People REALLY need to get away from the idea that IP laws that prevent you from copying somebody else's work inhibit innovation.
Strawman.
You apparently forgot that new technology is based on and interacts with existing technology. If someone patented the recording and playback of a signal that can be displayed as a visual image before the VCR was invented, is the VCR really innovative?
If you don't think so then you need to check your premises.
Bad IP laws prevent you from building on other people's work and that inhibits innovation.
Any guesses? I'm thinking somewhere in the high six figures. That's a *lot* of work on that inlet/fan. It's a pity the exhaust eyelets are such Rube Goldberg contraptions.
I'd be willing to bet that there's another seven figure to be spent before the first loon would be willing to take a ride in it. Depending on how well that money was spent, it could be me...:)
Regards,
Ross
I had a former collegue that just couldn't grasp the use of design patterns, and thus despised the concept. He also couldn't solve large scale programming problems and wasn't much of a software architect in general. Then, the book anti-patterns comes out which he latched onto as some sort of weapon against the evil design patterns.
Ya know, I'll bet he loved the "Golden Hammer" antipattern. For those in the cheap seats: the golden hammer antipattern observes that people who get a shiny new tool tend to look at all new problems as if the tool can solve them. I.e. if the only tool you've got is a hammer, all of your problems start to look like nails.
This particular application of this anti-pattern (as a universal pattern debunking argument) is particularly ironic.
Drugs are bad because the user hurts people to get their drugs. Legalizing drugs won't stop people from committing other crimes to buy them.
What kind of crack are you smoking? Must be pretty good stuff. When cocaine costs what prescription morphine costs (a close relative and about 1/20th the street price of clean coke) you'll only have to steal 1/20th as much stuff to get your fix.
That's a lot less crime from addicts stealing to get their next fix. And that's just one of the positive side effects from legalization to say nothing of reduced addiction rates, reduced o/d events, drugs sales generating tax revenue, money saved on smaller police departments, etc.
crank is methamphetamines (think super strong diet pills or ritalin which we soak unruly children with), while heroin is a narcotic in the same family with morphine and coedine.
Except that the number of addicts has gone up (as a proportion of population) under prohibition. Take a look at Amsterdam now. The number of users per 100,000 population is identical to the US and the number of addicts is lower. All that with free clean needles, legal marijuana, and a much lower police presence and enforcement effort.
One of the neat and interesting effects of prohibition is that by making the drugs scarce and much more costly, users are forced to use higher risk means of injesting the drugs to maximize the number of highs per unit of drug.
The higher risk is of addiction and overdose. Crack will get you more highs than freebase which will get you more highs than snorting which will get you more highs than spiking your whiskey and the risk of addiction follows the same curve. The amount of THC in marijuana can be directly correlated with enforcement effort. The more the police crack down the stronger the pot gets. When the police settle back, so does the pot. Far be it from me to imply causation.
If the real goal is of the war on some drugs is to reduce the number of users or addicts, the current effort is an abject failure and should be immediately abandoned as too embarassing for anyone to pretend it's doing any good.
Luckily for some, the goal of the war on some drugs is actually to make certain that substantial police forces around the country (and world) have constant employment and legislators have regular excuses to pass more laws eroding your Constitutional rights (you did know that the 4th Amendment is all but gone WRT search and seizure?).
It's not that disinfo is a source of reliable information or not. I have both disinfo books, and each one is about half-filled with comments I've written in the margin about the crack smoking that the author is doing. The other half is some really impressive outside-the-box thinking and writing.
The thing is, disinfo is the only source of a lot of information that happens to be unpalatable to the mainstream press. Read this particular article and judge it on it's merits. Specifically, judge it on it's bibliography and the merits of the group of people who are quoted and cited.
I agree that much of what disinfo has to offer is crap. The fact that 5% of disinfo is fantastically useful stuff that you won't get anywhere else is why I'll put up with the effort of filtering through the crap.
BTW, if your remark that disinfo is usually crap is an argument that the FDA rule really is a good one, that's called an "ad hominem" argumentative fallacy. Just for future reference.
The FDA rule that you quote is an extremely weak rule, is the only FDA rule on the topic of preventing BSE, and the serious problems with it are discussed in some depth in the article.
The fact is that the FDA has not yet banned animal protien in animal food products and really really needs to before things get bad. As in awful.
Chronic Wasting Disease, Mad Cow Disease, and Creutzfeldt-Jakob Disease are all forms of Transmissible Spongiform Encephalopathies (TSE's) and they really ought to frighten you.
The parts that ought to frighten you don't necessarily seem that bad until all of the factors are taken in at once:
1) total incurability of infected people/animals.
2) near indestructability of prions (1100F for hours, etc.)
3) ability of TSE's to cross species (scrapie in sheep, BSE in cattle, CJD in people, TME in mink, PSE in pigs, etc.) and it's all the same group of diseases. They differ in the speed that they cause damage, but that's about it.
4) The US meat/poultry industry practice of rendering slaughterhouse remains and *DOWNER CATTLE* into feed for other animals and poultry. This rendering process always includes brain and spinal cord tissue in the resulting product.
Basically, if the US meat industry hasn't found BSE in cattle, it's because it doesn't want to. The fact that downer cattle are never checked for BSE should piss just about everyone off. When Dr. Richard Marsh at the University of Wisconsin injected US cattle with TME infected US mink tissues, the cattle didn't act like the British cattle, they simply collapsed, looking like any other downer cow.
The US industry takes those downer cows, never checks to see what might have brought them down, grinds them up, brains and all, and feeds them to chickens, pigs, other cattle.
The scariest part is that slower forms of CJD (the human disease) look exactly like Alzheimer's and other forms of progressive dementia. In a Yale study, 6 of 46 Alzheimer's patients (13%!) were CJD positive at autopsy.
CWD (deer, elk, etc.) is almost certainly picked up from raiding contaminated feed meant for livestock. At least, that's my marginally informed position on the topic. It has to be injested somehow and it's a distorted animal protien so these wild herbivorous animals have to be consuming animal proteins to get sick.
The European Union has now banned all animal products in livestock feed, but the US FDA resists this simple and absolutely necessary step to halt the progress of the perfect pathogen throughout the United States.
An article that does a much better job of describing these problems and substantiating these arguments is at: "mad cows and englishmen". I hope it worries you and that you tell someone else about it. Even better, tell your congresscritter about it and what you think about it.
The individual is the best arbiter of their own mental "health". If this little old lady is better able to function in her daily life because she uses a game to help her deal with her grief, more power to her.
It's a rather scary fact that so many people (just like the parent poster) are willing to concede determination of their capacity as a moral agent to an insular group whose degrees are based on the softest of soft science and the worst metaethical foundation I have heard of.
Mental illness is a determination that the "ill" does not have the capacity to make moral decisions, excusing their resulting behavior as amoral and placing them under the total control of the psychiatric industry. I don't buy it for one minute. The reasons for their decisions may not make sense to you and I, but the reasons are still there. Mental illness is the worst of a paternalistic assumption that norms can be enforced by the majority. When wrongs are committed, they should be punished. The rest is crap.
Any other set of assumptions leads to dangerous conclusions like preventing retarted people from having romantic relationships, drugging children in classrooms to make them more compliant, the dehumanization of our elderly in retirement homes, etc., etc.
I use all of these things (Windows XP, Office XP (with autosave), Mozilla, Gnome, OpenOffice, Mozilla (again), and do heavy duty enterprise server side software development), with anti-aliased fonts turned on everywhere and a 1600x1200 screen. I have a 850MHz P3 laptop, and they all scream, even on that absolutely beautiful screen. The only time I'm bothered by latency is during application load and I blame the 5400rpm hard drive, not the CPU.
I haven't bought a game in a while, so we'll see what that's like when I get Star Wars Galaxies next month, but I suspect that, like everything else, I'll get acceptable performance a notch or two down from the highest level of detail and I won't notice the compromise at all during regular game play.
I have no intention of replacing this laptop unless it breaks and cannot be fixed. In fact, I'm about to buy some laptops for the developers in my little startup company, and I'm probably going to go looking for used Thinkpads that I can put under an IBM maintenance contract because they are so capable (and one like my A21p is only about $1000 +$200/year maintenance).
Those fabs are likely to be foundries for companies that design marketable CPU's. And where will these designs come from? The same places that are currently designing marketable CPU's with the addition of some additional startups right here in the US.
The culture of the far east doesn't have what it takes to forge ahead into brave unknowns and consistently produce risky and excellent designs. That culture excels at honing, or incrementally improving existing designs and products. It's as if the whole culture was run by committee on every scale.
Back to the point, if China builds huge fabs, the most profitable things they'll make are chips designed right here in the US whose companies use the Chinese fabs as outsourced fabrication (foundries). Worry about them? I'm counting on them. Those huge fabs will make the chip startup company a possibility again by dramatically lowering barriers to entry.
But you forgot to include amortized marketing and development costs in your price. And for the XBOX effort, those are substantial.
Microsoft just had to go and use a TV instead of a PC monitor as the primary display, so that's not a video card you're going to pick up for $15. Custom motherboard, custom chipset?, custom physical interfaces for peripherals. All of that costs money. Though those original spec hard drives are getting pretty darned cheap at this point. And marketing...? Are you kidding? They're dropping hundreds of millions on marketing.
So your argument is that there is no marginal loss on each XBOX sold. Fine, that's probably true. But they can still be losing money on XBOX that can be characterized as $X per unit when you take into account all of those other ways that microsoft has had to spend money to get this thing in the hands of consumers. And when you look at MS's business unit financials, that's exactly what you see.
You realize that all current rockets are inherently unstable in the way you describe? I.e. the thrust is all at one end and has to be carefully controlled to keep it pushing through the Cg. If anything, this rocket is substantially more stable than the Saturn V precisely because the Cp is at least close to the Cg (and so small errors in thrust direction act on a small moment arm and induce a correspondingly small rotation).
Tug-type rockets (think two thrusters on a frame at the top of the payload with their exhausts directed down past the payload, just like James Bond's jetpack) will be practical and enormously easier to control when they operate entirely in outer space. The need to be aerodynamic puts a kink in their use as a lift vehicle (don't think NASA didn't consider it).
Start with the simplest manifold, which is a sphere. That only needs four colors. Which should make sense because maps are 2d projections of a mostly spherical surface and maps only need four colors.
Now add a topological "handle" to the sphere (poke a hole in it and make a torus or donut). Now you need six colors, though there's no proof of that, only a counter example for five.
It turns out that for each "handle" or tunnel in the manifold you need zero, one, or two more colors. If the next handle doesn't interact (stick two donuts together at the edge), then you don't need more colors. If you poke another hole in the donut from the edge into the middle, you do need another color.
Another way to think about it. Take a map that requires four colors to draw. Now add a tunnel in place between two areas of the map that have the same color and say that the border between the two endpoints is halfway through the tunnel (so that two areas with the same color touch). You'll have to add another color to print that map now. Repeat for five colors, then six, etc. You can always add another tunnel that will force two regions of the same color to be in contact, so for an arbitrarily complex manifold, you can't specify how many colors it will take.
If I was to design a combat spacecraft control system, you would hear the other spacecraft exploding because the system would note that the spacecraft was exploding and would make the noises in your headset.
I probably wouldn't make the bridge chair shake however.
Any developer who says they're getting 25 hours of on-task time a week is either in the top 0.1% or smoking crack.
Developers getting 3 hours of on-task time a day are doing just fine as long as the code they write during those 15 hours a week is well designed and well tested and...
The reality for developers is, in a 40 hour week: meetings, trips to the restroom, casual conversations, personal errands and phone calls, getting settled in, sitting back and massaging the wrists, etc... All prevent you from working a full 40 hours in a week. And if you try, you'll burn out. Taking breaks is a good thing.
The best you can do as a dev manager is: work to make sure that communication is effective (vertical and horizontal), team and individual goals are clear, and the time spent on-task is synchronized so that people can work together most of the time and aren't interrupted by other people who aren't on-task.
Still, if you get 20 hours of on-task work a week fro each employee (only 50% of the time spent in the office), you're going to have rather amazing results (because your team will be in the top 5% in high tech according to SEI research).
So stop griping about 3-5 hours a day. That's the range from normal to exceptional. Expecting 8 hours a day is about as sane as being upset because you aren't using your laser printer to it's full monthly capacity.
A deliberate decision to create a culture where people are valued can be made at the top (CEO, board, etc.).
Companies that show loyalty as part of a culture of giving a rat's ass about each other tend to get fanatical loyalty in return. Especially in this day and age when so many employers don't meet people's need to have a place where they know they're respected and can feel a bit safer than an unlit alley at 2:00am.
Plenty of management won't see it because they're deeply cynical and project that cynicism onto others, plenty of employees won't see it because they simply have no idea what loyalty means. But that's why culture fit is fully as important as skills on the "interview goals" list (for both employee and employer).
Hell, this afternoon, I'm going to start attending a bartender class. I can live on $2500-$3000 a month and I'll have the time to work on my own software projects without risk that my work will be entrapped by an overzealous non-compete interpretation or otherwise greedy ex-employer.
That, and I'll meet a whole lot more women as a bartender than as a sit in a cube and code geek.:)
But what the customer needs *will* change over time, and if your ability to adjust to that change isn't there, you aren't going to meet the needs of your customer with the result of your effort.
There are lots of ways to manage change. Some work better than others. Allowing customers to throw out the spec every three months doesn't work. Holding a customer to every detail in a spec written two years ago doesn't work either. Something in between is the answer.
Agile processes work very well to manage change by giving your customer lots of visibility into the process and giving them a lot of information to make change decisions. How important is this new feature you say you need? Is it more important that these five things we are planning to work on next week?
That visibility into what's going on also automatically manages the customer expectations extraordinarily effectively. You want to see a happy customer, I'll show you a customer who has very accurate scheduling numbers, strong feature commitments, and high confidence that if their needs change, that what we're planning to deliver can change to meet those needs.
XP is really one of a larger group of "Agile" methodoligies.
The basic tenet of agile development is to throw out the big waterfall and shrink the development cycle down to lots of tiny little cycles. The big benefit here is to eliminate the "convergence" system integration step. After a certain point, the product is always ready to ship so the cost of "premature convergence" is eliminated and you get lots and lots of opportunities to reorganize requirements. You'll have a list of a few hundred things to do, pick the most important two or three, get them done, put it in front of customers (who will give you more things to do), repeat.
Feature creep is automatically managed because you also get lots of chances to see how well your estimation is working and when marketing makes new demands, they get to help decide where those new demands get put into the todo list. When someone has a chance to immediately understand the impact of a new feature, it makes it much easier to understand whether or not this is really a "gotta have" or not.
XP won't make sense unless you buy books and push them under people's noses, and even then, some won't like it because it appears to demand a complete upheaval of what you're currently doing. Be sure to emphasize that many of the practices can improve your system when applied individually. You don't need the whole thing to be better.
If you get "Rapid Development" by Steve McConnell, an older authority on development processes, most of the processes in the chapters on dev process are agile. Evolving prototype, etc. There are lots of ways to present this that won't hit the buttons of change averse people.
XP itself really is fairly extreme. There are lots of ways to improve a process that don't have to toally disrupt a culture but which can really improve the way things get done.
The SEI (Software Engineering Institude, Carnegie Mellon) TSP (Team Software Process) research numbers indicate that two good people workinmg together will spend 15% less time writing fully tested code than those same two developers working independently. This number does not just count the amount of time spent in the coding phase, but the total time spent getting the code to work.
The big gain comes with the automatic code review which will dramatically reduce the number of errors introduced during the coding stage (that later have to be removed during compiling, testing, and after deployment) and improves the tests by providing another perspective who will suggest test cases that the other would not have thought of alone. The earlier these things happen the better. With pair programming, they're happening immediately.
Finally, there's a social aspect to pair programming that keeps the intensity up. If you're allowed to go off to your desk and do your work alone, most people will take lots of mini breaks, read a little slashdot, take care of a few bills, etc. With pair programming, you're forced to respect another person's time and that directly encourages better time management, even from developers who don't have good time management.
Don't pretend that you can spend all day doing pair programming, it's exhausting. But if you treat it like a series of one-on-one meetings and make time between pair programming sessions to take care of other things that need doing, it works amazingly well.
Then, you also get cross training issues. The junior codes, the senior sits back and helps them become familiar with the system and getting things done in it.
Now, there are ways to do pair programming badly. One of the worst is senior develops, junior sits back bored to tears (in Senior/Junior pairings, the Junior developers must be coding most, but not all, of the time). Also, big differences in typing speed can cause hair pulling and destroyed nerves, so anyone can't just pair with anyone on most teams (though this is a good excuse to get typing training for slower team members).
In the end, however, pair programming is supported by real world numbers. If you have a chance, take a PSP (Personal Software Process) class. During the class, much of what seems like foolishness will become wisdom.
Slashdot Mirror
You apparently forgot that new technology is based on and interacts with existing technology. If someone patented the recording and playback of a signal that can be displayed as a visual image before the VCR was invented, is the VCR really innovative?
If you don't think so then you need to check your premises.
Bad IP laws prevent you from building on other people's work and that inhibits innovation.
Regards, Ross
Any guesses? I'm thinking somewhere in the high six figures. That's a *lot* of work on that inlet/fan. It's a pity the exhaust eyelets are such Rube Goldberg contraptions. I'd be willing to bet that there's another seven figure to be spent before the first loon would be willing to take a ride in it. Depending on how well that money was spent, it could be me... :)
Regards,
Ross
Ya know, I'll bet he loved the "Golden Hammer" antipattern. For those in the cheap seats: the golden hammer antipattern observes that people who get a shiny new tool tend to look at all new problems as if the tool can solve them. I.e. if the only tool you've got is a hammer, all of your problems start to look like nails.
This particular application of this anti-pattern (as a universal pattern debunking argument) is particularly ironic.
Regards,
Ross
What kind of crack are you smoking? Must be pretty good stuff. When cocaine costs what prescription morphine costs (a close relative and about 1/20th the street price of clean coke) you'll only have to steal 1/20th as much stuff to get your fix.
That's a lot less crime from addicts stealing to get their next fix. And that's just one of the positive side effects from legalization to say nothing of reduced addiction rates, reduced o/d events, drugs sales generating tax revenue, money saved on smaller police departments, etc.
Regards, Rosscrank is methamphetamines (think super strong diet pills or ritalin which we soak unruly children with), while heroin is a narcotic in the same family with morphine and coedine.
Regards,
Ross
Except that the number of addicts has gone up (as a proportion of population) under prohibition. Take a look at Amsterdam now. The number of users per 100,000 population is identical to the US and the number of addicts is lower. All that with free clean needles, legal marijuana, and a much lower police presence and enforcement effort.
One of the neat and interesting effects of prohibition is that by making the drugs scarce and much more costly, users are forced to use higher risk means of injesting the drugs to maximize the number of highs per unit of drug.
The higher risk is of addiction and overdose. Crack will get you more highs than freebase which will get you more highs than snorting which will get you more highs than spiking your whiskey and the risk of addiction follows the same curve. The amount of THC in marijuana can be directly correlated with enforcement effort. The more the police crack down the stronger the pot gets. When the police settle back, so does the pot. Far be it from me to imply causation.
If the real goal is of the war on some drugs is to reduce the number of users or addicts, the current effort is an abject failure and should be immediately abandoned as too embarassing for anyone to pretend it's doing any good.
Luckily for some, the goal of the war on some drugs is actually to make certain that substantial police forces around the country (and world) have constant employment and legislators have regular excuses to pass more laws eroding your Constitutional rights (you did know that the 4th Amendment is all but gone WRT search and seizure?).
Regards,
Ross
In this case, PSE refers to Porcine Spongiform Encephalopathy, which is most definitely a TSE.
Regards,
Ross
Dime store psychology thrown about without invitation is a sure sign of a closet control freak. :)
Regards,
Ross
It's not that disinfo is a source of reliable information or not. I have both disinfo books, and each one is about half-filled with comments I've written in the margin about the crack smoking that the author is doing. The other half is some really impressive outside-the-box thinking and writing.
The thing is, disinfo is the only source of a lot of information that happens to be unpalatable to the mainstream press. Read this particular article and judge it on it's merits. Specifically, judge it on it's bibliography and the merits of the group of people who are quoted and cited.
I agree that much of what disinfo has to offer is crap. The fact that 5% of disinfo is fantastically useful stuff that you won't get anywhere else is why I'll put up with the effort of filtering through the crap.
BTW, if your remark that disinfo is usually crap is an argument that the FDA rule really is a good one, that's called an "ad hominem" argumentative fallacy. Just for future reference.
Regards,
Ross
Read the article:
"mad cows and englishmen".
The FDA rule that you quote is an extremely weak rule, is the only FDA rule on the topic of preventing BSE, and the serious problems with it are discussed in some depth in the article.
The fact is that the FDA has not yet banned animal protien in animal food products and really really needs to before things get bad. As in awful.
Regards,
Ross
The parts that ought to frighten you don't necessarily seem that bad until all of the factors are taken in at once:
1) total incurability of infected people/animals.
2) near indestructability of prions (1100F for hours, etc.)
3) ability of TSE's to cross species (scrapie in sheep, BSE in cattle, CJD in people, TME in mink, PSE in pigs, etc.) and it's all the same group of diseases. They differ in the speed that they cause damage, but that's about it.
4) The US meat/poultry industry practice of rendering slaughterhouse remains and *DOWNER CATTLE* into feed for other animals and poultry. This rendering process always includes brain and spinal cord tissue in the resulting product.
Basically, if the US meat industry hasn't found BSE in cattle, it's because it doesn't want to. The fact that downer cattle are never checked for BSE should piss just about everyone off. When Dr. Richard Marsh at the University of Wisconsin injected US cattle with TME infected US mink tissues, the cattle didn't act like the British cattle, they simply collapsed, looking like any other downer cow.
The US industry takes those downer cows, never checks to see what might have brought them down, grinds them up, brains and all, and feeds them to chickens, pigs, other cattle.
The scariest part is that slower forms of CJD (the human disease) look exactly like Alzheimer's and other forms of progressive dementia. In a Yale study, 6 of 46 Alzheimer's patients (13%!) were CJD positive at autopsy.
CWD (deer, elk, etc.) is almost certainly picked up from raiding contaminated feed meant for livestock. At least, that's my marginally informed position on the topic. It has to be injested somehow and it's a distorted animal protien so these wild herbivorous animals have to be consuming animal proteins to get sick.
The European Union has now banned all animal products in livestock feed, but the US FDA resists this simple and absolutely necessary step to halt the progress of the perfect pathogen throughout the United States.
An article that does a much better job of describing these problems and substantiating these arguments is at: "mad cows and englishmen". I hope it worries you and that you tell someone else about it. Even better, tell your congresscritter about it and what you think about it.
Regards, Ross
I ran out of moderator points yesterday, or you'd be +1 insightful...
Regards,
Ross
The individual is the best arbiter of their own mental "health". If this little old lady is better able to function in her daily life because she uses a game to help her deal with her grief, more power to her.
:)
It's a rather scary fact that so many people (just like the parent poster) are willing to concede determination of their capacity as a moral agent to an insular group whose degrees are based on the softest of soft science and the worst metaethical foundation I have heard of.
Mental illness is a determination that the "ill" does not have the capacity to make moral decisions, excusing their resulting behavior as amoral and placing them under the total control of the psychiatric industry. I don't buy it for one minute. The reasons for their decisions may not make sense to you and I, but the reasons are still there. Mental illness is the worst of a paternalistic assumption that norms can be enforced by the majority. When wrongs are committed, they should be punished. The rest is crap.
Any other set of assumptions leads to dangerous conclusions like preventing retarted people from having romantic relationships, drugging children in classrooms to make them more compliant, the dehumanization of our elderly in retirement homes, etc., etc.
Sorry bout that. Rant over
Regards,
Ross
I use all of these things (Windows XP, Office XP (with autosave), Mozilla, Gnome, OpenOffice, Mozilla (again), and do heavy duty enterprise server side software development), with anti-aliased fonts turned on everywhere and a 1600x1200 screen. I have a 850MHz P3 laptop, and they all scream, even on that absolutely beautiful screen. The only time I'm bothered by latency is during application load and I blame the 5400rpm hard drive, not the CPU.
I haven't bought a game in a while, so we'll see what that's like when I get Star Wars Galaxies next month, but I suspect that, like everything else, I'll get acceptable performance a notch or two down from the highest level of detail and I won't notice the compromise at all during regular game play.
I have no intention of replacing this laptop unless it breaks and cannot be fixed. In fact, I'm about to buy some laptops for the developers in my little startup company, and I'm probably going to go looking for used Thinkpads that I can put under an IBM maintenance contract because they are so capable (and one like my A21p is only about $1000 +$200/year maintenance).
Regards,
Ross
Those fabs are likely to be foundries for companies that design marketable CPU's. And where will these designs come from? The same places that are currently designing marketable CPU's with the addition of some additional startups right here in the US.
The culture of the far east doesn't have what it takes to forge ahead into brave unknowns and consistently produce risky and excellent designs. That culture excels at honing, or incrementally improving existing designs and products. It's as if the whole culture was run by committee on every scale.
Back to the point, if China builds huge fabs, the most profitable things they'll make are chips designed right here in the US whose companies use the Chinese fabs as outsourced fabrication (foundries). Worry about them? I'm counting on them. Those huge fabs will make the chip startup company a possibility again by dramatically lowering barriers to entry.
Regards,
Ross
But you forgot to include amortized marketing and development costs in your price. And for the XBOX effort, those are substantial.
Microsoft just had to go and use a TV instead of a PC monitor as the primary display, so that's not a video card you're going to pick up for $15. Custom motherboard, custom chipset?, custom physical interfaces for peripherals. All of that costs money. Though those original spec hard drives are getting pretty darned cheap at this point. And marketing...? Are you kidding? They're dropping hundreds of millions on marketing.
So your argument is that there is no marginal loss on each XBOX sold. Fine, that's probably true. But they can still be losing money on XBOX that can be characterized as $X per unit when you take into account all of those other ways that microsoft has had to spend money to get this thing in the hands of consumers. And when you look at MS's business unit financials, that's exactly what you see.
Regards,
Ross
You realize that all current rockets are inherently unstable in the way you describe? I.e. the thrust is all at one end and has to be carefully controlled to keep it pushing through the Cg. If anything, this rocket is substantially
more stable than the Saturn V precisely because the Cp is at least close to the Cg (and so small errors in thrust direction act on a small moment arm and induce a correspondingly small rotation).
Tug-type rockets (think two thrusters on a frame at the top of the payload with their exhausts directed down past the payload, just like James Bond's jetpack) will be practical and enormously easier to control when they operate entirely in outer space. The need to be aerodynamic puts a kink in their use as a lift vehicle (don't think NASA didn't consider it).
Regards,
Ross
Start with the simplest manifold, which is a sphere. That only needs four colors. Which should make sense because maps are 2d projections of a mostly spherical surface and maps only need four colors.
:)
Now add a topological "handle" to the sphere (poke a hole in it and make a torus or donut). Now you need six colors, though there's no proof of that, only a counter example for five.
It turns out that for each "handle" or tunnel in the manifold you need zero, one, or two more colors. If the next handle doesn't interact (stick two donuts together at the edge), then you don't need more colors. If you poke another hole in the donut from the edge into the middle, you do need another color.
Another way to think about it. Take a map that requires four colors to draw. Now add a tunnel in place between two areas of the map that have the same color and say that the border between the two endpoints is halfway through the tunnel (so that two areas with the same color touch). You'll have to add another color to print that map now. Repeat for five colors, then six, etc. You can always add another tunnel that will force two regions of the same color to be in contact, so for an arbitrarily complex manifold, you can't specify how many colors it will take.
Wait, that actually is a proof.
Regards,
Ross
If I was to design a combat spacecraft control system, you would hear the other spacecraft exploding because the system would note that the spacecraft was exploding and would make the noises in your headset.
I probably wouldn't make the bridge chair shake however.
Regards,
Ross
Any developer who says they're getting 25 hours of on-task time a week is either in the top 0.1% or smoking crack.
Developers getting 3 hours of on-task time a day are doing just fine as long as the code they write during those 15 hours a week is well designed and well tested and...
The reality for developers is, in a 40 hour week: meetings, trips to the restroom, casual conversations, personal errands and phone calls, getting settled in, sitting back and massaging the wrists, etc... All prevent you from working a full 40 hours in a week. And if you try, you'll burn out. Taking breaks is a good thing.
The best you can do as a dev manager is: work to make sure that communication is effective (vertical and horizontal), team and individual goals are clear, and the time spent on-task is synchronized so that people can work together most of the time and aren't interrupted by other people who aren't on-task.
Still, if you get 20 hours of on-task work a week fro each employee (only 50% of the time spent in the office), you're going to have rather amazing results (because your team will be in the top 5% in high tech according to SEI research).
So stop griping about 3-5 hours a day. That's the range from normal to exceptional. Expecting 8 hours a day is about as sane as being upset because you aren't using your laser printer to it's full monthly capacity.
Regards,
Ross
A deliberate decision to create a culture where people are valued can be made at the top (CEO, board, etc.).
Companies that show loyalty as part of a culture of giving a rat's ass about each other tend to get fanatical loyalty in return. Especially in this day and age when so many employers don't meet people's need to have a place where they know they're respected and can feel a bit safer than an unlit alley at 2:00am.
Plenty of management won't see it because they're deeply cynical and project that cynicism onto others, plenty of employees won't see it because they simply have no idea what loyalty means. But that's why culture fit is fully as important as skills on the "interview goals" list (for both employee and employer).
Regards,
Ross
Hell, this afternoon, I'm going to start attending a bartender class. I can live on $2500-$3000 a month and I'll have the time to work on my own software projects without risk that my work will be entrapped by an overzealous non-compete interpretation or otherwise greedy ex-employer.
:)
That, and I'll meet a whole lot more women as a bartender than as a sit in a cube and code geek.
Regards,
Ross
But what the customer needs *will* change over time, and if your ability to adjust to that change isn't there, you aren't going to meet the needs of your customer with the result of your effort.
There are lots of ways to manage change. Some work better than others. Allowing customers to throw out the spec every three months doesn't work. Holding a customer to every detail in a spec written two years ago doesn't work either. Something in between is the answer.
Agile processes work very well to manage change by giving your customer lots of visibility into the process and giving them a lot of information to make change decisions. How important is this new feature you say you need? Is it more important that these five things we are planning to work on next week?
That visibility into what's going on also automatically manages the customer expectations extraordinarily effectively. You want to see a happy customer, I'll show you a customer who has very accurate scheduling numbers, strong feature commitments, and high confidence that if their needs change, that what we're planning to deliver can change to meet those needs.
Regards,
Ross
XP is really one of a larger group of "Agile" methodoligies.
The basic tenet of agile development is to throw out the big waterfall and shrink the development cycle down to lots of tiny little cycles. The big benefit here is to eliminate the "convergence" system integration step. After a certain point, the product is always ready to ship so the cost of "premature convergence" is eliminated and you get lots and lots of opportunities to reorganize requirements. You'll have a list of a few hundred things to do, pick the most important two or three, get them done, put it in front of customers (who will give you more things to do), repeat.
Feature creep is automatically managed because you also get lots of chances to see how well your estimation is working and when marketing makes new demands, they get to help decide where those new demands get put into the todo list. When someone has a chance to immediately understand the impact of a new feature, it makes it much easier to understand whether or not this is really a "gotta have" or not.
XP won't make sense unless you buy books and push them under people's noses, and even then, some won't like it because it appears to demand a complete upheaval of what you're currently doing. Be sure to emphasize that many of the practices can improve your system when applied individually. You don't need the whole thing to be better.
If you get "Rapid Development" by Steve McConnell, an older authority on development processes, most of the processes in the chapters on dev process are agile. Evolving prototype, etc. There are lots of ways to present this that won't hit the buttons of change averse people.
XP itself really is fairly extreme. There are lots of ways to improve a process that don't have to toally disrupt a culture but which can really improve the way things get done.
Regards,
Ross
The SEI (Software Engineering Institude, Carnegie Mellon) TSP (Team Software Process) research numbers indicate that two good people workinmg together will spend 15% less time writing fully tested code than those same two developers working independently. This number does not just count the amount of time spent in the coding phase, but the total time spent getting the code to work.
The big gain comes with the automatic code review which will dramatically reduce the number of errors introduced during the coding stage (that later have to be removed during compiling, testing, and after deployment) and improves the tests by providing another perspective who will suggest test cases that the other would not have thought of alone. The earlier these things happen the better. With pair programming, they're happening immediately.
Finally, there's a social aspect to pair programming that keeps the intensity up. If you're allowed to go off to your desk and do your work alone, most people will take lots of mini breaks, read a little slashdot, take care of a few bills, etc. With pair programming, you're forced to respect another person's time and that directly encourages better time management, even from developers who don't have good time management.
Don't pretend that you can spend all day doing pair programming, it's exhausting. But if you treat it like a series of one-on-one meetings and make time between pair programming sessions to take care of other things that need doing, it works amazingly well.
Then, you also get cross training issues. The junior codes, the senior sits back and helps them become familiar with the system and getting things done in it.
Now, there are ways to do pair programming badly. One of the worst is senior develops, junior sits back bored to tears (in Senior/Junior pairings, the Junior developers must be coding most, but not all, of the time). Also, big differences in typing speed can cause hair pulling and destroyed nerves, so anyone can't just pair with anyone on most teams (though this is a good excuse to get typing training for slower team members).
In the end, however, pair programming is supported by real world numbers. If you have a chance, take a PSP (Personal Software Process) class. During the class, much of what seems like foolishness will become wisdom.
Regards,
Ross