I think that you're overestimating the value of support contracts and other open source based business plans. Sure, there's some money to be made there, but it's a latecomer to the open source party and only a tiny piece of the puzzle.
Open source works because all of us are smarter than one of us. Programmers naturally look for preexisting solutions to problems, because it enables them to get to the thing that they really want to do faster. And they'll naturally return the favor when they can. It's just politeness to contribute bug fixes.
This model has serious, serious flaws. There will always be more takers than givers. But the good news is that distribution is cheap, so one giver can support hundreds of freeloaders.
Other problems are harder. Many of the contributions take place against the background of a standard closed-source project, where the management doesn't mind participating in open source as long as the real product development remains proprietary. A utopian pure open-source environment will fail; the whole thing works as well as it does only because the economics of redistribution are so cheap.
There are many other issues which are not easy to work around, and that's what this guy is really getting at: open source can't promote the non-fun stuff, like good user interfaces and (for the most part) QA. Certain crucial pieces of infrastructure (Apache, Linux kernel) have so many people banging on them that they get QA'ed anyway, and they're so integral to other money-making schemes that it ends up being in some people's interests to do the work anyway. But away from those projects the software gets buggier and buggier, and you'd have to pay people to make them less buggy.
So in the end there's money to be made in the standard business model, which is actually what JBoss is using. The difference is that some of the software they develop "leaks" around the edges into open source, because that's their way of playing nice with other people doing the same thing. The more core something is, the more effective it is to share your work and to use the shares in return; the system supports the freeloaders.
The real money is in doing specific work for specific customers, of which "support contracts" are only a trivial part. "Support contracts" are really just another name for "closed-source, proprietary software" built on top of the open source. And that's just business as usual.
As programmers, we'll share because it's fun and we'll share because we're a community that likes to help each other out. That's at the end of the day; from 9 to 5 we'll continue to write software the way it's always been done, for the same economic reasons: you have to pay people to develop the boring stuff and the stuff that involves knowing the subject domain. The kernel and Apache mean you don't have to know about anything except computers. If you want to build a ticket reservation site or a pharmaceutical database, you actually have to know something outside of computers, and that always costs money.
The answer, I suspect, is almost certainly "yes", but it's not very interesting.
The way the Game of Life implements a Turing machine is extremely complicated, so it's not interesting as a way to make a Turing machine. It's interesting because the result that it can be done at all is surprising. It's interesting that a simple game gives rise to the full power of the TM. It also gives you a new way to study the basic properties of the TM: anything you prove about Life can be applied to the TM.
This game he's describing is more complicated, so it's therefore less interesting that it gives rise to a TM. And because it's so complex, it's hard to prove anything about it, so it's hard to learn anything new about TMs.
For that matter I'm having a hard time getting enthused about the whole notion of the Jecology game. If it can be demonstrated that it models the world with some fidelity we might learn something. If it makes pretty pictures, that's also of some value. But I doubt we'll learn much about the nature of cellular automata; that's best done in simple machines.
But is there enough money in providing BPL to rural users to make it worth the investment? Although you're piggybacking off of the initial investments made in universal service decades ago, you still have to make significant investments at each station with relatively few users per station.
Most of the costs will be passed on to the users, but the overhead will still be significant. And they do have a fallback plan, which is plain old modem. It's not broadband, but there comes a point where they say, "I'd rather pay AOL $20/month for slow service than $100/month to the BPL provider for fast service." That sets the upper limit of what they'd pay. If that's less than the overhead plus profit, they just don't get BPL.
So unless Google is making some sort of charitable donation, I'd need to see more numbers before I'd say that Google's BPL investment is to get rural users.
Well, sure you'd rather win something worth $150k than $9,900, which is what 10,000 songs costs. They're not claiming it's worth untold millions.
The marginal cost to Apple may be small, but 10,000 songs are still 10,000 songs, and they're worth something to you, assuming you listen to music. The music's pretty tangible.
The marginal cost may be zero, but the amortized cost is not. They can't give you the next song without having paid for the servers, the software, the music studios, the advertising, etc.
If your goal is not so much to have stuff as to make make sure you're socking it to the guys giving you the prize, well, I'm sure Apple is very sorry. If you write to jobs@apple.com, he'll send you a gift certificate to be redeemed for a free iTunes song ($.99 processing fee applies.)
(And just for reference, contests with a house as a prize rarely actually give away a house. If you read the rules closely, you'll find that you can take cash instead. Nearly everybody takes the cash.)
I'd feel better about it if I trusted the proficiency test.
Tests are a very rough measure of your skill. They're used to broadly separate candidates into maybe-acceptable and useless. You wouldn't make your decision based on it. You have to interview the person, and you can tell better from that than from the test whether he's any good or not. The tests are good only to weed out the obviously unacceptable candidates before you schedule an interview.
I've taken some of these, and sometimes they're an insult; they ask about easily-looked-up trivia. And there's a difference between solving problems and answering riddles. I don't much care for tests that are nominally testing my "lateral thinking", because I hate the idea of losing a job because I didn't get the joke.
Without seeing what this test looked like I can't support or condemn the guy. But let's just say that for some tests, yeah, I'd consider myself above it. Especially if I was invited.
Slight addendum: Cyc doesn't do first-order logic. Actually, I can't tell you what the precise semantics of CycL are, and as far as I can tell, neither can they. It's got all sorts of nonmonotonic stuff crammed in.
Nor would I say it's based on "pretty sound knowledge-base-engineering". It's better than mindpixel (a LOT better than mindpixel), but there's a lot of crud in the Cyc knowledge base.
So on a scale from Microsoft Bob to Google, Cyc is still well to the left of center; there are some good ideas that haven't really panned out yet (or, increasingly likely, ever). But mindpixel is way off the chart.
It's pretty prohibitive. Many power lines aren't even insulated, much less caged. Among other things, the weight of the cage would mean more towers to support them, and that's big-time money.
My 85-90% figure comes from the book Freakonomics, which is actually rather scholarly despite the title, and is based on work by economist Steven Levitt.
There are a number of differences between the bagel scenario and the coffee scenario, not the least of which is that students are far poorer than office workers. Then again, the data say that rich executives had a lower compliance rate than ordinary office workers.
They claimed that they wanted Gator not for its popup software but for the personalization.
This sounds fishy to me. Microsoft doesn't want their products to look bad, and they know that people hate Gator's popup ads, at least in their present form.
Microsoft shouldn't be protecting old Gator products. They want to take it and modify it to be at least as invasive but less obvious about it. You should expect Gator to be arriving on your system via Windows update, less obnoxious but a lot harder to remove (and almost certainly called by a new name).
I mostly just call it "people". Set out a sign next to a box of bagels saying, "Bagels: $1" and about 85-90% of the people will pay for the bagels. And 10-15% won't.
In other words, it doesn't take a corporation for people to act dishonestly. It just takes an opportunity.
The CD format is hopelessly unprotected, but it's also got a huge installed base. If you want people to change you're going to have to convince them that it's an upgrade: higher bit rate, better sepration, liner notes and titles. Then they can slip in what they really want: copy protection. That's the primary goal; the other features are the spoonful of sugar to make the bitter pill go down.
So both properties (features and copy protection) are the "real" reason. The former are the reason for consumers to upgrade; the latter is why the industry makes the upgrade available.
Actually, I think only a scientist would find that joke funny. A nonscientist is just going to give you dumb stares.
It's an exaggeration for comic effect. Every good scientist recognizes in it some silly oversimplification he or she has made at some point in the past, because that's the way science gets done. The nonscientist doesn't get it precisely because he doesn't know how science gets done.
Public misunderstanding of science has little to do with the jokes of scientists poking fun at themselves. In fact, one of the best ways to convince the general public that science is simply wrong is to be humorless about it. Lighten up.
There are some differences between this and Passport. For one thing there's no wallet component. That limits its usefulness, but it also limits the damage (which in turn means that we don't need to be quite so paranoid about it.)
For another it's a distributed (kinda) system rather than a centralized one. The system is actually quite clever; it's basically a way for you to set up your own ID system. You can "shop around" for ID providers. All you really need to own is the URL. If your ID provider goes out of business, you can just get a new one.
True, I don't expect it to go much beyond blogging to start with. But great standards often start as "Hey, let's try this" and then they expand. Yeah, there's a lot of bitching afterwards about "Oh, they didn't think to incorporate X obvious feature" but standards often do much better for organic growth rather than having a central authority's fiat first.
It's a kind of weird inverse vaporware: the code exists, and runs, but you can't know where it is. Ordinary vaporware you know where it is (inside the offices at Duke Nukem Forever) but you have no idea what it looks like.
Captcha solves a different problem. Captcha proves that you're a human (more or less). OpenID proves that you are you. That doesn't prove that you're a human; it just proves that you know a password. But since you're the only one who knows that password, you're uniquely you and you don't have to create a separate account on each system you visit.
So it's a convenience for users, not to prevent spammers. This does have spam implications: you can blacklist/whitelist ID servers and you don't have to give your email to every site you visit, but it's not really about preventing spam. It's about simplifying the mass of passwords and accounts you have.
The only motivation I can detect for Open Id is to save people FIVE SECONDS by logging into a new forum, website
I'd have thought the motivation was to limit the number of separate accounts you need. Having a billion accounts running around is a massive security nightmare. Either you're using the same password everywhere (and telling every web site owner your password) or you're wandering around with a notebook of thousands of passwords.
Firefox won't remember your password if the computer is a public terminal, or if you use multiple computers (e.g. at home and at work.)
No, this isn't the ultimate solution (which involves encryption, a portable very strong crypto key time-based challenge-response, and perhaps biometrics), but it could be a good half-measure.
The US is already in on the pact, along with the EU. But the plan doesn't say "Do not spam"; it's the beginning of a process for fighting spam.
Yeah, you know spam when you see it, but it's a little harder when you tell a country to filter every outgoing email. What's required is a lot more complex: a mechanism for coordinating, tracking, and aggregating complaints; for tracking down offenders across countries without violating sovereignty; resolving problems with private individuals in foreign countries.
Just defining "spam" is incredibly hard; look at the way the US government botched it. They called it "unsolicited commercial email", which is wrong. Not all spam is commercial, and not all unsolicited commercial email is necessarily spam.
This is just the beginning and it's going to be a long time before this yields any actual spam reduction.
Invertebrates also use cholesterol, partly to keep things flowing in cold water. Shrimp are famously high in cholesterol.
Land-based invertebrates, on the other hand, don't need the cholesterol to keep things flowing. So bugs are low in cholesterol. If that is good news for your diet, well, let's just say I don't have any recipes for you.
So am I right in translating this link to mean he's pressing charges to make someone shut up that was obstructing his freedom of speech?
Actually, yeah. He wants people to stop making threats against him on Slashdot. It seems he made a few enemies somehow involved in Novell and the SCO thing. I did read the thing and still am not sure.
I love the "bell and whistle" logo they've applied to the story. Very appropriate, especially since the review discussess little besides the eye candy.
The law in this case defies easy summary, but the gist is that when you get a trademark, you're really getting the right not to have confusingly similar things taking your business. The definition of "confusingly similar" is vague, and depends on a few zillion common law cases.
So when he applied "stealth" to sporting goods, it meant you couldn't come out with similar products with similar names. Again, "similar" is up to courts to decide. If you've got "stealth" bowling balls, are "stealth" skis too close? What about golf balls?
Clearly it doesn't apply to airplanes, and honestly I don't know why Northrop didn't just invite him to court. Except that court is expensive, even when you win, and the fact that they paid him $10 is a lot less than the retainer on their lawyers. It's disgusting, I know.
I don't know why they ceded any rights to him at all without knowing more facts about the case. They imply that they weren't seeking their own trademarks, but may have continued to use the name. But a news article isn't a legal brief, so it's hard to tell.
So to answer your question: the word "stealth" has been around forever, but trademarks don't apply only to made-up words. You can take a common word and trademark it for your application, which allows you to keep other people from muscling in on your good idea by confusing people with a similarly-named product. It's the sort of thing that would be accomplished by politeness if you could depend on that. For everything else, there are lawyers.
Slashdot Mirror
I think that you're overestimating the value of support contracts and other open source based business plans. Sure, there's some money to be made there, but it's a latecomer to the open source party and only a tiny piece of the puzzle.
Open source works because all of us are smarter than one of us. Programmers naturally look for preexisting solutions to problems, because it enables them to get to the thing that they really want to do faster. And they'll naturally return the favor when they can. It's just politeness to contribute bug fixes.
This model has serious, serious flaws. There will always be more takers than givers. But the good news is that distribution is cheap, so one giver can support hundreds of freeloaders.
Other problems are harder. Many of the contributions take place against the background of a standard closed-source project, where the management doesn't mind participating in open source as long as the real product development remains proprietary. A utopian pure open-source environment will fail; the whole thing works as well as it does only because the economics of redistribution are so cheap.
There are many other issues which are not easy to work around, and that's what this guy is really getting at: open source can't promote the non-fun stuff, like good user interfaces and (for the most part) QA. Certain crucial pieces of infrastructure (Apache, Linux kernel) have so many people banging on them that they get QA'ed anyway, and they're so integral to other money-making schemes that it ends up being in some people's interests to do the work anyway. But away from those projects the software gets buggier and buggier, and you'd have to pay people to make them less buggy.
So in the end there's money to be made in the standard business model, which is actually what JBoss is using. The difference is that some of the software they develop "leaks" around the edges into open source, because that's their way of playing nice with other people doing the same thing. The more core something is, the more effective it is to share your work and to use the shares in return; the system supports the freeloaders.
The real money is in doing specific work for specific customers, of which "support contracts" are only a trivial part. "Support contracts" are really just another name for "closed-source, proprietary software" built on top of the open source. And that's just business as usual.
As programmers, we'll share because it's fun and we'll share because we're a community that likes to help each other out. That's at the end of the day; from 9 to 5 we'll continue to write software the way it's always been done, for the same economic reasons: you have to pay people to develop the boring stuff and the stuff that involves knowing the subject domain. The kernel and Apache mean you don't have to know about anything except computers. If you want to build a ticket reservation site or a pharmaceutical database, you actually have to know something outside of computers, and that always costs money.
The answer, I suspect, is almost certainly "yes", but it's not very interesting.
The way the Game of Life implements a Turing machine is extremely complicated, so it's not interesting as a way to make a Turing machine. It's interesting because the result that it can be done at all is surprising. It's interesting that a simple game gives rise to the full power of the TM. It also gives you a new way to study the basic properties of the TM: anything you prove about Life can be applied to the TM.
This game he's describing is more complicated, so it's therefore less interesting that it gives rise to a TM. And because it's so complex, it's hard to prove anything about it, so it's hard to learn anything new about TMs.
For that matter I'm having a hard time getting enthused about the whole notion of the Jecology game. If it can be demonstrated that it models the world with some fidelity we might learn something. If it makes pretty pictures, that's also of some value. But I doubt we'll learn much about the nature of cellular automata; that's best done in simple machines.
But is there enough money in providing BPL to rural users to make it worth the investment? Although you're piggybacking off of the initial investments made in universal service decades ago, you still have to make significant investments at each station with relatively few users per station.
Most of the costs will be passed on to the users, but the overhead will still be significant. And they do have a fallback plan, which is plain old modem. It's not broadband, but there comes a point where they say, "I'd rather pay AOL $20/month for slow service than $100/month to the BPL provider for fast service." That sets the upper limit of what they'd pay. If that's less than the overhead plus profit, they just don't get BPL.
So unless Google is making some sort of charitable donation, I'd need to see more numbers before I'd say that Google's BPL investment is to get rural users.
Well, sure you'd rather win something worth $150k than $9,900, which is what 10,000 songs costs. They're not claiming it's worth untold millions.
The marginal cost to Apple may be small, but 10,000 songs are still 10,000 songs, and they're worth something to you, assuming you listen to music. The music's pretty tangible.
The marginal cost may be zero, but the amortized cost is not. They can't give you the next song without having paid for the servers, the software, the music studios, the advertising, etc.
If your goal is not so much to have stuff as to make make sure you're socking it to the guys giving you the prize, well, I'm sure Apple is very sorry. If you write to jobs@apple.com, he'll send you a gift certificate to be redeemed for a free iTunes song ($.99 processing fee applies.)
(And just for reference, contests with a house as a prize rarely actually give away a house. If you read the rules closely, you'll find that you can take cash instead. Nearly everybody takes the cash.)
I'd feel better about it if I trusted the proficiency test.
Tests are a very rough measure of your skill. They're used to broadly separate candidates into maybe-acceptable and useless. You wouldn't make your decision based on it. You have to interview the person, and you can tell better from that than from the test whether he's any good or not. The tests are good only to weed out the obviously unacceptable candidates before you schedule an interview.
I've taken some of these, and sometimes they're an insult; they ask about easily-looked-up trivia. And there's a difference between solving problems and answering riddles. I don't much care for tests that are nominally testing my "lateral thinking", because I hate the idea of losing a job because I didn't get the joke.
Without seeing what this test looked like I can't support or condemn the guy. But let's just say that for some tests, yeah, I'd consider myself above it. Especially if I was invited.
Slight addendum: Cyc doesn't do first-order logic. Actually, I can't tell you what the precise semantics of CycL are, and as far as I can tell, neither can they. It's got all sorts of nonmonotonic stuff crammed in.
Nor would I say it's based on "pretty sound knowledge-base-engineering". It's better than mindpixel (a LOT better than mindpixel), but there's a lot of crud in the Cyc knowledge base.
So on a scale from Microsoft Bob to Google, Cyc is still well to the left of center; there are some good ideas that haven't really panned out yet (or, increasingly likely, ever). But mindpixel is way off the chart.
It's pretty prohibitive. Many power lines aren't even insulated, much less caged. Among other things, the weight of the cage would mean more towers to support them, and that's big-time money.
My 85-90% figure comes from the book Freakonomics, which is actually rather scholarly despite the title, and is based on work by economist Steven Levitt.
There are a number of differences between the bagel scenario and the coffee scenario, not the least of which is that students are far poorer than office workers. Then again, the data say that rich executives had a lower compliance rate than ordinary office workers.
They claimed that they wanted Gator not for its popup software but for the personalization.
This sounds fishy to me. Microsoft doesn't want their products to look bad, and they know that people hate Gator's popup ads, at least in their present form.
Microsoft shouldn't be protecting old Gator products. They want to take it and modify it to be at least as invasive but less obvious about it. You should expect Gator to be arriving on your system via Windows update, less obnoxious but a lot harder to remove (and almost certainly called by a new name).
I mostly just call it "people". Set out a sign next to a box of bagels saying, "Bagels: $1" and about 85-90% of the people will pay for the bagels. And 10-15% won't.
In other words, it doesn't take a corporation for people to act dishonestly. It just takes an opportunity.
So which is it?
Both. It just depends on who you are.
The CD format is hopelessly unprotected, but it's also got a huge installed base. If you want people to change you're going to have to convince them that it's an upgrade: higher bit rate, better sepration, liner notes and titles. Then they can slip in what they really want: copy protection. That's the primary goal; the other features are the spoonful of sugar to make the bitter pill go down.
So both properties (features and copy protection) are the "real" reason. The former are the reason for consumers to upgrade; the latter is why the industry makes the upgrade available.
There are still Gopher servers?
Cool! I'd utterly forgotten about Gopher.
And if one fails, you can just swap it out!
Actually, I think only a scientist would find that joke funny. A nonscientist is just going to give you dumb stares.
It's an exaggeration for comic effect. Every good scientist recognizes in it some silly oversimplification he or she has made at some point in the past, because that's the way science gets done. The nonscientist doesn't get it precisely because he doesn't know how science gets done.
Public misunderstanding of science has little to do with the jokes of scientists poking fun at themselves. In fact, one of the best ways to convince the general public that science is simply wrong is to be humorless about it. Lighten up.
There are some differences between this and Passport. For one thing there's no wallet component. That limits its usefulness, but it also limits the damage (which in turn means that we don't need to be quite so paranoid about it.)
For another it's a distributed (kinda) system rather than a centralized one. The system is actually quite clever; it's basically a way for you to set up your own ID system. You can "shop around" for ID providers. All you really need to own is the URL. If your ID provider goes out of business, you can just get a new one.
True, I don't expect it to go much beyond blogging to start with. But great standards often start as "Hey, let's try this" and then they expand. Yeah, there's a lot of bitching afterwards about "Oh, they didn't think to incorporate X obvious feature" but standards often do much better for organic growth rather than having a central authority's fiat first.
It's a kind of weird inverse vaporware: the code exists, and runs, but you can't know where it is. Ordinary vaporware you know where it is (inside the offices at Duke Nukem Forever) but you have no idea what it looks like.
Captcha solves a different problem. Captcha proves that you're a human (more or less). OpenID proves that you are you. That doesn't prove that you're a human; it just proves that you know a password. But since you're the only one who knows that password, you're uniquely you and you don't have to create a separate account on each system you visit.
So it's a convenience for users, not to prevent spammers. This does have spam implications: you can blacklist/whitelist ID servers and you don't have to give your email to every site you visit, but it's not really about preventing spam. It's about simplifying the mass of passwords and accounts you have.
Put all your eggs in one basket, then make sure it's a really, really good basket.
It seems to me that "Hey, you can actually go out and download X" is news, even when "Hey, I've got an idea for X" was already news.
The only motivation I can detect for Open Id is to save people FIVE SECONDS by logging into a new forum, website
I'd have thought the motivation was to limit the number of separate accounts you need. Having a billion accounts running around is a massive security nightmare. Either you're using the same password everywhere (and telling every web site owner your password) or you're wandering around with a notebook of thousands of passwords.
Firefox won't remember your password if the computer is a public terminal, or if you use multiple computers (e.g. at home and at work.)
No, this isn't the ultimate solution (which involves encryption, a portable very strong crypto key time-based challenge-response, and perhaps biometrics), but it could be a good half-measure.
The US is already in on the pact, along with the EU. But the plan doesn't say "Do not spam"; it's the beginning of a process for fighting spam.
Yeah, you know spam when you see it, but it's a little harder when you tell a country to filter every outgoing email. What's required is a lot more complex: a mechanism for coordinating, tracking, and aggregating complaints; for tracking down offenders across countries without violating sovereignty; resolving problems with private individuals in foreign countries.
Just defining "spam" is incredibly hard; look at the way the US government botched it. They called it "unsolicited commercial email", which is wrong. Not all spam is commercial, and not all unsolicited commercial email is necessarily spam.
This is just the beginning and it's going to be a long time before this yields any actual spam reduction.
Invertebrates also use cholesterol, partly to keep things flowing in cold water. Shrimp are famously high in cholesterol.
Land-based invertebrates, on the other hand, don't need the cholesterol to keep things flowing. So bugs are low in cholesterol. If that is good news for your diet, well, let's just say I don't have any recipes for you.
So am I right in translating this link to mean he's pressing charges to make someone shut up that was obstructing his freedom of speech?
Actually, yeah. He wants people to stop making threats against him on Slashdot. It seems he made a few enemies somehow involved in Novell and the SCO thing. I did read the thing and still am not sure.
I love the "bell and whistle" logo they've applied to the story. Very appropriate, especially since the review discussess little besides the eye candy.
The law in this case defies easy summary, but the gist is that when you get a trademark, you're really getting the right not to have confusingly similar things taking your business. The definition of "confusingly similar" is vague, and depends on a few zillion common law cases.
So when he applied "stealth" to sporting goods, it meant you couldn't come out with similar products with similar names. Again, "similar" is up to courts to decide. If you've got "stealth" bowling balls, are "stealth" skis too close? What about golf balls?
Clearly it doesn't apply to airplanes, and honestly I don't know why Northrop didn't just invite him to court. Except that court is expensive, even when you win, and the fact that they paid him $10 is a lot less than the retainer on their lawyers. It's disgusting, I know.
I don't know why they ceded any rights to him at all without knowing more facts about the case. They imply that they weren't seeking their own trademarks, but may have continued to use the name. But a news article isn't a legal brief, so it's hard to tell.
So to answer your question: the word "stealth" has been around forever, but trademarks don't apply only to made-up words. You can take a common word and trademark it for your application, which allows you to keep other people from muscling in on your good idea by confusing people with a similarly-named product. It's the sort of thing that would be accomplished by politeness if you could depend on that. For everything else, there are lawyers.