As I recall, colds and flu mostly evolve in domesticated animals (such as pigs) in the Far East. Humans get them from the animals, and pass them around the world.
This being the case, you wouldn't necessarily expect super-strains to arise in humans, since the natural reservoir of animals won't be getting the drugs. Instead, the diseases would be effectively prevented from making the species-hop.
I have a pretty cool job. I use mathematics and computers to price options. I have a PhD in math, and this job (known as being a "quant") is reasonably common in financial firms of nearly all sizes. It has a lot to do with computer science -- we use many bits of the discipline known as "numerical analysis".
By making a few basic assumptions about the underlying stock price, it is possible to derive an equation (solvable by computer) expressing the option value in terms of a few simple terms:
1. The current stock price
2. Interest rates 3. The option strike 4. Dividends expected (yeah, right) 5. The "volatility"
The volatility is sort of how much the stock might be expected to vary in price.
Depending on the complexity of the option involved, some of this information may be unreliable or even unavailable, or more information might be needed (e.g. if the stock is in a foreign company but you measure wealth in dollars). But the spirit remains the same. For the most basic options, the orginal analysis is due to Black and Scholes, and was worth a Nobel in Economics (which Merton got to share, leading indirectly to his embarassment at LCTM). For a look at the simplest version of all this, check here. A more thorough reference is Hull's book.
Now, unfortunately, some of those "basic" assumptions I talked about are quite clearly false in the case of the most interesting companies issuing options. In particular, you can't sell the stock short, the price is likely subject to jumps (rather than moving continuously), and the volatility is nearly unknowable. So for a start-up, the analysis is useless. But for big companies, it works quite well.
Ultimately, I would say that if you are considering options issued by a market listed company with a reasonably large daily trading volume, you can get valuable information from such analysis. But for a small start up, it's pretty worthless, and you should stick with the scenario analyses advocated by other posts.
This doesn't have much to do with software config, but one thing to keep in mind is that your visual acuity drops with age. I'm not talking about the focus problems that necessitate reading glasses, but rather the effective resolution of your retinas. Take a piece of paper and draw some parallel lines 1mm apart on it. How far away can you distinguish those lines, even with perfect focus? Whatever answer you got, it will decrease with age. The conclusion I draw from this is that a laptop is not a very good solution for older folks. You want a BIG monitor set to 800x600, so as to make those pixels nice and big. I noticed this when I was helping my grandfather surf the web on my mom's laptop. His eyes had a lot of trouble distinguishing a lot of the tiny user interface elements that I take for granted. (Anybody remember the old single-pixel HFS indicator on the Mac Finder?)
This comment is pretty much dead on. I used to teach a lot of college-level math back when I was a grad student -- diff eqns, calc, whatever. Very few of the problems I assigned my students could be done more easily with an HP calculator. On the other hand, for more applied disciplines, like physics, chem, engineering, and so on, you will find an RPN calculator invaluable!
Why RPN? Well, others have addressed this question, too, but from my perspective it does a couple things for you: (1) it is a faster, more compact way of keying in a problem, and (2) once you "get it" you will think about the formulae in more logical ways. For some reason, using RPN on a calculator helps you do calculations in your head better, too. Probably has to do with keeping the distributive law foremost in your mind.
Although a good calculator is invaluable for the sciences outside of math, for math itself you might want to consider something like Mathematica, MatLab, etc. I use these programs professionally now, and they are really excellent. Forgive my pedantry in reminding you to *learn* the material, too. it can be tempting to let Mathematica do your Calc homework for you. That will put you at a terrible disadvantage in later classes. In math, one learns very little from studying for tests, and a lot from making mistakes on homework.
I was on the Broadview teams in the late 80's. Maybe the definition of "team" was a little looser back then. You didn't have to list everybody who worked on the Hunt (I don't know if you do now), so I'm counting everybody who helped out, even if they just spent a few minutes clipping their nails (one item was an ENTIRE CUP of nail clippings). Add in a few parents and siblings, and you get well over a hundered people. We would get about 25 not doing anything else that weekend, and 30-40 going to the judging to help out with the "group" activities.
I was never a road-tripper, but I had gathered from the Maroon article that they had tried to make you guys sleep on the road trip this time. Maybe that was just prophylactic propaganda. BTW in my day, the Web didn't exist, and we had to do research the hard way;-) I remember a question about the Ira J. Whatever memorial walkway and who Ira J. was. We ended up bugging his widow. Doh!
I always considered the sweet talkers the most important part of the team...I'm not good at it, but it's amazing what they can convince somebody to do.
Damn, I'd moderate this post up, if it wan't in reply to one of my own. It's good to hear from a real team member from a modern winning team.
Back in my day, when Broadview and Model UN were winning, we used an Appletalk network and a Filemaker database to keep track. Let's just say I'm impressed that you guys did so well with lower tech, and somewhat surprised that IT has regressed in the affair since 1990. Maybe we had overkill.
Our winning teams had about 125 people helping out in one way or another, about 25 not doing anything else that weekend, and 5 or so going completely without sleep. Some things were different (some incentive to steal things, more dangerous road trips, fatigue-wise, no Olympics, and less nudity). I'm curious, did you not assign a couple skilled sweet-talkers specifically to, well, the sweet-talking? Not my cup of tea, of course. I was usually on the "covert" team doing things that typically involved avoiding security personnel.
Well, as I understand it, plutonium is not dangerous for its radioactivity. The main trouble is that it is chemically poisonous. Like lead, but worse. So while all the idiots go running around worrying about the radioactivity, the real danger remains unacknowledged. IIRC a few pounds could poison the water supply of a city.
The Tayltslin/Caldicott article, BTW, repeats this mistake, in addressing the dangers of plutonium only in terms of radioactivity. In that sense it's every bit as much trash as the paranoid environmentalist crap it's trying to refute.
Incidentally, the students made just a few atoms of plutonium, not enough to be dangerous to anybody.
I went to U of Chicago, and was there when the Scavenger Hunt started. Our team lost badly my first year, took first prize the next two years, and took second my last year. I haven't gone since '91, but I still follow it a little.
As pointed out elsewhere here, the nuclear reactor is entirely believable. I might point out that the chief danger from plutonium in small amounts is not its radioactivity, but its poisonousness. Even 1 gram could kill a heck of a lot of people.
Like the article hints, you don't win these days without setting up a LAN and using some database technology. There are too many items to track otherwise. The big items everyone remembers, but there are lots of little 5- and 10-pointers, like ostrich eggs or whatever, that tend to get forgotten. Read the list, and you'll be amazed, but some teams get almost everything within 72 hours.
Back in my day, we got an airplane, a one-ton animal, a telephone pole, and a marching band. We found a collector of turn of the century train cars from the Chicago Elevated to loan us an El car, but we couldn't get permission from the city to take such a heavy load on the streets.
Teams that really contend for the top prizes are made up of 100 or more people, of whom at least 25 must be willing to dedicate the ENTIRE 72 hour period to collecting and building. Usually, you specialize, putting your smooth talkers onto tasks like wheedling Olympic medalists into loaning their medals, your skilled researchers onto finding the answers to obscure questions like the location and population of Waldo, your exhibitionists in the latex paint-on pants, etc.
I have never done anything more fun in my entire life.
A few years ago, there was an article in The Economist about the economics of the recording industry. Basically, their take on it was that the introduction of CD's had greatly increased the variety of music available. As we all know, even though CD's are cheaper to make than vinyl ever was, you still pay more for your music. Of course, the physical medium costs little in either case: 10% or so of the total consumer price.
The Economist's take on this was that, in effect, the extra cost of CD's was subsidizing a greater variety of music. Getting everyone to pay $15 for that Nirvana CD meant that the industry had a little cash to use on experimenting with . Vinyl had found an equilibrium point with a much more mainstream-oriented music culture. CD's, having an initially higher price in the 80's that was never really reduced, eventually came in far greater variety (they backed this up with stats -- vinyl never had nearly the number of titles that CD's got).
What interests me is, what economic equilibrium will be found by digital distribution? I like a lot of pretty obscure music, so I am a big winner from a system that overcharges the Celine Dion fans in order to take a chance on promoting smaller artists! But Celine Dion doesn't benefit from that -- she would certainly prefer NOT to subsidize them. If she can distribute directly, she can choose not to.
Ideally, those smaller artists will be able to promote themselves, at least enough to catch my attention. I can then pay whatever premium necessary to get their music, and Celine Dion can sell her own music directly to the fans for less. That may also eliminate lots of gross inefficiencies in the sort of indirect subsidies the CD medium has -- record companies are really pretty inefficient at turning big-artist profits into small-artist promotion.
Potentially, what will happen instead is that small interesting artists, deprived of any marketing savvy from a system that can now get away with just squeezing whatever profits it can from the big names, will disappear into obscurity and the noise of the net. That would make me sad.
Is this consistent with common carrier status?
on
ISP Sues Spammer
·
· Score: 1
Although I hate spam as much as anybody, this seems to be the wrong front on which to fight it. I think common carrier status for ISP's is a Good Thing. Suing a subscriber based on the contents of their usage implies an interest in that content. Which an ISP shouldn't have.
Can anybody think of a good phone company analogy?
It looks to me like this will just call attention to the fact that SQL Server is far slower, whatever the factor. Nobody's going to think about the hardware involved. They should wait to take the challenge until they can somehow be _faster_ than Oracle. Heh. If that ever happens.
Well, Leibniz did a better job with (1) notation and (2) promotion. Newton's actual mathematics was better. But Newton was a kook, and didn't want to publish until Leibniz started to publish equivalent stuff, at which point he got really pissed off, published, and claimed priority in a bitter dispute.
I see this whole thing as a winning situation, both for Apple and the computing community. Apple gets the obvious benefits of other smart people examining the code, with a reasonably small risk of their published code being turned into an Apple-killer of some kind. The community gets another OS for which they have source -- a big help not just for writing OS's, but also for writing apps. I can't tell you how many times we wished we could dive into the OpenStep source while debugging our apps.
This latter point (which I am not the first slashdotter to bring up) could be one of the strongest. An OS will not be successful in the broad market without good apps, and the ability to make more. The source gives developers a lot of confidence in stability and behavior of the underlying OS -- good for new development. The OpenStep platform was always a real pleasure to develop for, because of the superb tools and interface. (Mac OS 8, Win and Linux/X are relative bears by comparison, at least for GUI apps). This move strengthens a strong hand.
Apple's GUI is consistent and beautiful. This is something nearly every other *nix lacks, and is an especial weakness of Linux/X. You can make Linux/X look pretty decent. But consistent? Forget it. Anyone reading this can probably deal fine with inconsistent app interfaces. End users can't and won't.
So a good GUI in turn attracts end users, who (Microsoft's case shows) are very symbiotic with app developers. Is this enough for Apple to start to really challenge Microsoft? I don't think so, but it helps.
OS X is a good, stable implementation of BSD and Mach. NeXT was not just sitting on their thumbs, waiting for the money to roll in -- they did real work on the OS. (Enough to take over OS development at Apple when Apple bought them) There's a lot of work they did for OpenStep that the community hasn't yet gotten 'round to doing for open source OS's. If I had to choose an OS for, say, a dedicated screen trading app (I work in the finance industry) today, I would choose OS X. For a webserver, probably FreeBSD or Linux.
About the build environment: I would expect that most folk will have to do a lot o' tweaking to get any of this newly released code to build. Obj C isn't the issue, since gcc handles it fine (/bin/cc on Apple/NeXT has always been based on it). It's all the other cruft. Anything but an OS X machine is going to have something in the "wrong" place.
Although it would probably be pretty easy for Apple to support lots of hardware architectures (OpenStep ran on Intel, Motorola, HP-PA, SPARC, and [unofficially] PPC) I think other people's arguments about Apple needing to make their money off the hardware are valid. They won't keep people from recompiling the underlying OS on other architectures, perhaps, but it will be a long time before you see the GUI on anything but their own hardware. That's OK with me. Last week, they had open sourced nothing from OS X. This week, we've got lots of it, whatever the quibbles about licensing. The world is a better place.
OK, so who loses from this? Nobody but Microsoft. And maybe some other commercial Unix outfits, like Sun. Who benefits? Well, I count
Apple developers: obvious reasons.
Apple: more developers, better apps, maybe some free OS work.
Developers of other OS's: examples to peruse, ideas to copy, a benchmark.
Slashdot Mirror
As I recall, colds and flu mostly evolve in domesticated animals (such as pigs) in the Far East. Humans get them from the animals, and pass them around the world.
This being the case, you wouldn't necessarily expect super-strains to arise in humans, since the natural reservoir of animals won't be getting the drugs. Instead, the diseases would be effectively prevented from making the species-hop.
By making a few basic assumptions about the underlying stock price, it is possible to derive an equation (solvable by computer) expressing the option value in terms of a few simple terms: The volatility is sort of how much the stock might be expected to vary in price.
Depending on the complexity of the option involved, some of this information may be unreliable or even unavailable, or more information might be needed (e.g. if the stock is in a foreign company but you measure wealth in dollars). But the spirit remains the same. For the most basic options, the orginal analysis is due to Black and Scholes, and was worth a Nobel in Economics (which Merton got to share, leading indirectly to his embarassment at LCTM). For a look at the simplest version of all this, check here. A more thorough reference is Hull's book.
Now, unfortunately, some of those "basic" assumptions I talked about are quite clearly false in the case of the most interesting companies issuing options. In particular, you can't sell the stock short, the price is likely subject to jumps (rather than moving continuously), and the volatility is nearly unknowable. So for a start-up, the analysis is useless. But for big companies, it works quite well.
Ultimately, I would say that if you are considering options issued by a market listed company with a reasonably large daily trading volume, you can get valuable information from such analysis. But for a small start up, it's pretty worthless, and you should stick with the scenario analyses advocated by other posts.
This doesn't have much to do with software config, but
one thing to keep in mind is that your visual acuity drops with
age. I'm not talking about the focus problems that necessitate reading glasses,
but rather the effective resolution of your retinas. Take a piece of paper and draw some parallel lines 1mm apart on it. How far away can you distinguish
those lines, even with perfect focus? Whatever answer you got, it will decrease with age.
The conclusion I draw from this is that a laptop is not a very good solution
for older folks. You want a BIG monitor set to 800x600, so as to make those pixels nice and big.
I noticed this when I was helping my grandfather surf the web on my mom's laptop. His eyes had a lot of trouble distinguishing a lot of the tiny user interface elements that I take for granted. (Anybody remember the old single-pixel HFS indicator on the Mac Finder?)
Why RPN? Well, others have addressed this question, too, but from my perspective it does a couple things for you: (1) it is a faster, more compact way of keying in a problem, and (2) once you "get it" you will think about the formulae in more logical ways. For some reason, using RPN on a calculator helps you do calculations in your head better, too. Probably has to do with keeping the distributive law foremost in your mind.
Although a good calculator is invaluable for the sciences outside of math, for math itself you might want to consider something like Mathematica, MatLab, etc. I use these programs professionally now, and they are really excellent. Forgive my pedantry in reminding you to *learn* the material, too. it can be tempting to let Mathematica do your Calc homework for you. That will put you at a terrible disadvantage in later classes. In math, one learns very little from studying for tests, and a lot from making mistakes on homework.
I was never a road-tripper, but I had gathered from the Maroon article that they had tried to make you guys sleep on the road trip this time. Maybe that was just prophylactic propaganda. BTW in my day, the Web didn't exist, and we had to do research the hard way
I always considered the sweet talkers the most important part of the team...I'm not good at it, but it's amazing what they can convince somebody to do.
It's good to hear from a real team member from a modern winning team.
Back in my day, when Broadview and Model UN were winning, we used an Appletalk network and a Filemaker database to keep track. Let's just say I'm impressed that you guys did so well with lower tech, and somewhat surprised that IT has regressed in the affair since 1990. Maybe we had overkill.
Our winning teams had about 125 people helping out in one way or another, about 25 not doing anything else that weekend, and 5 or so going completely without sleep. Some things were different (some incentive to steal things, more dangerous road trips, fatigue-wise, no Olympics, and less nudity). I'm curious, did you not assign a couple skilled sweet-talkers specifically to, well, the sweet-talking? Not my cup of tea, of course. I was usually on the "covert" team doing things that typically involved avoiding security personnel.
Congrats on your work!
The Tayltslin/Caldicott article, BTW, repeats this mistake, in addressing the dangers of plutonium only in terms of radioactivity. In that sense it's every bit as much trash as the paranoid environmentalist crap it's trying to refute.
Incidentally, the students made just a few atoms of plutonium, not enough to be dangerous to anybody.
As pointed out elsewhere here, the nuclear reactor is entirely believable. I might point out that the chief danger from plutonium in small amounts is not its radioactivity, but its poisonousness. Even 1 gram could kill a heck of a lot of people.
Like the article hints, you don't win these days without setting up a LAN and using some database technology. There are too many items to track otherwise. The big items everyone remembers, but there are lots of little 5- and 10-pointers, like ostrich eggs or whatever, that tend to get forgotten. Read the list, and you'll be amazed, but some teams get almost everything within 72 hours.
Back in my day, we got an airplane, a one-ton animal, a telephone pole, and a marching band. We found a collector of turn of the century train cars from the Chicago Elevated to loan us an El car, but we couldn't get permission from the city to take such a heavy load on the streets.
Teams that really contend for the top prizes are made up of 100 or more people, of whom at least 25 must be willing to dedicate the ENTIRE 72 hour period to collecting and building. Usually, you specialize, putting your smooth talkers onto tasks like wheedling Olympic medalists into loaning their medals, your skilled researchers onto finding the answers to obscure questions like the location and population of Waldo, your exhibitionists in the latex paint-on pants, etc.
I have never done anything more fun in my entire life.
A few years ago, there was an article in The Economist about the economics of the recording industry. Basically, their take on it was that the introduction of CD's had greatly increased the variety of music available. As we all know, even though CD's are cheaper to make than vinyl ever was, you still pay more for your music. Of course, the physical medium costs little in either case: 10% or so of the total consumer price.
The Economist's take on this was that, in effect, the extra cost of CD's was subsidizing a greater variety of music. Getting everyone to pay $15 for that Nirvana CD meant that the industry had a little cash to use on experimenting with . Vinyl had found an equilibrium point with a much more mainstream-oriented music culture. CD's, having an initially higher price in the 80's that was never really reduced, eventually came in far greater variety (they backed this up with stats -- vinyl never had nearly the number of titles that CD's got).
What interests me is, what economic equilibrium will be found by digital distribution? I like a lot of pretty obscure music, so I am a big winner from a system that overcharges the Celine Dion fans in order to take a chance on promoting smaller artists! But Celine Dion doesn't benefit from that -- she would certainly prefer NOT to subsidize them. If she can distribute directly, she can choose not to.
Ideally, those smaller artists will be able to promote themselves, at least enough to catch my attention. I can then pay whatever premium necessary to get their music, and Celine Dion can sell her own music directly to the fans for less. That may also eliminate lots of gross inefficiencies in the sort of indirect subsidies the CD medium has -- record companies are really pretty inefficient at turning big-artist profits into small-artist promotion.
Potentially, what will happen instead is that small interesting artists, deprived of any marketing savvy from a system that can now get away with just squeezing whatever profits it can from the big names, will disappear into obscurity and the noise of the net. That would make me sad.
Can anybody think of a good phone company analogy?
You assembly coders make me laugh. When you want *real* speed, it's time to burn a custom chip. Slashdot could run on my pager!
Forget the PA-RISC machines, HP needs to ship these guys some Pilots!
It looks to me like this will just call attention to the fact that SQL Server is far slower, whatever the factor. Nobody's going to think about the hardware involved. They should wait to take the challenge until they can somehow be _faster_ than Oracle. Heh. If that ever happens.
I'm seeing some parallels here....
This latter point (which I am not the first slashdotter to bring up) could be one of the strongest. An OS will not be successful in the broad market without good apps, and the ability to make more. The source gives developers a lot of confidence in stability and behavior of the underlying OS -- good for new development. The OpenStep platform was always a real pleasure to develop for, because of the superb tools and interface. (Mac OS 8, Win and Linux/X are relative bears by comparison, at least for GUI apps). This move strengthens a strong hand.
Apple's GUI is consistent and beautiful. This is something nearly every other *nix lacks, and is an especial weakness of Linux/X. You can make Linux/X look pretty decent. But consistent? Forget it. Anyone reading this can probably deal fine with inconsistent app interfaces. End users can't and won't.
So a good GUI in turn attracts end users, who (Microsoft's case shows) are very symbiotic with app developers. Is this enough for Apple to start to really challenge Microsoft? I don't think so, but it helps.
OS X is a good, stable implementation of BSD and Mach. NeXT was not just sitting on their thumbs, waiting for the money to roll in -- they did real work on the OS. (Enough to take over OS development at Apple when Apple bought them) There's a lot of work they did for OpenStep that the community hasn't yet gotten 'round to doing for open source OS's. If I had to choose an OS for, say, a dedicated screen trading app (I work in the finance industry) today, I would choose OS X. For a webserver, probably FreeBSD or Linux.
About the build environment: I would expect that most folk will have to do a lot o' tweaking to get any of this newly released code to build. Obj C isn't the issue, since gcc handles it fine (/bin/cc on Apple/NeXT has always been based on it). It's all the other cruft. Anything but an OS X machine is going to have something in the "wrong" place.
Although it would probably be pretty easy for Apple to support lots of hardware architectures (OpenStep ran on Intel, Motorola, HP-PA, SPARC, and [unofficially] PPC) I think other people's arguments about Apple needing to make their money off the hardware are valid. They won't keep people from recompiling the underlying OS on other architectures, perhaps, but it will be a long time before you see the GUI on anything but their own hardware. That's OK with me. Last week, they had open sourced nothing from OS X. This week, we've got lots of it, whatever the quibbles about licensing. The world is a better place.
OK, so who loses from this? Nobody but Microsoft. And maybe some other commercial Unix outfits, like Sun. Who benefits? Well, I count
Apple developers: obvious reasons.
Apple: more developers, better apps, maybe some free OS work.
Developers of other OS's: examples to peruse, ideas to copy, a benchmark.
- Brian K. Boonstra, Ph.D.