> The people I've heard from who complain > about Linux not being preinstalled by OEMs > aren't complaining about paying for Linux; > they're complaining about paying for Windows.
However, since the cost of Windows is a cost they bear while attempting to acquire Linux, they are still complaining about the *market* for Linux.
You'll probably need to think about that one for a while before you finally realise you just plain don't understand economics in the first place.
> As for your other comments, c has significance > in special relativity aside from being the speed > of light in vacuo.
What a cute little straw man. Trouble is, I'm talking about the academic context, where precise calculations are made while assuming zero gravity and perfect vacuum to demonstrate an understanding of certain basic concepts.
Likewise, in an academic context, perfect competition is useful to demonstrate that you understand how to calculate all the appropriate curves.
If you get all bound up in "that would never happen", imagine how CS majors feel when their instructors have them implement an object heirarchy that ultimately produces several breeds of dog with different barks. What a senseless waste of time. In fifteen years of software development, I have never had to implement anything remotely similar to this stupidity, and yet it *continues* to be taught year after year.
> Also, do markets even have stable equilibria in > the first place, never mind approaching or > achieving them?
Equilibrium is the point in any market where supply equals demand.
Depending on how you define "have" and "stable", that either means all markets have it, or no markets have it.
> It's only recently that you have a lot of people > who could *really* freely choose what OS to use.
I don't see your scenario as representing freedom. If they are interested in cheap solutions, there is an economic factor which restricts their choice. One might even go so far as to say these people are being forced to choose Linux, whether they want it or not.
> So when principles of micro texts > discuss perfect competition and > market equilibrium, they're only > engaging in wishful thinking?
No, they're discussing an extreme case that never actually happens. Like, for example, when a physics textbook talks about the speed of light in a perfect vacuum. There is no such thing as a perfect vacuum. It is a convenience that simplifies the situation for purposes of academic discussion.
Likewise, there is no such thing as perfect competition, and there is no such thing as market equlibrium except in the most fleeting sense. You do, of course, have to pass *through* market equilibrium when you move from too much supply to too much demand. It just does not persist long enough to actually affect pricing in any meaningful sense.
> As for buyers being dissatisfied > with market equilibrium,
We don't have market equilibrium. Anywhere. See above.
> are you saying that people will > object to paying $5 for Fedora > Core 5 (or downloading it)?
They DO. Haven't you noticed this?
First they buy a PC and complain that they couldn't get it preinstalled. They can, from the right OEMs, but they don't want to try that hard.
Then they go out to a store and complain that they can't pick it up off the shelf. They can, at the right stores, but they don't want to try that hard.
Then they look at the web site and complain that they don't want to wait six weeks. They can get overnight shipping, if they want, but they don't want to spend that much.
Then they look at the ISO downloads and complain that they don't want to spend hours downloading it. And when someone else just hands them the disks, they complain that they have to reboot and set their BIOS to boot off the DVD drive.
Meanwhile, not having Linux installed, they go around complaining that they "have" to use Windows. Whinge, whinge, whinge. Pardon me if I just want them to STFU.
That's your opportunity for increasing market share. And as far as I'm concerned, you're welcome to it.
I recall discussing it with company management at CSC in 1994. So not only was it on the table, it was being taken seriously.
Trouble is, it was a bad idea then, and it's still a bad idea. There's a massive difference between hiring an administrative staff that knows Windows and one that knows Linux. If you're hiring a dozen people, it makes about a quarter million dollar difference annually, and you're also hiring from a much smaller pool of available applicants. When you consider that using Linux instead of Windows is only saving you about $5,000 a year in licensing fees... well, that's just plain stupid.
> There is a saying "It takes 20 years to make > an overnight success."
That saying is about people. Software generally becomes obsolete in twenty years. Some people claim this shows how great Linux is, but strangely enough, the people who didn't like Linux ten years ago *still* don't like it... for pretty much the same reasons. So I don't really see Linux making any headway.
> It seems like Linux is the only desktop os > that has ever gained marketshare against > Microsoft's monopoly.
This appearance is generated through unverifiable means. When someone buys a PC with Windows installed, installs Linux as a dual-boot option, and then never uses the Linux boot... which I have seen done many times... is that desktop market share?
I'll bet they count it. I'll also bet the surveys on Linux market share are largely conducted among a self-selected group that is likely to use Linux anyway. But let's pretend they're a standard distribution anyway, okay? It will make Linux look good.
Lying is lying, and wrong is wrong. Your intentions may be good, but we all know what's paved with those.
It is in perfect competition, which is precisely what makes it a special case: the marginal cost intersects the average cost curve at its lowest point. This is not a situation you would normally encounter in the real world, but it is theoretically possible - very much like travelling at the speed of light.
> When someone else comes along with something new,
New? Linux is about 15 years old.
> it is always compared to the existing system to > see if it measures up.
So... if Linux has been unable to measure up sufficiently to take significant desktop share over the past fifteen years, what exactly makes you think it's going to suddenly become important?
> Solid state memory is able to hold as much, in > smaller spaces, and is more flexible.
And a *damn* sight more expensive. I used solid state disk drives when they cost tens of thousands of dollars for a few hundred megabytes. They were great, but honestly - just not worth the money.
> What about the assumptions of perfect competition?
The purpose of perfect competition is to illustrate the concept of...
> What about the concept of market equilibrium?
Market equilibrium occurs when the supply of a good exactly matches the demand for a good. Given perfect competition, which as already stated cannot exist in the real world, market equilibrium occurs immediately and persists indefinitely. The result of this is that everyone charges the exact same price, which coincidentally is exactly what it costs them to produce the good in the first place.
However, market equilibrium is bad for everyone involved. The buyer cannot find a bargain, and the seller cannot make a profit. So it is in everyone's best interest to *destroy* market equilibrium wherever it threatens to arise, effectively guaranteeing that it never does. This implies that it is also in everyone's best interest to destroy any possibility of perfect competition, as well.
> How long do markets stay in equilibrium?
As long as supply and demand continue to track one another exactly, which may for all practical purposes be assumed to mean "never".
The open source movement often appears to idealise perfect competition and market equilibrium, incidentally. This is precisely why their rhetoric is flawed: their goal is not in anyone's best interest. They claim otherwise primarily by redefining their terms in the middle of the argument; if you say that the open source vision of the industry means you can't make any money writing software, they respond that you can make money documenting and supporting the software. Which is not writing software, so they haven't really answered the question. But pay no attention to the man behind the curtain.
> It is the computer science grads that live > in the real world and the economic grads > that are stuck in fantasy land.
It is interesting to watch people try to claim that their particular fantasy, in which they are intimately embroiled, is in fact the real world. It is, of course, EVERYONE ELSE who is lost in fantasy. After all, you can look right at it and tell immediately that it's not the same as the "reality" in which you live.
CS grads see one side of the world, economists see another. They're both looking at the same world. The problem lies in the question of how they respond to it.
> we have to deal with an imperfect world
So do economists. However, the economist knows that this imperfect world is not anybody's fault, and cannot be fixed. The CS grad, on the other hand, rants about how it would all be different if only people would do this or that or the other. If only the market behaved more rationally! If only human nature were radically different! If only everyone were more like ME!
> Economics is the classic discipline based on > assumption and wishful thinking.
I don't see a whole lot of wishful thinking in economics. There's a very strong streak of inevitability and fatalism throughout the field; things unfold this way no matter what you do, and trying to fix it will generally just make things worse. Whenever this aphorism is tested by some well-meaning jackass, behold! It makes things worse.
But without actual PROOF, of course, this massive mountain of evidence doesn't mean squat... so it can be ignored.
Now, THAT looks like wishful thinking. But maybe it's just me.
> someone should do the math, and calculate > how many jobs we need to create in the U.S. > to achieve 100% employment and outsource > enough jobs to create those jobs.
This is one reason why economics has been gaining and CS has been declining.
Fifteen people graduate college majoring in either economics or computer science. Three of them majored in economics. Therefore, for every economics graduate, there are four CS graduates.
So if you convince five more people to major in economics, does that therefore convince twenty more people to major in CS?
To an economist, it is ludicrous to suggest such a thing. They understand that correlation is not causation, and that something can be the result of human action while simultaneously *never* being the result of human desire.
To a CS graduate, this is an interesting proposal that bears examination, because they live in a world of pure theory and mathematical idealism. The only way you can know something is to prove it mathematically, and economics - while amazingly accurate - is every bit as unproven and anecdotal as astrology.
The word is beginning to notice that the latter group tends to waste more money on things that will "obviously" never work. Unfortunately, that latter group ignores the "obviously" because every once in a while they figure out that it *does* in fact work... if you do it THIS way.
Both groups are essential, but we're constantly trying to eliminate one in favor of the other.
> In my experience, this has amost always > been Microsoft's fault.
This has never happened. The fictional example is about a change in the standards which prevents existing browsers from displaying compliant documents. No such change has ever been made to the HTML standard.
However, when changes have been made to a web standard and a particular browser does not support them - like, for example, the "first-word" or "first-letter" pseudo-elements in CSS not being supported in IE6 - nobody goes to the standards body and demands that the standards be altered to accurately reflect what IE is displaying. They go to Microsoft and demand that these optional elements be fully and completely supported, because they're still *in* the CSS standard. That's what you should do. You complain to your client software's manufacturer, because you want the standard supported.
But strangely, when Microsoft changes a standard they own and control, and Samba does not support it - the Samba team are not the ones taking the heat. Microsoft are clearly to blame, because they are the ones who created the standard that Samba is not following. The simple fact that they have no obligation whatsoever to support the Samba team is irrelevant, because Samba is cool and Microsoft are teh suck.
The open source community is, of course, fully within their rights to behave inconsistently and irrationally. They are, however, not allowed to behave this way *and* deny doing so... unless they elect to identify themselves as a religion.
> Their discussions are of course > conducted on open mailing lists > and Microsoft could easily establish > this co-operation any time they wanted
This effectively amounts to a demand that Samba should be a Microsoft product. Let's look at a series of events.
1. A protocol is designed. 2. A client is built to implement the protocol. 3. The protocol is changed by its designers. 4. The client no longer works.
Whose fault is it that the client no longer works? Should you go to the protocol developers and complain that your client doesn't work anymore? What should they do about it? Should they revert to the previous protocol version because your client doesn't work? Should they volunteer their time and energy to fix your client?
Imagine that you can't view web pages anymore, because of some change made to the HTML standard. Who is to blame for this? Is it the standards body's fault? Or is it exclusively the responsibility of the people who make your browser?
> Why are people still using IE, even > the most uneducated users must have > heard of alternative browsers by now.
I have very little difficulty with most of these "bugs" that get reported. They don't happen to me. The OSS devotees with whom I am acquainted are frequently claiming I lead some kind of charmed life.
The fact is, most people do not practice secure browsing habits, which is a problem in ANY browser. There is nothing on the planet that can protect you from clicking a link to sucker.we-steal-your-money.com if you're stupid enough to do it. IE gives me more than enough ability to determine where this link really leads, and since I check the link before clicking it, I don't click those links. The idea that OSS advocates consider this some sort of abnormal magical protection is truly frightening.
So why do I use IE? Because security is a process, not a product. I use IE safely. Other browsers often try to "protect" me by concealing the information I could use to make my own decision, making the decision for me instead. Some products actively prevent me from making my own decision, deliberately overriding my requests because they simply can't believe I might actually *want* to do what I said. This isn't security, it's infantilism. You can keep it. I'll use something that does what I tell it, even when it thinks I'm crazy. Like IE.
Once upon a time, we used to say "UNIX doesn't stop you from doing stupid things, because that would stop you from doing clever things". We criticised Windows for preventing the user from doing clever things. Microsoft listened, and changed their stance significantly. So now we're criticising Windows for letting you do stupid things. How ironic.
When the Samba developers do not maintain close communication with Microsoft about potential changes to roaming profiles and implement support for those changes, it is not Microsoft's fault when things break. It is the Samba team's fault.
When you cannot get support for your PDC because it is running a non-Microsoft platform, that is not Microsoft's fault. It is your platform provider's fault.
The problem in this scenario is that you are relying on software which has been provided by volunteers, and those volunteers are not providing the level of support you expect.
Go complain to the volunteers. When they say it's Microsoft's fault, call that the bullshit it is and demand better from this "superior" community. When they can't deliver, and they can't, maybe you'll revisit the question of why people use Microsoft software. Maybe you'll even think about it with something remotely resembling actual intelligence.
Want to see my math? Microsoft licensing fees make up less than 0.25% of my annual operating budget. One day of downtime represents more than 0.27% of my annual operations. So if I can save just one day a year of downtime by using an all-Microsoft solution, I win.
> Cringely seems to be making a mountain out > of a molehill (and I'm not a MS fanboy).
I agree, and I'm a Microsoft Partner.
Basically, the question they were facing is a very real question. Whenever you have several people involved in a startup and one of them faces health issues, it is highly important to create a contingency plan in the event of their death.
A common version of such a plan is to take out life insurance on the individual which pays sufficient benefits to purchase his share of the company outright. This generally gives his heirs something in excess of the cash value of the shares, which makes them happy, and keeps the shares themselves within the company family - which makes it happy.
The basics: make the patent holder prove the validity of the patent when challenged, and pay for any challenge they lose. If they win, the challenger pays.
This makes it expensive to lose a patent challenge no matter which side you take, so nobody will want to lose one. A patent will be applied for only when the company feels confident it will not lose a challenge, and challenged only when the challenger feels confident that the patent is invalid. Over time, the system heals itself.
Patents serve a very useful purpose. It is possible to develop something over many years of hard work which is considered impossible by the industry, but that suddenly becomes completely obvious upon publication of your solution. Copyright will protect YOUR solution, but if someone else writes his own software to use the same solution, copyright doesn't help. Trademark doesn't help. Your only possible protection is a patent.
Development, whether in software or hardware, is a front-loaded process. You spend a lot of time and money and effort on it before you have anything at all. The patent system is what protects you from making that investment and having someone else duplicate it; since he doesn't have that investment, his overhead is lower, and he can charge a lower price. This effectively forces you, the inventor, out of the market for your own product.
This is wrong. We all know this is wrong. And the patent system, broken as it is, remains the only thing we have to prevent this.
> No matter how smart you are, or how smart you think you are, > you are never, never so smart that you are always right and > don't have to listen to others.
But he does listen. It's just a waste of time.
Bill Boss: "We need a solution to our supply shortage."
Joe Idiot: "I've been thinking about that, and what's really going on there is that we just don't have enough of what we need."
Well, DUH.
But the original poster is playing the wrong game. The boss doesn't want to help you find a solution. He wants you to demonstrate sufficient understanding of the problem that he can trust you to find the solution on your own. It's a contest for who best understands the problem. You need to take small steps forward until the boss is satisfied that you understand and goes away. Then you solve the problem by yourself, without bothering the boss about it.
And that's the difference between being intelligent and being smart. Smart makes you successful. Intelligent just makes you weird.
I never said "pure technical merit". I said "better". Technical merit isn't the only thing that makes a product better, which is precisely why OS/2 and Netscape were left scratching their heads in Microsoft's wake. They thought being the technically superior product meant they would win. Unfortunately, most people don't quite understand "technically superior", and they go with whatever looks good. Netscape's default browser color was a drab medium gray, while IE's was gleaming white. The difference in contrast alone spelled "better" to many people... especially older people. And that wasn't even remotely a technical superiority; it was just one line initialising the default background color.
But it made a much bigger difference to a lot more people than any kind of standards compliance.
> the bottom line is that they are MY kids, > and its MY decision.
Bingo.
My son Logan shot his first pedestrian in GTA:SA when he was eight months old.
I have zero intention of censoring content. He needs to learn how to handle it. I will help him learn that, and I believe the earlier he learns to handle it, the better.
If other children come over to the house, different rules apply. I will speak with their parents to see where *they* want the line drawn at my house. I will speak with my son about how other children aren't always allowed to see and read and do the same things he is.
And, of course, if my son proves unable to handle something... a little censorship might actually be necessary. I'm opposed to it, but I'm not stupidly crusading for an ideal against all sense and reason.
> How about instead of looking down on > youth for acting the way you admittedly > acted, try to remember why you acted > that way and understand them.
I don't care why they act that way. I care THAT they act that way. I have a right to require specific behavior at my place of business. I can post a "no loitering" sign. I can post a "no smoking" sign. I can even post a "no public bathrooms" sign. If I can forbid bathroom usage at my place of business, I think I can forbid pretty much anything I want. So if I don't like the way you behave, I can tell you not to behave that way.
And even if you aren't really a threat, I don't know that. How do you tell the difference between the Trenchcoat Mafia at Columbine and the average punks at any high school? You don't. You can't. They are indistinguishable, even though they are indeed very different. And that's why we treat all the "outsider" kids in public school like felons.
If you want to fix some sort of moral outrage, I suggest you start there. Keeping kids from loitering in front of stores is a silly cause to fight; fight the system that is destroying our creative original thinkers before they have the educational foundation to start thinking anything valuable. The eventual cost to society is simply inestimable.
I actually see evidence of a different perception driving companies' adoption of Linux.
"What he did."
Let's say that fifty companies are all thinking about adopting Linux. Making a smart decision about this takes time and money. But if you just wait, sooner or later one of them will make a decision. If you then make the same decision, you don't have to spend time or money to make that decision. You simply copy what they've done, and it will be almost like you spent as much time and money as they did.
There are subtleties, of course, like choice of distribution and usage within the company. My company uses a basic philosophy that any server inside the firewall is Windows, and any server outside the firewall is Linux. If all you see is that we took down our Windows web servers and replaced them with Linux, you might get the wrong idea and start trying to replace your internal servers with Linux. That's not what we did, so you're not getting the same benefits we get. (It may, of course, turn out to be a Big Win. Even a blind dog finds a bone sometimes.)
The flaw in this thinking is that you don't control the decision-making process. You don't know whether your competition made a long detailed study of Linux v. Windows in the general sense, or compared the performance of two file servers for a month, or just got drunk and flipped a coin. You are basing your decision purely and simply on the belief that your competition is smarter than you are. This belief usually indicates that there are things you are already doing wrong, and you know it. Copying someone else's right move will not magically undo all your previous bad ones.
This isn't new or unique; companies have *always* blindly copied the work of other companies, so they don't have to come up with their own ideas. Coming up with ideas is HARD, dammit. But it seems to me that you might get better results asking "what is my competition doing WRONG, and how can we do it right?"
That's what Microsoft is usually asking. There are two things that count: first and best. It's better to be first, because once you're first, nobody else can come along later and be first; if you're not the first, you *have* to be the best.
Microsoft rarely produces the first of anything. Do the math.
Slashdot Mirror
> The people I've heard from who complain
> about Linux not being preinstalled by OEMs
> aren't complaining about paying for Linux;
> they're complaining about paying for Windows.
However, since the cost of Windows is a cost they bear while attempting to acquire Linux, they are still complaining about the *market* for Linux.
You'll probably need to think about that one for a while before you finally realise you just plain don't understand economics in the first place.
> As for your other comments, c has significance
> in special relativity aside from being the speed
> of light in vacuo.
What a cute little straw man. Trouble is, I'm talking about the academic context, where precise calculations are made while assuming zero gravity and perfect vacuum to demonstrate an understanding of certain basic concepts.
Likewise, in an academic context, perfect competition is useful to demonstrate that you understand how to calculate all the appropriate curves.
If you get all bound up in "that would never happen", imagine how CS majors feel when their instructors have them implement an object heirarchy that ultimately produces several breeds of dog with different barks. What a senseless waste of time. In fifteen years of software development, I have never had to implement anything remotely similar to this stupidity, and yet it *continues* to be taught year after year.
> Also, do markets even have stable equilibria in
> the first place, never mind approaching or
> achieving them?
Equilibrium is the point in any market where supply equals demand.
Depending on how you define "have" and "stable", that either means all markets have it, or no markets have it.
> It's only recently that you have a lot of people
> who could *really* freely choose what OS to use.
I don't see your scenario as representing freedom. If they are interested in cheap solutions, there is an economic factor which restricts their choice. One might even go so far as to say these people are being forced to choose Linux, whether they want it or not.
> So when principles of micro texts
> discuss perfect competition and
> market equilibrium, they're only
> engaging in wishful thinking?
No, they're discussing an extreme case that never actually happens. Like, for example, when a physics textbook talks about the speed of light in a perfect vacuum. There is no such thing as a perfect vacuum. It is a convenience that simplifies the situation for purposes of academic discussion.
Likewise, there is no such thing as perfect competition, and there is no such thing as market equlibrium except in the most fleeting sense. You do, of course, have to pass *through* market equilibrium when you move from too much supply to too much demand. It just does not persist long enough to actually affect pricing in any meaningful sense.
> As for buyers being dissatisfied
> with market equilibrium,
We don't have market equilibrium. Anywhere. See above.
> are you saying that people will
> object to paying $5 for Fedora
> Core 5 (or downloading it)?
They DO. Haven't you noticed this?
First they buy a PC and complain that they couldn't get it preinstalled. They can, from the right OEMs, but they don't want to try that hard.
Then they go out to a store and complain that they can't pick it up off the shelf. They can, at the right stores, but they don't want to try that hard.
Then they look at the web site and complain that they don't want to wait six weeks. They can get overnight shipping, if they want, but they don't want to spend that much.
Then they look at the ISO downloads and complain that they don't want to spend hours downloading it. And when someone else just hands them the disks, they complain that they have to reboot and set their BIOS to boot off the DVD drive.
Meanwhile, not having Linux installed, they go around complaining that they "have" to use Windows. Whinge, whinge, whinge. Pardon me if I just want them to STFU.
That's your opportunity for increasing market share. And as far as I'm concerned, you're welcome to it.
> This is relatively recent (last 4-5 years?)
I recall discussing it with company management at CSC in 1994. So not only was it on the table, it was being taken seriously.
Trouble is, it was a bad idea then, and it's still a bad idea. There's a massive difference between hiring an administrative staff that knows Windows and one that knows Linux. If you're hiring a dozen people, it makes about a quarter million dollar difference annually, and you're also hiring from a much smaller pool of available applicants. When you consider that using Linux instead of Windows is only saving you about $5,000 a year in licensing fees... well, that's just plain stupid.
> There is a saying "It takes 20 years to make
> an overnight success."
That saying is about people. Software generally becomes obsolete in twenty years. Some people claim this shows how great Linux is, but strangely enough, the people who didn't like Linux ten years ago *still* don't like it... for pretty much the same reasons. So I don't really see Linux making any headway.
> It seems like Linux is the only desktop os
> that has ever gained marketshare against
> Microsoft's monopoly.
This appearance is generated through unverifiable means. When someone buys a PC with Windows installed, installs Linux as a dual-boot option, and then never uses the Linux boot... which I have seen done many times... is that desktop market share?
I'll bet they count it. I'll also bet the surveys on Linux market share are largely conducted among a self-selected group that is likely to use Linux anyway. But let's pretend they're a standard distribution anyway, okay? It will make Linux look good.
Lying is lying, and wrong is wrong. Your intentions may be good, but we all know what's paved with those.
It is in perfect competition, which is precisely what makes it a special case: the marginal cost intersects the average cost curve at its lowest point. This is not a situation you would normally encounter in the real world, but it is theoretically possible - very much like travelling at the speed of light.
> When someone else comes along with something new,
New? Linux is about 15 years old.
> it is always compared to the existing system to
> see if it measures up.
So... if Linux has been unable to measure up sufficiently to take significant desktop share over the past fifteen years, what exactly makes you think it's going to suddenly become important?
> Solid state memory is able to hold as much, in
> smaller spaces, and is more flexible.
And a *damn* sight more expensive. I used solid state disk drives when they cost tens of thousands of dollars for a few hundred megabytes. They were great, but honestly - just not worth the money.
> What about the assumptions of perfect competition?
The purpose of perfect competition is to illustrate the concept of...
> What about the concept of market equilibrium?
Market equilibrium occurs when the supply of a good exactly matches the demand for a good. Given perfect competition, which as already stated cannot exist in the real world, market equilibrium occurs immediately and persists indefinitely. The result of this is that everyone charges the exact same price, which coincidentally is exactly what it costs them to produce the good in the first place.
However, market equilibrium is bad for everyone involved. The buyer cannot find a bargain, and the seller cannot make a profit. So it is in everyone's best interest to *destroy* market equilibrium wherever it threatens to arise, effectively guaranteeing that it never does. This implies that it is also in everyone's best interest to destroy any possibility of perfect competition, as well.
> How long do markets stay in equilibrium?
As long as supply and demand continue to track one another exactly, which may for all practical purposes be assumed to mean "never".
The open source movement often appears to idealise perfect competition and market equilibrium, incidentally. This is precisely why their rhetoric is flawed: their goal is not in anyone's best interest. They claim otherwise primarily by redefining their terms in the middle of the argument; if you say that the open source vision of the industry means you can't make any money writing software, they respond that you can make money documenting and supporting the software. Which is not writing software, so they haven't really answered the question. But pay no attention to the man behind the curtain.
> It is the computer science grads that live
> in the real world and the economic grads
> that are stuck in fantasy land.
It is interesting to watch people try to claim that their particular fantasy, in which they are intimately embroiled, is in fact the real world. It is, of course, EVERYONE ELSE who is lost in fantasy. After all, you can look right at it and tell immediately that it's not the same as the "reality" in which you live.
CS grads see one side of the world, economists see another. They're both looking at the same world. The problem lies in the question of how they respond to it.
> we have to deal with an imperfect world
So do economists. However, the economist knows that this imperfect world is not anybody's fault, and cannot be fixed. The CS grad, on the other hand, rants about how it would all be different if only people would do this or that or the other. If only the market behaved more rationally! If only human nature were radically different! If only everyone were more like ME!
> Economics is the classic discipline based on
> assumption and wishful thinking.
I don't see a whole lot of wishful thinking in economics. There's a very strong streak of inevitability and fatalism throughout the field; things unfold this way no matter what you do, and trying to fix it will generally just make things worse. Whenever this aphorism is tested by some well-meaning jackass, behold! It makes things worse.
But without actual PROOF, of course, this massive mountain of evidence doesn't mean squat... so it can be ignored.
Now, THAT looks like wishful thinking. But maybe it's just me.
> someone should do the math, and calculate
> how many jobs we need to create in the U.S.
> to achieve 100% employment and outsource
> enough jobs to create those jobs.
This is one reason why economics has been gaining and CS has been declining.
Fifteen people graduate college majoring in either economics or computer science. Three of them majored in economics. Therefore, for every economics graduate, there are four CS graduates.
So if you convince five more people to major in economics, does that therefore convince twenty more people to major in CS?
To an economist, it is ludicrous to suggest such a thing. They understand that correlation is not causation, and that something can be the result of human action while simultaneously *never* being the result of human desire.
To a CS graduate, this is an interesting proposal that bears examination, because they live in a world of pure theory and mathematical idealism. The only way you can know something is to prove it mathematically, and economics - while amazingly accurate - is every bit as unproven and anecdotal as astrology.
The word is beginning to notice that the latter group tends to waste more money on things that will "obviously" never work. Unfortunately, that latter group ignores the "obviously" because every once in a while they figure out that it *does* in fact work... if you do it THIS way.
Both groups are essential, but we're constantly trying to eliminate one in favor of the other.
> In my experience, this has amost always
> been Microsoft's fault.
This has never happened. The fictional example is about a change in the standards which prevents existing browsers from displaying compliant documents. No such change has ever been made to the HTML standard.
However, when changes have been made to a web standard and a particular browser does not support them - like, for example, the "first-word" or "first-letter" pseudo-elements in CSS not being supported in IE6 - nobody goes to the standards body and demands that the standards be altered to accurately reflect what IE is displaying. They go to Microsoft and demand that these optional elements be fully and completely supported, because they're still *in* the CSS standard. That's what you should do. You complain to your client software's manufacturer, because you want the standard supported.
But strangely, when Microsoft changes a standard they own and control, and Samba does not support it - the Samba team are not the ones taking the heat. Microsoft are clearly to blame, because they are the ones who created the standard that Samba is not following. The simple fact that they have no obligation whatsoever to support the Samba team is irrelevant, because Samba is cool and Microsoft are teh suck.
The open source community is, of course, fully within their rights to behave inconsistently and irrationally. They are, however, not allowed to behave this way *and* deny doing so... unless they elect to identify themselves as a religion.
> Their discussions are of course
> conducted on open mailing lists
> and Microsoft could easily establish
> this co-operation any time they wanted
This effectively amounts to a demand that Samba should be a Microsoft product. Let's look at a series of events.
1. A protocol is designed.
2. A client is built to implement the protocol.
3. The protocol is changed by its designers.
4. The client no longer works.
Whose fault is it that the client no longer works? Should you go to the protocol developers and complain that your client doesn't work anymore? What should they do about it? Should they revert to the previous protocol version because your client doesn't work? Should they volunteer their time and energy to fix your client?
Imagine that you can't view web pages anymore, because of some change made to the HTML standard. Who is to blame for this? Is it the standards body's fault? Or is it exclusively the responsibility of the people who make your browser?
> Did you try the POC exploit in TFA?
Yes, and it works in IE7 too.
> The whole point of it is that you see a javascript link
You don't find JavaScript links immediately suspicious?
> Why are people still using IE, even
> the most uneducated users must have
> heard of alternative browsers by now.
I have very little difficulty with most of these "bugs" that get reported. They don't happen to me. The OSS devotees with whom I am acquainted are frequently claiming I lead some kind of charmed life.
The fact is, most people do not practice secure browsing habits, which is a problem in ANY browser. There is nothing on the planet that can protect you from clicking a link to sucker.we-steal-your-money.com if you're stupid enough to do it. IE gives me more than enough ability to determine where this link really leads, and since I check the link before clicking it, I don't click those links. The idea that OSS advocates consider this some sort of abnormal magical protection is truly frightening.
So why do I use IE? Because security is a process, not a product. I use IE safely. Other browsers often try to "protect" me by concealing the information I could use to make my own decision, making the decision for me instead. Some products actively prevent me from making my own decision, deliberately overriding my requests because they simply can't believe I might actually *want* to do what I said. This isn't security, it's infantilism. You can keep it. I'll use something that does what I tell it, even when it thinks I'm crazy. Like IE.
Once upon a time, we used to say "UNIX doesn't stop you from doing stupid things, because that would stop you from doing clever things". We criticised Windows for preventing the user from doing clever things. Microsoft listened, and changed their stance significantly. So now we're criticising Windows for letting you do stupid things. How ironic.
Let me make this perfectly clear.
When the Samba developers do not maintain close communication with Microsoft about potential changes to roaming profiles and implement support for those changes, it is not Microsoft's fault when things break. It is the Samba team's fault.
When you cannot get support for your PDC because it is running a non-Microsoft platform, that is not Microsoft's fault. It is your platform provider's fault.
The problem in this scenario is that you are relying on software which has been provided by volunteers, and those volunteers are not providing the level of support you expect.
Go complain to the volunteers. When they say it's Microsoft's fault, call that the bullshit it is and demand better from this "superior" community. When they can't deliver, and they can't, maybe you'll revisit the question of why people use Microsoft software. Maybe you'll even think about it with something remotely resembling actual intelligence.
Want to see my math? Microsoft licensing fees make up less than 0.25% of my annual operating budget. One day of downtime represents more than 0.27% of my annual operations. So if I can save just one day a year of downtime by using an all-Microsoft solution, I win.
> Cringely seems to be making a mountain out
> of a molehill (and I'm not a MS fanboy).
I agree, and I'm a Microsoft Partner.
Basically, the question they were facing is a very real question. Whenever you have several people involved in a startup and one of them faces health issues, it is highly important to create a contingency plan in the event of their death.
A common version of such a plan is to take out life insurance on the individual which pays sufficient benefits to purchase his share of the company outright. This generally gives his heirs something in excess of the cash value of the shares, which makes them happy, and keeps the shares themselves within the company family - which makes it happy.
See, some of us *can* think rationally.
I keep talking about this, but nobody listens.
http://www.darklock.com/blog?p=55
The basics: make the patent holder prove the validity of the patent when challenged, and pay for any challenge they lose. If they win, the challenger pays.
This makes it expensive to lose a patent challenge no matter which side you take, so nobody will want to lose one. A patent will be applied for only when the company feels confident it will not lose a challenge, and challenged only when the challenger feels confident that the patent is invalid. Over time, the system heals itself.
Patents serve a very useful purpose. It is possible to develop something over many years of hard work which is considered impossible by the industry, but that suddenly becomes completely obvious upon publication of your solution. Copyright will protect YOUR solution, but if someone else writes his own software to use the same solution, copyright doesn't help. Trademark doesn't help. Your only possible protection is a patent.
Development, whether in software or hardware, is a front-loaded process. You spend a lot of time and money and effort on it before you have anything at all. The patent system is what protects you from making that investment and having someone else duplicate it; since he doesn't have that investment, his overhead is lower, and he can charge a lower price. This effectively forces you, the inventor, out of the market for your own product.
This is wrong. We all know this is wrong. And the patent system, broken as it is, remains the only thing we have to prevent this.
It still needs to be fixed. I've blogged about this a little: http://www.darklock.com/blog/?p=55
That video rocked.
"I'm getting drunk! Are there any girls there?!"
> No matter how smart you are, or how smart you think you are,
> you are never, never so smart that you are always right and
> don't have to listen to others.
But he does listen. It's just a waste of time.
Bill Boss: "We need a solution to our supply shortage."
Joe Idiot: "I've been thinking about that, and what's really going on there is that we just don't have enough of what we need."
Well, DUH.
But the original poster is playing the wrong game. The boss doesn't want to help you find a solution. He wants you to demonstrate sufficient understanding of the problem that he can trust you to find the solution on your own. It's a contest for who best understands the problem. You need to take small steps forward until the boss is satisfied that you understand and goes away. Then you solve the problem by yourself, without bothering the boss about it.
And that's the difference between being intelligent and being smart. Smart makes you successful. Intelligent just makes you weird.
Nobody on the Microsoft side has ever suggested that Linus or ESR or RMS is the devil.
Oh, wait, I did. But it's not like I was serious.
I never said "pure technical merit". I said "better". Technical merit isn't the only thing that makes a product better, which is precisely why OS/2 and Netscape were left scratching their heads in Microsoft's wake. They thought being the technically superior product meant they would win. Unfortunately, most people don't quite understand "technically superior", and they go with whatever looks good. Netscape's default browser color was a drab medium gray, while IE's was gleaming white. The difference in contrast alone spelled "better" to many people... especially older people. And that wasn't even remotely a technical superiority; it was just one line initialising the default background color.
But it made a much bigger difference to a lot more people than any kind of standards compliance.
> the bottom line is that they are MY kids,
> and its MY decision.
Bingo.
My son Logan shot his first pedestrian in GTA:SA when he was eight months old.
I have zero intention of censoring content. He needs to learn how to handle it. I will help him learn that, and I believe the earlier he learns to handle it, the better.
If other children come over to the house, different rules apply. I will speak with their parents to see where *they* want the line drawn at my house. I will speak with my son about how other children aren't always allowed to see and read and do the same things he is.
And, of course, if my son proves unable to handle something... a little censorship might actually be necessary. I'm opposed to it, but I'm not stupidly crusading for an ideal against all sense and reason.
> How about instead of looking down on
> youth for acting the way you admittedly
> acted, try to remember why you acted
> that way and understand them.
I don't care why they act that way. I care THAT they act that way. I have a right to require specific behavior at my place of business. I can post a "no loitering" sign. I can post a "no smoking" sign. I can even post a "no public bathrooms" sign. If I can forbid bathroom usage at my place of business, I think I can forbid pretty much anything I want. So if I don't like the way you behave, I can tell you not to behave that way.
And even if you aren't really a threat, I don't know that. How do you tell the difference between the Trenchcoat Mafia at Columbine and the average punks at any high school? You don't. You can't. They are indistinguishable, even though they are indeed very different. And that's why we treat all the "outsider" kids in public school like felons.
If you want to fix some sort of moral outrage, I suggest you start there. Keeping kids from loitering in front of stores is a silly cause to fight; fight the system that is destroying our creative original thinkers before they have the educational foundation to start thinking anything valuable. The eventual cost to society is simply inestimable.
You're right.
I *should* be able to get these updates for Windows in one central and well-defined place, and Microsoft isn't doing it.
Which leads one to wonder...
I see exactly why Microsoft doesn't provide MySQL and PHP updates in their update services. That only makes sense.
I see exactly why MySQL and PHP updates are provided in Linux update services. That only makes sense.
But why doesn't ANYONE out there run an update service for *Windows* that provides MySQL and PHP updates?
In fact, why don't the Linux update services *already* run corollary update services for Windows computers?
Looks to me like *everyone's* dropping the ball here.
I actually see evidence of a different perception driving companies' adoption of Linux.
"What he did."
Let's say that fifty companies are all thinking about adopting Linux. Making a smart decision about this takes time and money. But if you just wait, sooner or later one of them will make a decision. If you then make the same decision, you don't have to spend time or money to make that decision. You simply copy what they've done, and it will be almost like you spent as much time and money as they did.
There are subtleties, of course, like choice of distribution and usage within the company. My company uses a basic philosophy that any server inside the firewall is Windows, and any server outside the firewall is Linux. If all you see is that we took down our Windows web servers and replaced them with Linux, you might get the wrong idea and start trying to replace your internal servers with Linux. That's not what we did, so you're not getting the same benefits we get. (It may, of course, turn out to be a Big Win. Even a blind dog finds a bone sometimes.)
The flaw in this thinking is that you don't control the decision-making process. You don't know whether your competition made a long detailed study of Linux v. Windows in the general sense, or compared the performance of two file servers for a month, or just got drunk and flipped a coin. You are basing your decision purely and simply on the belief that your competition is smarter than you are. This belief usually indicates that there are things you are already doing wrong, and you know it. Copying someone else's right move will not magically undo all your previous bad ones.
This isn't new or unique; companies have *always* blindly copied the work of other companies, so they don't have to come up with their own ideas. Coming up with ideas is HARD, dammit. But it seems to me that you might get better results asking "what is my competition doing WRONG, and how can we do it right?"
That's what Microsoft is usually asking. There are two things that count: first and best. It's better to be first, because once you're first, nobody else can come along later and be first; if you're not the first, you *have* to be the best.
Microsoft rarely produces the first of anything. Do the math.