Yeah, I had one of those friends, who was such a control freak of his C64, that I basically rode my bike to his house just to watch him play. I think one of the reasons I am still involved in computers is because I'd have to come home to my ailing TI-99 and make it do something new and interesting. Computer graphics were a mystery to my friend, but I was playing with sprites on my machine, and although it wasn't as cool as his store-bought stuff (I made an asteroids-type game... that, er... crashed reliably about 30 seconds in), it was mine. A word to the wise: it's no fun playing your own dungeon-crawlers. You already know how to win.
But the C64 undoubtedly had some cool games. Sid Meier's Pirates! was and still is one of the best (Xbox's version was a very good remake, IMHO), and Prince of Persia was just mindblowingly awesome. Guillotines! I think that was the first game that actually make me tremble with anxiety. Thankfully, my cousin owned one of these machines, and was so fed up with trying to use it (he could never reliably punch in the LOAD commands) that I was able to log some time on one.
Big media companies are the only ones who have the resources to police their works. Do you think the FBI would even notice if people started illegally distributing my software? Even if I made a stink about it, claimed that this was 'millions of dollars of loss' to me, do you think they would care? Now what if Disney does the same thing? The FBI listens to Disney.
Sure, I agree, copyright is a good thing. But what we have now is insane. Copyright lasts for the lifetime of the author, plus 70 years. That's essentially in perpetuity, because the public that is alive today, and likely their children and even grandchildren, will never see those works enter the public domain. I can buy an argument that works should remain copyrighted for the lifetime of the author (although I personally do not believe it should be this way), but an additional seventy years? Who does this benefit? Let's see... something that never dies... er, not vampires... Highlander? Oh, right! A big media company!
I don't argue for people making an honest living. But when you've got the government pulling Gestapo shit for the big hitters, that's not honest. That's fascist.
GPS really gives people a false sense of confidence. I spend a lot of time in the backcountry, and I have gotten lost many times. I say "lost", but that's not really true-- I put myself in this situation for the purpose of exploration. GPS devices aren't really all that useful. You still need a compass*, and you still need to be familiar with basic backwoods travel and orienteering. Dead reckoning is the single most important orienteering skill a person can have. What do you do when your GPS batteries die? This often happens faster than you expect in cold weather.
When you are in unfamiliar territory, you tend to make irrational decisions, even when you have proper orienteering gear. Especially when you are worried about time constraints, or your hiking partner, or so on. There's a real art to learning how to calm yourself down, think rationally, and move on in these situations. The fight-or-flight instinct is a powerful force.
That said, GPS is a great invention. But it doesn't help you if you don't know how to read a topo map and apply that knowledge to where you are currently standing.
*I understand that many GPS devices come with digital compasses now. But I still wouldn't rely on it-- bring it compass; it will almost always work. Also, know direction-finding in good weather. Try to know some general rules about orienteering in bad weather. Know about prevailing winds, and so on.
Part of the problem is that very few professional designers (e.g., print designers) are also cognizant enough about software design to explain to a GIMP developer what the problems are. I am only vaguely aware of some of these issues; my main problem wihth the GIMP boils down to palettes not being truly floating. I can live with the right-click-to-do-anything mode. But designers need to do some additional things: full control of color space (RAW, CMYK, spot colors), they need color space histograms, they need to be able to easily produce color separations, they need to be able to easily put together batch transformations, they need some sophisticated marquis tools, they need to be able to mix raster and vector art, have access to Postscript, TrueType, and now OpenType fonts, and so on. I do not know if these things are available in the GIMP, because I am not a designer, nor do I regularly use the GIMP. But that's just to point out that designers do need specialized tools.
Also-- the GIMP is ugly to these people. As a programmer, I find certain elements very elegant. Most designers couldn't care less that you can write GIMP plugins in Perl; they're irritated that when they double-click on an image in InDesign, it doesn't automatically come up for editing in the GIMP. Some of these issues are petty-- but in my experience, designers tend to work in a more immediate-feeling, less rational domain. Appearance is important-- after all, they're paid because they have a highly developed aesthetic sense. Programmers are so horrendously square to many of them.
Adobe's stuff is expensive, and in many cases overrated, but they have indeed put a lot of thought into the workflow (e.g., integration between DTP app and photo editor) and appearance, and for a lot of people, this is all that matters. These facts combined with certain mailing list posts that lead me to believe that some GIMP developers suffer from a bit of a 'high horse' syndrome lead me to believe that this is why GIMP is not universally accepted, despite the fact that it is a highly capable application.
Right, but we must remember that the primary tension in science is that there are, practically speaking, an infinite number of hypotheses for any given phenomena. How do we decide which one best describes our phenomena? We pick the one that seems the best. In order to justify this decision, we use what we have already-- our scientific knowledge and our analytical tools, but in the end, it is a feeling that guides us. Analytical tools, like logic, can make wrong conclusions based on false premises, and scientific knowledge is often invalidated in the course of study. Words like "elegant" and "powerful" are how we make ourselves feel better about this process, but how else would we do it?
The point is that, really, we have no foolproof way of knowing. So the Flying Spaghetti Monster and intelligent design should not be 'taught' in school science curriculum, but they should be discussed. They are not scientific, and that is demonstrated easily. We should not be afraid to expose people to objectionable hypotheses, as long as we give them our reasons for rejecting them. Wikipedia articles should not turn away from alternative explanations-- it should take them head-on, and if an explanation is objectionable for some reason, that reason should be noted. This is still a balancing act, as you note.
I never said anything about open source. That was your inference-- go back and read my post. Now, it turns out that my personal opinion is that the good open source tends to do better on security-- in general-- than even the good proprietary stuff. There are some exceptions. Both Sun and IBM have invested some real brainpower thinking about how to do security right, and you can't get much more proprietary than Sun and IBM (or at least Sun and IBM of yore-- they're changing now). There are probably many other examples as well.
You miss one very important difference between ActiveX and XPI. XPI has access to the Mozilla application. ActiveX has access to the entire operating system. So a security flaw in ActiveX has much deeper security implications. A user's mistake with XPI is mitigated by the fact that XPI has no access to the OS-- at least, not directly.
Anyhow, my point is that "we" are in a state of poor computer security, because the dominant platform was not designed with security in mind. There are many other products out there where security is not an afterthought, and where core functionality is not determined by a marketing department.
BTW-- we use "obscure products" like OpenBSD, Linux, AIX, VMS, and MacOS X server to run our multi-billion dollar business. Windows has its place, but that role diminishes with every passing day.
then I would be inclined to agree with Mr. Ranum's points. But the fact is that there are lots of people out there working on Real Security. Let's see, there's OpenBSD's work to integrate cryptography as a system service, there's Neils Provos' work on systrace, there's GCC's ProPolice stack-smashing protection, there's OpenBSD's write XOR execute protection (which, BTW, Windows now has to some small extent), there are phishing mitigation features in Firefox, there are Free implementations of good authentication systems (e.g., MIT Kerberos, Heimdal), lots of programs now ship with sane defaults (ala Postfix and qmail), there are safe-string libraries of all license stripes, and on and on and on! The fact that Microsoft apparently does not use their own safe-string implementation is indicative of the problem here. Microsoft writes crap. If you want systems where security is a real concern, it's easy to find it. That's not to say that those systems are "secure"-- security is always a work in progress-- but to say that "our responses to those problems also remain the same" is disingenuous. Projects like OpenBSD (among many others mentioned above) have attempted to identify entire classes of problems, and solve them on the big-picture level instead of doing the patch-a-week thing.
He also said that "This device could make discoveries that are Earth-shattering". I think it's pretty clear why AMS is getting canned. We like the Earth in one piece!
Another important to think about is that it is never too late to go back to school. As a parent, you're probably keenly aware of time constraints, and that going back will be difficult, and costly. But now that I'm in an undergrad degree program for the second time (I have a philosophy Bachelor's; doing CS this time around), I have to say that I am constantly inspired by the older students in my program. Many of them have children, and some are even retired. These people typically run circles around the first-time students, and they're more likely to put the teacher on the spot for non-performance. I mean, we're paying for it this time!
Anyhow, I think the consensus is, once you get back into the swing of it, you'll find that the sacrifices become easier. Even if that means working from 9-5, going to class from 6-9, and then spending your entire weekend doing homework.
On the other hand, there are a number of experienced Exchange admins, myself being one, who could give you pages of information about why Exchange sucks (which is to say that Exchange has great features but terrible design). I would be very interested in an open-source groupware application that did all of the things that Exchange does, and more importantly, could start off as a drop-in replacement. So I think this question, and the following discussion, is worthwhile. Yeah, "some people" will complain. So what?
The thing about the military, they really are like the "dogs of war", namely, that there are a lot of unintended consequences. They are a wild, destructive force. The military is, of course, getting better at becoming more precise, but it is still not precise. E.g., aereal bombardment has come a long way since, say, the bombing of Munich, but missed targets are still common, and collateral damage is unavoidable. Because of this, my opinion is that you need to regard war as a necessary option, but last option. The threat of war is often enough to coerce a foe into doing the "right thing". Of course, when you're already in war, as you mention, you need to leave it to the people who are skilled in using that force. This is why it is essential to provide an objective-- without one, the military will keep on going until someone else steps in to stop it. This is the lesson that we should have learned from Vietnam.
You are right, and the article hints at this problem somewhat toward the end. The author set up "Brainology" clinics with schoolteachers and schoolchildren to talk about developing brainpower. Clearly, this is the kind of encouragement that these children should be getting, but more importantly, the fact that it had such a profound effect means that schoolteachers weren't aware of it. And I think that the educational system is abyssimal in this regard-- they do not follow scientific finding in education, either because of bureaucracy or ignorance (or both), and even worse, when they do attempt to taylor learning based on scientific evidence, they get the science wrong.
Clearly, you picked up this skill despite your environment. So there's a balance that must be reached. I recall a gym teacher in elementary school who was determined to believe I was a non-athlete because of my performance in baseball (bad), despite the fact that I was way ahead of my peers in terms of fitness (I played a lot of hockey). He would consign me to sit on the bench with "the nerds", while the "real athletes" played. What an asshole. Anyway, I ended up making varsity Cross Country my freshman year, and went on to become league MVP (undefeated runner) as a junior. I hold two course records at my high school, and they've remained standing for the past decade.
It's amazing what kids are capable of. I've always believed (and my opinion has not waved as I've become an adult) that adults are almost always the problem. The "problem" part of a "problem child" does not exist solely within the child-- it's a property of that child's relationship with others. Sometimes the problem needs to be recified with the child. Sometimes the problem needs to be rectified with the adult.
The danger is not that your children will fail, and have permanently damaged egos-- the danger is that your child will never experience failure, and thus learn the important skill of picking up the pieces and moving on. Parents naturally want to save their children from the suffering that comes from defeat (e.g., the track race on field day, the art competition, spelling bee, science fair, etc.), but this is an important experience, and one that they will eventually have, regardless of how much parents shelter them. I would much rather have my child feel crushed because he lost the Boy Scout knot-tying competition than have his first failure be at that new job out of college. The young adult who knows ego management will be in a much better position to dust himself off and carry on than the college grad who takes failure as a sign of permanent inability.
Last night's On Point featured Tom Perkins, the venture capitalist who funded Netscape, Google, AOL, and so on, and he said something that struck me-- he said that he has failed often, but that his successes outnumber his failures. He also said that his firm has a reputation of betting on the entrepeneur who has failed once before. The entrepeneur who fails, learns from it, and tries again is the kind of guy he wants to invest in.
Are you serious? Here's a conscientious programmer-- he cares about what happens to the company if someone else finds this code. He's not saying that he wants to rat the company out to their competitors; he explicitly stated that he wanted to talk to management about this. Having people around like this is a good thing, and I'd consider him a model employee.
Now it may turn out that the best thing for him to do is to keep his mouth shut. You and I obviously have very different takes on this, but I believe that it is important for him to have asked the question. Perhaps what he should do is ask management to clarify their position on unattributed code posted in forums. That would 1) solve the proble, 2) let management know that, hey, this guy cares, and, more importantly 3) it would let the lead programmer off the hook. Good career moves all around. If he goes about this diplomatically, and the lead programmer is vindictive-- well, this isn't a good place to work anyhow.
I know that we love to pretend that all software design decisions should be made on technical merit, but the reality is that most of them are not. There are always external issues. That's life.
It's a failing of BIND that it won't talk some arbitrary competing directory protocol? Are you serious? If this is a problem for you, write a gateway program. I'm quite happy leaving fluff like that out of BIND. I want it to do DNS, period. If you haven't noticed, DNS is quite complicated enough without having to worry about everyone else's protocol-of-the-month. If it were up to me, BIND would have fewer features. E.g., secure zone transfers can be accomplished better using SSH. I/AXFR should die.
Skip the RAID. RAID is for availability-- you don't need high availability at home. What you want is a fast, easy backup procedure. RAID arrays increase your likelihood of failure-- you have more disks-- the difference being that failure no longer [necessarily] equals downtime. Done right, it is expensive. This is worth it if downtime costs you more than RAID does. If you're using RAID in a machine which requires you to power off and disassemble the machine to replace the disk-- you're wasting your money. That is, unless you're running RAID 0, in which case, you're not doing anything at all to make your data more reliable-- just faster.
Since you seem to want to go the NAS route, you should just set up a file server. Get into the habit of making backups. You can automate this process somewhat. If you go the Linux or BSD route, there are lots of good, free utilities out there to help you with this. On my fileserver, I use "dump", and I dump the whole disk to another one, and for certain data, I use rsync. On my Mac at home, I use SuperDuper. My girlfriend uses Acronis. I have multiple backup sets, since disks are cheap. At work, where backups are very important, we use tape-- but that's another thing like RAID. Tape is more reliable than disk, but slow and expensive. Do you really need it?
Medicine will continue to be a non-science no matter how hard the community tries. This is untrue. While there are many questions which science is ill prepared to investigate, this is not one of them. Science cannot answer questions about metaphysics, mysticism, and so on, because those subjects have nothing for the tools of science (thesis, antithesis, and synthesis) to work on.
Medicine is complex. But that doesn't stop or discourage scientists. The world is complex. Science has always, and will always, face this issue. Medicine is a perfect subject for the application of science. Do physicists give up because certain things are not directly observable? Those working in public health have to work with what they're given.
I would much rather my doctor give me advice based on years of compounded peer-reviewed research than an opinion based on anecdote. Because, without science-- that's what you're talking about.
True, but-- we're talking about a spreadsheet. How much number-crunching capability do people need? I suspect that anyone whose computational requirements are sufficient enough to require real numerical horsepower already understands the limitations of floating-point math. Very few normal Excel users do, and I think, for them, a program that meets their expectations-- that is, a program that does arbitrary precision decimal math, accurately-- is more important than speed.
Read carefully
Although the data is still being analysed, researchers at Boston University believe they can correctly identify the sound Mr Ramsay's brain is imagining some 80% of the time.
In the next few weeks, a computer will start the task of translating his thoughts into sounds.
"We hope it will be a breakthrough," says Joe Wright of Neural Signals, which has helped develop the technology. While this is indeed promising, and I hope that this 'unlocks' this poor fellow, this 'unlocking' has not happened yet. Hopefully, when they are able to decipher these signals, he's not saying, "Kill me" over and over again.
This stirs up a pocket of long-unused brain tissue... I remember that one of my favorite games for the Sega Master System was Ghostbusters. And it was so much better than the Nintendo version that my friends had (I can't remember why I thought so). Never could quite beat Zuul, though.
SMS was a great console. I always get weird looks from people when I bring it up, but it seemed years ahead of the Nintendo at the time. I had a game that used 3-D glasses! Never saw that on the Nintendo.
The important question is, why is Excel using floating-point for Math? There are plenty of decimal math libraries out there, and I'm sure Microsoft could bang out a pretty good one themselves if they cared enough to do so.
I ran my personal web/email server (OpenBSD/Apache/Postfix/Courier IMAP) on a VIA C3 system for several years. Nice machine, relatively low power (added ~$7-10 to my electricity bill per month). Unfortunately, it died without warning one day. Motherboard is completely nonresponsive-- even with another power supply. Although it was connected to a UPS, I suppose I can't rule out power issues as being the cause. I've since replaced it with a throwaway PIII machine from the office, and that machine seems to fill the same role just as well-- including being a hell of a lot faster.
I'd give VIA another shot, only I can pretty much get an endless supply of old PIIIs from work.
NEU is a private university. But I agree with your main point, and I think that even private universities should contribute to the public good. Software patents seem antithetical to that idea.
Slashdot Mirror
Yeah, I had one of those friends, who was such a control freak of his C64, that I basically rode my bike to his house just to watch him play. I think one of the reasons I am still involved in computers is because I'd have to come home to my ailing TI-99 and make it do something new and interesting. Computer graphics were a mystery to my friend, but I was playing with sprites on my machine, and although it wasn't as cool as his store-bought stuff (I made an asteroids-type game... that, er... crashed reliably about 30 seconds in), it was mine. A word to the wise: it's no fun playing your own dungeon-crawlers. You already know how to win.
But the C64 undoubtedly had some cool games. Sid Meier's Pirates! was and still is one of the best (Xbox's version was a very good remake, IMHO), and Prince of Persia was just mindblowingly awesome. Guillotines! I think that was the first game that actually make me tremble with anxiety. Thankfully, my cousin owned one of these machines, and was so fed up with trying to use it (he could never reliably punch in the LOAD commands) that I was able to log some time on one.
Big media companies are the only ones who have the resources to police their works. Do you think the FBI would even notice if people started illegally distributing my software? Even if I made a stink about it, claimed that this was 'millions of dollars of loss' to me, do you think they would care? Now what if Disney does the same thing? The FBI listens to Disney.
Sure, I agree, copyright is a good thing. But what we have now is insane. Copyright lasts for the lifetime of the author, plus 70 years. That's essentially in perpetuity, because the public that is alive today, and likely their children and even grandchildren, will never see those works enter the public domain. I can buy an argument that works should remain copyrighted for the lifetime of the author (although I personally do not believe it should be this way), but an additional seventy years? Who does this benefit? Let's see... something that never dies... er, not vampires... Highlander? Oh, right! A big media company!
I don't argue for people making an honest living. But when you've got the government pulling Gestapo shit for the big hitters, that's not honest. That's fascist.
Three-part harmony is OK, but I'd rather just distribute sunglasses and be done with it.
GPS really gives people a false sense of confidence. I spend a lot of time in the backcountry, and I have gotten lost many times. I say "lost", but that's not really true-- I put myself in this situation for the purpose of exploration. GPS devices aren't really all that useful. You still need a compass*, and you still need to be familiar with basic backwoods travel and orienteering. Dead reckoning is the single most important orienteering skill a person can have. What do you do when your GPS batteries die? This often happens faster than you expect in cold weather.
When you are in unfamiliar territory, you tend to make irrational decisions, even when you have proper orienteering gear. Especially when you are worried about time constraints, or your hiking partner, or so on. There's a real art to learning how to calm yourself down, think rationally, and move on in these situations. The fight-or-flight instinct is a powerful force.
That said, GPS is a great invention. But it doesn't help you if you don't know how to read a topo map and apply that knowledge to where you are currently standing.
*I understand that many GPS devices come with digital compasses now. But I still wouldn't rely on it-- bring it compass; it will almost always work. Also, know direction-finding in good weather. Try to know some general rules about orienteering in bad weather. Know about prevailing winds, and so on.
Part of the problem is that very few professional designers (e.g., print designers) are also cognizant enough about software design to explain to a GIMP developer what the problems are. I am only vaguely aware of some of these issues; my main problem wihth the GIMP boils down to palettes not being truly floating. I can live with the right-click-to-do-anything mode. But designers need to do some additional things: full control of color space (RAW, CMYK, spot colors), they need color space histograms, they need to be able to easily produce color separations, they need to be able to easily put together batch transformations, they need some sophisticated marquis tools, they need to be able to mix raster and vector art, have access to Postscript, TrueType, and now OpenType fonts, and so on. I do not know if these things are available in the GIMP, because I am not a designer, nor do I regularly use the GIMP. But that's just to point out that designers do need specialized tools.
Also-- the GIMP is ugly to these people. As a programmer, I find certain elements very elegant. Most designers couldn't care less that you can write GIMP plugins in Perl; they're irritated that when they double-click on an image in InDesign, it doesn't automatically come up for editing in the GIMP. Some of these issues are petty-- but in my experience, designers tend to work in a more immediate-feeling, less rational domain. Appearance is important-- after all, they're paid because they have a highly developed aesthetic sense. Programmers are so horrendously square to many of them.
Adobe's stuff is expensive, and in many cases overrated, but they have indeed put a lot of thought into the workflow (e.g., integration between DTP app and photo editor) and appearance, and for a lot of people, this is all that matters. These facts combined with certain mailing list posts that lead me to believe that some GIMP developers suffer from a bit of a 'high horse' syndrome lead me to believe that this is why GIMP is not universally accepted, despite the fact that it is a highly capable application.
Right, but we must remember that the primary tension in science is that there are, practically speaking, an infinite number of hypotheses for any given phenomena. How do we decide which one best describes our phenomena? We pick the one that seems the best. In order to justify this decision, we use what we have already-- our scientific knowledge and our analytical tools, but in the end, it is a feeling that guides us. Analytical tools, like logic, can make wrong conclusions based on false premises, and scientific knowledge is often invalidated in the course of study. Words like "elegant" and "powerful" are how we make ourselves feel better about this process, but how else would we do it?
The point is that, really, we have no foolproof way of knowing. So the Flying Spaghetti Monster and intelligent design should not be 'taught' in school science curriculum, but they should be discussed. They are not scientific, and that is demonstrated easily. We should not be afraid to expose people to objectionable hypotheses, as long as we give them our reasons for rejecting them. Wikipedia articles should not turn away from alternative explanations-- it should take them head-on, and if an explanation is objectionable for some reason, that reason should be noted. This is still a balancing act, as you note.
I never said anything about open source. That was your inference-- go back and read my post. Now, it turns out that my personal opinion is that the good open source tends to do better on security-- in general-- than even the good proprietary stuff. There are some exceptions. Both Sun and IBM have invested some real brainpower thinking about how to do security right, and you can't get much more proprietary than Sun and IBM (or at least Sun and IBM of yore-- they're changing now). There are probably many other examples as well.
You miss one very important difference between ActiveX and XPI. XPI has access to the Mozilla application. ActiveX has access to the entire operating system. So a security flaw in ActiveX has much deeper security implications. A user's mistake with XPI is mitigated by the fact that XPI has no access to the OS-- at least, not directly.
Anyhow, my point is that "we" are in a state of poor computer security, because the dominant platform was not designed with security in mind. There are many other products out there where security is not an afterthought, and where core functionality is not determined by a marketing department.
BTW-- we use "obscure products" like OpenBSD, Linux, AIX, VMS, and MacOS X server to run our multi-billion dollar business. Windows has its place, but that role diminishes with every passing day.
then I would be inclined to agree with Mr. Ranum's points. But the fact is that there are lots of people out there working on Real Security. Let's see, there's OpenBSD's work to integrate cryptography as a system service, there's Neils Provos' work on systrace, there's GCC's ProPolice stack-smashing protection, there's OpenBSD's write XOR execute protection (which, BTW, Windows now has to some small extent), there are phishing mitigation features in Firefox, there are Free implementations of good authentication systems (e.g., MIT Kerberos, Heimdal), lots of programs now ship with sane defaults (ala Postfix and qmail), there are safe-string libraries of all license stripes, and on and on and on! The fact that Microsoft apparently does not use their own safe-string implementation is indicative of the problem here. Microsoft writes crap. If you want systems where security is a real concern, it's easy to find it. That's not to say that those systems are "secure"-- security is always a work in progress-- but to say that "our responses to those problems also remain the same" is disingenuous. Projects like OpenBSD (among many others mentioned above) have attempted to identify entire classes of problems, and solve them on the big-picture level instead of doing the patch-a-week thing.
He also said that "This device could make discoveries that are Earth-shattering". I think it's pretty clear why AMS is getting canned. We like the Earth in one piece!
Another important to think about is that it is never too late to go back to school. As a parent, you're probably keenly aware of time constraints, and that going back will be difficult, and costly. But now that I'm in an undergrad degree program for the second time (I have a philosophy Bachelor's; doing CS this time around), I have to say that I am constantly inspired by the older students in my program. Many of them have children, and some are even retired. These people typically run circles around the first-time students, and they're more likely to put the teacher on the spot for non-performance. I mean, we're paying for it this time!
Anyhow, I think the consensus is, once you get back into the swing of it, you'll find that the sacrifices become easier. Even if that means working from 9-5, going to class from 6-9, and then spending your entire weekend doing homework.
On the other hand, there are a number of experienced Exchange admins, myself being one, who could give you pages of information about why Exchange sucks (which is to say that Exchange has great features but terrible design). I would be very interested in an open-source groupware application that did all of the things that Exchange does, and more importantly, could start off as a drop-in replacement. So I think this question, and the following discussion, is worthwhile. Yeah, "some people" will complain. So what?
The thing about the military, they really are like the "dogs of war", namely, that there are a lot of unintended consequences. They are a wild, destructive force. The military is, of course, getting better at becoming more precise, but it is still not precise. E.g., aereal bombardment has come a long way since, say, the bombing of Munich, but missed targets are still common, and collateral damage is unavoidable. Because of this, my opinion is that you need to regard war as a necessary option, but last option. The threat of war is often enough to coerce a foe into doing the "right thing". Of course, when you're already in war, as you mention, you need to leave it to the people who are skilled in using that force. This is why it is essential to provide an objective-- without one, the military will keep on going until someone else steps in to stop it. This is the lesson that we should have learned from Vietnam.
You are right, and the article hints at this problem somewhat toward the end. The author set up "Brainology" clinics with schoolteachers and schoolchildren to talk about developing brainpower. Clearly, this is the kind of encouragement that these children should be getting, but more importantly, the fact that it had such a profound effect means that schoolteachers weren't aware of it. And I think that the educational system is abyssimal in this regard-- they do not follow scientific finding in education, either because of bureaucracy or ignorance (or both), and even worse, when they do attempt to taylor learning based on scientific evidence, they get the science wrong.
Clearly, you picked up this skill despite your environment. So there's a balance that must be reached. I recall a gym teacher in elementary school who was determined to believe I was a non-athlete because of my performance in baseball (bad), despite the fact that I was way ahead of my peers in terms of fitness (I played a lot of hockey). He would consign me to sit on the bench with "the nerds", while the "real athletes" played. What an asshole. Anyway, I ended up making varsity Cross Country my freshman year, and went on to become league MVP (undefeated runner) as a junior. I hold two course records at my high school, and they've remained standing for the past decade.
It's amazing what kids are capable of. I've always believed (and my opinion has not waved as I've become an adult) that adults are almost always the problem. The "problem" part of a "problem child" does not exist solely within the child-- it's a property of that child's relationship with others. Sometimes the problem needs to be recified with the child. Sometimes the problem needs to be rectified with the adult.
The danger is not that your children will fail, and have permanently damaged egos-- the danger is that your child will never experience failure, and thus learn the important skill of picking up the pieces and moving on. Parents naturally want to save their children from the suffering that comes from defeat (e.g., the track race on field day, the art competition, spelling bee, science fair, etc.), but this is an important experience, and one that they will eventually have, regardless of how much parents shelter them. I would much rather have my child feel crushed because he lost the Boy Scout knot-tying competition than have his first failure be at that new job out of college. The young adult who knows ego management will be in a much better position to dust himself off and carry on than the college grad who takes failure as a sign of permanent inability.
Last night's On Point featured Tom Perkins, the venture capitalist who funded Netscape, Google, AOL, and so on, and he said something that struck me-- he said that he has failed often, but that his successes outnumber his failures. He also said that his firm has a reputation of betting on the entrepeneur who has failed once before. The entrepeneur who fails, learns from it, and tries again is the kind of guy he wants to invest in.
Are you serious? Here's a conscientious programmer-- he cares about what happens to the company if someone else finds this code. He's not saying that he wants to rat the company out to their competitors; he explicitly stated that he wanted to talk to management about this. Having people around like this is a good thing, and I'd consider him a model employee.
Now it may turn out that the best thing for him to do is to keep his mouth shut. You and I obviously have very different takes on this, but I believe that it is important for him to have asked the question. Perhaps what he should do is ask management to clarify their position on unattributed code posted in forums. That would 1) solve the proble, 2) let management know that, hey, this guy cares, and, more importantly 3) it would let the lead programmer off the hook. Good career moves all around. If he goes about this diplomatically, and the lead programmer is vindictive-- well, this isn't a good place to work anyhow.
I know that we love to pretend that all software design decisions should be made on technical merit, but the reality is that most of them are not. There are always external issues. That's life.
MST3K made me laugh so hard, I would get asthma attacks. Why did you do this to me?
It's a failing of BIND that it won't talk some arbitrary competing directory protocol? Are you serious? If this is a problem for you, write a gateway program. I'm quite happy leaving fluff like that out of BIND. I want it to do DNS, period. If you haven't noticed, DNS is quite complicated enough without having to worry about everyone else's protocol-of-the-month. If it were up to me, BIND would have fewer features. E.g., secure zone transfers can be accomplished better using SSH. I/AXFR should die.
Skip the RAID. RAID is for availability-- you don't need high availability at home. What you want is a fast, easy backup procedure. RAID arrays increase your likelihood of failure-- you have more disks-- the difference being that failure no longer [necessarily] equals downtime. Done right, it is expensive. This is worth it if downtime costs you more than RAID does. If you're using RAID in a machine which requires you to power off and disassemble the machine to replace the disk-- you're wasting your money. That is, unless you're running RAID 0, in which case, you're not doing anything at all to make your data more reliable-- just faster.
Since you seem to want to go the NAS route, you should just set up a file server. Get into the habit of making backups. You can automate this process somewhat. If you go the Linux or BSD route, there are lots of good, free utilities out there to help you with this. On my fileserver, I use "dump", and I dump the whole disk to another one, and for certain data, I use rsync. On my Mac at home, I use SuperDuper. My girlfriend uses Acronis. I have multiple backup sets, since disks are cheap. At work, where backups are very important, we use tape-- but that's another thing like RAID. Tape is more reliable than disk, but slow and expensive. Do you really need it?
Medicine is complex. But that doesn't stop or discourage scientists. The world is complex. Science has always, and will always, face this issue. Medicine is a perfect subject for the application of science. Do physicists give up because certain things are not directly observable? Those working in public health have to work with what they're given.
I would much rather my doctor give me advice based on years of compounded peer-reviewed research than an opinion based on anecdote. Because, without science-- that's what you're talking about.
True, but-- we're talking about a spreadsheet. How much number-crunching capability do people need? I suspect that anyone whose computational requirements are sufficient enough to require real numerical horsepower already understands the limitations of floating-point math. Very few normal Excel users do, and I think, for them, a program that meets their expectations-- that is, a program that does arbitrary precision decimal math, accurately-- is more important than speed.
In the next few weeks, a computer will start the task of translating his thoughts into sounds.
"We hope it will be a breakthrough," says Joe Wright of Neural Signals, which has helped develop the technology. While this is indeed promising, and I hope that this 'unlocks' this poor fellow, this 'unlocking' has not happened yet. Hopefully, when they are able to decipher these signals, he's not saying, "Kill me" over and over again.
This stirs up a pocket of long-unused brain tissue... I remember that one of my favorite games for the Sega Master System was Ghostbusters. And it was so much better than the Nintendo version that my friends had (I can't remember why I thought so). Never could quite beat Zuul, though.
SMS was a great console. I always get weird looks from people when I bring it up, but it seemed years ahead of the Nintendo at the time. I had a game that used 3-D glasses! Never saw that on the Nintendo.
The important question is, why is Excel using floating-point for Math? There are plenty of decimal math libraries out there, and I'm sure Microsoft could bang out a pretty good one themselves if they cared enough to do so.
I ran my personal web/email server (OpenBSD/Apache/Postfix/Courier IMAP) on a VIA C3 system for several years. Nice machine, relatively low power (added ~$7-10 to my electricity bill per month). Unfortunately, it died without warning one day. Motherboard is completely nonresponsive-- even with another power supply. Although it was connected to a UPS, I suppose I can't rule out power issues as being the cause. I've since replaced it with a throwaway PIII machine from the office, and that machine seems to fill the same role just as well-- including being a hell of a lot faster.
I'd give VIA another shot, only I can pretty much get an endless supply of old PIIIs from work.
NEU is a private university. But I agree with your main point, and I think that even private universities should contribute to the public good. Software patents seem antithetical to that idea.