The the end user, DRM just means they have the ability to play media on their computer that they otherwise couldn't.
Granting for a moment that the user won't be able to play materials on his computer without DRM, this is not a reason to put DRM into the operating system. It's not optimal from the user's standpoint, nor is it really optimal from the copyright holder's standpoint. It's only optimal from the standpoint of the party that controls the operating system.
What benefit to the end user would there be in Microsoft not implementing DRM infrastructure and therefore denying them the ability to play DRM-encumbered content on their computers ?
The benefit is a simpler, more efficient operating system for everything other than playing HD video. In any case you are conflating the two issues. DRM doesn't have to go into the operating system. If people really need it it'd ideally be built into TVs and monitors using a transferable hardware key -- like a GSM SIM.
Not everyone agrees that a free market is what's best for society.
Agreed, however you don't have to pick addlepated examples like crypto-communists to illustrate the point.
The market maximizes the efficiency of resource to the production of excludable benefits. If you want an economic system to produce iPods for around $100-$200, the market is the thing. If you want a system that will ensure breathable air and drinkable water for everyone, it's not. Not without some tinkering.
It seems that a large portion of the public feels that corporations have far too much power and that free market has failed.
And why would that be? I think it's because people have been sold a bill of ideological goods. The power -- the inclination of The Market to grant every human wish has been grossly oversold. Just as its benefits and evils were once grossly oversold.
And now comes the backlash, which of course means medicines worse than the diseases they're supposed to cure.
Again and again, it has struck me that people have been taught to view The Market like God: a benevolent, infallible, personal God who cares about you personally. In truth, what The Market is, is a machine, and a remarkably good one, for the efficient distribution of resources in certain, widely applicable situations. It doesn't know or care who you are or give a fig for human welfare at all. With sufficiently advanced machines, it could well go on after efficiently eliminating the human race.
A milling machine can create all kinds of useful an beneficial things, but if you stick your hand in the wrong place the machine will unthinkingly rip it off. Electricity is a huge benefit to humanity, but if you put a penny in the circuit breaker box, in the course of nature it will burn your house down. If you let individuals externalize costs in a market system, or if you don't find ways of internalizing benefits, The Market will tear your society to shreds.
Electricity is a good example. It would be absurd to outlaw electricity because it causes fires. We would miss the benefits of electricity, one of which is a reduced incidence of fire as things like candles and oil lanterns have fallen into disuse. The Market causes social problems, but not as many as it solves.
That doesn't mean we have to accept everything The Market does as right, or even (as some do) define right by what the Market does. We should let it work unregulated where regulation is trying to do something The Market is inherently good at. Most ideas for manipulating oil prices are counter productive, reducing both the incentive to produce and the incentive to conserve.
On the other hand if the market doesn't charge somebody for their pollution, we should feel free to step in. In some cases, we should outlaw the pollution. In other cases, we can create a market for pollution credits. The latter is also a form of regulation, it just doesn't feel like one.
They did have youth before you were born you know.
The word "hip" has been in the American lexicon since probably before 1900; it became widely known in the 40s. But it took off in the 50s, an era of such material wealth and intellectual conformity that anybody with any individuality was hip. Later on, "hip" got confused with "cool", and deliberately so. Anything desirable will be used as a badge for marketing. But "hip" ("hep" if you prefer) is the one thing that can't be mass produced, which is why you young 'uns think its a joke.
I agree, the UAC prompts are the result of programs designed to get admin rights at any time; I was thinking more of WFP.
With respect to MAFIAA, that's exactly my point. They didn't pay to have it put there. MS put it there in order to attract them as business partners.
MS doesn't sell to users, it sells to people who buy for the users. This is more of the same. The DRM isn't there for the users, it is to own the market for people who want to sell to the users.
Well, to be a geek you have to really, really, really be into something that most people find pointless, incomprehensible, or dull. To be a geek subculture, you have to be organized around something of that nature.
It follows that while many MBAs may be geeks, the MBA subculture is not a geek subculture. The last time I checked, making money had fairly obvious popular appeal.
"Cool" is in the eye of the beholder. There's another term that entered youth culture through jazz, with roots that go all the way back to Mother Africa. The word "hip" comes from a West African word "hep", mean "one who knows."
To be cool, you have to attract the admiration of others. To be hip you must possess knowledge not available to the public at large. For example, I had a friend who'd walk into a certain restaurant on a Friday night and get immediately seated. Even if they had a line waiting, they'd see him at the back of the line and immediately usher him from to a table. That was sort of cool. But it wasn't hip. His secret was available to anybody: you just had to eat there five times a week.
Now many years ago there used to be a restaurant in my neighborhood that opened at midnight and closed at 6:00am. It catered to an eclectic mix of insomniacs, workers leaving the night shift or going to the graveyard shift, musicians hanging out after their gigs, and vampirish denizens of the night (this was back before anybody had heard of the "goth" subculture).
Being a regular at that place made you hip.
I'd say the very definition of "geek" would be "hip" without being recognizably "cool" to most people. Slinging a mean soldering iron makes you hip to electronics, but cool only to your electronics geek buddies.
Well, I think you're asking a good question. I don't even have a problem with UAC; I think it's a good idea actually.
The biggest problem with Vista is that Microsoft was not up front about what you really needed to run it; many "Vista Ready" machines -- weren't. On top of that, Vista went out without drivers for a lot of things, which is a lot of the point of using Windows over something better -- better hardware support.
I think Vista is largely a mixed bag; it was released beta quality, which (in part) probably contributes to its heroic resource demands on hardware. Even the early MacOS 10 releases were pretty inefficient. But with respect to beta software being released as production ready, I'd the same thing about Ubuntu Hardy. It's not really release quality IMHO. However, its easier to take a few lumps on an upgrade if it is (a) free and (b) optional.
And that, I think, is a big part of the reason for Vista hatred. People have decided they don't like riding the upgrade merry-go-round. They got to the point they felt like they could live with XP; they'd probably pay good money for an improved XP. What they got was something which was not as radical as intended (no WinFS), but sufficiently radical to be noticeably rocky and resource intensive. Some of the changes in Vista are unqualified improvements, some of the changes are defensible with implementation faults (UAC and Windows File Protection), and some are there to support Microsoft's agenda alone (DRM).
We may be in an era where customers don't want to be dragged kicking and screaming into a vendor's vision for the future. They'd rather see consistent, incremental improvements. Even the minor changes Microsoft makes in situations like this are starting to piss people off, like renaming control panel applets.
People may not be happy about having to pay for MacOS upgrades, but they're getting incremental improvements on a known quantity. Likewise, I think Ubuntu Hardy is a bit rocky, but the changes are intended to be much the same: incremental improvements on a known quantity. And it's asymptotically approaching that point.
Well, they're probably using some kind of hash based document fingerprinting anyway. Ignoring low entropy characteristics of a word when calculating the fingerprint makes sense, because you can always go back and take it into account once you've eliminated 99.999999999% of the documents on the Internet.
OK, office formats obviously must be open. Where possible, there should be more than one vendor supplying compliant software, and that format should be documented so we can access our documents in the future.
But what about other kinds of information?
In some cases there are good and open formats. For example, for geographic information, we can use Geography Markup Language. It doesn't make sense to use GML as a working format, but as an archival format it makes a lot of sense. Public GIS data should be convertible to, and made available in GML.
Now, what about relational databases? Surely there is much important public information in relational systems, from Access through Oracle, and surely that information should be archivable in an open format. Furthermore, it makes even less sense than in GIS to use the same format for archiving as it does for operational use.
Does anybody know of an open standard for relational data? Would it make more sense to specify an XML schema for each relational schema, or to have a generic XML schema for relational data?
Well, the big problem isn't with the Debian site; it's with the aspect ratio of most monitors these days. Flowing to fill the full horizontal width of the screen is actually a bad idea for such a text-heavy site, because such long lines are hard to read.
Look at the Debian home page in a browser window that is narrowed to allow about 7-10 words per line in the main text, and it looks -- nice. Not coincidentally, the Ubuntu site squeezes about ten words across into the main text.
I'm sitting here on a laptop with a screen that's designed for watching wide screen movies, but it'd be better for me to rotate it 90 degrees if I'm reading text.
Well, getting there in back is not "possessing". And the "law" in this case only applies to determining the ownership of things that are unquestionably the property of some individual. It does not apply to common property, whose ownership is not disputed. I can't erect a tent and picket fence on a quarter acre of Central Park and claim I own it by right of possession.
The real intellectual underpinning of property is utility. In the State of Nature, there is no private property, and therefore anybody can use anything. It's perfectly reasonable to own a herd of pigs; you're having a herd doesn't stop anybody from hunting the wild pigs in the forest. And if you take an individual pig from the wild herd, you aren't depriving anybody else of the use of that herd, because there are still plenty of other wild pigs more or less like the one you harvested left for others.
This has been called by some the "Lockean Proviso", after John Locke's Second Treatise of Government. Basically it says its permissible to claim something as your property if it doesn't deprive everyone else. When you take something out of the public domain, there must be "enough, and as good, left in common for others." What you are proposing is that the first person to take a pig in the forest herd should be able to put up a fence around the forest and claim the entire herd has his own. That's not permissible under the Lockean theory of private property, because it deprives others of what had been theirs, albeit in common.
It's questionable whether the privatization of the Moon could be justified under the Lockean proviso in any case.
By modern standards, England of Locke's day was underpopulated: five and a half million for all of England, versus today's seven and a half million in London alone. And there were the vast, unexplored North American continent. It was unimaginable that the resources of the Earth could possibly be used up. For example, it was unthinkable that the population of a fish species in the sea might be depleted. Humanity lived off the surplus of ecosystems in equilibrium. It made complete sense, then, to colonize and privatize the vast, unbounded wilderness, because it didn't harm anybody except a few aborigines, who'd be compensated for the loss of their land with the priceless treasure of Christianity.
Although the Moon doesn't have aborigines (as far as we know), it's hard to view any reservoir of resources as endless, given the technology of the twenty-first century. Furthermore, planetary (including lunar) resource exploitation is limited by the cost moving mass from one planetary surface to another. If there were vast gold deposits on the Moon, it would hardly be worth going there to get them, because by the time you brought back enough gold to turn a profit, gold would probably be cheaper than iron.
The thing that makes the most sense to bring back from other planets, at least in this stage of technology, is information, which has the highest possible value to mass ratio. Denying scientific access to others is self-defeating. The only economic reason to do so is if you can snag something of unique importance, in which case you don't leave "as much and as good" to others.
Complaining about the bureaucracy with co-workers can be very therapeutic.
Actually, it's not therapueutic.
Pop psychology has adopted a model of the human mind from the early days of the industrial revolution: the steam engine. You correct a dangerously overheated boiler by "letting off steam". You can't fix an overstressed mind that way.
The human mind is something for which we don't have an exact mechanical analog yet, but it certainly doesn't work such a simplistic way. True, you feel better after "letting off steam" by complaining, but you would feel better after doing anything else you found pleasant and companionable.
The truth is that complaining about your situation only reinforces your thinking about it. In return for some modest short term relief you saddle yourself with a tiny bit more of long term burden. So complaining is not therapeutic, even if it makes you feel better. What would be therapeutic is developing alternative ways of thinking about and acting in your current situation. This might make you feel worse in the short term but reduce some of the stress burden you carry in the long term.
ok but there's not tons of old main frames running still?
No, there's not lots of old mainframes running still. But there are probably more new mainframes running than when computers were exclusively located in data centers. Back on the day, your chances of working directly with a mainframe, given that you worked with computers, was 1.0; now it's probably more like 0.001. But there's a lot more people working with computers.
Well, a knack for engineering is something not everybody has. It's also something of economic value. Therefore it is natural for potential students, who really don't know anything about the job market (much less the future job market) to view engineering as a path to upward mobility.
Whether it is a ladder to upward mobility depends on where you are starting from. The engineering ladder starts at socioeconomic bedrock and extends to the upper reaches of the middle of the middle class.The investment banker's ladder starts roughly where the engineer's ladder ends, then ascends into orbit.
If your primary concern is getting rich, and you have family and social connections extending into the boardrooms of major financial institutions, you'd be nuts to become an engineer. You want to get on the banking director career path. On the other hand, if the only reason you can afford college is that you've got a scholarship and you work too many hours on your work-study to network with the elite social clubs, engineering is a good path to get you solidly into the middle class, from which you can become an entrepreneur.
You've never worked with somebody with an engineering degree whose engineering knowledge was bogus?
Lucky you.
It's not "Humanities, bad. Engineering good." It's about good schools and bad schools, and even more so about good students and bad students. You take the pick of the litter from a top drawer school, and put them in a situation where they have to learn, and they'll learn.
Back when I started in this business, almost nobody had any formal computer training. I had a friend from high school who got a degree in humanities from Yale, and a few years later was designing logic circuits. Another friend who did pre-law U Chicago is now a senior engineer at IBM. Granted, these are not ordinary people; they were extremely gifted people who could have mastered anything they set their mind to. They just got the tech bug late in their college careers.
Of course, this was in the pioneering days of computing. Just about anybody who was willing to apply got an interview. It was a tough slog finding people, but sometimes you got a gem. A brilliant person with a humanities degree can apply himself for a couple of years on the job and become a brilliant engineer. An ordinary person with an engineering degree, after a couple years on the job, is still an ordinary person.
I'm not saying engineering education is irrelevant, of course. As you go farther down the talent scale, the more important having a thoroughly guided education is. I think that there are just fewer self-selected engineering students in the world, so there isn't quite as much room for bad programs (although on-line education is opening up new vistas for mediocrity).
There are people out there, who measuring yourself against intellectually is a humbling experience; some of them are in humanities programs. It's comforting to look at somebody who knows things you don't, and dismiss those things as unimportant. But it's not intellectually honest. How many times as engineers do we face this kind of intellectual discrimination? Condescension is a two-way street.
It's funny how so many of our problems turn out to be interconnected.
The chances are, you could live closer to work, but you'd have to compromise. You or your spouse might not have the same choice of jobs. You'd have to accept a smaller home, probably a smaller income. On the other hand, you'd have more free time, you'd get more exercise, you'd be closer to many recreational opportunities.
The lifestyle you live now has compromises. You spend more time in the car, then you have to spend time in the gym, so you have less leisure time. On the other hand, you have more surplus income, so you take nicer vacations. You live in a development that is not designed for people to spend their time in, but you probably have one car for every driver.
I'm not saying you are a bad or unvirtuous person for not choosing to live a simpler life with less time spent servicing your commute and your house, I'm just pointing out that nobody has to live the suburban subdivision lifestyle. It's a choice that you make. If you made a radically different choice, you'd have different problems. So many harried people feel victimized by their lifestyle choices, because they can't imagine being happy living any other way. In fact, psychologists who study happiness think that our general level of happiness depends primarily on our character, not our circumstances.
This reflects, I think, the amazing adaptability of human beings. Virtually anything could be changed in your lifestyle, and you would rise (or fall as the case may be) to the occasion. You could win the lottery and never have to work again, but sooner or later you'll end up roughly where you are now on the happiness scale. As human beings, we're bad at imagining change, or initiating change, but we're masters at accommodating it.
Still, the one thing you never get more of is time. I think most people should consider this more in their choices. You might run out of money, but you definitely will run out of time.
The uncanny valley works both ways though. The closer to human you get, the more the person perceiving looks back towards the robotic edge of the valley. A few steps into the valley from the robotic side keeps the person focused on the far edge.
The trick is not to try to make the robot human like; the trick is to give the robot behaviors that are purposeful and reassuring, and let human imagination bridge the gap.
How many military personnel have developed almost human feelings towards machines who they have "shared" peril with, and which they've trusted their lives to? Ships, aircraft, even bomb removal robots receive real affection.
For that matter, how many people have positive feelings about the Mars rovers Spirit and Opportunity. They seem -- plucky.
Awareness of the human tendency towards anthropomorphizing machines is probably useful to a designer, so long as he doesn't exploit it in an obvious way. The more determined, supportive, and flexible the behavior of the robot, the more a person can come to feel trust and liking towards it, so long as it doesn't do anything (or be anything) which is obviously intended to manipulate.
Just what I want when I'm pinned down by sniper fire. How about some lemon-scented napkins?
Which turns out to be just the thing MyGyver needs; he shreds the napkins into fine paper dust, stuffs some steel tubing ripped from the robot's chassis with it, banging each end shut with debris. A pair of wires from the robot's power harness leads into the tubing.
He then touches the wires to the robot's battery, and the homemade bomb fizzles. Fortunately, the robot's Li-ion battery is constructed from Chinese made knock-off cells. Lacking the safety mechanisms of the genuine Japanese made cells, they provide a satisfying explosion results when the battery is shorted. This solves MyGyver's trapped-by-sniper problem in a rather permanent way.
I wonder if this technology couldn't be used to induce creativity by disrupting habitual thought patterns temporarily.
If you could map the patterns of neuron firing during problem solving, you could turn off inhibitory patterns, allowing wild, crazy, unconventional ideas a little more time to form. Most of the result would be garbage, of course, but sometimes creativity requires letting a not-quite-right idea a little temporary leeway.
You wouldn't want to use this all the time, only after you'd put a significant amount of work into a problem and had fallen into a rut.
You're making the classic engineering mistake: mis-defining the problem.
Disrupting the speech centers of the brain does not preempt attempts at communications. And you need communication; it's just that men, left to themselves, would communicate by passing terse status messages: "I'm hungry"; "I'm angry"; "I'm going to sleep"; "I want sex."
Women send the same status messages, but they seem to gain satisfaction out of the process itself. Therefore they send messages in steganographic form: the basis status messages are there, wrapped all kinds of other data which do not require your immediate action. It pays to pay at least some attention; she may start an "I want sex" status message by telling you that her sister's neighbor's aunt is going in for a gall stone operation.
The wise man knows that he should celebrate the differences between the sexes if he wants to celebrate the difference between the sexes.
Therefore, it is best to cultivate the skill of appearing mildly interested and engaged, making reflexive, non-committal listening responses, and paying just enough attention to pick out any cues that indicate something that requires immediate action. It's a lot like driving, actually. You get that sixth sense for when somebody is going to cut you off, or roll into an intersection without coming to a stop. It's not magic, it's practice.
Maybe a better way of saying this is that open source programmers aren't better programmers than closed source ones.
But nobody ever said open source programmers are better. The argument is that open source software gets continually better from a user's perspective. If it doesn't for enough users, somebody else gives them what they want. If you aren't happy with SUSE's direction, you can go to RHEL and vice versa without creating a lot of fuss. Chances are somebody is taking the same basic building blocks and putting them together more like you want them.
In any case, I'm not impressed by including the Widows kernel to Unix kernels. It's not very useful to compare a microkernel to a monolithic kernel without including enough fo the modules that go around it to implement the same functionality. They should compare it to Mach instead.
Here's a thought. The IBM PC was introduced in 1981 at a cost of $3045 -- about $7570 in current dollars. PC-DOS, which granted is an IBM product, but we'll use that for a stand-in for MS-DOS for now, PC-DOS set you back $40. That's $100 in current dollars.
Vista Home lists at $129, but sells for a street price of around -- wait for it -- $100.
But chances are you ain't going to run it on a $7500 computer. Let's stipulate that a reasonably usable computer will set you back, display and all about $1200 (I know you can get by spending a lot less). So today you are spending 8% for your operating system, as opposed to 1.3% in 1981.
Now, let's think about a $200 laptop with a $3 operating system. You are spending -- wait for it -- about 1.5% on the operating system.
Coincidence? Hard to say. What's clear is that Microsoft is charging the same amount of constant dollars for operating systems as when PCs were starting the adoption curve in 1981. They are proposing to charge roughly the same proportion of system price in places where computer adoption is just getting started.
What's confusing to you is that you're mixing up three different classes of things: non-profit organizations; non-profit organizations with a charitable purpose; and non-profit organizations with a charitable purpose that qualify for Federal tax exemption under section 501(c) 3 of the tax code. Each of these succeeding classes is smaller than the previous one.
Even under 501(c), there are twenty seven other sections under which a non-profit can qualify for some degree of tax exemption. Veterans organizations qualify under 501(c)19, for example. Not all non-profits are charitable (e.g. private clubs); not all charitable organizations are tax exempt; not all tax exempt organizations are exempt under 501(c)3.
But even for 501(c)3s, the analysis stands: you must make a profit. The profit goes into next year's budget, or into the endowment. You can't distribute the profit for the private benefit of a set of "owners", say the board or people who control the board.
Of course, it's not hard to get around this limitation. I could tell you stories that would would shock you, and they're not even the worst things that happen out there. Charity attracts the best and worst of humanity's character, and there is plenty of room for the worst to flourish. No politician is going to go after bad charities, because the rogues and cheaters in charity are well connected and quite expert at taking care of themselves. And no politician wants to be known as the scourge of charities, even though culling the bad ones would be a great service to the good ones.
What really pisses me off is that including XP on these things will increase the cost directly and indirectly ($3+$7) a total of 10% of the target $100 price of the laptop.
Well, that's beside the point, since the target is not reachable yet. It's more like a 5% increase. And maybe not that. If they sell waaay more laptops, it may end up being a wash, or even cheaper, because of larger volume purchases of components.
The shelf life of the original vision was always limited; it was based on the idea that there was no hardware appropriate for, and affordable to, developing countries. While the appropriate is still up for grabs, affordable is just a matter of time. A hundred dollar laptop in a thousand dollar laptop world is dramatic. A two hundred dollar laptop in a world with four hundred dollar laptops is less so. Granted the Eepc doesn't have the battery life needed, but the hardware dimension of the digital divide continues to narrow every year.
Ivan Krstic's rant is actually quite insightful. He's pissed at Negroponte, as well as the other people who are pissed at Negroponte, because they're having the wrong argument.
The vision that got everyone excited was to put education and collaboration tools into the hands of students who didn't have them before. Worrying about adding $7 to the cost of the hardware is silly, when you don't have any means to actually track the distribution of that hardware. If you ship a thousand units, and only a hundred make it into the hands of the intended users, you've just paid $200,000 to deploy 100 laptops, or $2000/laptop.
It's not an either/or question, but it's a little like one. The project is engulfed in this huge controversy of $7-$10, while it is not yet dealing with the $1800 question. The problem is that we've lost focus on the educational mission.
The Windows issue is a total side show. The real problem is about "resources", which is a polite way to say "money". Worrying about $10 per unit is the kind of thing that in business I call a "problem we'd like to have". The real question is whether you've really enabled your focus customers to have that problem.
The XO would make a fine Xubuntu or DSL workstation. So why develop Sugar at all? That's a bigger question than whether Sugar should run on Windows. It's obviously a nice idea to reinvent the GUI, but is that the best use of project resources? Why not develop all the collaboration and educational tools as open source, and let anybody who wants run it on Linux or port it to Windows or MacOS?
Well, the short answer is that a new, education centric user interface is a nice thing to have. But is it really the biggest obstacle to the vision that could be removed with the "resources" that have been devoted to it? Charities frequently run on ego as much as idealism; when you look at them closely, it's often hard to assemble the big pictures from the pixels.
OLPC has done the world a great service, by forcing manufacturers to get into the low end game. The existence of this game is good for impoverished users. It's also good for Linux. OLPC has changed the landscape, and it would probably be a good thing if it reoriented itself to accomplish its mission in that landscape.
When it comes to doing it "for real", things look a lot different than people imagined up front.
"Non profit" just means they'll have a zero budget balance, i.e., no money to share after the year is up.
That's not correct. You can no more run a non-profit without a surplus (in other words a "profit") than you can any other enterprise. It'd be too financially risky to give yourself no slack, and too financially irresponsible to spend your slack wildly at the end of the year.
If you've ever looked at a non-profit financial statement, the difference from a for-profit is that "Owner's Equity" on the balance sheet is called "Retained Earnings". And that indicates the fundamental difference, which is not so much a matter of how the organization budgets (although that is somewhat different), or the kinds of revenue raising activities it undertakes (which is less different than you might think), as it is purpose. For-profit enterprises exist to generate value for, then distribute that value to, the owners. Non-profit enterprises exist to perform a mission, although that can be to create value for some target beneficiaries.
Just as for-profit enterprises feel they need a mission to generate profit efficiently, non-profit enterprise need profit to pursue their mission effectively. If you run out of cash, or if the creditors are beating down the door, you can't change the world.
The mission of a non-profit is usually charitable or educational, but not necessarily. A non-profit can be formed for the private benefit of the people creating it, for example some types of cooperatives. The "Best Western" hotel organization in the United States is a non-profit cooperative. The REI outdoor sporting goods stores are a non-profit cooperative that is nearly indistinguishable from a for-profit; the difference is that the dividends paid to members are based on the members' purchases. It is not a reward for investment, it is a repayment for spending more than the minimum than could be charged sustainably.
And, in the end, it is all about sustainability. A "mission", for a for-profit business, is a necessary evil. You could generate revenue in a completely opportunistic way, and it often pays to be somewhat opportunistic, but ultimately no organization can be good at everything, nor can it court everyone as customers.
Profit, for the non-profit enterprise, is likewise a necessary evil. OLPC could charge less for each PC, and get more into the hands of students as long as their cash held out which would not be for long.
So, in many ways, you run a charity (which is what we are talking about here) just the same as business. Oh, you have people who just give you money, but most of that money is what is called "encumbered"; it's no different from being a consultancy that gets an up-front payment for some service they are going to provide. You don't book it as income until the work is done.
This means you consider exactly the same factors a business does when you make a strategic decision. The difference is this: in a push-comes-to-shove scenario, you choose maximizing mission over maximizing profit. For you, the profit is there to support the mission; for a business it is the other way around.
Slashdot Mirror
Granting for a moment that the user won't be able to play materials on his computer without DRM, this is not a reason to put DRM into the operating system. It's not optimal from the user's standpoint, nor is it really optimal from the copyright holder's standpoint. It's only optimal from the standpoint of the party that controls the operating system.
The benefit is a simpler, more efficient operating system for everything other than playing HD video. In any case you are conflating the two issues. DRM doesn't have to go into the operating system. If people really need it it'd ideally be built into TVs and monitors using a transferable hardware key -- like a GSM SIM.
Next encrypt it with a one-time pad. Then burn all copies of the pad.
Agreed, however you don't have to pick addlepated examples like crypto-communists to illustrate the point.
The market maximizes the efficiency of resource to the production of excludable benefits. If you want an economic system to produce iPods for around $100-$200, the market is the thing. If you want a system that will ensure breathable air and drinkable water for everyone, it's not. Not without some tinkering.
And why would that be? I think it's because people have been sold a bill of ideological goods. The power -- the inclination of The Market to grant every human wish has been grossly oversold. Just as its benefits and evils were once grossly oversold.
And now comes the backlash, which of course means medicines worse than the diseases they're supposed to cure.
Again and again, it has struck me that people have been taught to view The Market like God: a benevolent, infallible, personal God who cares about you personally. In truth, what The Market is, is a machine, and a remarkably good one, for the efficient distribution of resources in certain, widely applicable situations. It doesn't know or care who you are or give a fig for human welfare at all. With sufficiently advanced machines, it could well go on after efficiently eliminating the human race.
A milling machine can create all kinds of useful an beneficial things, but if you stick your hand in the wrong place the machine will unthinkingly rip it off. Electricity is a huge benefit to humanity, but if you put a penny in the circuit breaker box, in the course of nature it will burn your house down. If you let individuals externalize costs in a market system, or if you don't find ways of internalizing benefits, The Market will tear your society to shreds.
Electricity is a good example. It would be absurd to outlaw electricity because it causes fires. We would miss the benefits of electricity, one of which is a reduced incidence of fire as things like candles and oil lanterns have fallen into disuse. The Market causes social problems, but not as many as it solves.
That doesn't mean we have to accept everything The Market does as right, or even (as some do) define right by what the Market does. We should let it work unregulated where regulation is trying to do something The Market is inherently good at. Most ideas for manipulating oil prices are counter productive, reducing both the incentive to produce and the incentive to conserve.
On the other hand if the market doesn't charge somebody for their pollution, we should feel free to step in. In some cases, we should outlaw the pollution. In other cases, we can create a market for pollution credits. The latter is also a form of regulation, it just doesn't feel like one.
They did have youth before you were born you know.
The word "hip" has been in the American lexicon since probably before 1900; it became widely known in the 40s. But it took off in the 50s, an era of such material wealth and intellectual conformity that anybody with any individuality was hip. Later on, "hip" got confused with "cool", and deliberately so. Anything desirable will be used as a badge for marketing. But "hip" ("hep" if you prefer) is the one thing that can't be mass produced, which is why you young 'uns think its a joke.
I agree, the UAC prompts are the result of programs designed to get admin rights at any time; I was thinking more of WFP.
With respect to MAFIAA, that's exactly my point. They didn't pay to have it put there. MS put it there in order to attract them as business partners.
MS doesn't sell to users, it sells to people who buy for the users. This is more of the same. The DRM isn't there for the users, it is to own the market for people who want to sell to the users.
Well, to be a geek you have to really, really, really be into something that most people find pointless, incomprehensible, or dull. To be a geek subculture, you have to be organized around something of that nature.
It follows that while many MBAs may be geeks, the MBA subculture is not a geek subculture. The last time I checked, making money had fairly obvious popular appeal.
"Cool" is in the eye of the beholder. There's another term that entered youth culture through jazz, with roots that go all the way back to Mother Africa. The word "hip" comes from a West African word "hep", mean "one who knows."
To be cool, you have to attract the admiration of others. To be hip you must possess knowledge not available to the public at large. For example, I had a friend who'd walk into a certain restaurant on a Friday night and get immediately seated. Even if they had a line waiting, they'd see him at the back of the line and immediately usher him from to a table. That was sort of cool. But it wasn't hip. His secret was available to anybody: you just had to eat there five times a week.
Now many years ago there used to be a restaurant in my neighborhood that opened at midnight and closed at 6:00am. It catered to an eclectic mix of insomniacs, workers leaving the night shift or going to the graveyard shift, musicians hanging out after their gigs, and vampirish denizens of the night (this was back before anybody had heard of the "goth" subculture).
Being a regular at that place made you hip.
I'd say the very definition of "geek" would be "hip" without being recognizably "cool" to most people. Slinging a mean soldering iron makes you hip to electronics, but cool only to your electronics geek buddies.
Well, I think you're asking a good question. I don't even have a problem with UAC; I think it's a good idea actually.
The biggest problem with Vista is that Microsoft was not up front about what you really needed to run it; many "Vista Ready" machines -- weren't. On top of that, Vista went out without drivers for a lot of things, which is a lot of the point of using Windows over something better -- better hardware support.
I think Vista is largely a mixed bag; it was released beta quality, which (in part) probably contributes to its heroic resource demands on hardware. Even the early MacOS 10 releases were pretty inefficient. But with respect to beta software being released as production ready, I'd the same thing about Ubuntu Hardy. It's not really release quality IMHO. However, its easier to take a few lumps on an upgrade if it is (a) free and (b) optional.
And that, I think, is a big part of the reason for Vista hatred. People have decided they don't like riding the upgrade merry-go-round. They got to the point they felt like they could live with XP; they'd probably pay good money for an improved XP. What they got was something which was not as radical as intended (no WinFS), but sufficiently radical to be noticeably rocky and resource intensive. Some of the changes in Vista are unqualified improvements, some of the changes are defensible with implementation faults (UAC and Windows File Protection), and some are there to support Microsoft's agenda alone (DRM).
We may be in an era where customers don't want to be dragged kicking and screaming into a vendor's vision for the future. They'd rather see consistent, incremental improvements. Even the minor changes Microsoft makes in situations like this are starting to piss people off, like renaming control panel applets.
People may not be happy about having to pay for MacOS upgrades, but they're getting incremental improvements on a known quantity. Likewise, I think Ubuntu Hardy is a bit rocky, but the changes are intended to be much the same: incremental improvements on a known quantity. And it's asymptotically approaching that point.
Well, they're probably using some kind of hash based document fingerprinting anyway. Ignoring low entropy characteristics of a word when calculating the fingerprint makes sense, because you can always go back and take it into account once you've eliminated 99.999999999% of the documents on the Internet.
Nice, nick, by the way.
OK, office formats obviously must be open. Where possible, there should be more than one vendor supplying compliant software, and that format should be documented so we can access our documents in the future.
But what about other kinds of information?
In some cases there are good and open formats. For example, for geographic information, we can use Geography Markup Language. It doesn't make sense to use GML as a working format, but as an archival format it makes a lot of sense. Public GIS data should be convertible to, and made available in GML.
Now, what about relational databases? Surely there is much important public information in relational systems, from Access through Oracle, and surely that information should be archivable in an open format. Furthermore, it makes even less sense than in GIS to use the same format for archiving as it does for operational use.
Does anybody know of an open standard for relational data? Would it make more sense to specify an XML schema for each relational schema, or to have a generic XML schema for relational data?
Well, the big problem isn't with the Debian site; it's with the aspect ratio of most monitors these days. Flowing to fill the full horizontal width of the screen is actually a bad idea for such a text-heavy site, because such long lines are hard to read.
Look at the Debian home page in a browser window that is narrowed to allow about 7-10 words per line in the main text, and it looks -- nice. Not coincidentally, the Ubuntu site squeezes about ten words across into the main text.
I'm sitting here on a laptop with a screen that's designed for watching wide screen movies, but it'd be better for me to rotate it 90 degrees if I'm reading text.
Well, getting there in back is not "possessing". And the "law" in this case only applies to determining the ownership of things that are unquestionably the property of some individual. It does not apply to common property, whose ownership is not disputed. I can't erect a tent and picket fence on a quarter acre of Central Park and claim I own it by right of possession.
The real intellectual underpinning of property is utility. In the State of Nature, there is no private property, and therefore anybody can use anything. It's perfectly reasonable to own a herd of pigs; you're having a herd doesn't stop anybody from hunting the wild pigs in the forest. And if you take an individual pig from the wild herd, you aren't depriving anybody else of the use of that herd, because there are still plenty of other wild pigs more or less like the one you harvested left for others.
This has been called by some the "Lockean Proviso", after John Locke's Second Treatise of Government. Basically it says its permissible to claim something as your property if it doesn't deprive everyone else. When you take something out of the public domain, there must be "enough, and as good, left in common for others." What you are proposing is that the first person to take a pig in the forest herd should be able to put up a fence around the forest and claim the entire herd has his own. That's not permissible under the Lockean theory of private property, because it deprives others of what had been theirs, albeit in common.
It's questionable whether the privatization of the Moon could be justified under the Lockean proviso in any case.
By modern standards, England of Locke's day was underpopulated: five and a half million for all of England, versus today's seven and a half million in London alone. And there were the vast, unexplored North American continent. It was unimaginable that the resources of the Earth could possibly be used up. For example, it was unthinkable that the population of a fish species in the sea might be depleted. Humanity lived off the surplus of ecosystems in equilibrium. It made complete sense, then, to colonize and privatize the vast, unbounded wilderness, because it didn't harm anybody except a few aborigines, who'd be compensated for the loss of their land with the priceless treasure of Christianity.
Although the Moon doesn't have aborigines (as far as we know), it's hard to view any reservoir of resources as endless, given the technology of the twenty-first century. Furthermore, planetary (including lunar) resource exploitation is limited by the cost moving mass from one planetary surface to another. If there were vast gold deposits on the Moon, it would hardly be worth going there to get them, because by the time you brought back enough gold to turn a profit, gold would probably be cheaper than iron.
The thing that makes the most sense to bring back from other planets, at least in this stage of technology, is information, which has the highest possible value to mass ratio. Denying scientific access to others is self-defeating. The only economic reason to do so is if you can snag something of unique importance, in which case you don't leave "as much and as good" to others.
Actually, it's not therapueutic.
Pop psychology has adopted a model of the human mind from the early days of the industrial revolution: the steam engine. You correct a dangerously overheated boiler by "letting off steam". You can't fix an overstressed mind that way.
The human mind is something for which we don't have an exact mechanical analog yet, but it certainly doesn't work such a simplistic way. True, you feel better after "letting off steam" by complaining, but you would feel better after doing anything else you found pleasant and companionable.
The truth is that complaining about your situation only reinforces your thinking about it. In return for some modest short term relief you saddle yourself with a tiny bit more of long term burden. So complaining is not therapeutic, even if it makes you feel better. What would be therapeutic is developing alternative ways of thinking about and acting in your current situation. This might make you feel worse in the short term but reduce some of the stress burden you carry in the long term.
No, there's not lots of old mainframes running still. But there are probably more new mainframes running than when computers were exclusively located in data centers. Back on the day, your chances of working directly with a mainframe, given that you worked with computers, was 1.0; now it's probably more like 0.001. But there's a lot more people working with computers.
Well, a knack for engineering is something not everybody has. It's also something of economic value. Therefore it is natural for potential students, who really don't know anything about the job market (much less the future job market) to view engineering as a path to upward mobility.
Whether it is a ladder to upward mobility depends on where you are starting from. The engineering ladder starts at socioeconomic bedrock and extends to the upper reaches of the middle of the middle class.The investment banker's ladder starts roughly where the engineer's ladder ends, then ascends into orbit.
If your primary concern is getting rich, and you have family and social connections extending into the boardrooms of major financial institutions, you'd be nuts to become an engineer. You want to get on the banking director career path. On the other hand, if the only reason you can afford college is that you've got a scholarship and you work too many hours on your work-study to network with the elite social clubs, engineering is a good path to get you solidly into the middle class, from which you can become an entrepreneur.
You've never worked with somebody with an engineering degree whose engineering knowledge was bogus?
Lucky you.
It's not "Humanities, bad. Engineering good." It's about good schools and bad schools, and even more so about good students and bad students. You take the pick of the litter from a top drawer school, and put them in a situation where they have to learn, and they'll learn.
Back when I started in this business, almost nobody had any formal computer training. I had a friend from high school who got a degree in humanities from Yale, and a few years later was designing logic circuits. Another friend who did pre-law U Chicago is now a senior engineer at IBM. Granted, these are not ordinary people; they were extremely gifted people who could have mastered anything they set their mind to. They just got the tech bug late in their college careers.
Of course, this was in the pioneering days of computing. Just about anybody who was willing to apply got an interview. It was a tough slog finding people, but sometimes you got a gem. A brilliant person with a humanities degree can apply himself for a couple of years on the job and become a brilliant engineer. An ordinary person with an engineering degree, after a couple years on the job, is still an ordinary person.
I'm not saying engineering education is irrelevant, of course. As you go farther down the talent scale, the more important having a thoroughly guided education is. I think that there are just fewer self-selected engineering students in the world, so there isn't quite as much room for bad programs (although on-line education is opening up new vistas for mediocrity).
There are people out there, who measuring yourself against intellectually is a humbling experience; some of them are in humanities programs. It's comforting to look at somebody who knows things you don't, and dismiss those things as unimportant. But it's not intellectually honest. How many times as engineers do we face this kind of intellectual discrimination? Condescension is a two-way street.
It's funny how so many of our problems turn out to be interconnected.
The chances are, you could live closer to work, but you'd have to compromise. You or your spouse might not have the same choice of jobs. You'd have to accept a smaller home, probably a smaller income. On the other hand, you'd have more free time, you'd get more exercise, you'd be closer to many recreational opportunities.
The lifestyle you live now has compromises. You spend more time in the car, then you have to spend time in the gym, so you have less leisure time. On the other hand, you have more surplus income, so you take nicer vacations. You live in a development that is not designed for people to spend their time in, but you probably have one car for every driver.
I'm not saying you are a bad or unvirtuous person for not choosing to live a simpler life with less time spent servicing your commute and your house, I'm just pointing out that nobody has to live the suburban subdivision lifestyle. It's a choice that you make. If you made a radically different choice, you'd have different problems. So many harried people feel victimized by their lifestyle choices, because they can't imagine being happy living any other way. In fact, psychologists who study happiness think that our general level of happiness depends primarily on our character, not our circumstances.
This reflects, I think, the amazing adaptability of human beings. Virtually anything could be changed in your lifestyle, and you would rise (or fall as the case may be) to the occasion. You could win the lottery and never have to work again, but sooner or later you'll end up roughly where you are now on the happiness scale. As human beings, we're bad at imagining change, or initiating change, but we're masters at accommodating it.
Still, the one thing you never get more of is time. I think most people should consider this more in their choices. You might run out of money, but you definitely will run out of time.
The uncanny valley works both ways though. The closer to human you get, the more the person perceiving looks back towards the robotic edge of the valley. A few steps into the valley from the robotic side keeps the person focused on the far edge.
The trick is not to try to make the robot human like; the trick is to give the robot behaviors that are purposeful and reassuring, and let human imagination bridge the gap.
How many military personnel have developed almost human feelings towards machines who they have "shared" peril with, and which they've trusted their lives to? Ships, aircraft, even bomb removal robots receive real affection.
For that matter, how many people have positive feelings about the Mars rovers Spirit and Opportunity. They seem -- plucky.
Awareness of the human tendency towards anthropomorphizing machines is probably useful to a designer, so long as he doesn't exploit it in an obvious way. The more determined, supportive, and flexible the behavior of the robot, the more a person can come to feel trust and liking towards it, so long as it doesn't do anything (or be anything) which is obviously intended to manipulate.
Which turns out to be just the thing MyGyver needs; he shreds the napkins into fine paper dust, stuffs some steel tubing ripped from the robot's chassis with it, banging each end shut with debris. A pair of wires from the robot's power harness leads into the tubing.
He then touches the wires to the robot's battery, and the homemade bomb fizzles. Fortunately, the robot's Li-ion battery is constructed from Chinese made knock-off cells. Lacking the safety mechanisms of the genuine Japanese made cells, they provide a satisfying explosion results when the battery is shorted. This solves MyGyver's trapped-by-sniper problem in a rather permanent way.
I wonder if this technology couldn't be used to induce creativity by disrupting habitual thought patterns temporarily.
If you could map the patterns of neuron firing during problem solving, you could turn off inhibitory patterns, allowing wild, crazy, unconventional ideas a little more time to form. Most of the result would be garbage, of course, but sometimes creativity requires letting a not-quite-right idea a little temporary leeway.
You wouldn't want to use this all the time, only after you'd put a significant amount of work into a problem and had fallen into a rut.
You're making the classic engineering mistake: mis-defining the problem.
Disrupting the speech centers of the brain does not preempt attempts at communications. And you need communication; it's just that men, left to themselves, would communicate by passing terse status messages: "I'm hungry"; "I'm angry"; "I'm going to sleep"; "I want sex."
Women send the same status messages, but they seem to gain satisfaction out of the process itself. Therefore they send messages in steganographic form: the basis status messages are there, wrapped all kinds of other data which do not require your immediate action. It pays to pay at least some attention; she may start an "I want sex" status message by telling you that her sister's neighbor's aunt is going in for a gall stone operation.
The wise man knows that he should celebrate the differences between the sexes if he wants to celebrate the difference between the sexes.
Therefore, it is best to cultivate the skill of appearing mildly interested and engaged, making reflexive, non-committal listening responses, and paying just enough attention to pick out any cues that indicate something that requires immediate action. It's a lot like driving, actually. You get that sixth sense for when somebody is going to cut you off, or roll into an intersection without coming to a stop. It's not magic, it's practice.
Well, for a certain definition of "quality".
Maybe a better way of saying this is that open source programmers aren't better programmers than closed source ones.
But nobody ever said open source programmers are better. The argument is that open source software gets continually better from a user's perspective. If it doesn't for enough users, somebody else gives them what they want. If you aren't happy with SUSE's direction, you can go to RHEL and vice versa without creating a lot of fuss. Chances are somebody is taking the same basic building blocks and putting them together more like you want them.
In any case, I'm not impressed by including the Widows kernel to Unix kernels. It's not very useful to compare a microkernel to a monolithic kernel without including enough fo the modules that go around it to implement the same functionality. They should compare it to Mach instead.
Here's a thought. The IBM PC was introduced in 1981 at a cost of $3045 -- about $7570 in current dollars. PC-DOS, which granted is an IBM product, but we'll use that for a stand-in for MS-DOS for now, PC-DOS set you back $40. That's $100 in current dollars.
Vista Home lists at $129, but sells for a street price of around -- wait for it -- $100.
But chances are you ain't going to run it on a $7500 computer. Let's stipulate that a reasonably usable computer will set you back, display and all about $1200 (I know you can get by spending a lot less). So today you are spending 8% for your operating system, as opposed to 1.3% in 1981.
Now, let's think about a $200 laptop with a $3 operating system. You are spending -- wait for it -- about 1.5% on the operating system.
Coincidence? Hard to say. What's clear is that Microsoft is charging the same amount of constant dollars for operating systems as when PCs were starting the adoption curve in 1981. They are proposing to charge roughly the same proportion of system price in places where computer adoption is just getting started.
What's confusing to you is that you're mixing up three different classes of things: non-profit organizations; non-profit organizations with a charitable purpose; and non-profit organizations with a charitable purpose that qualify for Federal tax exemption under section 501(c) 3 of the tax code. Each of these succeeding classes is smaller than the previous one.
Even under 501(c), there are twenty seven other sections under which a non-profit can qualify for some degree of tax exemption. Veterans organizations qualify under 501(c)19, for example. Not all non-profits are charitable (e.g. private clubs); not all charitable organizations are tax exempt; not all tax exempt organizations are exempt under 501(c)3.
But even for 501(c)3s, the analysis stands: you must make a profit. The profit goes into next year's budget, or into the endowment. You can't distribute the profit for the private benefit of a set of "owners", say the board or people who control the board.
Of course, it's not hard to get around this limitation. I could tell you stories that would would shock you, and they're not even the worst things that happen out there. Charity attracts the best and worst of humanity's character, and there is plenty of room for the worst to flourish. No politician is going to go after bad charities, because the rogues and cheaters in charity are well connected and quite expert at taking care of themselves. And no politician wants to be known as the scourge of charities, even though culling the bad ones would be a great service to the good ones.
Well, that's beside the point, since the target is not reachable yet. It's more like a 5% increase. And maybe not that. If they sell waaay more laptops, it may end up being a wash, or even cheaper, because of larger volume purchases of components.
The shelf life of the original vision was always limited; it was based on the idea that there was no hardware appropriate for, and affordable to, developing countries. While the appropriate is still up for grabs, affordable is just a matter of time. A hundred dollar laptop in a thousand dollar laptop world is dramatic. A two hundred dollar laptop in a world with four hundred dollar laptops is less so. Granted the Eepc doesn't have the battery life needed, but the hardware dimension of the digital divide continues to narrow every year.
Ivan Krstic's rant is actually quite insightful. He's pissed at Negroponte, as well as the other people who are pissed at Negroponte, because they're having the wrong argument.
The vision that got everyone excited was to put education and collaboration tools into the hands of students who didn't have them before. Worrying about adding $7 to the cost of the hardware is silly, when you don't have any means to actually track the distribution of that hardware. If you ship a thousand units, and only a hundred make it into the hands of the intended users, you've just paid $200,000 to deploy 100 laptops, or $2000/laptop.
It's not an either/or question, but it's a little like one. The project is engulfed in this huge controversy of $7-$10, while it is not yet dealing with the $1800 question. The problem is that we've lost focus on the educational mission.
The Windows issue is a total side show. The real problem is about "resources", which is a polite way to say "money". Worrying about $10 per unit is the kind of thing that in business I call a "problem we'd like to have". The real question is whether you've really enabled your focus customers to have that problem.
The XO would make a fine Xubuntu or DSL workstation. So why develop Sugar at all? That's a bigger question than whether Sugar should run on Windows. It's obviously a nice idea to reinvent the GUI, but is that the best use of project resources? Why not develop all the collaboration and educational tools as open source, and let anybody who wants run it on Linux or port it to Windows or MacOS?
Well, the short answer is that a new, education centric user interface is a nice thing to have. But is it really the biggest obstacle to the vision that could be removed with the "resources" that have been devoted to it? Charities frequently run on ego as much as idealism; when you look at them closely, it's often hard to assemble the big pictures from the pixels.
OLPC has done the world a great service, by forcing manufacturers to get into the low end game. The existence of this game is good for impoverished users. It's also good for Linux. OLPC has changed the landscape, and it would probably be a good thing if it reoriented itself to accomplish its mission in that landscape.
When it comes to doing it "for real", things look a lot different than people imagined up front.
That's not correct. You can no more run a non-profit without a surplus (in other words a "profit") than you can any other enterprise. It'd be too financially risky to give yourself no slack, and too financially irresponsible to spend your slack wildly at the end of the year.
If you've ever looked at a non-profit financial statement, the difference from a for-profit is that "Owner's Equity" on the balance sheet is called "Retained Earnings". And that indicates the fundamental difference, which is not so much a matter of how the organization budgets (although that is somewhat different), or the kinds of revenue raising activities it undertakes (which is less different than you might think), as it is purpose. For-profit enterprises exist to generate value for, then distribute that value to, the owners. Non-profit enterprises exist to perform a mission, although that can be to create value for some target beneficiaries.
Just as for-profit enterprises feel they need a mission to generate profit efficiently, non-profit enterprise need profit to pursue their mission effectively. If you run out of cash, or if the creditors are beating down the door, you can't change the world.
The mission of a non-profit is usually charitable or educational, but not necessarily. A non-profit can be formed for the private benefit of the people creating it, for example some types of cooperatives. The "Best Western" hotel organization in the United States is a non-profit cooperative. The REI outdoor sporting goods stores are a non-profit cooperative that is nearly indistinguishable from a for-profit; the difference is that the dividends paid to members are based on the members' purchases. It is not a reward for investment, it is a repayment for spending more than the minimum than could be charged sustainably.
And, in the end, it is all about sustainability. A "mission", for a for-profit business, is a necessary evil. You could generate revenue in a completely opportunistic way, and it often pays to be somewhat opportunistic, but ultimately no organization can be good at everything, nor can it court everyone as customers.
Profit, for the non-profit enterprise, is likewise a necessary evil. OLPC could charge less for each PC, and get more into the hands of students as long as their cash held out which would not be for long.
So, in many ways, you run a charity (which is what we are talking about here) just the same as business. Oh, you have people who just give you money, but most of that money is what is called "encumbered"; it's no different from being a consultancy that gets an up-front payment for some service they are going to provide. You don't book it as income until the work is done.
This means you consider exactly the same factors a business does when you make a strategic decision. The difference is this: in a push-comes-to-shove scenario, you choose maximizing mission over maximizing profit. For you, the profit is there to support the mission; for a business it is the other way around.