acceptance. If anything, I suspect there's some hidden desire, as probably happens in the regular publishing industry too, to punish those who go outside the system for not having followed the rules. And that kind of thing seems a less good side of the peer review process.
Great point, succinctly expressed. I totally agree. This is related to a point I was going to make: It annoys me that one cannot resell bandwidth. The notion that one person having 3 people in the house can grab bandwidth for all of those people at one price, but three people living separately have to buy 3 separate services seems unfair. In practice, it means that lots of people cheat and get away with it, while the people who don't cheat are charged a premium (or, more specifically: several premiums) for operating to the letter of the rule. For quite a while, I've been pushing the need for Universal Business Access, and have only just recently written it up, but it relates to that.
Email has a similar kind of issue, where not paying is more expensive than paying, since we all pay for spam due to email being free, and the cost of that spam is certainly way higher than the cost that email itself would be--except to the spammers.
Can you say what axioms you think are violated by what they did? I see what they did as subsetting, not as changing the rules. I'd like to understand better otherwise.
In general, floating point from the outset doesn't even guarantee x+1-1=x, so it's hard to say that someone trying to make subsets of floating point is ill-advised in doing so. Floating point takes risks and for you to say:
seems a hair's breadth from a religious rather than a computational statement.
I don't personally acknowledge the concept legitimacy of the unadorned word "works". I prefer to use the terminology "works for " and to judge "works" in a specific context. The problem with using it in general is that it presupposes a one-mode-fits-all paradigm, and I don't think that's a good assumption.
I'd rather equip people to think, to use their brains, and that means equipping them to resist phrases like "look, we've worked this out and it works so you don't have to think about it."
One of the most empowering things that I ever learned about floating point was when I saw someone with a great knowledge of floating point say "I don't understand floating point." I suddenly became conscious that one could "not know" for reasons that are wise, not just reasons that are uneducated. I have come to understand that a little healthy fear and respect of what is really nothing more than a calculated gamble goes a long way. I do not understand floating point. I also doubt many do, and especially I doubt that many people who claim to understand it do. I'm sure a few do, but not as many as are intended to be users of it. That's not to say I don't think people should study it--it's to say I think they really need to know a lot to really appreciate its subtleties. And it's also to say that I think teaching them "it works" is not the path to that understanding.
I don't really disagree with your premise of wanting algebra to work out. I just feel like I may disagree with what you may be saying follows from that. But feel free to set me straight.
I don't disagree with this kind of statement you made though:
Absolutely I think people should be educated in this kind of subtlety, even if they are only going to use such algorithms, so they will be good shoppers.
I appreciate seeing this cross-reference--thanks for posting it.
However, I do disagree with your conclusion. The regions of non-conformance that Excel cites are issues that are commonly discussed and sometimes actively disputed, and in any case that often require very skilled understanding to navigate. Although I've not been in on any internal discussions at Microsoft about how they organize their product line and who they intend to sell to, I feel confident in saying that Excel's target customer is not typifically a physics student. This is a product for business, and the mathematics of binary-coded decimal (BCD) seems more appropriate to its intended end, even if they don't actually use BCD in practice, probably for efficiency or convenience. But the banking community is not wrong in preferring to do things in variance from IEEE's floating point standard either.
Note that Microsoft has not assigned an incompatible meaning to the situations on which they don't conform (although it might not be unreasonable if they did), it has merely said that it doesn't support some cases. I actually think that even the brief passage you cite offers credible justifications for their decisions. (I'm frankly surprised that the things they do are classified as non-conformances. There are other standards where the decision to simply not implement part of a standard and to signal an error in that case is considered a subset and therefore to be strictly conforming. Maybe there is some subtle way they deviate, or maybe they're just being careful in classifying what they did as non-conforming when it's really a judgment call so they aren't accused of hiding the issue.)
But back to the point of this thread, there are two take-home points:
(1) Excel does not seem to me an appropriate first choice as a tool for doing math or science, and especially as a tool for learning about that. A program in those areas should pick one or more tools that accommodate the notation, computational power, library base, and other factors required to support serious computations.
(2) Students of math and science should be taught not that there is one particular way that numerical computation is done, but rather should be taught why fixed point / BCD is different than floating point / IEEE 754 so that they can make informed decisions about which tools suffice in which circumstances. For that matter, I'd also like them to understand the difference between numeric computation and symbolic computation, and why each of those has its place. School ought not be about dispensing dogma. It ought to be about preparing people to be good decision makers as circumstances change.
A solid understanding of the nature of floating point numbers wouldn't hurt either. For example, something like David Goldberg's What Every Computer Scientist Should Know About Floating-Point Arithmetic.
As to the language not being important, I don't know that that's entirely true. Each language offers a choice of types and some sets of choices are more instructive than others. I personally think Lisp or Scheme are good teaching candidates because they offer arbitrary precision integers and rational numbers in addition to basic floating point number types so that it's easy to see side-by-side the trade-offs being made between correctness on one hand and space/speed on the other hand that go along with choices in this regard.
Plus, if you go the Scheme route, you get teaching materials focusing on good abstraction like Sussman and Wisdom's Structure and Interpretation of Classical Mechanics , the proper companion to Abelson and Sussman's popular CS text Structure and Interpretation of Computer Programs !
<tongue-in-cheek>
Nope, you've pretty much enumerated the entire space of things anyone would want to make out of software and certainly that they might have wanted to receive money for. Moreover, now that you've prodded me to think on it further, I realize there's no chance that anyone would ever have wanted to make software of some other kind than you've enumerated and yet has then declined to make such software because of the likelihood that an immediate free-software knockoff made by a free software "sniper" might undercut their investment. So, upon reflection, I stand corrected. Sorry for any confusion.
</tongue-in-cheek>
By the way, the point of my original post was about Google, and my mention of free software was only incidental. I've replied here in fun so as not to ignore your post, but I plan not to carry this further and have declined my otherwise-automatic karma bonus on this post as a self-imposed penalty for drifting even this far off topic.
I voluntarily discounted my karma bonus to make this reply, which is really just a courtesy response and won't advance the debate.
You're right that the effect is not total. However, some free software projects have killed some proprietary software efforts. It's a subjective matter to decide how significant that effect is--I happen to rate the negatives more highly than some other people.
Anyway, my intent was not to open the whole free software debate, it was merely to allude to it, so I'll agree to disagree with you on this point rather than re-do the entire debate here. I wasn't intending to troll about free software, only to make an allusion about free software to those who, like me, believe that effect to be significant. If you're not one of them, and you think I'm nuts for thinking as I do, I guess I can live with that for today. We'll debate the issue of free software another time in another forum in a more serious way, I'm quite sure.
Yes and no. After all, I did say "If Google becomes the standard of mail," so that's of course a hypothetical. I didn't say it had happened. I was hypothesizing about an effect that could happen by applying observations about a trend in another area. I hardly think it's jumping the gun to speculate. All speculation jumps the gun. That's the point of speculation, to be ahead of the gun.
And the mistake you are making is that you think that things that competition does not value are not of value. Competition is not sentient and functions only when proper assignments of dollar value have been created. It is possible to badly assign dollar values to things and in so doing to create situations in which competition will kill things of value as, for example, it might be argued is occurring with things like global warming, where a competitive market is seeking various earth-unfriendly situations because the dollar value of having a world that continues to function correctly has not been appropriately assigned. So please don't imply that the God of Competition is all-knowing about what is good and right and proper in the world without adequately investigating the complete assignment of human value to dollar value.
I'll take a neutral point: Let's say we believe that there is intrinsic human decency value to health care, and yet there are companies in the world who don't pay it. What we will find (and are finding) is that the market will seek supply from those who are in the market at the least cost. In some cases, we can suppose, there's the theoretical possibility that these remote locations to which jobs and businesses are moving offer equivalent health care and other standard of living to what the US offers, and perhaps in those cases what it's saying is that the US simply charges more than it needs to in order to provide these things. But certainly it is the case that in some cases, one reason the cost is lower in those other places is that there is not health care (or not at the same level) or there is not environmental regulation (or not at the same level). Competition will squeeze that out as a luxury because the person buying the sneakers or the iPod or the toaster or whatever doesn't think they're buying someone health care or buying the world the right to survive climate change--they just selfishly want cheaper products without regard to any of that.
And so if I own a company where I would like to offer better health care to my employees, I may not be able to afford to because my competition does not and I cannot compete without offering the lowest of what anyone else does or that cost of my "indulging" my employees with good health care will show through to my product. But that is not to say that that health care had no value. The problem is that in a layered product world, value occurs at different levels and value at one level is not always tolerated at another, so that platitudes like your (repeated for emphasis)
do not hold up. The fact is that if I have a company where my employees really will be "happier" using a commercial piece of software internally, but if it won't crank out a better end-product, the competitive marketplace rates the value of the employee happiness at zero. And so you're quite right that the market will crank that out.
But what you're wrong about is that there is no loss in the process. Because in the end, taken to its extreme, competition is capable of cranking out all fun and happiness from all the world in the name of serving people better. But if no one is having any fun or happiness, to what end is that?
I think we should build soci
One of the things I don't like about free software is that it basically pays for itself off the profits of an unrelated industry, eliminating competition in an otherwise viable industry because someone can afford to offer the service for free as a loss leader to other business.
A thing that is especially troublesome is that not only does it basically make it so that no one can afford to be in the business area (software development for money) competing with the free thing (software given away for nothing), but also no one can afford not to use the free thing because the cost of the luxury of buying an alternative brand will be exposed by the market as superfluous if passed along to end users.
It seems to me that if this becomes a trend, it will be the effective continuation of that paradigm shift by Google into another area, and that the logical continuation of this, by analogy, would be that not only can no one afford to compete with Google and other agencies giving away free mail but no one will be able to afford not to use Google's mail.
That would be sad if it turns out that there are reasons why using Google's mail is not a good idea... such as, for example, concerns about privacy.
If Google becomes the standard of mail, the problem is that it can afford to add incidental services in parity with any nuisance it causes, making it impossible for would-be competitors to match on a value-point by value-point basis even if they find a way that should theoretically be able to compete.
Well, if I understand your situation correctly, I suspect the reason your employer can pay you is that the software is not free, and so that's why they have money to pay you. My guess is that you'd notice it if your employer had no money coming in from that software. From your "POV", that could look like unemployment.
If that's all you (and probably others) aspire to in terms of design, that explains why I don't like this paradigm and you don't see the problem. To me, anything that involves adding a widget or module to Apache is not program design, it's program maintenance. To have programming reduced to this is like telling an would-be orchestral composer there's plenty of work available writing advertising jingles.