...but if N-Korea has nukes, it will not be invaded by the US, ergo no US troops next door to China.
Which just furthers my point. If the Chinese goverment really worries about an invasion of U.S. soliders, they are seriously deluded.
Heck, in WWII, the planed invasion of Japan, a beaten, firebombed nation (although not nuked yet) with less than a twentieth the population of China now, was estimated to cause a MILLION or more allied casualties. That was one of the biggest arguments that was given to nuke Japan. Even if we take nukes off the table, a plain old-invasion of China would make WWII casualty figures look like chump change.
I see why the North Koreans would really like nukes, but it still doesn't make sense to me from China's point of view. I would place stronger odds on the Koreas' uniting sometime in the next 50 years than not. Language, culture and blood are much stronger long-term ties than country. Then, they'll have a unified, nuclear armed Korea right on their border. Which will likely encourage a nuclear armed Japan (which they might do much sooner as a result of North Korea's tests anyway). Is encouraging a nuclear arms race on your borders really a good idea? I can't see how having a few more heavily armed, somewhat paranoid neighbors benefits you.
why is the USA even bothering to defend North Korea? Since half of South Korea hates the USA and the other half riots at the prospect of having more open trade on their side, one has to ask, why is the USA in Korea at all? Right away, if North Korea and South Korea destroyed each other, it would be better for American car companies. We wouldn't have as many Hyundais and Kias running around the USA.
Perhaps statements like this are part of the reason why some South Koreans 'hate' the US?
I agree with your post about the U.S. needing to get out of the world-running business. But - your casual statement regarding the extermination of 70+ million people only in terms of positive impact to U.S. car companies is not helpful to your argument, since people may assume you are a ghoul, which means they won't take your otherwise good idea as seriously as they should.
When China finally wakes up and realizes that having a somewhat unstable next door neighbor armed with nukes is a bad idea, this sort of thing will stop - North Korea survives only because China keeps giving them tons of aid.
Perhaps the North Koreans are interested in China's continued aid supplies over the long term? As in, after they get a credible, deliverable weapon, 'If you stop the gravy train, we take out Hong Kong, even if we're glass 8 minutes later. That whole "we don't like the west" thing was just so you would let us build nukes.'
I really don't get China's motivations. Once the nuclear genie is out, they won't be able to stuff it back in. It's like the U.S.A. helping Haiti to get nukes because they don't like Cuba. Does it not occur to the Chinese govt. that once North Korea has a real nuclear capability, they could aim it anywhere they so wished?
There are only weather forecasters. Climate science is not science, that would require testability and we don't have anything to test with. When weather forecasters start chucking millions of tonnes of sulphate aerosols (or whatever) into the atmosphere, then it will be science.
That implies that any methodology that can not perform artifical experiments is not a science. That would exclude large swaths of biology, specifically those involving evolution, from the title 'science'. That's a bit extreme, don't you think? As long as a methodology makes testable predictions that can and will be used to evaluate the usefulness of the model that made those predictions, it seems like it deserves the title 'science'.
For instance, the first verification of General Relativity was done by measuring the bending of distant starlight when it skimmed the surface of the sun. This could only be done during a solar eclipse. We knew of no way to arrange circumstances into a 'test' that could be done in a lab, at will, (at the time) so we had to wait for nature to provide one (a solar eclipse) in 1922. Does the fact that nature arranged the circumstances detract from the validity of the test?
If a climate model makes specific predictions, that 20 years later become true, it increases the probability that the model is an accurate predictor. Because we can't manipulate the atmosphere as you suggest, unfortunately, we have to do it the slow way - it will take decades before we really know whether our models are accurate. But to suggest that climatology is not science because it can't be experimentally verified in a lab is a bit extreme.
Actually, I realized you probably were questioning the first part of parents' assertions more than the last:
If at any point in recorded history, you proposed that the earth was flat, the overwhelming majority of people thought you were a nutjob.
Yeah, that's a bit more questionable, isn't it? There has been casually observable evidence for the Earth's roundness in certain places (shorelines, where one has an opportunity to see a ship vanish over the horizion hull first, rather than just get too small to see), and if one knows the mechanism causing a lunar eclipse, the always circular shadow of the earth strongly suggests it's spherical.
OTOH, the geometry of the lunar eclipses was not well understood by most 'ordinary' people (even today), many didn't live on seashores, and historically, most people were illiterate and unable to read descriptions about these things. So, it is pretty unlikely 'the overwhelming majority of people' would think you were a nutter for proposing the earth was flat.
Perhaps the parent exaggerated a bit, and meant that throughout recorded history, the idea that the earth was round instead of flat was known by educated people of the times, and accepted as true by a significant proportion.
Please provide reputable, verifiable evidence of the information in your post, preferably from multiple sources.
You can do your own googling for that. But to give you some confidence that the parent is not making things up, consider what the casual name for Native Americans is - 'Indians'.
That's because Columbus believed, until his dying day, that he had landed in a group of islands off of Asia/India, not discovered a new continent - hence, the term 'Indians' for Native Americans. Columbus was not a fool, and he knew how far he had sailed on his journeys. His interpertation though, was that he had sailed across a much smaller ocean, on a much smaller planet (I think he estimated a circumference of 18,000 miles vs. the true figure of 25,000 miles), and hit the other side of Eurasia.
anachronism |É(TM)ËnakrÉ(TM)ËOEnizÉ(TM)m|
noun
a thing belonging or appropriate to a period other than that in which it exists, esp. a thing that is conspicuously old-fashioned.
While it's always easy to bash politicians for doing something meaningless, to some extent that is inevitable in a system where powers of legislation are separated from powers of execution. Individual Congressfolk are not the ones hiring and firing the school chancellors, teachers, admins, etc., nor do they set the property tax rates (the most common way local school systems are funded in the USA) or school budgets. The effects they can have on education are at the broadest level only, like federal budget suppliments, standards setttings before such suppliments can be recieved (see the NCLB...), etc.
The reality is in the USA, the primary education system is a highly local affair, with standards set by local sensibilities. This is one reason you keep seeing movements to push creation 'science' (it doesn't deserve the word unless surrounded in scare quotes) pass or nearly pass school boards in Pennsylvania, Oklahoma, Texas, and Kansas, to name a few. In the USA as a whole, about a third of the population rejects evolution outright (see National Geographic or a gazillion other polls). The evolution deniers are not evenly distributed across the USA, but more clumped. Certain congressional districts, I'm guessing, have half of their voters (or more) who would vote against a congressperson who declared that they would vote for a bill to promote science standards if the mandatory teaching of evolution was a part of those standards. With a constituency like that, the odds of passing (at the federal level) a significant, science-standards based education bill are slim. Remember, 1/3 of the USA population rejects evolution outright. That's a lot of people. So, instead, we get National Pi Day.
So, look on the bright side. Now that we have a National Pi Day, maybe we don't have to worry about attempts to legislatively redefine pi any more....
It's true... one of my friends is an IP attorney for one of the big labels. That didn't stop him from "borrowing" my hard drive full of music. A friend of my wife works for a different big media company, again as an attorney. Her husband actually uses her company laptop for p2p, to the constant admonishment of his wife.
It's about money, not morals. The two IP lawyers that I speak about are not monsters and they like p2p just as much as the rest of us do. But they are lawyers and will do whatever is in the best interest of their clients/employers, just as ethics dictates.
[double-take]
This word you use, ethics. I do not think it means what you think it means.
I mean, in my silly little world, 'ethics', among other things, means doing unto others as you would have them do unto you, and I take it your friend doesn't want to be sued into financial oblivion. So let's forget me and my silly definition. The dictionary definition of ethics is:
Moral principles that govern a person's or group's behavior.
You got it right when you said it's about money, not morals. I just question how you can claim they have ethics without morals, since one seems to require the other. Call a spade a spade - when your job entails bankrupting random file-sharers while you yourself engage in the same behavior, that sounds like a casual abuse of power and a willful blindness to the hypocracy of one's own actions at best. Either that, or a 'morality' where your friend believes that laws are only meant to be enforced against other people, or the fact that you were ordered to do it by your boss excuses you of any unethical consequences. Sounds like qualities human monsters in the past have had in spades, even if they are great dinner company and will water your plants for you while you are on vacation.
I don't at all mean this as a dig at you. I'm simply trying to point out that people are wholes - I don't think there is such a thing as an 'ethics free' zone while at work, and I would have a hard time being friends with someone like that.
Have you looked at one? I mean, have you *even tried* reading from one? A lot of people have bad things to say about the Kindle. But the vast majority of those people have never even seen one with their own eyes. Let alone picked one up and seriously used it.
Yes, for about half an hour I had a chance to play around with and read from one at a friend's home, who happens to be a book editor and a Kindle enthusiast. It's easier to read from than a computer screen. It's worse to read from than the printed page. As I implied with my 60 Hz comment, since 60 Hz CRTs drive some people batty and others not at all, it bothers me enough that I find it unpleasant. For some people, like you, it may be sufficient. For her? She's still on the fence, and she reads books for a living.
Allows me to put any files, from any creators, in any format I want to read on it, WITHOUT MANDATORY DRM. Pdfs, rtfs, plain text, Word docs, whatever.
Yeah, it does that. Next hand waving excuse...
Wrong. The only support that is built in for non-DRM content is mobipocket books, and plain text. No pdfs, Word docs, and rtf - although an 'experimental' converter is available for pdfs, and if you feel like e-mailing your docs to Amazon, for Word and HTML docs. For me, that's too much of a PITA. I like dragging the files I want to read on it over, and not have to convert them. I guess this issue is not a problem for you. It is for me.
Allows me to have access to the underlying OS as I see fit.
Oh, I see the problem now. You don't want an e-book reader. You want a toy. A gadget you can play with. Well they sell the development kit for 3k if you want.
I use my Kindle as a reader. I put books on it, then, I read them. Ta da. It's made a great replacement for paper books.
Yes. How astute of you, reading what I wrote and noticing I want what I want! I want, for me, not anyone else, including you, a general, multi-function device, that can be used to do many things. I don't feel like carrying around an e-book reader, a notebook computer, a phone, an address book, a calendar, and lord knows what else. The fewer things I have to carry around in my daily life, the better.
One reason I generally dislike zealots of all stripes is they instantly assume what's good for them is good for every one else. The Kindle is good enough, for you. I don't think it's good enough for me. I didn't predict its demise in the marketplace, I didn't say Kindle buyers were stupid, I simply gave a list of reasons why I won't be buying one. I even said 'Yet' in the subject of my post, and said 'git off my lawn' to give the hint to the clueful that I realize I may wind up being behind the adoption curve because of my personal preferences.
What do you contribute to the discussion? You write back a nasty response with factual errors and incorrect assumptions, denigrating my reasons because they are not yours. Kind of a waste, don't you think? Unless you get off on being contemptuous, sarcastic, and wrong...
Who thinks we'll see a Microsoft Linux distro in the future?
It would be a very smart move for Microsoft; they could offload a tremendous amount of development pain, use their influence where it would do some good (make the hardware vendors cough up the specs), etc.
I don't think it will happen, though, because Microsoft has a deep, deep infection of ego. Large corporations, particularly successful ones, tend to get this at some point. Such an infection may only kill off your innovation and usability - monopoly and political power can prop up your balance sheets for a very long time. But it can blind you to real, oncoming competitive threats, because the idea that someone else can be better than you is organizationally inconcievable. It's rarely curable, although a near-death experience (assuming it doesn't lead to a full-dead experience!) can do the trick.
Free Software products will inevitably prevail over proprietary options because it is the moral and intelligent decision.
Man, pass me a glass of whatever YOU are drinking...
I agree with the sentiment of what you wrote, but I work for the airlines. Intelligence and morality in decision-making here is as common as snowstorms in Aruba. I suspect it's not that common in the rest of the world either, judging from the available evidence....
I'd really like for you to be right, but... just because an idea or concept is the right and intelligent thing has very little bearing on its ultimate success. It also heavily depends on the tenacity and charisma of its proponents, the weakness and misteps of its opponents, and a healtly dose of luck. And whenever I hear someone declare with great confidence that an idea will win because of its inherent rightness... well, good luck with that.
Why should I buy an e-book reader? Show me one with the following abilities:
1) Has 600 DPI+, not 167 DPI. 167 DPI is about as pleasant as a 60 Hz CRT.
2) Allows me to put any files, from any creators, in any format I want to read on it, WITHOUT MANDATORY DRM. Pdfs, rtfs, plain text, Word docs, whatever.
3) Allows me to have access to the underlying OS as I see fit.
The ONLY significant thing the Kindle has over a netbook is a (slightly) more pleasant screen - e-ink vs. LCD and 167 DPI vs. 100 DPI. They both still suck for long extended reads, but I can access the internet, read any docs I feel like, do some coding or play simple games on a netbook. Perhaps someone should stick an e-ink screen on a netbook?
In the meanwhile, I'm keeping my books. And git off my lawn.
It IS interesting... and there are many areas of progress where a little applied automation can go quite far. I'm just trying to point out that the actual mechanical acts of taking off from an arbitrary airport, climbing, crusing, approaching, and landing at an arbitrary airport, are NOT the hardest problems - they are pretty close to solved today, in the sense that we know how to build and deploy a system that would do these things...
The harder problem is of the kind that bedevils game designers and other programmers who are trying to implement AI. We still don't have a good handle on how to give computers 'good judgement', or handle risk analysis for previously unknown, or highly complex situations. To date, the most successful UAV's have human pilots to make the judgement calls.
I'm actually more interested in seeing fully automated driving than flying. When someone deploys a car where you can get in, punch in a destination, and have it safely transport you there, I'll know flying is about to go the same way....
Of course, this doesn't even touch upon when it will be cost-effective to go that way!
Well, by that definition, there is not a 'real' autopilot in the world. Did you even bother to READ the wikipedia article you posted? Let me summarize the stats for you:
All I can conclude from these numbers, spread over 50 years, from piston to turboprop to turbofan powered airplanes, ditched in winter, summer, rivers, oceans, bays, and every other concevable condition, is that a) Ditchings are very rare, and b) It's more likely than not some people are going to croak.
In addition, the survival rates say NOTHING about how difficult it is to pull off a ditching, particularly for this theoretical 'real' autopilot that doesn't exist yet. Every ditching here was done by human pilots, not autopilots.
Think of it this way. How do you program an autopilot to ditch? When is it a good idea, and when is it not? For instance, PanAm 943 lost two of its four engines. But rather than ditching immediately, it circled around the USCGC Pontchartrain until dawn, THEN ditched - plucking people out of the ocean in daylight is far easier than at night. Would you have been able to program an autopilot to make that decision?
Another issue - river landings, which were 3 of the 11 ditchings listed above. There is usually very little margin for error landing on a narrow river. The 95% confidence radius of position (sometimes called ANP, for Actual Navigational Performance) most modern airliners is typically about.3 nautical miles in typical conditions. That is FAR too low to execute a river landing if the river is less than a mile wide! In addition, how do you avoid ships by autopilot?
The realities of ditching are far more complex than you allow for.
One question that comes to my mind: could advanced autopilot tech lead to more ubiquitous personal aircraft?
I don't really know anything about it, but I've always assumed that one of the big hurdles preventing us from having "flying cars" (by which I don't necessarily mean an actual car, but something lots of individuals could buy and fly under casual circumstances) is the difficulty of learning to fly safely. If you could program a destination and have the entire trip flown by an autopilot, from takeoff to landing, would that help the situation?
A little, but not much. The main difficulty in learning to fly safely has less to do with physical skill and far more to do with good judgement. Airplanes still are rather marginal beasts, in the sense that they can easily be flown into situations that are rather hard to get out of. Let's assume that we have invented and installed a magical autopilot system that does what you described. What situations will it help with, and which ones will it not help with?
IT WILL HELP WHEN...
- A pilot gets lost. Hopefully this problem will go away.
- A pilot flies into unexpected IFR (instrument) flying conditions. Press the button, and the airplane magically keeps itself upright and lands them somewhere safe. This is probably the biggest gain in safety right here.
- A pilot 'loses it' - gets airsick, incapacitated, etc.
IT WON'T HELP WHEN...
- The pilot doesn't put enough fuel on for the flying they want to do. This is, sadly, more common than you would think.
- The pilot flies into thunderstorms, freezing rain, etc. Bad weather that can knock a plane from the sky has always existed, and pilots keep flying into it. Our abilities to automatically predict and avoid such weather is still very, very limited.
- The pilot doesn't maintain the aircraft or magic autopilot system, and something breaks at an inoppurtune time.
- The pilot overloads the aircraft with too much stuff, or balances it badly. The predictable (but unknown to the autopilot) lack of performance that will result could be deadly.
- And, of course, engine failures requiring off-airport landings. The recent USAir landing in the Hudson illustrates where a skilled human is most valuable.
What you're referring to is Category III autolanding, which in CAT III C has no decision height but instead the aircraft can land completely on its own (30 m in CAT III A and 15 m in B, IIRC). More landings are done that way than not - and all landings if the weather is bad since autopilots do a much better job than humans.
These statements are not quite correct, and reflect a lack of actual experience with such approaches. Generally, there are both visibility and ceiling requirements set for such approaches. The typical visibilty minima for Cat IIIA is 600 feet/175 m RVR and the typical decision height [DH] is 50 feet above touchdown zone height. Cat IIIC is typically 'zero-zero' - no ceiling or visibility minimums are estabilished. Interestingly, the ability to do Cat IIIC landings revolves around rollout control systems. Most autopilots steer via the ailerons, not the rudder, so once the main gear are on the ground and wings level, the autopilot can no longer steer via ailerons, and a human has to take over via rudder and ground steering. For a human to take over, he/she has to be able to see a little bit! Hence, the need for the autopilot to have a rollout control system if you want truly 'blind' landings....
However, to say, 'more landings are done that way than not' is completely inaccurate. The vast majority of landings by commerical aircraft are done by hand. You forget that in order to even be authorized to autoland, the ILS critical area on the ground must be cleared, a step most control towers are loathe to do unless necessary, as these critical areas frequently encompass heavily used ground areas. This critical area clearence is one of the reasons the maximum arrival rate for most commerical airports drops precipitiously when low IFR conditions exist.
Also, to say the autopilot does a better job than humans at landing is inaccurate - it does a different job than a human. In conditions of low/unusual visibility, with a poor transition between instrument and visual flying, the autopilot generally performs better, since it relies on radio signals and not visual cues. This is only one, somewhat uncommon, landing profile. Autopilots generally do worse at dealing with windshear and other rapid changes in windspeed. This fact is typically encoded into aircraft flight manuals in the limitations section, where autolands are prohibited when the surface winds exceed fairly low parameters, far lower than the aircraft is capable of (and frequently does) operating in. On the 737 at my airline, the max headwind/x-wind/tailwind components allowed for autoland are 25/15/10 knots, respectively - limits that are exceeded by actual conditions quite frequently.
Also, many commerical airports do not have Cat III approaches, and are unlikely to get them any time soon...
I guess my point is that, while the technology exists, and is used, it's also quite limited and impractical for common day-to-day ops as it currently stands... I have no doubt it will happen someday, but 'someday' is decades away presently.
Besides, there is something even harder than flying an airplane - taxiing it!;-)
Besides, think about it, do you really give one tenth of one shit about most of the people you know? Don't fucking like, of course not. Only a few of those people are true friends who would actually help you out of a bind at cost to themselves. The rest of them could get hit by a bus tomorrow and the only way it would affect your life is if you received an invitation to the service.
Actually, yes, I do - I've thought about the company I keep quite a lot in life. I came to the conclusion that most people are worth giving a shit about, even if they don't happen to have the same hobbies as you, same tastes as you, same politics as you, etc. I find that most people, given a decent opportunity, will be decent people. I also find, that if you expect someone to screw you over, it will be noticed, no matter how subtle you think you are, and a self-fulfilling prophesy will result.
Hence, my point above - if we treated the living with as much charity as the dead, perhaps we would appreciate their positive contributions, and they would be encouraged to be more positive, and our lives would be a bit better for it.
So, perhaps you should think about it - rather than think, 'Only a few of those people are true friends who would actually help you out of a bind at cost to themselves', perhaps you should try thinking, 'Hey, I could help out this aquantence move - he really is in a bind - but there goes my weekend. Oh well, there will be more weekends.' You might make a new friend. You might learn that this person isn't worth being friends with. You might be pleasantly suprised to discover that the person has the same attitude as you do, and make a really close friend. You might just get some excerise, see some interesting furniture, and see what someone else's tastes are. You might meet a cute girl at your friend's new place who's impressed by your easy-going generosity. Or you might sleep a bit better that night, knowing something useful was done.
Try it out! Good sig, BTW....
Well, he's not here to read the comments.
on
Roland Piquepaille Dies
·
· Score: 3, Insightful
I always wondered about the following question:
Why is it we only show up for peoples' weddings and funerals? Do they have to hook up psuedo-permanently with a gal/guy or be in a casket before we'll take the time out of our busy lives to see them?
So, for all those who miss him in requiem - treat the living, give them as much charity, forgive their trespasses, and appreciate their good qualities while they are still around to hear it.
I would like to switch from incandescents to CFL's or LED's, but one big thing stops me - I like being able to dim my lights to variable levels, depending on a whole host of factors. I know the basics of electricity, like what the difference is between a volt, an amp, an ohm, a watt, etc., and how to evalutate DC circuit diagrams using Kirchoff's laws. I vaguely understand AC, and how it interacts with inductances, capacitors, etc...
So - can someone explain why it's not possible (or at least very difficult) to build a dimmable CFL or LED? I think I understand how dimmers work with incandescent bulbs - they increase the resistance in the circuit, and decrease the voltage drop in the bulb, as well as the current, leading to substantially less power expended, from P = IV.
Just to make sure I've got it right: A 120 W bulb on a 120 volt circuit must have a current of 1 amp through it because P = IV. (I'm ignoring the fact that we are dealing with AC rather than DC, but if I understand correctly, the principle is the same). Therefore, the resistance of the bulb is (from V/I=R) 120 V / 1 A = 120 Ohms. If I put an additional 120 Ohm resistor in series with our bulb, and assuming the voltage from the mains and the resistance of the bulb doesn't change, the overall resistance of the circuit is 120 ohms + 120 ohms = 240 ohms, and therefore the current through the circuit is 120 V / 240 ohms = 0.5 amps. The voltage drops across the bulb and resistor must sum to 120 V, and since they have the same resistance, the voltage drops across each must be equal - therefore they must each be 60 V. Thus the bulb only uses 60 V * 0.5 A = 30 Watts - 1/4 of the previous amount. Of course, the dimmer provides the other 60 V of drop, and the same 0.5 A is going through it as well, so another 30 W is wasted in the dimmer. This 30 W heats up the filament in the bulb much less than 120 W, so the filament glows much more dimly and much more shifted towards the infrared. An incandescent bulb would work whether the circuit was AC or DC.
How does one vary the brightness of a given CFL or LED? And if it's not in a direct fashion, by reducing the voltage available to the CFL/LED by sticking a resistor in the circuit, is there some way through clever electronics to sense the change in voltage, and convert that to a change in whatever it is that varies the brightness of a CFL/LED?
It's ostensibly about calculus - but it's really about how to think with crystal clarity and minimal assumptions and develop a vast field (in-joke) of conclusions from this. It was the first truly deep mathematics book I read, and I think it's a worthwhile step for anyone who thinks for a living.
There is a concept that is mentioned from time to time in mathematics education, called 'mathematical maturity'. It's a hard concept to define, and hard to measure with a test. My own, personal, idea about it is the level when you have to learn mathematics that extends beyond what your intuition has reached.
It is a difficult bridge to cross, and most people (in my experience) stop learning mathematics when their intuition about it ceases to carry them further. But it is an amazing bridge to cross, and once you do, you start to see how axioms can start to shape your intuition - it's a breathtaking experience (at least it was for me).
I crossed that bridge, a long time ago, thanks to a great mathematics professor, Dr. Swiatek, and that book. Like many of the greatest books, it takes you on a journey - it tells you a story. It brings you along on a voyage through the landscape of what a number really is, and what comes as a result. I can't recommend it enough if you are determined to learn what mathematics is really about.
Software purchases seem to make people forget everything they know about economics. For any other business-critical purchase, the first question asked is 'who is the second source for this product?' For software, people seem to forget to ask this question.
Economics got nothing to do with it. People, and consequentally many businesses, will suffer almost any risk to avoid a learning experience.
Story: A friend of a friend was working on her PhD thesis, in Microsoft Word, on WinXP, on a four year old Dell laptop. Her machine had literally thousands of hours of her work on it - I think most people would agree that the files on that laptop are critical to her career.
One day, the machine stops booting. It just goes to the recovery mode screen of XP, and that's it. She comes crying that years of work is on this machine. No backups, of course.
I'm a nice human. I pull the hard drive, mount it in a Debian box, pull off all the files I can, wipe it, reinstall XP, put it back in the laptop, reinstall her apps. An afternoon of work.
Well, she is very grateful. Extremely grateful - offers to take me out for a nice dinner, etc. I say don't worry about it. Then she asks the question - 'How can I prevent this from happening again?'
This is where things go downhill. I suggest, first, backing up her files regularly. But she doesn't want to pay for an external hard drive. I offer her an old one I have lying around, gathering dust. I explain all you have to do is plug it in on a regular basis, and either copy your files to it, or get a backup utility to automate the process. She demurrs, and claims she's get a flash drive or something. Okay... I suggest getting a Mac, with a copy of Microsoft Office, to reduce her exposure to malware and reduce to probabilites of having to reinstall the system in a year. No, that would cost too much. I suggest putting Ubuntu and Open Office, to reduce exposure to malware and get a little more life out of the Dell. No, she doesn't want to learn something new.
She's isn't a dummy - she's earned her PhD from Columbia University in the meanwhile. She's fully capable of learning new things. But she has no interest, heck, negative interest, in learning something about how to use a computer other than what she already knows.
If a brilliant, educated, well-to-do, young person has this much reluctance to learn something new, properly backup her thesis, or reduce the probabilites of malware - what hope is there of convincing 'average' business-people to make the right choice about business-critical decisions about sole-source vendors, backups, malware and spyware?
I wish that was an isolated story - but I have similar ones about businesses that PREFER a sole-source vendor, because then they never have to learn anything new - even if it costs them $10M a year. It's not smart, it doesn't make business sense, and someday a competitor will eat their lunch, but the power of intertia in human nature is strong.
I've found Macports to be fantastic. It has one, small, consistent, and easily fixable problem - when it has to download and build multiple packages, sometimes it errors out with a missing package complaint, even thought the package in question was apparently built and installed. Just rerun the same Macports command, and it then 'notices' the built and installed package it missed before, and proceeds on it's way. Django, PostgreSQL, and a whole host of python 2.5 libs I have installed specifically with Macports, without problems. Hope this helps a but...
This is one of the major issues that keeps me on Linux.
Despite the fact you can get almost anything to run on OS X eventually, for most software it's much much harder to get up to date software versions then "apt-get install fizzbuzz" on Ubuntu or debian testing.
sudo port install fizzbuzz
Occsaionally, I've seen Macports fail to finish a build - you just repeat the same command, and then it works fine. Python 2.5.2, WxWidgets + WxPython, matplotlib, mysql... all work flawlessly through Macports. It's an underappreciated gem.
The most annoying non-unixy thing with OS X is the NetInfo garbage...
You know that NetInfo is gone as of 10.5, right? That's over a year old now. Open Directory is now used for authentication and authorization. Try man dscl, and you'll find a whole bunch of useful info on it.
Slashdot Mirror
Which just furthers my point. If the Chinese goverment really worries about an invasion of U.S. soliders, they are seriously deluded.
Heck, in WWII, the planed invasion of Japan, a beaten, firebombed nation (although not nuked yet) with less than a twentieth the population of China now, was estimated to cause a MILLION or more allied casualties. That was one of the biggest arguments that was given to nuke Japan. Even if we take nukes off the table, a plain old-invasion of China would make WWII casualty figures look like chump change.
I see why the North Koreans would really like nukes, but it still doesn't make sense to me from China's point of view. I would place stronger odds on the Koreas' uniting sometime in the next 50 years than not. Language, culture and blood are much stronger long-term ties than country. Then, they'll have a unified, nuclear armed Korea right on their border. Which will likely encourage a nuclear armed Japan (which they might do much sooner as a result of North Korea's tests anyway). Is encouraging a nuclear arms race on your borders really a good idea? I can't see how having a few more heavily armed, somewhat paranoid neighbors benefits you.
Perhaps statements like this are part of the reason why some South Koreans 'hate' the US?
I agree with your post about the U.S. needing to get out of the world-running business. But - your casual statement regarding the extermination of 70+ million people only in terms of positive impact to U.S. car companies is not helpful to your argument, since people may assume you are a ghoul, which means they won't take your otherwise good idea as seriously as they should.
When China finally wakes up and realizes that having a somewhat unstable next door neighbor armed with nukes is a bad idea, this sort of thing will stop - North Korea survives only because China keeps giving them tons of aid.
Perhaps the North Koreans are interested in China's continued aid supplies over the long term? As in, after they get a credible, deliverable weapon, 'If you stop the gravy train, we take out Hong Kong, even if we're glass 8 minutes later. That whole "we don't like the west" thing was just so you would let us build nukes.'
I really don't get China's motivations. Once the nuclear genie is out, they won't be able to stuff it back in. It's like the U.S.A. helping Haiti to get nukes because they don't like Cuba. Does it not occur to the Chinese govt. that once North Korea has a real nuclear capability, they could aim it anywhere they so wished?
That implies that any methodology that can not perform artifical experiments is not a science. That would exclude large swaths of biology, specifically those involving evolution, from the title 'science'. That's a bit extreme, don't you think? As long as a methodology makes testable predictions that can and will be used to evaluate the usefulness of the model that made those predictions, it seems like it deserves the title 'science'.
For instance, the first verification of General Relativity was done by measuring the bending of distant starlight when it skimmed the surface of the sun. This could only be done during a solar eclipse. We knew of no way to arrange circumstances into a 'test' that could be done in a lab, at will, (at the time) so we had to wait for nature to provide one (a solar eclipse) in 1922. Does the fact that nature arranged the circumstances detract from the validity of the test?
If a climate model makes specific predictions, that 20 years later become true, it increases the probability that the model is an accurate predictor. Because we can't manipulate the atmosphere as you suggest, unfortunately, we have to do it the slow way - it will take decades before we really know whether our models are accurate. But to suggest that climatology is not science because it can't be experimentally verified in a lab is a bit extreme.
Yeah, that's a bit more questionable, isn't it? There has been casually observable evidence for the Earth's roundness in certain places (shorelines, where one has an opportunity to see a ship vanish over the horizion hull first, rather than just get too small to see), and if one knows the mechanism causing a lunar eclipse, the always circular shadow of the earth strongly suggests it's spherical.
OTOH, the geometry of the lunar eclipses was not well understood by most 'ordinary' people (even today), many didn't live on seashores, and historically, most people were illiterate and unable to read descriptions about these things. So, it is pretty unlikely 'the overwhelming majority of people' would think you were a nutter for proposing the earth was flat.
Perhaps the parent exaggerated a bit, and meant that throughout recorded history, the idea that the earth was round instead of flat was known by educated people of the times, and accepted as true by a significant proportion.
You can do your own googling for that. But to give you some confidence that the parent is not making things up, consider what the casual name for Native Americans is - 'Indians'.
That's because Columbus believed, until his dying day, that he had landed in a group of islands off of Asia/India, not discovered a new continent - hence, the term 'Indians' for Native Americans. Columbus was not a fool, and he knew how far he had sailed on his journeys. His interpertation though, was that he had sailed across a much smaller ocean, on a much smaller planet (I think he estimated a circumference of 18,000 miles vs. the true figure of 25,000 miles), and hit the other side of Eurasia.
Hmmm... we can do better at coining a new word.
Let's start with an anachronism:
anachronism |É(TM)ËnakrÉ(TM)ËOEnizÉ(TM)m| noun a thing belonging or appropriate to a period other than that in which it exists, esp. a thing that is conspicuously old-fashioned.
Now, let's combine that with an acronym:
anachronism + acronym = anachronym! Sounds better...
While it's always easy to bash politicians for doing something meaningless, to some extent that is inevitable in a system where powers of legislation are separated from powers of execution. Individual Congressfolk are not the ones hiring and firing the school chancellors, teachers, admins, etc., nor do they set the property tax rates (the most common way local school systems are funded in the USA) or school budgets. The effects they can have on education are at the broadest level only, like federal budget suppliments, standards setttings before such suppliments can be recieved (see the NCLB...), etc.
The reality is in the USA, the primary education system is a highly local affair, with standards set by local sensibilities. This is one reason you keep seeing movements to push creation 'science' (it doesn't deserve the word unless surrounded in scare quotes) pass or nearly pass school boards in Pennsylvania, Oklahoma, Texas, and Kansas, to name a few. In the USA as a whole, about a third of the population rejects evolution outright (see National Geographic or a gazillion other polls). The evolution deniers are not evenly distributed across the USA, but more clumped. Certain congressional districts, I'm guessing, have half of their voters (or more) who would vote against a congressperson who declared that they would vote for a bill to promote science standards if the mandatory teaching of evolution was a part of those standards. With a constituency like that, the odds of passing (at the federal level) a significant, science-standards based education bill are slim. Remember, 1/3 of the USA population rejects evolution outright. That's a lot of people. So, instead, we get National Pi Day.
So, look on the bright side. Now that we have a National Pi Day, maybe we don't have to worry about attempts to legislatively redefine pi any more....
[double-take]
This word you use, ethics. I do not think it means what you think it means.
I mean, in my silly little world, 'ethics', among other things, means doing unto others as you would have them do unto you, and I take it your friend doesn't want to be sued into financial oblivion. So let's forget me and my silly definition. The dictionary definition of ethics is:
Moral principles that govern a person's or group's behavior.
You got it right when you said it's about money, not morals. I just question how you can claim they have ethics without morals, since one seems to require the other. Call a spade a spade - when your job entails bankrupting random file-sharers while you yourself engage in the same behavior, that sounds like a casual abuse of power and a willful blindness to the hypocracy of one's own actions at best. Either that, or a 'morality' where your friend believes that laws are only meant to be enforced against other people, or the fact that you were ordered to do it by your boss excuses you of any unethical consequences. Sounds like qualities human monsters in the past have had in spades, even if they are great dinner company and will water your plants for you while you are on vacation.
I don't at all mean this as a dig at you. I'm simply trying to point out that people are wholes - I don't think there is such a thing as an 'ethics free' zone while at work, and I would have a hard time being friends with someone like that.
Yes, for about half an hour I had a chance to play around with and read from one at a friend's home, who happens to be a book editor and a Kindle enthusiast. It's easier to read from than a computer screen. It's worse to read from than the printed page. As I implied with my 60 Hz comment, since 60 Hz CRTs drive some people batty and others not at all, it bothers me enough that I find it unpleasant. For some people, like you, it may be sufficient. For her? She's still on the fence, and she reads books for a living.
Wrong. The only support that is built in for non-DRM content is mobipocket books, and plain text. No pdfs, Word docs, and rtf - although an 'experimental' converter is available for pdfs, and if you feel like e-mailing your docs to Amazon, for Word and HTML docs. For me, that's too much of a PITA. I like dragging the files I want to read on it over, and not have to convert them. I guess this issue is not a problem for you. It is for me.
Yes. How astute of you, reading what I wrote and noticing I want what I want! I want, for me, not anyone else, including you, a general, multi-function device, that can be used to do many things. I don't feel like carrying around an e-book reader, a notebook computer, a phone, an address book, a calendar, and lord knows what else. The fewer things I have to carry around in my daily life, the better.
One reason I generally dislike zealots of all stripes is they instantly assume what's good for them is good for every one else. The Kindle is good enough, for you. I don't think it's good enough for me. I didn't predict its demise in the marketplace, I didn't say Kindle buyers were stupid, I simply gave a list of reasons why I won't be buying one. I even said 'Yet' in the subject of my post, and said 'git off my lawn' to give the hint to the clueful that I realize I may wind up being behind the adoption curve because of my personal preferences.
What do you contribute to the discussion? You write back a nasty response with factual errors and incorrect assumptions, denigrating my reasons because they are not yours. Kind of a waste, don't you think? Unless you get off on being contemptuous, sarcastic, and wrong...
It would be a very smart move for Microsoft; they could offload a tremendous amount of development pain, use their influence where it would do some good (make the hardware vendors cough up the specs), etc.
I don't think it will happen, though, because Microsoft has a deep, deep infection of ego. Large corporations, particularly successful ones, tend to get this at some point. Such an infection may only kill off your innovation and usability - monopoly and political power can prop up your balance sheets for a very long time. But it can blind you to real, oncoming competitive threats, because the idea that someone else can be better than you is organizationally inconcievable. It's rarely curable, although a near-death experience (assuming it doesn't lead to a full-dead experience!) can do the trick.
Man, pass me a glass of whatever YOU are drinking...
I agree with the sentiment of what you wrote, but I work for the airlines. Intelligence and morality in decision-making here is as common as snowstorms in Aruba. I suspect it's not that common in the rest of the world either, judging from the available evidence....
I'd really like for you to be right, but... just because an idea or concept is the right and intelligent thing has very little bearing on its ultimate success. It also heavily depends on the tenacity and charisma of its proponents, the weakness and misteps of its opponents, and a healtly dose of luck. And whenever I hear someone declare with great confidence that an idea will win because of its inherent rightness... well, good luck with that.
Why should I buy an e-book reader? Show me one with the following abilities:
1) Has 600 DPI+, not 167 DPI. 167 DPI is about as pleasant as a 60 Hz CRT.
2) Allows me to put any files, from any creators, in any format I want to read on it, WITHOUT MANDATORY DRM. Pdfs, rtfs, plain text, Word docs, whatever.
3) Allows me to have access to the underlying OS as I see fit.
The ONLY significant thing the Kindle has over a netbook is a (slightly) more pleasant screen - e-ink vs. LCD and 167 DPI vs. 100 DPI. They both still suck for long extended reads, but I can access the internet, read any docs I feel like, do some coding or play simple games on a netbook. Perhaps someone should stick an e-ink screen on a netbook?
In the meanwhile, I'm keeping my books. And git off my lawn.
It IS interesting... and there are many areas of progress where a little applied automation can go quite far. I'm just trying to point out that the actual mechanical acts of taking off from an arbitrary airport, climbing, crusing, approaching, and landing at an arbitrary airport, are NOT the hardest problems - they are pretty close to solved today, in the sense that we know how to build and deploy a system that would do these things...
The harder problem is of the kind that bedevils game designers and other programmers who are trying to implement AI. We still don't have a good handle on how to give computers 'good judgement', or handle risk analysis for previously unknown, or highly complex situations. To date, the most successful UAV's have human pilots to make the judgement calls.
I'm actually more interested in seeing fully automated driving than flying. When someone deploys a car where you can get in, punch in a destination, and have it safely transport you there, I'll know flying is about to go the same way....
Of course, this doesn't even touch upon when it will be cost-effective to go that way!
Well, by that definition, there is not a 'real' autopilot in the world. Did you even bother to READ the wikipedia article you posted? Let me summarize the stats for you:
.3 nautical miles in typical conditions. That is FAR too low to execute a river landing if the river is less than a mile wide! In addition, how do you avoid ships by autopilot?
15-Jan-2009: USAir 1549 - 0% fatal
06-Aug-2005: Tuninter 1153 - 49% fatal
16-Jan-2002: Guruda 421 - 2% fatal
23-Nov-1996: Ethiopian 961 - 71% fatal
02-May-1970: ALM 980 - 37% fatal
21-Aug-1963: Aeroflot ??? - 0% fatal
04-Oct-1960: Eastern 375 - 86% fatal
??-Oct-1956: PanAm 943 - 0% fatal
??-Apr-1956: Northwest 2 - 13% fatal
19-Jun-1954: Swissair ??? - 25% fatal
16-Apr-1952: ??? - 0% fatal
All I can conclude from these numbers, spread over 50 years, from piston to turboprop to turbofan powered airplanes, ditched in winter, summer, rivers, oceans, bays, and every other concevable condition, is that a) Ditchings are very rare, and b) It's more likely than not some people are going to croak.
In addition, the survival rates say NOTHING about how difficult it is to pull off a ditching, particularly for this theoretical 'real' autopilot that doesn't exist yet. Every ditching here was done by human pilots, not autopilots.
Think of it this way. How do you program an autopilot to ditch? When is it a good idea, and when is it not? For instance, PanAm 943 lost two of its four engines. But rather than ditching immediately, it circled around the USCGC Pontchartrain until dawn, THEN ditched - plucking people out of the ocean in daylight is far easier than at night. Would you have been able to program an autopilot to make that decision?
Another issue - river landings, which were 3 of the 11 ditchings listed above. There is usually very little margin for error landing on a narrow river. The 95% confidence radius of position (sometimes called ANP, for Actual Navigational Performance) most modern airliners is typically about
The realities of ditching are far more complex than you allow for.
A little, but not much. The main difficulty in learning to fly safely has less to do with physical skill and far more to do with good judgement. Airplanes still are rather marginal beasts, in the sense that they can easily be flown into situations that are rather hard to get out of. Let's assume that we have invented and installed a magical autopilot system that does what you described. What situations will it help with, and which ones will it not help with?
IT WILL HELP WHEN...
- A pilot gets lost. Hopefully this problem will go away.
- A pilot flies into unexpected IFR (instrument) flying conditions. Press the button, and the airplane magically keeps itself upright and lands them somewhere safe. This is probably the biggest gain in safety right here.
- A pilot 'loses it' - gets airsick, incapacitated, etc.
IT WON'T HELP WHEN...
- The pilot doesn't put enough fuel on for the flying they want to do. This is, sadly, more common than you would think.
- The pilot flies into thunderstorms, freezing rain, etc. Bad weather that can knock a plane from the sky has always existed, and pilots keep flying into it. Our abilities to automatically predict and avoid such weather is still very, very limited.
- The pilot doesn't maintain the aircraft or magic autopilot system, and something breaks at an inoppurtune time.
- The pilot overloads the aircraft with too much stuff, or balances it badly. The predictable (but unknown to the autopilot) lack of performance that will result could be deadly.
- And, of course, engine failures requiring off-airport landings. The recent USAir landing in the Hudson illustrates where a skilled human is most valuable.
Food for thought.
These statements are not quite correct, and reflect a lack of actual experience with such approaches. Generally, there are both visibility and ceiling requirements set for such approaches. The typical visibilty minima for Cat IIIA is 600 feet/175 m RVR and the typical decision height [DH] is 50 feet above touchdown zone height. Cat IIIC is typically 'zero-zero' - no ceiling or visibility minimums are estabilished. Interestingly, the ability to do Cat IIIC landings revolves around rollout control systems. Most autopilots steer via the ailerons, not the rudder, so once the main gear are on the ground and wings level, the autopilot can no longer steer via ailerons, and a human has to take over via rudder and ground steering. For a human to take over, he/she has to be able to see a little bit! Hence, the need for the autopilot to have a rollout control system if you want truly 'blind' landings....
;-)
However, to say, 'more landings are done that way than not' is completely inaccurate. The vast majority of landings by commerical aircraft are done by hand. You forget that in order to even be authorized to autoland, the ILS critical area on the ground must be cleared, a step most control towers are loathe to do unless necessary, as these critical areas frequently encompass heavily used ground areas. This critical area clearence is one of the reasons the maximum arrival rate for most commerical airports drops precipitiously when low IFR conditions exist.
Also, to say the autopilot does a better job than humans at landing is inaccurate - it does a different job than a human. In conditions of low/unusual visibility, with a poor transition between instrument and visual flying, the autopilot generally performs better, since it relies on radio signals and not visual cues. This is only one, somewhat uncommon, landing profile. Autopilots generally do worse at dealing with windshear and other rapid changes in windspeed. This fact is typically encoded into aircraft flight manuals in the limitations section, where autolands are prohibited when the surface winds exceed fairly low parameters, far lower than the aircraft is capable of (and frequently does) operating in. On the 737 at my airline, the max headwind/x-wind/tailwind components allowed for autoland are 25/15/10 knots, respectively - limits that are exceeded by actual conditions quite frequently.
Also, many commerical airports do not have Cat III approaches, and are unlikely to get them any time soon...
I guess my point is that, while the technology exists, and is used, it's also quite limited and impractical for common day-to-day ops as it currently stands... I have no doubt it will happen someday, but 'someday' is decades away presently.
Besides, there is something even harder than flying an airplane - taxiing it!
Actually, yes, I do - I've thought about the company I keep quite a lot in life. I came to the conclusion that most people are worth giving a shit about, even if they don't happen to have the same hobbies as you, same tastes as you, same politics as you, etc. I find that most people, given a decent opportunity, will be decent people. I also find, that if you expect someone to screw you over, it will be noticed, no matter how subtle you think you are, and a self-fulfilling prophesy will result.
Hence, my point above - if we treated the living with as much charity as the dead, perhaps we would appreciate their positive contributions, and they would be encouraged to be more positive, and our lives would be a bit better for it.
So, perhaps you should think about it - rather than think, 'Only a few of those people are true friends who would actually help you out of a bind at cost to themselves', perhaps you should try thinking, 'Hey, I could help out this aquantence move - he really is in a bind - but there goes my weekend. Oh well, there will be more weekends.' You might make a new friend. You might learn that this person isn't worth being friends with. You might be pleasantly suprised to discover that the person has the same attitude as you do, and make a really close friend. You might just get some excerise, see some interesting furniture, and see what someone else's tastes are. You might meet a cute girl at your friend's new place who's impressed by your easy-going generosity. Or you might sleep a bit better that night, knowing something useful was done.
Try it out! Good sig, BTW....
I always wondered about the following question:
Why is it we only show up for peoples' weddings and funerals? Do they have to hook up psuedo-permanently with a gal/guy or be in a casket before we'll take the time out of our busy lives to see them?
So, for all those who miss him in requiem - treat the living, give them as much charity, forgive their trespasses, and appreciate their good qualities while they are still around to hear it.
R.I.P. anyway.
I would like to switch from incandescents to CFL's or LED's, but one big thing stops me - I like being able to dim my lights to variable levels, depending on a whole host of factors. I know the basics of electricity, like what the difference is between a volt, an amp, an ohm, a watt, etc., and how to evalutate DC circuit diagrams using Kirchoff's laws. I vaguely understand AC, and how it interacts with inductances, capacitors, etc...
So - can someone explain why it's not possible (or at least very difficult) to build a dimmable CFL or LED? I think I understand how dimmers work with incandescent bulbs - they increase the resistance in the circuit, and decrease the voltage drop in the bulb, as well as the current, leading to substantially less power expended, from P = IV.
Just to make sure I've got it right: A 120 W bulb on a 120 volt circuit must have a current of 1 amp through it because P = IV. (I'm ignoring the fact that we are dealing with AC rather than DC, but if I understand correctly, the principle is the same). Therefore, the resistance of the bulb is (from V/I=R) 120 V / 1 A = 120 Ohms. If I put an additional 120 Ohm resistor in series with our bulb, and assuming the voltage from the mains and the resistance of the bulb doesn't change, the overall resistance of the circuit is 120 ohms + 120 ohms = 240 ohms, and therefore the current through the circuit is 120 V / 240 ohms = 0.5 amps. The voltage drops across the bulb and resistor must sum to 120 V, and since they have the same resistance, the voltage drops across each must be equal - therefore they must each be 60 V. Thus the bulb only uses 60 V * 0.5 A = 30 Watts - 1/4 of the previous amount. Of course, the dimmer provides the other 60 V of drop, and the same 0.5 A is going through it as well, so another 30 W is wasted in the dimmer. This 30 W heats up the filament in the bulb much less than 120 W, so the filament glows much more dimly and much more shifted towards the infrared. An incandescent bulb would work whether the circuit was AC or DC.
How does one vary the brightness of a given CFL or LED? And if it's not in a direct fashion, by reducing the voltage available to the CFL/LED by sticking a resistor in the circuit, is there some way through clever electronics to sense the change in voltage, and convert that to a change in whatever it is that varies the brightness of a CFL/LED?
I'll recommend a book that's really not CS, but is the most elegant, crystal-clear exposition of developing ideas from axioms I have ever read:
Calculus by Spivak
It's ostensibly about calculus - but it's really about how to think with crystal clarity and minimal assumptions and develop a vast field (in-joke) of conclusions from this. It was the first truly deep mathematics book I read, and I think it's a worthwhile step for anyone who thinks for a living.
There is a concept that is mentioned from time to time in mathematics education, called 'mathematical maturity'. It's a hard concept to define, and hard to measure with a test. My own, personal, idea about it is the level when you have to learn mathematics that extends beyond what your intuition has reached.
It is a difficult bridge to cross, and most people (in my experience) stop learning mathematics when their intuition about it ceases to carry them further. But it is an amazing bridge to cross, and once you do, you start to see how axioms can start to shape your intuition - it's a breathtaking experience (at least it was for me).
I crossed that bridge, a long time ago, thanks to a great mathematics professor, Dr. Swiatek, and that book. Like many of the greatest books, it takes you on a journey - it tells you a story. It brings you along on a voyage through the landscape of what a number really is, and what comes as a result. I can't recommend it enough if you are determined to learn what mathematics is really about.
Economics got nothing to do with it. People, and consequentally many businesses, will suffer almost any risk to avoid a learning experience.
Story: A friend of a friend was working on her PhD thesis, in Microsoft Word, on WinXP, on a four year old Dell laptop. Her machine had literally thousands of hours of her work on it - I think most people would agree that the files on that laptop are critical to her career.
One day, the machine stops booting. It just goes to the recovery mode screen of XP, and that's it. She comes crying that years of work is on this machine. No backups, of course.
I'm a nice human. I pull the hard drive, mount it in a Debian box, pull off all the files I can, wipe it, reinstall XP, put it back in the laptop, reinstall her apps. An afternoon of work.
Well, she is very grateful. Extremely grateful - offers to take me out for a nice dinner, etc. I say don't worry about it. Then she asks the question - 'How can I prevent this from happening again?'
This is where things go downhill. I suggest, first, backing up her files regularly. But she doesn't want to pay for an external hard drive. I offer her an old one I have lying around, gathering dust. I explain all you have to do is plug it in on a regular basis, and either copy your files to it, or get a backup utility to automate the process. She demurrs, and claims she's get a flash drive or something. Okay... I suggest getting a Mac, with a copy of Microsoft Office, to reduce her exposure to malware and reduce to probabilites of having to reinstall the system in a year. No, that would cost too much. I suggest putting Ubuntu and Open Office, to reduce exposure to malware and get a little more life out of the Dell. No, she doesn't want to learn something new.
She's isn't a dummy - she's earned her PhD from Columbia University in the meanwhile. She's fully capable of learning new things. But she has no interest, heck, negative interest, in learning something about how to use a computer other than what she already knows.
If a brilliant, educated, well-to-do, young person has this much reluctance to learn something new, properly backup her thesis, or reduce the probabilites of malware - what hope is there of convincing 'average' business-people to make the right choice about business-critical decisions about sole-source vendors, backups, malware and spyware?
I wish that was an isolated story - but I have similar ones about businesses that PREFER a sole-source vendor, because then they never have to learn anything new - even if it costs them $10M a year. It's not smart, it doesn't make business sense, and someday a competitor will eat their lunch, but the power of intertia in human nature is strong.
I've found Macports to be fantastic. It has one, small, consistent, and easily fixable problem - when it has to download and build multiple packages, sometimes it errors out with a missing package complaint, even thought the package in question was apparently built and installed. Just rerun the same Macports command, and it then 'notices' the built and installed package it missed before, and proceeds on it's way. Django, PostgreSQL, and a whole host of python 2.5 libs I have installed specifically with Macports, without problems. Hope this helps a but...
sudo port install fizzbuzz
Occsaionally, I've seen Macports fail to finish a build - you just repeat the same command, and then it works fine. Python 2.5.2, WxWidgets + WxPython, matplotlib, mysql... all work flawlessly through Macports. It's an underappreciated gem.
You know that NetInfo is gone as of 10.5, right? That's over a year old now. Open Directory is now used for authentication and authorization. Try man dscl, and you'll find a whole bunch of useful info on it.