Well, that's the same motivation: an institution that wants to gain market share. However if users don't like that feature, the developers who agree with them will fork.
It's a bit like how quantum computing is supposed to work. Only so many developers can work on a project before diminishing returns sets in; a good controversy can be addressed by a good schism, where the question can be resolved by two more optimally sized teams.
What's surprising is that it doesn't happen more often.
I'm talking about the problem of finding the economically optimal assignment of feature to product. Since a feature is either in or not in a product, it's an integer programming, assuming we can express marginal profits as linear equations, which would be the easiest case.
It's kind of a flaw that is endemic to the commercial software development model. This is not to say that that model is useless or F/OSS doesn't have its own problems.
The root of the problem is how we "add value" to a piece of software. Since with F/OSS, software development has a service model, you mainly add value by adding services: documentation, support, consulting. You can't "add value" by adding features to the software, at least if you try to you only get paid once for doing so.
A proprietary software developer can get paid multiple times for adding a piece of value into the software. For software that is sold, this is driven by market segmentation. The least pernicious form of this is the ubiquitous "bronze/silver/gold" model where they try to maximize their return from cheapskates, pragmatists and spendthrifts respectively. If you are cheapskate who needs a feature in the "gold" edition, you're out of luck. In the worst case, it drives a bewildering proliferation of "products", as vendors try to find the division of features that maximizes their returns (which is an instance of the NP-Complete "integer programming problem", only approximations are practical). From a customers standpoint, it sometimes looks like a whirlwind has picked up all the features and dropped them into random pigeonholes.
The "value adding" imperative still applies to free as in free beer proprietary software. In such cases, the developer still is looking to get paid, only in different coin, e.g. control of formats and the market power that comes with it. Adobe benefits from PDF being a non-proprietary format because it encourages adoption, but it is risky because they wouldn't benefit if they did not control the dominant implementations of PDF technology. And they try very hard, I think, to have the best implementations, which leads to the old problem of adding value by adding features. The hope is that by adding features nobody has asked for, when those features are missing from a different implementation, that implementation will be seen as less complete and polished. I think this often works, but it leads to this kind of blowback siutation: security flaws introduced to users systems along with features the user never asked for.
Well, personally I have found Vista more _stable_ than recent versions of Ubuntu. I use Ubuntu because (a) the kinds of tools I currently need are more available and free as in beear and (b) I prefer Gnome to Vista and (c) the kinds of performance issues that matter most to me seem to be a bit better on Linux.
The stability issues I've had with Ubuntu are most likely related to ACPI. In a sense, this is not an issue of OS quality per se but the quality of the ACPI standard, which requires somebody (the manufacturer of the laptop, in the case of Windows, the user in the case of Linux) to figure out how to get ACPI features to work with an operating system.
On the other hand, I've had gut wrenchingly bad performance out of Vista, but the only crash was of the sound system, which poured white noise out of the speakers but left all the other system functions working perfectly. This is a sign that Vista is in some ways better architected than earlier version of Windows, where such a fault would almost certainly have required a hard restart.
Vista on a mainframe might solve what is one of the worst faults of Windows: braindead I/O. Mainframes aren't so much about computing power (although they typically have plenty), they're about really awesome I/O bandwidth. Certainly, we wouldn't be worrying about the kind of hardware and hardware driver faults that bedevil every operating system in one way or the other.
The important thing to remember is that operating systems, as we knew them in the 80s and 90s, don't matter as much as they used to. The focus of application architecture these days is the network. Getting worked up over "operating systems" is getting to be like having big fights over machine instruction set architetures. Sure they matter -- on a certain level -- but a much of their relative merits and problems are thoroughly contained, insulated from applications through multiple layers of abstraction.
Enterprises are very conservative; they like to stick with what works. This move is really just like mainframe systems provide support for ancient character terminal protocols and such. It's not a target for new development, it's more a way of keeping applications running in their current form indefinitely, like a fly trapped in amber.
Come now. Nobody seriously things we should send bricks to the moon. Initially we'll send any structures we need, but any reasonable moon industrialization or colonization will require using local materials. But the first step in using local materials on a large scale will certainly involve moving a great deal of mass, probably far more mass than we'd need to significantly increase our initial capacity there.
That's the point: you need to spend mass to save mass. Sending giant solar furnaces to the Moon would be a huge investment that would almost certainly not pay off on the scale we could contemplate in the immediate future, because we don't have enough investment in the other things we'd need to exploit that. Debris berm building robots sounds about right.
An interesting thought occurs though. One solution to the mass problem is simply patience. Since the Von Neumann approach is based on exploiting exponential growth, but growth doesn't have to be fast at the outset. Suppose you put ten robots on the moon that could scavenge enough material to make one robot in ten years. After the first ten years, you'd have 11 robots. After a hundred years, you'd have about 25 robots. After a 1000 years, you'd have almost 14,000 robots. After 2000 years, you'd have 190 million robots.
It seems likely to me that you'd need to move a lot of mass to bootstrap a Von Neumann style self-replicating system. Assuming you have a "large enough parabolic mirror" is a bit like the proverbial assumption you have a can opener.
This is especially true because we don't know if there are richly concentrated deposits of all materials that are useful in building robots on the Moon, although we have reason to doubt it. Most of the interesting things we think we might get from the moon involve processing huge quantities of material. Things like helium, or water. Prospecting meteor craters might yield a number of interesting things like iridium, and could in theory yield a lot of iron, although getting at that iron brings us back to the bootstrapping problem. We'd have to find a way to do without plastic or rubber though.
Ultimately, it's possible. After all, humans transformed the Earth and we aren't much bigger than a lawnmower; the difference is everything was in place to support humans replicating more humans when we arrived here. Getting to anything like a significant self-supporting industrial capability on the Moon is going to involve countless numbers of small, not so spectacular steps that don't seem like much progress at all. Demonstrating that crude, non-self-replicating robots made with terrestrial materials can do useful structural tasks is a logical step in a direction of a number of goals that are far, far off.
There's case law where lawyers have tried exactly what you are suggesting and, guess what? Sometimes they win, sometimes they lose.
So, yes, I am telling you that a half-decent lawyer, even a pretty damned good one, isn't going to be able talk you out of any kind of copyright trouble you might dream up for yourself. The problem with "clever arguments" by "half-decent" lawyers is that good lawyers have heard all of them before and can kick the crap out of them in their sleep. That's why you check with a lawyer before to beard the copyright lion.
DRM is a different issue, but related in that it provides people in the exclusive possession of material to which they don't own exclusive copyright to exert a measure of copyright-like control over that material.
I daresay that any image Picasso stole from Pinocchio would be so radically revised as to escape the Mouse's copyright.
That's simply not true. Yes, transformation is a necessary condition for claiming fair use. But it is not sufficient. Certain kinds of transformation where there is a recognized public interest (e.g. for parody or critical commentary) are fair use, but degree of transformation is not sufficient defense in itself, it only shows that the transformation was sufficient serves its claimed purpose. If the purpose of the transformation is not protected, then the degree of transformation isn't the deciding factor, but the degree of appropriation.
In any case, the purpose of the artist, which is artistic merit, has no legal force at all.
OK, I've got this keyboard and a display and, and, I can take them apart and use one without the other. I might even be able to use a keyboard that wasn't the one that came with the original...
Now this seems hauntingly familiar.... Where have I seen a form factor that offers this kind of mix and match modularity?
All joking aside, what matters is your data and your tools for manipulating the data. I think it makes sense to give your data and at least some of the software a distinct physical module, like the case of desktop computer but smaller and detachable.
I hesitate to say this, because of the heinous crimes committed in the name of what I'm about to mention, but I'm almost tempted to think of this as the model-view-controller pattern applied to form factor. I should be able to detach my model and use it anyplace I please with whatever peripherals I please.
Well, real estate is not an essential ingredient in creating new real estate, whereas the ideas and expressions are essential ingredients in creating new ideas and new expression.
Therefore even if there is some kind of moral imperative to respect intellectual property rights of creators, the nature of that property and those rights are logically different in nature because all intellectual property exists by virtue of treating previous creations as less than absolute property.
In any case, we don't give real property owners absolute right to control public use of their property either. This is place where public abuses of private rights do from time to time occur, such as stretching utility easements into different kinds of intrusive uses. But that kind of recognition of public interests even in private real estate has legitimate uses and justification.
What he means is great artists do something new with old materials, whether those are themes, motifs, styles, images -- whatever.
However, in a world where art is perpetually governed by technologically and legally enforceable restrictions, the freedom of artists to make fundamentally new works out of old materials is greatly curtailed.
Also consider that the way lawyers enforce their clients' rights is to overreach. In effect they carve out new, extralegal rights for their clients, at least so far as use by those who can't afford even the prospect of litigation. So an IP protection dominated culture would likely be moribund. At it currently stands, copyright is a pragmatic bargain, and the effect of stretching the copyright holder's authority a bit is to tweak the bargain slightly. The effect of stretching a copyright holder's authority indefinitely is a totally different kettle of fish. When culture is distributed exclusively in DRM forms, and defeating DRM is illegal per se regardless of the legal status of the protected content, then culture will cease to have much contemporary relevance as its vocabulary will be limited corporately sponsored and licensed expression, or to works from the distant past.
Well, you spotted an important point. Note that I say that private contractual arrangements are probably indispensable.
I'm quite aware that an absolute dichotomy between the cultural good and private enterprise is unsupportable. What I'm saying is that the greater cultural good cannot be entrusted to private contractual decisions alone, which is considerably less broad an assertion.
To abandon a trademark, it isn't enough to stop using it. You've got to stop using it with the intent of never using it again. If they discontinued the product with an press release saying, "this is the last 'netbook' branded product we'll ever sell," then they'd have abandoned the trademark at that instant. However nobody does that sort of thing.
On the other hand, you just can't be sly about whether you want to use a trademark, hoping to catch somebody in an indiscretion at a far future date. IIRC, three years without use or attempt to defend is considered prima facie evidence of abandonment.
Now, I believe Psion discountined its last netbook in 2005; however "discontinued" only means that they aren't selling them through the normal wholesale channels. My understanding is that they continued to sell netbooks direct to some corporate accounts for as much as a year later. Perhaps they were liquidating inventory. Also, IIRC, in 2006 Psion filed documents with the FTC indicating its intent to continue the use of trademark. In either case the clock would run out on the trademark some time this year.
Now onto who is virtuous and wronged and who is evil and taking advantage.
It's quite plausible to believe that Intel thought Psion had met the three year criteria for abandonment. It's also quite plausible that Psion had every intent to abandon the trademark, until Intel made it valuable again. This doesn't mean that HP did its homework properly, nor does Intel screwing up mean Psion would win in court (although IANAL). We could imagine that during discovery, Intel could prove that Psion hand no real intent of ever using that brand again, and only started to make trouble when Intel's use of the term made it popular. However, won't happen because these things tend to get discreetly settled and the details thoroughly buried. Maybe that would indicate that Psion was just shaking down Intel, or maybe it means that they decided the cash offer was worth more than than the brand. We'll never know.
Again, IANAL, but if this were referred to me in my part time capacity as philosopher-king, I'd enforce the C&D but grant no damages to Psion over what would be an understandable technical error. If Psion didn't at least announce a new 'netBook' branded product this year, I'd rule the word as public domain.
I actually think Psion might have a case here. They registered the "netBook" trademark years ago, and have been selling "netBook" branded subnotebooks continually over the last decade.
At this point, I think "Netbook" has become generic, but not through a failure of Psion to protect its trademark. Intel, while not selling a device of its own, improperly appropriated Psion's trademark for its own commercial ends (selling Atom processors to subnotebook manufacturers among others). If Sun decided that "J2EE Server" wasn't going to win any adopters, and decided to call them "Websphere servers", that would be clearly be improper, even though they weren't using to refer specifically to their implementation.
Physicians, Begley says, must stop treatments that are rooted more in local medical culture than in medical science, embrace practices that have been shown scientifically to be superior to others, and ignore critics who paint CER as government control of doctors' decision-making.
All very good, except that medical science isn't up to the task of providing reliable diagnostic and treatment guidelines. We can see this because this is the kind of efficiency program the insurance companies try to impose.
For example, a family acquaintance has a fifth grader who is unusually athletic and physically mature for his age. He swims competitively and practices every day. If you look at him, he is obviously very lean and muscular, but because of his body mass index the insurance company insists he and his pediatrician implement a weight management plan.
Of course it is not beyond the capabilities of science to distinguish between lean body mass and fat. It's just extremely difficult for a committee to take scientific literature and come up with comprehensive, iron-clad guidelines for treatment. And that's what physicians need. When they have a patient, they need guidelines that are right for this patient, not 80% of patients or even 95% of patients.
I think a better approach would be to institute some kind of scientific peer review of treatment decisions between physicians. But of course health care is so damned expensive it has a hard time accommodating new mandates, even if they are good for long term cost control
This is an important point. It leads an irony worth considering, based on the well known principle that local optimizations don't lead to global optimization.
Suppose Alice has content that Bob wants. Very few people would feel that Alice ought to be compelled to provide Bob that content just be cause he wants it. So Bob has to offer Alice some incentive. Alice names a fancy price, which Bob declines to pay. So instead Bob and Alice come to a somewhat more complex arrangement. Bob agrees to pay Alice a smaller price, but restrict his uses to those he is most willing to pay for. From a libertarian viewpoint, this seems innocuous. Alice and Bob are happy, so that should be the end of that.
But Alice and Bob are happy only with respect to the direct effects of this transaction on their interests. They are not necessarily happy about the net effect of information being encumbered this way throughout society. For example, many software licenses forbid publishing benchmarks or reviews without approval. Is the world a better place when people only have a vendor's word for what a piece of software is capable of doing? Are buyers better off?
The rental thing sounds innocuous, but it has important consequences as well. Alice can probably maximize her revenue for her existing content by adopting a rental model. But culture depends on free re-use of ideas, both subtle reuses that don't fall within the scope of copyright law, and obvious reuses.
Disney, for example comes down hard on people who would reuse images from its version Pinnochio, even though that work is by now sixty-four years old. However, their 1940 movie makes free use of the original story by Carlo Collodi that, because it was published in 1883, was only fifty seven years old at the time. Disney would argue, correctly, that at the time their usage was legal whereas their movie is still protected by current copyright. But they can't make a utilitarian argument that in this case protection is for the greater common good. Nor can they reasonably say it represents morally necessary protection of a fundamental right of creators to perpetual intelletual property protection.
The important thing is that when information is controlled solely by private agreements, the net effect of all these local optimizations across society is not globally optimal for the parties living in that society. We can take a lesson from the popular music industry, which is creatively moribund and therefore financially vulnerable. They can blame "piracy" if they like, but if they were producing what consumers considered a good value they would be much better off. Yes, it is possible to download files of unknown quality and provenance for free, but when consumers have access to a good selection through convenient distribution at a fair price, they prefer it. The iTunes store proves this.
But even something like the iTunes store is not a long term solution. As Pablo Picasso said, "Bad artists copy. Great artists steal." Art and culture depend on artists making old works their own.
The progress of useful arts and sciences can't be left entirely to contractual or licensing arrangements between private individuals, as practically useful and even indispensable as those might be. Given the power of technology to restrict information, the law that makes those arrangements possible must also promote the continuing enrichment of the general intellectual welfare, if we are not to suffer dire long term consequences. Unfortunately, we live in a country where politicians aren't very interested in culture, or any aspect of the life of the mind. I was shocked to hear one politician last week mock the idea that studying the spread of venereal disease was a worthwhile use of money. I believe that this anti-intellectualism comes from being so rich for so long that we've come to believe that we can live by managing the prior accumulated wealth of generations of intellectual achievement.
If you recall anything about the days when MS-DOS, then Windows rose to prominence, you will recall that programming was an absolute nightmare, and attempts to make it easier were dreadful kludges that hid the mess just long enough that you could get yourself up to your neck in crap.
But programmers put up with it, and a generation of programmers grew up thinking that this is normal, in order to target the program where the users were.
Now, how do you get users? To a first approximation, of course applications (games) sell platform (console) sales, but what I think this guy is talking about is a different pardigm, the one that drove the adoption of Microsoft platforms in the 1980s: the killer app. There were better desktop platforms than DOS back in, say, 1983, but business customers didn't want to hear about them if they didn't run Lotus 1-2-3. Lotus was making money hand over fist, and so didn't have any immediate incentive to rock the boat, but in retrospect, it would have been worthwhile to support a second OS contender. It's always easier to be sensible in hindsight.
The same may go here. Although Sony might not have made the right decision with the PS3, it might not be as utterly brain-dead as it appears with benefit of hindsight. The reasoning might go, give the programmers hardware that can do all kinds of interesting new things, and while there'll be a lot of struggle, somebody will put those things together into something unlike anything anybody has seen before, and that will be a killer app.
With the benefit of hindsight, we can see certain things. First, you don't necessarily need esoteric technology to give people something they haven't seen and want. Nintendo had a hit by marrying a clever (but not really very complicated) controller to a technologically conservative console then setting the price at an impulse purchase friendly level. This will be a case study taught in business schools for years to come.
Second the idea that companies will spend highly expensive programmer time on figuring out new technology only works if there aren't marginally better uses of that time, like releasing the same game on more than one platform. In an industry where PS3 is not dominant, it can't command that kind of attention.
Finally, if nobody strikes the killer app mother-lode, the victory goes to the most agile company. If developers are ignoring the unique capabilities of the PS3 and targeting multiple platforms with more or less the same stuff, then that means that apps come out sooner and better on the easier to program platform. New developers choosing platforms to target initially will choose the ones with the most revenue and least cost first, then look at the more difficult ones once their marginal revenues drop. So all in all, if exotic architecture doesn't do wonders at the outset, it holds you back rather than catapulting you forward.
Again, this is easy to see in hindsight; in foresight it's much easier to get jazzed over something really cool and exotic. And we have to keep in mind, things might have turned out differently. All it would have taken was one programmer having just the right insight that would allow him to do something on the PS3 unlike anything anybody had seen yet. It's a hell of a gamble betting that kind of investment, but you have to remember the dream of every business is to control a market as its own private fiefdom.
I'd much rather be killed in the name of some fanatical religious ideology -- or even just plain xenophobic hatred -- than to perish in some tawdry pecuniary scheme. Wouldn't you?
to think of ways to set aside funds for small businesses; ways that would encourage them to adopt and develop open source software.
We might take the Depression era grants that went to artists to decorate many public spaces as a model. Modest grants to people who work in information technology to create freely redistributable solutions public informatics problems would have several important advantages.
(1) Such a plan maintains a domestic informatics workforce in the face of increased pressure to move jobs to low wage countries. Maintaining and increasing the skills of this workforce will make it attractive for industry to turn to it when the economy improves.
(2) Mandating interoperability with open, non-proprietary standards improves the competitiveness of the domestic IT industry, where businesses are too often driven by premature efforts to create some kind of market niche where they ar protected from competition.
(3) The grants should require that the small businesses have a plant to use the work to increase their capabilities, and particularly favor the development of new kinds of technology or application for technology.
(4) Focusing the stimulus money on small businesses allows technology bets to be spread across a greater variety of approaches; it is less likely to introduce what is in effect central planning into engineering decisions. It is also unlikely to strengthen the hand of one big player against all the others because of its skill at obtaining Federal money.
(5) Low margins and ready sweat equity will encourage greater adoption of free software.
(6) There is already a Federal mechanism for doing this; the Small Business Innovative Research grant program.
SBIR currently pays for mostly a lot of boondoggles, although even boondoggles if they are kept close enough to home can be stimulative. However, with the right requirements placed on grant applications the value created can be maximized -- important if we want to avoid the inflationary effects of stimulation. Favoring free software would mean that nearly any work done in an area results in public value. Even bad or poorly implemented systems contain lessons that can be studied and built upon; when the system is proprietary, those lessons die.
Often the problem with SBIR is that work doesn't really result in something that can be commercialized. Even if the system is good, often people can't market it. So perhaps the most politically effective way to do this is to require that if the developer does not make a commercial success of the software within a fixed period, that it be released under a free license.
The whole basis of copyright is to reward authors by giving them legal control over certain uses of their works. In the old days the scope of this was limited to copying.
However software isn't usable without copying, and so we have the end user license. The end user license has two functions (other than to protect the vendor from liability concerns): to allow the user to make the copies of the software necessary to use it, and to allow the vendor to monetize various kinds of uses of the software.
The whole idea of free (as in freedom) software is to minimize the restrictions placed on recipients of the software, both direct and downstream. But making money with software licensed is based on restricting uses of the software and charging for lifting certain of those restrictions.
So you have to consider your business model, and where the money (more precisely, where the sales) come from. There are many variations, such as work-for-hire on free software, support based models, hybrid models with free (from liberal but proprietary to open source) versions and proprietary editions with or without added features, on to the traditional model where you charge every time your customer tries to take a breath.
I'm not one of those people who think selling proprietary, closed software is the moral equivalent of child molestation. But you have to accept that inherent in your question is the idea you are making money from restricting your users' use of the software. You really ought to reframe your question then. It should be "How do I provide my customers with a great value?" You can make sure you interoperate well, conforming to standards as far as possible. You can have liberal copying and backup policies, and avoid onerous copy protection.
Perhaps most of all, providing good value for the money you are charging to remove restrictions on users means keeping things simple. 90% of the reason I favor free software most of the time is most proprietary software doesn't justify the time it takes to figure out what you are allowed to do or what you need to buy. Functionality is divided up into a dozen packages and given obscure names to make it hard to know what you paying for.
That, at least, is certainly not going to happen. He's not getting his messaging services from Verizon, after all, it's going to go through some kind of White House server, and so by law every jot and tiddle will be archived.
Notice I said, "in part". People who believe something as complex as a close election result as "a reason" are buying narratives constructed with an agenda in mind.
We could just as well say that it was tax increases after "read my lips" that did GHW Bush in. Or the deficit. Perot's candidacy was a result of that.
Barney Frank once mocked a Republican lawmaker who, after a measure backed by the Republican leadership soundly failed, observed that if only half the Democrats had voted with the measure it would have passed. Of course if the majority of your side vote one way, and half of the other side votes that way, then the measure wins. Votes are carried, and elections are won, in the margins.
It would only make sense to say Bush lost "because of Perot" if nobody who voted for Clinton would ever consider voting for Bush. That is true of some, even many of his supporters, but subsequent history shows that he was adept at winning voters who in the past had voted Republican and would again in the future. So while Perot was a factor in Bush's loss, he was certainly not more of a factor than Clinton, or Bush himself.
Here's something to ponder. Do presidents carry a wallet? What would they use it for? Surely, they never need to show ID. Why would they need cash, or credit cards? They're not going to take a spur of the moment side trip to Target to do some Christmas shopping.
George H. W. Bush lost the election to Bill Clinton in part because of an infamous incident where he visited a supermarket and was amazed to find that prices were scanned by laser. As president, and vice president for eight years before that, he had never once popped out to the supermarket for a gallon of milk.
This puts the Obama-Blackberry thing in a different perspective. Presidents spend their days wrapped up in a kind of cocoon that isolates them from what are the normal details of life for everyone else. Every mundane need is quietly taken care of without his having to ask. Every communication outside his immediate support system is elaborately screened, planned and orchestrated. So Obama's insistence on keeping "his" blackberry makes a kind of sense in this context; it reflects a desire to have some other channel of communication that isn't completely managed on his behalf by his usual staff.
Slashdot Mirror
Well, that's the same motivation: an institution that wants to gain market share. However if users don't like that feature, the developers who agree with them will fork.
It's a bit like how quantum computing is supposed to work. Only so many developers can work on a project before diminishing returns sets in; a good controversy can be addressed by a good schism, where the question can be resolved by two more optimally sized teams.
What's surprising is that it doesn't happen more often.
I'm talking about the problem of finding the economically optimal assignment of feature to product. Since a feature is either in or not in a product, it's an integer programming, assuming we can express marginal profits as linear equations, which would be the easiest case.
It's kind of a flaw that is endemic to the commercial software development model. This is not to say that that model is useless or F/OSS doesn't have its own problems.
The root of the problem is how we "add value" to a piece of software. Since with F/OSS, software development has a service model, you mainly add value by adding services: documentation, support, consulting. You can't "add value" by adding features to the software, at least if you try to you only get paid once for doing so.
A proprietary software developer can get paid multiple times for adding a piece of value into the software. For software that is sold, this is driven by market segmentation. The least pernicious form of this is the ubiquitous "bronze/silver/gold" model where they try to maximize their return from cheapskates, pragmatists and spendthrifts respectively. If you are cheapskate who needs a feature in the "gold" edition, you're out of luck. In the worst case, it drives a bewildering proliferation of "products", as vendors try to find the division of features that maximizes their returns (which is an instance of the NP-Complete "integer programming problem", only approximations are practical). From a customers standpoint, it sometimes looks like a whirlwind has picked up all the features and dropped them into random pigeonholes.
The "value adding" imperative still applies to free as in free beer proprietary software. In such cases, the developer still is looking to get paid, only in different coin, e.g. control of formats and the market power that comes with it. Adobe benefits from PDF being a non-proprietary format because it encourages adoption, but it is risky because they wouldn't benefit if they did not control the dominant implementations of PDF technology. And they try very hard, I think, to have the best implementations, which leads to the old problem of adding value by adding features. The hope is that by adding features nobody has asked for, when those features are missing from a different implementation, that implementation will be seen as less complete and polished. I think this often works, but it leads to this kind of blowback siutation: security flaws introduced to users systems along with features the user never asked for.
Well, personally I have found Vista more _stable_ than recent versions of Ubuntu. I use Ubuntu because (a) the kinds of tools I currently need are more available and free as in beear and (b) I prefer Gnome to Vista and (c) the kinds of performance issues that matter most to me seem to be a bit better on Linux.
The stability issues I've had with Ubuntu are most likely related to ACPI. In a sense, this is not an issue of OS quality per se but the quality of the ACPI standard, which requires somebody (the manufacturer of the laptop, in the case of Windows, the user in the case of Linux) to figure out how to get ACPI features to work with an operating system.
On the other hand, I've had gut wrenchingly bad performance out of Vista, but the only crash was of the sound system, which poured white noise out of the speakers but left all the other system functions working perfectly. This is a sign that Vista is in some ways better architected than earlier version of Windows, where such a fault would almost certainly have required a hard restart.
Vista on a mainframe might solve what is one of the worst faults of Windows: braindead I/O. Mainframes aren't so much about computing power (although they typically have plenty), they're about really awesome I/O bandwidth. Certainly, we wouldn't be worrying about the kind of hardware and hardware driver faults that bedevil every operating system in one way or the other.
The important thing to remember is that operating systems, as we knew them in the 80s and 90s, don't matter as much as they used to. The focus of application architecture these days is the network. Getting worked up over "operating systems" is getting to be like having big fights over machine instruction set architetures. Sure they matter -- on a certain level -- but a much of their relative merits and problems are thoroughly contained, insulated from applications through multiple layers of abstraction.
Enterprises are very conservative; they like to stick with what works. This move is really just like mainframe systems provide support for ancient character terminal protocols and such. It's not a target for new development, it's more a way of keeping applications running in their current form indefinitely, like a fly trapped in amber.
Come now. Nobody seriously things we should send bricks to the moon. Initially we'll send any structures we need, but any reasonable moon industrialization or colonization will require using local materials. But the first step in using local materials on a large scale will certainly involve moving a great deal of mass, probably far more mass than we'd need to significantly increase our initial capacity there.
That's the point: you need to spend mass to save mass. Sending giant solar furnaces to the Moon would be a huge investment that would almost certainly not pay off on the scale we could contemplate in the immediate future, because we don't have enough investment in the other things we'd need to exploit that. Debris berm building robots sounds about right.
An interesting thought occurs though. One solution to the mass problem is simply patience. Since the Von Neumann approach is based on exploiting exponential growth, but growth doesn't have to be fast at the outset. Suppose you put ten robots on the moon that could scavenge enough material to make one robot in ten years. After the first ten years, you'd have 11 robots. After a hundred years, you'd have about 25 robots. After a 1000 years, you'd have almost 14,000 robots. After 2000 years, you'd have 190 million robots.
It seems likely to me that you'd need to move a lot of mass to bootstrap a Von Neumann style self-replicating system. Assuming you have a "large enough parabolic mirror" is a bit like the proverbial assumption you have a can opener.
This is especially true because we don't know if there are richly concentrated deposits of all materials that are useful in building robots on the Moon, although we have reason to doubt it. Most of the interesting things we think we might get from the moon involve processing huge quantities of material. Things like helium, or water. Prospecting meteor craters might yield a number of interesting things like iridium, and could in theory yield a lot of iron, although getting at that iron brings us back to the bootstrapping problem. We'd have to find a way to do without plastic or rubber though.
Ultimately, it's possible. After all, humans transformed the Earth and we aren't much bigger than a lawnmower; the difference is everything was in place to support humans replicating more humans when we arrived here. Getting to anything like a significant self-supporting industrial capability on the Moon is going to involve countless numbers of small, not so spectacular steps that don't seem like much progress at all. Demonstrating that crude, non-self-replicating robots made with terrestrial materials can do useful structural tasks is a logical step in a direction of a number of goals that are far, far off.
There's case law where lawyers have tried exactly what you are suggesting and, guess what? Sometimes they win, sometimes they lose.
So, yes, I am telling you that a half-decent lawyer, even a pretty damned good one, isn't going to be able talk you out of any kind of copyright trouble you might dream up for yourself. The problem with "clever arguments" by "half-decent" lawyers is that good lawyers have heard all of them before and can kick the crap out of them in their sleep. That's why you check with a lawyer before to beard the copyright lion.
DRM is a different issue, but related in that it provides people in the exclusive possession of material to which they don't own exclusive copyright to exert a measure of copyright-like control over that material.
That's simply not true. Yes, transformation is a necessary condition for claiming fair use. But it is not sufficient. Certain kinds of transformation where there is a recognized public interest (e.g. for parody or critical commentary) are fair use, but degree of transformation is not sufficient defense in itself, it only shows that the transformation was sufficient serves its claimed purpose. If the purpose of the transformation is not protected, then the degree of transformation isn't the deciding factor, but the degree of appropriation.
In any case, the purpose of the artist, which is artistic merit, has no legal force at all.
OK, I've got this keyboard and a display and, and, I can take them apart and use one without the other. I might even be able to use a keyboard that wasn't the one that came with the original...
Now this seems hauntingly familiar .... Where have I seen a form factor that offers this kind of mix and match modularity?
All joking aside, what matters is your data and your tools for manipulating the data. I think it makes sense to give your data and at least some of the software a distinct physical module, like the case of desktop computer but smaller and detachable.
I hesitate to say this, because of the heinous crimes committed in the name of what I'm about to mention, but I'm almost tempted to think of this as the model-view-controller pattern applied to form factor. I should be able to detach my model and use it anyplace I please with whatever peripherals I please.
Well, real estate is not an essential ingredient in creating new real estate, whereas the ideas and expressions are essential ingredients in creating new ideas and new expression.
Therefore even if there is some kind of moral imperative to respect intellectual property rights of creators, the nature of that property and those rights are logically different in nature because all intellectual property exists by virtue of treating previous creations as less than absolute property.
In any case, we don't give real property owners absolute right to control public use of their property either. This is place where public abuses of private rights do from time to time occur, such as stretching utility easements into different kinds of intrusive uses. But that kind of recognition of public interests even in private real estate has legitimate uses and justification.
What he means is great artists do something new with old materials, whether those are themes, motifs, styles, images -- whatever.
However, in a world where art is perpetually governed by technologically and legally enforceable restrictions, the freedom of artists to make fundamentally new works out of old materials is greatly curtailed.
Also consider that the way lawyers enforce their clients' rights is to overreach. In effect they carve out new, extralegal rights for their clients, at least so far as use by those who can't afford even the prospect of litigation. So an IP protection dominated culture would likely be moribund. At it currently stands, copyright is a pragmatic bargain, and the effect of stretching the copyright holder's authority a bit is to tweak the bargain slightly. The effect of stretching a copyright holder's authority indefinitely is a totally different kettle of fish. When culture is distributed exclusively in DRM forms, and defeating DRM is illegal per se regardless of the legal status of the protected content, then culture will cease to have much contemporary relevance as its vocabulary will be limited corporately sponsored and licensed expression, or to works from the distant past.
Well, you spotted an important point. Note that I say that private contractual arrangements are probably indispensable.
I'm quite aware that an absolute dichotomy between the cultural good and private enterprise is unsupportable. What I'm saying is that the greater cultural good cannot be entrusted to private contractual decisions alone, which is considerably less broad an assertion.
I'm not sure you've got that right.
To abandon a trademark, it isn't enough to stop using it. You've got to stop using it with the intent of never using it again. If they discontinued the product with an press release saying, "this is the last 'netbook' branded product we'll ever sell," then they'd have abandoned the trademark at that instant. However nobody does that sort of thing.
On the other hand, you just can't be sly about whether you want to use a trademark, hoping to catch somebody in an indiscretion at a far future date. IIRC, three years without use or attempt to defend is considered prima facie evidence of abandonment.
Now, I believe Psion discountined its last netbook in 2005; however "discontinued" only means that they aren't selling them through the normal wholesale channels. My understanding is that they continued to sell netbooks direct to some corporate accounts for as much as a year later. Perhaps they were liquidating inventory. Also, IIRC, in 2006 Psion filed documents with the FTC indicating its intent to continue the use of trademark. In either case the clock would run out on the trademark some time this year.
Now onto who is virtuous and wronged and who is evil and taking advantage.
It's quite plausible to believe that Intel thought Psion had met the three year criteria for abandonment. It's also quite plausible that Psion had every intent to abandon the trademark, until Intel made it valuable again. This doesn't mean that HP did its homework properly, nor does Intel screwing up mean Psion would win in court (although IANAL). We could imagine that during discovery, Intel could prove that Psion hand no real intent of ever using that brand again, and only started to make trouble when Intel's use of the term made it popular. However, won't happen because these things tend to get discreetly settled and the details thoroughly buried. Maybe that would indicate that Psion was just shaking down Intel, or maybe it means that they decided the cash offer was worth more than than the brand. We'll never know.
Again, IANAL, but if this were referred to me in my part time capacity as philosopher-king, I'd enforce the C&D but grant no damages to Psion over what would be an understandable technical error. If Psion didn't at least announce a new 'netBook' branded product this year, I'd rule the word as public domain.
I actually think Psion might have a case here. They registered the "netBook" trademark years ago, and have been selling "netBook" branded subnotebooks continually over the last decade.
At this point, I think "Netbook" has become generic, but not through a failure of Psion to protect its trademark. Intel, while not selling a device of its own, improperly appropriated Psion's trademark for its own commercial ends (selling Atom processors to subnotebook manufacturers among others). If Sun decided that "J2EE Server" wasn't going to win any adopters, and decided to call them "Websphere servers", that would be clearly be improper, even though they weren't using to refer specifically to their implementation.
All very good, except that medical science isn't up to the task of providing reliable diagnostic and treatment guidelines. We can see this because this is the kind of efficiency program the insurance companies try to impose.
For example, a family acquaintance has a fifth grader who is unusually athletic and physically mature for his age. He swims competitively and practices every day. If you look at him, he is obviously very lean and muscular, but because of his body mass index the insurance company insists he and his pediatrician implement a weight management plan.
Of course it is not beyond the capabilities of science to distinguish between lean body mass and fat. It's just extremely difficult for a committee to take scientific literature and come up with comprehensive, iron-clad guidelines for treatment. And that's what physicians need. When they have a patient, they need guidelines that are right for this patient, not 80% of patients or even 95% of patients.
I think a better approach would be to institute some kind of scientific peer review of treatment decisions between physicians. But of course health care is so damned expensive it has a hard time accommodating new mandates, even if they are good for long term cost control
This is an important point. It leads an irony worth considering, based on the well known principle that local optimizations don't lead to global optimization.
Suppose Alice has content that Bob wants. Very few people would feel that Alice ought to be compelled to provide Bob that content just be cause he wants it. So Bob has to offer Alice some incentive. Alice names a fancy price, which Bob declines to pay. So instead Bob and Alice come to a somewhat more complex arrangement. Bob agrees to pay Alice a smaller price, but restrict his uses to those he is most willing to pay for. From a libertarian viewpoint, this seems innocuous. Alice and Bob are happy, so that should be the end of that.
But Alice and Bob are happy only with respect to the direct effects of this transaction on their interests. They are not necessarily happy about the net effect of information being encumbered this way throughout society. For example, many software licenses forbid publishing benchmarks or reviews without approval. Is the world a better place when people only have a vendor's word for what a piece of software is capable of doing? Are buyers better off?
The rental thing sounds innocuous, but it has important consequences as well. Alice can probably maximize her revenue for her existing content by adopting a rental model. But culture depends on free re-use of ideas, both subtle reuses that don't fall within the scope of copyright law, and obvious reuses.
Disney, for example comes down hard on people who would reuse images from its version Pinnochio, even though that work is by now sixty-four years old. However, their 1940 movie makes free use of the original story by Carlo Collodi that, because it was published in 1883, was only fifty seven years old at the time. Disney would argue, correctly, that at the time their usage was legal whereas their movie is still protected by current copyright. But they can't make a utilitarian argument that in this case protection is for the greater common good. Nor can they reasonably say it represents morally necessary protection of a fundamental right of creators to perpetual intelletual property protection.
The important thing is that when information is controlled solely by private agreements, the net effect of all these local optimizations across society is not globally optimal for the parties living in that society. We can take a lesson from the popular music industry, which is creatively moribund and therefore financially vulnerable. They can blame "piracy" if they like, but if they were producing what consumers considered a good value they would be much better off. Yes, it is possible to download files of unknown quality and provenance for free, but when consumers have access to a good selection through convenient distribution at a fair price, they prefer it. The iTunes store proves this.
But even something like the iTunes store is not a long term solution. As Pablo Picasso said, "Bad artists copy. Great artists steal." Art and culture depend on artists making old works their own.
The progress of useful arts and sciences can't be left entirely to contractual or licensing arrangements between private individuals, as practically useful and even indispensable as those might be. Given the power of technology to restrict information, the law that makes those arrangements possible must also promote the continuing enrichment of the general intellectual welfare, if we are not to suffer dire long term consequences. Unfortunately, we live in a country where politicians aren't very interested in culture, or any aspect of the life of the mind. I was shocked to hear one politician last week mock the idea that studying the spread of venereal disease was a worthwhile use of money. I believe that this anti-intellectualism comes from being so rich for so long that we've come to believe that we can live by managing the prior accumulated wealth of generations of intellectual achievement.
But how do you get developers?
If you recall anything about the days when MS-DOS, then Windows rose to prominence, you will recall that programming was an absolute nightmare, and attempts to make it easier were dreadful kludges that hid the mess just long enough that you could get yourself up to your neck in crap.
But programmers put up with it, and a generation of programmers grew up thinking that this is normal, in order to target the program where the users were.
Now, how do you get users? To a first approximation, of course applications (games) sell platform (console) sales, but what I think this guy is talking about is a different pardigm, the one that drove the adoption of Microsoft platforms in the 1980s: the killer app. There were better desktop platforms than DOS back in, say, 1983, but business customers didn't want to hear about them if they didn't run Lotus 1-2-3. Lotus was making money hand over fist, and so didn't have any immediate incentive to rock the boat, but in retrospect, it would have been worthwhile to support a second OS contender. It's always easier to be sensible in hindsight.
The same may go here. Although Sony might not have made the right decision with the PS3, it might not be as utterly brain-dead as it appears with benefit of hindsight. The reasoning might go, give the programmers hardware that can do all kinds of interesting new things, and while there'll be a lot of struggle, somebody will put those things together into something unlike anything anybody has seen before, and that will be a killer app.
With the benefit of hindsight, we can see certain things. First, you don't necessarily need esoteric technology to give people something they haven't seen and want. Nintendo had a hit by marrying a clever (but not really very complicated) controller to a technologically conservative console then setting the price at an impulse purchase friendly level. This will be a case study taught in business schools for years to come.
Second the idea that companies will spend highly expensive programmer time on figuring out new technology only works if there aren't marginally better uses of that time, like releasing the same game on more than one platform. In an industry where PS3 is not dominant, it can't command that kind of attention.
Finally, if nobody strikes the killer app mother-lode, the victory goes to the most agile company. If developers are ignoring the unique capabilities of the PS3 and targeting multiple platforms with more or less the same stuff, then that means that apps come out sooner and better on the easier to program platform. New developers choosing platforms to target initially will choose the ones with the most revenue and least cost first, then look at the more difficult ones once their marginal revenues drop. So all in all, if exotic architecture doesn't do wonders at the outset, it holds you back rather than catapulting you forward.
Again, this is easy to see in hindsight; in foresight it's much easier to get jazzed over something really cool and exotic. And we have to keep in mind, things might have turned out differently. All it would have taken was one programmer having just the right insight that would allow him to do something on the PS3 unlike anything anybody had seen yet. It's a hell of a gamble betting that kind of investment, but you have to remember the dream of every business is to control a market as its own private fiefdom.
Ummm. I have a question. How exactly does one cue something that is incessant?
Corrupt killer for money? Oh, that's so dreadful.
I'd much rather be killed in the name of some fanatical religious ideology -- or even just plain xenophobic hatred -- than to perish in some tawdry pecuniary scheme. Wouldn't you?
"Not cutting in yet"?
Um. Is that the flavor of the week?
to think of ways to set aside funds for small businesses; ways that would encourage them to adopt and develop open source software.
We might take the Depression era grants that went to artists to decorate many public spaces as a model. Modest grants to people who work in information technology to create freely redistributable solutions public informatics problems would have several important advantages.
(1) Such a plan maintains a domestic informatics workforce in the face of increased pressure to move jobs to low wage countries. Maintaining and increasing the skills of this workforce will make it attractive for industry to turn to it when the economy improves.
(2) Mandating interoperability with open, non-proprietary standards improves the competitiveness of the domestic IT industry, where businesses are too often driven by premature efforts to create some kind of market niche where they ar protected from competition.
(3) The grants should require that the small businesses have a plant to use the work to increase their capabilities, and particularly favor the development of new kinds of technology or application for technology.
(4) Focusing the stimulus money on small businesses allows technology bets to be spread across a greater variety of approaches; it is less likely to introduce what is in effect central planning into engineering decisions. It is also unlikely to strengthen the hand of one big player against all the others because of its skill at obtaining Federal money.
(5) Low margins and ready sweat equity will encourage greater adoption of free software.
(6) There is already a Federal mechanism for doing this; the Small Business Innovative Research grant program.
SBIR currently pays for mostly a lot of boondoggles, although even boondoggles if they are kept close enough to home can be stimulative. However, with the right requirements placed on grant applications the value created can be maximized -- important if we want to avoid the inflationary effects of stimulation. Favoring free software would mean that nearly any work done in an area results in public value. Even bad or poorly implemented systems contain lessons that can be studied and built upon; when the system is proprietary, those lessons die.
Often the problem with SBIR is that work doesn't really result in something that can be commercialized. Even if the system is good, often people can't market it. So perhaps the most politically effective way to do this is to require that if the developer does not make a commercial success of the software within a fixed period, that it be released under a free license.
The whole basis of copyright is to reward authors by giving them legal control over certain uses of their works. In the old days the scope of this was limited to copying.
However software isn't usable without copying, and so we have the end user license. The end user license has two functions (other than to protect the vendor from liability concerns): to allow the user to make the copies of the software necessary to use it, and to allow the vendor to monetize various kinds of uses of the software.
The whole idea of free (as in freedom) software is to minimize the restrictions placed on recipients of the software, both direct and downstream. But making money with software licensed is based on restricting uses of the software and charging for lifting certain of those restrictions.
So you have to consider your business model, and where the money (more precisely, where the sales) come from. There are many variations, such as work-for-hire on free software, support based models, hybrid models with free (from liberal but proprietary to open source) versions and proprietary editions with or without added features, on to the traditional model where you charge every time your customer tries to take a breath.
I'm not one of those people who think selling proprietary, closed software is the moral equivalent of child molestation. But you have to accept that inherent in your question is the idea you are making money from restricting your users' use of the software. You really ought to reframe your question then. It should be "How do I provide my customers with a great value?" You can make sure you interoperate well, conforming to standards as far as possible. You can have liberal copying and backup policies, and avoid onerous copy protection.
Perhaps most of all, providing good value for the money you are charging to remove restrictions on users means keeping things simple. 90% of the reason I favor free software most of the time is most proprietary software doesn't justify the time it takes to figure out what you are allowed to do or what you need to buy. Functionality is divided up into a dozen packages and given obscure names to make it hard to know what you paying for.
That, at least, is certainly not going to happen. He's not getting his messaging services from Verizon, after all, it's going to go through some kind of White House server, and so by law every jot and tiddle will be archived.
Notice I said, "in part". People who believe something as complex as a close election result as "a reason" are buying narratives constructed with an agenda in mind.
We could just as well say that it was tax increases after "read my lips" that did GHW Bush in. Or the deficit. Perot's candidacy was a result of that.
Barney Frank once mocked a Republican lawmaker who, after a measure backed by the Republican leadership soundly failed, observed that if only half the Democrats had voted with the measure it would have passed. Of course if the majority of your side vote one way, and half of the other side votes that way, then the measure wins. Votes are carried, and elections are won, in the margins.
It would only make sense to say Bush lost "because of Perot" if nobody who voted for Clinton would ever consider voting for Bush. That is true of some, even many of his supporters, but subsequent history shows that he was adept at winning voters who in the past had voted Republican and would again in the future. So while Perot was a factor in Bush's loss, he was certainly not more of a factor than Clinton, or Bush himself.
Here's something to ponder. Do presidents carry a wallet? What would they use it for? Surely, they never need to show ID. Why would they need cash, or credit cards? They're not going to take a spur of the moment side trip to Target to do some Christmas shopping.
George H. W. Bush lost the election to Bill Clinton in part because of an infamous incident where he visited a supermarket and was amazed to find that prices were scanned by laser. As president, and vice president for eight years before that, he had never once popped out to the supermarket for a gallon of milk.
This puts the Obama-Blackberry thing in a different perspective. Presidents spend their days wrapped up in a kind of cocoon that isolates them from what are the normal details of life for everyone else. Every mundane need is quietly taken care of without his having to ask. Every communication outside his immediate support system is elaborately screened, planned and orchestrated. So Obama's insistence on keeping "his" blackberry makes a kind of sense in this context; it reflects a desire to have some other channel of communication that isn't completely managed on his behalf by his usual staff.