I think the larger strategy here may be to use Linux in network devices like set-top boxes, etc.
Could be you're right. When I first heard that Microsoft was planning to use the Xbox not only as a game set but also as a Tivo-style playback unit, I began to see Microsoft's vision. A "total entertainment device" has huge sales potential. Consider: how expensive are a Tivo, a Digital Cable box, a DVD player, and a game set, when all purchased separately? Obviously, the answer is "a lot". A single digital fun device has a lot sales potential indeed.
It's vaguely alarming, really. If Microsoft put together something like that ahead of everyone else, they're all but assured market dominance.
You may very well be correct, at least for many different kinds of services. I suppose it's all a matter of degree. Say, for example, you have some giant check-printers that can put out 250,000 checks a month at very small per-check cost. If you're Verizon corporation or some other giant, it might make sense to own one of these. If you're not, it might make more sense to share. If pooling some set of critical resources (and then allowing the pooler of those resources to make a profit, let's not forget) doesn't make sense, then outsourcing doesn't make sense.
Just suppose that this transaction went through -- given the millions if not billions that AOLTW could piss away on legal fees, would this pose a serious challenge to the GPL?
No. If they made the GPL go away, they'd be convicting themselves of willful copyright violation. Nothing but the GPL gives them the right to use the source. Why would they want to do away with the one thing that gives them that right? That would be stupid.
How can it be cheaper for you to to hire the staff than it is for me to hire the staff?
Economies of scale. Efficiency of purpose. Multitasking off of like tasks.
What do I mean by all that, you ask?
Economies of scale: A company which does a lot of a particular thing realizes productivity gains in that thing do to fact that it does this thing a lot.
Efficiency of purpose: a company dedicated to doing a single thing can often generate specialists at that thing who share knowledge with eachother and therefore become effective at that thing.
Multitasking of like tasks: One person can do the same kind of thing several different times much better than they can do different kinds of things different times.
Whether or not this applies to Exchange and so forth I don't know, but it's certainly a proven model when it comes to, say, payroll.
It is not 'adoption by the masses' that I was saying, though, but 'delivered to the masses' that was significant.
I can see your distinction; whether or not it's relevant or not depends on perspective. You can't deliver to the masses if they react with "the good ole' fashion way is bettah, and neveryoumind anyway, cause that lectricity stuff is from the devil." Deliver implies receive, I'm quite sure you see.
The point of just one chapter of the book is that ideas and information flow relatively freely around societies, but only some of them decided to adapt, to change, to willfully embrace. Those cultures that do this dominate the world. Those that don't are forever destined to be trampled under the other's bootheels.
That's one of the reasons I think Japan is going to do amazingly well in the 21st century. Their culture just loves toys, and moreover, from afar they admire many things American.
Some other writer (who's name I can't recall) said that, just in the way that the latter half of the 20th century was dominated by American power, the first half of the 21st century would be dominated by American culture. What he meant was that cultures around the world are adopting for their own those things American that they find goodness in. They will, of course, make it uniquely their own.
Fortunately, economics is most assuredly not a zero some game. Instead, it's a funny game where the success of the various different players all increase the success of all involved.
If you were to take a contemporary of the 19th century and have them examine the living standards of a contemporary of the early 21st century, they would see a world of such abundance they would scarcely be able to believe it. Imagine, if you will, a world a century from now where manufacturing productivity expanded with the same magnitude as the expansion of productivity from the last century to this one. This roughtly describes a world with 200 times the manufacturing productivity that we have today. Things, once expensive, have negligible cost.
Economics is about scarce resources. So one has to ask: what in the next century will be scarce? Contact with real live human beings will be its own commodity, I suspect. Intellectual property will be a commodity, I also suspect. And note that no matter how much automation we develop, the need to have people there to make it all work properly seems to constantly increase, rather than decrease.
These are the forces you can expect to see at work over the next few decades and throughout the century. I won't speculate on how A.I. might transform all that. That's a long way off, still.
Your use of the word mass is exactly what I'm talking about: an invention can come and then go and be forgotten -- perhaps even sometimes never even recorded -- but when a culture actually adopts the technology -- in mass, as you put it -- this is actually the thing that makes the culture tread the path to cultural ascendancy.
For a very interesting treatise on this subject and other similar things which lead a culture to dominance, read the Pulitzer Prize winning Guns, Germs, and Steel. It's one of the most insightful books I've read in the last few years.
The defnse sector is a good business to be in right now. To wit: the sector has an endemic problem, where large amounts of management is dominated by older employees who will, of course, eventually have to retire. This naturally leads one to wonder where the next generation of DoD managers will come from.
As this tail expires over the next half decade, I see a demand rise, yep. Opportunity abounds.
The second was written a couple of years ago, and asserts that innovation and openness to change will keep the American programmer on top for years.
One of the ongoing memes in American culture is our eagerness for new technology and ideas. Many Americans falsely believe that many of the worlds most important inventions were actually invented here. They weren't. The vast majority of them were invented somewhere else first and then blithely ignored. What happened here is that the invention was adopted.
The continued presence or absence of our technological eagerness and flexible predicates our future success.
This is one of the reasons I think the Japanese will fly very high indeed across the 21st century. They have an appetite for technology that exceeds even our own.
There are many cultures world wide that have this appetite now. I firmly believe that this will quite reliably predict the success of these countries through the 21st century, mitigated of course by outside influences.
The converse is also true. Look at the cultures that repudiate technology; they're practically guaranteed to remain impoverished has-been countries which any of the dominate players could roll over on a whim.
N.B.: I'm not making any claim that America is the superior culture in this regard any more. I will say, however, that we are on the list.
At the beginning of the 20th century, the vast majority of workers in the U.S. were dedicated to agrarian jobs. Obviously, within a very short time period there was massive social change as the the majority of work shifted from agricultural pursuits to industrial pursuits where it peaked at over 60% in the mid 60%. During the early part of this period, there was much public grief as everyone complained how horrible it was that people were working in factories and the sort. There was much hysterionics as various alarmists talked about the disaster in the making.
By the year 2000, less than 2% of the U.S. population was dedicated to agricultural work. Agricultural producitivity expanded something like 200 fold during this period. With the wonderful, colorfully, jaundiced lens of hindsight, of course, we know this was no disaster.
Something similar is happening now. The 1960s saw the beginning of the decline of industry in the U.S., and it's been steadily decreasing ever since.
Service jobs are beginning to rule the day, and -- just like the early 1900's -- hysterionic alarmists are espousing their doomsday predictions (n.b.: I'm not accusing the author of the book of this, just a general observation).
A close examination of the tranformation, however, yields the information that the very fastest growing sections of the service sector are the professional services. We are quickly becoming a society where specialized knowledge rules the day. Lawyers, physicians, engineers, hell even the mechanics and secretaries are workers who need to understand computers and computing.
I'm not sure where I'm going with all this, except to point out that by 2100 and most likely a lot sooner very few people will be in jobs directly attached to manufacturing. We'll be one giant service economy.
At my first job I was hired to program, but I knew so much more about computers than those in the environment, I ended up doing far more systems selection than programming. You know: buy this, download that, integrate the other damn thing.
But of course you're generally correct, and what I just described is the role of most good programmers everywhere. We don't so much as "write programs" as we do "solve problems," as it were. My job these days always seems to be writing a few clever algorithms which transform a complex task into a coherent intermediate representation and writing glue that knows how to make this representation work with any variety of other things, commercial, freeware, internally adapted, and otherwise.
Only in my hobby time do I actually get to write innovative new programs.
You may not be fired for consulting a coworker, but if you take a coworker's worker and then claim you did it yourself, you'd certainly better cover your bases.
Case in point: You get seriously ill and you need state of the art medication. It cost some drug company 40 cents to manufacture it per dose. The drug company charges 60 dollars per dose to the consumer. The research was paid for with public funds.
While obviously as the initiator of this subthread I agree with you in spirit, I should like to point out you really need to rethink how you think about the cost of drug research. Suppose I were to accept your figure of 40 cents manufacturing cost and $60 dollars per item sale value. Very well. But you're crafting an example which doesn't include the amortized cost of producing the product. For one, manufacturing cost is hardly the only cost of producing a drug. It costs nearly a billion dollars U.S. to make it through the FDA NDA process. Further, for every drug that makes it through the process, there exist plethora of them that do not. Each of these other failing drugs cause losses to the company's bottom line which end up being amortized across the successful products. So all is not as it seems.
The occasional lucky company may do very well. They may have more winners than losers, and end up being the next Pfizer or Genentech. This might, over the short term, demonstrate stellar profits and stock market performance. But that's only over the short term. The dynamics and uncertainties of the business practically guarantee that this will all eventually fall out in the wash. It's just too risky of a business with too many unknowns to make any guarantees. And even more to the point, for every lucky company there most certainly exists unlucky ones.
What I'm trying to say here is that when you look at a single company and then focus in on a single one of their products (how about Viagra?), it's quite easy to become rather myopic about it. This near sightedness will make you miss the forest for the trees if you're not careful.
All this, of course, strays from the central point. Research paid for by the people ought to belong to the people. It's our money and what it buys ought to be our knowledge.
The kicker is, just about everyone at my place of work agrees with me. But everything still stalls. I've espoused this very opinion quite loudly in public, and I get lots of head nodding. The problem seems to be that no one, not even senior management, really seems to think that they can do what is ethical if it conflicts with corporate interests.
What's more interesting about this is that our own history has proven this to be objectively untrue. Other projects in the same or other departments which have shared information with the public and open up and don't fight with the other government contractors on large efforts always end up as top players in their niche. There seems to be this meme that goes "if they're so confident in their work they're willing to give it away and show everyone else, damn they must be good."
Um, that's not what I meant, though. It's quite common for the contractors to retain exclusive commerical use, got it? It's a normal practice, and, IMO, an abusive one. It betrays the public trust.
It's a single-source point of failure and accessible by default to the owner of the computer and routinely touched by all manner of software. I feel confident in predicting that every windows user is likely to eventually encounter registry corruption of the sort that will eventually cause their computer to misbehave quite badly to the point of requiring a professional to fix it unless they are an expert computer user. Apple's desktop file, but more importantly the way it could autoregenerate itself with reliably good outcomes would serve as a much better model. While the desktop file had its own downsides, that aspect of the environment functioned well and did so consistently. It's one of the few things about using Macs that I still miss after all these years of having grown in different directions.
Now, if the funding is exclusively for developing a cool program...
In fact, that's just exactly what I was talking about: government-funded software research where the software is per se part of the contract deliverable. It's already required to be released to any other government representative or designee, it simply isn't. The rules are flouted with passive resistance, foot dragging, and so forth. Putting software into an open repository would simply make the players obey already existing laws and nip the foot-dragging in the bud.
There's one current exception in current contract dejure, however, and that's this: government contractors researching software for the government are allowed to retain their software for commercial use. I disagree with this and believe it should be done away with. There is already good precedent for this. A copyrightable material produced by the government becomes part of the public domain. Government-sponsored contract software development should follow this model.
Note that these are two distinctly different issues, which work best admittedly when paired together (a government-access only Source Forge look-alike would somewhat defeat the purpose).
I can see very little benefit to the public for the current blackhole model, where literally a hundred (and potentially hundreds) of millions of dollars in software development infrastructure is blackholed annually.
It's real travesty and, in my opinion, betrays the public trust.
While I'm with ya, consider: if a University becomes profitable, it doesn't require as much funding from the public rolls. Certainly you can see how politicians might like that.
I've been working in the government funded software research and development community for a decade and I must have seen as much as $50 million or more of the Public's dollars go into the giant black hole of software research.
One of the major problems of goverment funded research, even when it is contractually bound to be open for government-related inspection and use (which most of it is) is that the various players all jealously guard their turf. This includes other contractors who, even when legitimately approached for copies of the source which they are contractually bound to give you, curiously develop problems getting messages, getting back to you, shipping you source, and providing you access. You'd think it would end there, but no.
The government players themselves jealously guard their turf. Since there is similar and even duplicative work funded across DoD and government, government reps have no desire at all to share. They view the other similar projects as competitive and worry that if one of them gets the upper hand, their own project will be unfunded as redundant or irrelevant. This creates a situation where the government players -- those who are supposed to be working for the Public Trust -- instead drag their feet and use passive resistance in giving up software to even those who are allowed to see it, such as other members of goverment or government contractors working on the government's behalf.
The end result of all of this is that enormous sums of software gets locked up in boxes and never sees the light of day. About the only person who actively looks at the source is the original contractor. For research efforts, its understandable and reasonable that a research project doesn't result in a piece of software that's used by either no one or the very few. However, what's not not reasonable is that the information itself is effectively vaporized.
This is a completely execrable situation and grossly violates the Public Trust. Not only is the system vastly wasteful of the public dollar, it likewise violates the entire basis of public research: the open sharing of information.
For some time now, a sort of jewel in my mind's eye has been glimmering, and it goes like this:
All goverment software development, with the exception of sensitive projects, should be forced into placement under open license into a high profile source repostory such as Source Forge. This, for every government project, would be the primary CVS repository of the project. Project developers would insert code here and be subject to detailed public scrutiny with default anonymous CVS read access.
In my opinion this would blow open the doors of enormous amounts of software development and be of enormous benefit to the general public. Consider how neatly nipped in the bud all the beaureacratic foot-dragging would be. Intermediaries? None. You want the source code? CLICK.
This should be the new standard of non-classified government software development. The money belongs to the People, dammit! So should the research.
I don't know why people keep focusing on market share. Market share doesn't matter. It's market size that matters, as it's the size (and not share) which dictates economies of scale and ultimately how much cash your receive as a result of your business activities. If the entire world wide community of computer purchasers quadrupled tommorrow, but Apple's market size did not increase, their share would drop to 1/4 of its current value. Plenty of companies, however, are perfectly viable with very small market shares, particularly when they satisfy niche markets -- as does Apple.
True, part of the reason that win95 died so much wasn't MSes fault. It was bad software downloaded from the highspeed access, corrupting the registry and whatnot.
You're not exactly winning any friends by mentioning the horrifying registry. It's an abortion, and it is all Microsoft's fault. They invented the thing.
However, remaining silent doesn't solve the problem. Somebody has to speak up.
Public name-calling (e.g., "Asshole" was the exact word) is a sign of immaturity. But then again, not that suprising. As one poster aptly put it, "kernel developers have egos the size of small countries."
Slashdot Mirror
I think the larger strategy here may be to use Linux in network devices like set-top boxes, etc.
Could be you're right. When I first heard that Microsoft was planning to use the Xbox not only as a game set but also as a Tivo-style playback unit, I began to see Microsoft's vision. A "total entertainment device" has huge sales potential. Consider: how expensive are a Tivo, a Digital Cable box, a DVD player, and a game set, when all purchased separately? Obviously, the answer is "a lot". A single digital fun device has a lot sales potential indeed.
It's vaguely alarming, really. If Microsoft put together something like that ahead of everyone else, they're all but assured market dominance.
C//
You may very well be correct, at least for many different kinds of services. I suppose it's all a matter of degree. Say, for example, you have some giant check-printers that can put out 250,000 checks a month at very small per-check cost. If you're Verizon corporation or some other giant, it might make sense to own one of these. If you're not, it might make more sense to share. If pooling some set of critical resources (and then allowing the pooler of those resources to make a profit, let's not forget) doesn't make sense, then outsourcing doesn't make sense.
C//
Just suppose that this transaction went through -- given the millions if not billions that AOLTW could piss away on legal fees, would this pose a serious challenge to the GPL?
No. If they made the GPL go away, they'd be convicting themselves of willful copyright violation. Nothing but the GPL gives them the right to use the source. Why would they want to do away with the one thing that gives them that right? That would be stupid.
C//
How can it be cheaper for you to to hire the staff than it is for me to hire the staff?
Economies of scale. Efficiency of purpose. Multitasking off of like tasks.
What do I mean by all that, you ask?
Economies of scale: A company which does a lot of a particular thing realizes productivity gains in that thing do to fact that it does this thing a lot.
Efficiency of purpose: a company dedicated to doing a single thing can often generate specialists at that thing who share knowledge with eachother and therefore become effective at that thing.
Multitasking of like tasks: One person can do the same kind of thing several different times much better than they can do different kinds of things different times.
Whether or not this applies to Exchange and so forth I don't know, but it's certainly a proven model when it comes to, say, payroll.
C//
It is not 'adoption by the masses' that I was saying, though, but 'delivered to the masses' that was significant.
:)
I can see your distinction; whether or not it's relevant or not depends on perspective. You can't deliver to the masses if they react with "the good ole' fashion way is bettah, and neveryoumind anyway, cause that lectricity stuff is from the devil." Deliver implies receive, I'm quite sure you see.
The point of just one chapter of the book is that ideas and information flow relatively freely around societies, but only some of them decided to adapt, to change, to willfully embrace. Those cultures that do this dominate the world. Those that don't are forever destined to be trampled under the other's bootheels.
That's one of the reasons I think Japan is going to do amazingly well in the 21st century. Their culture just loves toys, and moreover, from afar they admire many things American.
Some other writer (who's name I can't recall) said that, just in the way that the latter half of the 20th century was dominated by American power, the first half of the 21st century would be dominated by American culture. What he meant was that cultures around the world are adopting for their own those things American that they find goodness in. They will, of course, make it uniquely their own.
Fortunately, economics is most assuredly not a zero some game. Instead, it's a funny game where the success of the various different players all increase the success of all involved.
These are decidedly nice rules.
C//
C//
If you were to take a contemporary of the 19th century and have them examine the living standards of a contemporary of the early 21st century, they would see a world of such abundance they would scarcely be able to believe it. Imagine, if you will, a world a century from now where manufacturing productivity expanded with the same magnitude as the expansion of productivity from the last century to this one. This roughtly describes a world with 200 times the manufacturing productivity that we have today. Things, once expensive, have negligible cost.
Economics is about scarce resources. So one has to ask: what in the next century will be scarce? Contact with real live human beings will be its own commodity, I suspect. Intellectual property will be a commodity, I also suspect. And note that no matter how much automation we develop, the need to have people there to make it all work properly seems to constantly increase, rather than decrease.
These are the forces you can expect to see at work over the next few decades and throughout the century. I won't speculate on how A.I. might transform all that. That's a long way off, still.
C//
1. Mass electrification, ...
Your use of the word mass is exactly what I'm talking about: an invention can come and then go and be forgotten -- perhaps even sometimes never even recorded -- but when a culture actually adopts the technology -- in mass, as you put it -- this is actually the thing that makes the culture tread the path to cultural ascendancy.
For a very interesting treatise on this subject and other similar things which lead a culture to dominance, read the Pulitzer Prize winning Guns, Germs, and Steel. It's one of the most insightful books I've read in the last few years.
C//
Then there's the condescending view all government employees have of contractors.
:)
That's "blood sucking government contractor" to you, bud.
C//
The defnse sector is a good business to be in right now. To wit: the sector has an endemic problem, where large amounts of management is dominated by older employees who will, of course, eventually have to retire. This naturally leads one to wonder where the next generation of DoD managers will come from.
As this tail expires over the next half decade, I see a demand rise, yep. Opportunity abounds.
C//
The second was written a couple of years ago, and asserts that innovation and openness to change will keep the American programmer on top for years.
One of the ongoing memes in American culture is our eagerness for new technology and ideas. Many Americans falsely believe that many of the worlds most important inventions were actually invented here. They weren't. The vast majority of them were invented somewhere else first and then blithely ignored. What happened here is that the invention was adopted.
The continued presence or absence of our technological eagerness and flexible predicates our future success.
This is one of the reasons I think the Japanese will fly very high indeed across the 21st century. They have an appetite for technology that exceeds even our own.
There are many cultures world wide that have this appetite now. I firmly believe that this will quite reliably predict the success of these countries through the 21st century, mitigated of course by outside influences.
The converse is also true. Look at the cultures that repudiate technology; they're practically guaranteed to remain impoverished has-been countries which any of the dominate players could roll over on a whim.
N.B.: I'm not making any claim that America is the superior culture in this regard any more. I will say, however, that we are on the list.
C//
Get ready to see your programming job get exported to India and China.
...by 2015 computers will be fast enough that point-and-drool paint-by-numbers tools...
Inherent in this are several flawed assumptions.
Flawed assumption number one is that as these countries advance to 2015, they themselves won't be demanding IT labor.
Flawed assumption number two is that by 2015 local companies won't continue to prefer to work with people locally and personally.
This is likewise flawed. All evidence is to the contrary; the faster computers get, the more programmers seem to be employed.
C//
At the beginning of the 20th century, the vast majority of workers in the U.S. were dedicated to agrarian jobs. Obviously, within a very short time period there was massive social change as the the majority of work shifted from agricultural pursuits to industrial pursuits where it peaked at over 60% in the mid 60%. During the early part of this period, there was much public grief as everyone complained how horrible it was that people were working in factories and the sort. There was much hysterionics as various alarmists talked about the disaster in the making.
By the year 2000, less than 2% of the U.S. population was dedicated to agricultural work. Agricultural producitivity expanded something like 200 fold during this period. With the wonderful, colorfully, jaundiced lens of hindsight, of course, we know this was no disaster.
Something similar is happening now. The 1960s saw the beginning of the decline of industry in the U.S., and it's been steadily decreasing ever since.
Service jobs are beginning to rule the day, and -- just like the early 1900's -- hysterionic alarmists are espousing their doomsday predictions (n.b.: I'm not accusing the author of the book of this, just a general observation).
A close examination of the tranformation, however, yields the information that the very fastest growing sections of the service sector are the professional services. We are quickly becoming a society where specialized knowledge rules the day. Lawyers, physicians, engineers, hell even the mechanics and secretaries are workers who need to understand computers and computing.
I'm not sure where I'm going with all this, except to point out that by 2100 and most likely a lot sooner very few people will be in jobs directly attached to manufacturing. We'll be one giant service economy.
C//
At my first job I was hired to program, but I knew so much more about computers than those in the environment, I ended up doing far more systems selection than programming. You know: buy this, download that, integrate the other damn thing.
But of course you're generally correct, and what I just described is the role of most good programmers everywhere. We don't so much as "write programs" as we do "solve problems," as it were. My job these days always seems to be writing a few clever algorithms which transform a complex task into a coherent intermediate representation and writing glue that knows how to make this representation work with any variety of other things, commercial, freeware, internally adapted, and otherwise.
Only in my hobby time do I actually get to write innovative new programs.
C//
C//
No. Iocane powder.
It's odorless, tasteless, and completely undetectable.
C//
You may not be fired for consulting a coworker, but if you take a coworker's worker and then claim you did it yourself, you'd certainly better cover your bases.
C//
Case in point: You get seriously ill and you need state of the art medication. It cost some drug company 40 cents to manufacture it per dose. The drug company charges 60 dollars per dose to the consumer. The research was paid for with public funds.
While obviously as the initiator of this subthread I agree with you in spirit, I should like to point out you really need to rethink how you think about the cost of drug research. Suppose I were to accept your figure of 40 cents manufacturing cost and $60 dollars per item sale value. Very well. But you're crafting an example which doesn't include the amortized cost of producing the product. For one, manufacturing cost is hardly the only cost of producing a drug. It costs nearly a billion dollars U.S. to make it through the FDA NDA process. Further, for every drug that makes it through the process, there exist plethora of them that do not. Each of these other failing drugs cause losses to the company's bottom line which end up being amortized across the successful products. So all is not as it seems.
The occasional lucky company may do very well. They may have more winners than losers, and end up being the next Pfizer or Genentech. This might, over the short term, demonstrate stellar profits and stock market performance. But that's only over the short term. The dynamics and uncertainties of the business practically guarantee that this will all eventually fall out in the wash. It's just too risky of a business with too many unknowns to make any guarantees. And even more to the point, for every lucky company there most certainly exists unlucky ones.
What I'm trying to say here is that when you look at a single company and then focus in on a single one of their products (how about Viagra?), it's quite easy to become rather myopic about it. This near sightedness will make you miss the forest for the trees if you're not careful.
All this, of course, strays from the central point. Research paid for by the people ought to belong to the people. It's our money and what it buys ought to be our knowledge.
The kicker is, just about everyone at my place of work agrees with me. But everything still stalls. I've espoused this very opinion quite loudly in public, and I get lots of head nodding. The problem seems to be that no one, not even senior management, really seems to think that they can do what is ethical if it conflicts with corporate interests.
What's more interesting about this is that our own history has proven this to be objectively untrue. Other projects in the same or other departments which have shared information with the public and open up and don't fight with the other government contractors on large efforts always end up as top players in their niche. There seems to be this meme that goes "if they're so confident in their work they're willing to give it away and show everyone else, damn they must be good."
In spite of this, the contrary instincts rule.
C//
Um, that's not what I meant, though. It's quite common for the contractors to retain exclusive commerical use, got it? It's a normal practice, and, IMO, an abusive one. It betrays the public trust.
C//
And what's wrong with the registry?
It's a single-source point of failure and accessible by default to the owner of the computer and routinely touched by all manner of software. I feel confident in predicting that every windows user is likely to eventually encounter registry corruption of the sort that will eventually cause their computer to misbehave quite badly to the point of requiring a professional to fix it unless they are an expert computer user. Apple's desktop file, but more importantly the way it could autoregenerate itself with reliably good outcomes would serve as a much better model. While the desktop file had its own downsides, that aspect of the environment functioned well and did so consistently. It's one of the few things about using Macs that I still miss after all these years of having grown in different directions.
C//
Now, if the funding is exclusively for developing a cool program...
In fact, that's just exactly what I was talking about: government-funded software research where the software is per se part of the contract deliverable. It's already required to be released to any other government representative or designee, it simply isn't. The rules are flouted with passive resistance, foot dragging, and so forth. Putting software into an open repository would simply make the players obey already existing laws and nip the foot-dragging in the bud.
There's one current exception in current contract dejure, however, and that's this: government contractors researching software for the government are allowed to retain their software for commercial use. I disagree with this and believe it should be done away with. There is already good precedent for this. A copyrightable material produced by the government becomes part of the public domain. Government-sponsored contract software development should follow this model.
Note that these are two distinctly different issues, which work best admittedly when paired together (a government-access only Source Forge look-alike would somewhat defeat the purpose).
I can see very little benefit to the public for the current blackhole model, where literally a hundred (and potentially hundreds) of millions of dollars in software development infrastructure is blackholed annually.
It's real travesty and, in my opinion, betrays the public trust.
C//
...and it is stopping many people to do real work in platforms that require so.
How? They can't set their CVSROOT? LOL.
Oh, you mean they want to keep it to themselves. Yeah, I gotcha. Well then, they shouldg get money from someone other than the public.
C//
While I'm with ya, consider: if a University becomes profitable, it doesn't require as much funding from the public rolls. Certainly you can see how politicians might like that.
C//
I've been working in the government funded software research and development community for a decade and I must have seen as much as $50 million or more of the Public's dollars go into the giant black hole of software research.
One of the major problems of goverment funded research, even when it is contractually bound to be open for government-related inspection and use (which most of it is) is that the various players all jealously guard their turf. This includes other contractors who, even when legitimately approached for copies of the source which they are contractually bound to give you, curiously develop problems getting messages, getting back to you, shipping you source, and providing you access. You'd think it would end there, but no.
The government players themselves jealously guard their turf. Since there is similar and even duplicative work funded across DoD and government, government reps have no desire at all to share. They view the other similar projects as competitive and worry that if one of them gets the upper hand, their own project will be unfunded as redundant or irrelevant. This creates a situation where the government players -- those who are supposed to be working for the Public Trust -- instead drag their feet and use passive resistance in giving up software to even those who are allowed to see it, such as other members of goverment or government contractors working on the government's behalf.
The end result of all of this is that enormous sums of software gets locked up in boxes and never sees the light of day. About the only person who actively looks at the source is the original contractor. For research efforts, its understandable and reasonable that a research project doesn't result in a piece of software that's used by either no one or the very few. However, what's not not reasonable is that the information itself is effectively vaporized.
This is a completely execrable situation and grossly violates the Public Trust. Not only is the system vastly wasteful of the public dollar, it likewise violates the entire basis of public research: the open sharing of information.
For some time now, a sort of jewel in my mind's eye has been glimmering, and it goes like this:
All goverment software development, with the exception of sensitive projects, should be forced into placement under open license into a high profile source repostory such as Source Forge. This, for every government project, would be the primary CVS repository of the project. Project developers would insert code here and be subject to detailed public scrutiny with default anonymous CVS read access.
In my opinion this would blow open the doors of enormous amounts of software development and be of enormous benefit to the general public. Consider how neatly nipped in the bud all the beaureacratic foot-dragging would be. Intermediaries? None. You want the source code? CLICK.
This should be the new standard of non-classified government software development. The money belongs to the People, dammit! So should the research.
C//
I don't know why people keep focusing on market share. Market share doesn't matter. It's market size that matters, as it's the size (and not share) which dictates economies of scale and ultimately how much cash your receive as a result of your business activities. If the entire world wide community of computer purchasers quadrupled tommorrow, but Apple's market size did not increase, their share would drop to 1/4 of its current value. Plenty of companies, however, are perfectly viable with very small market shares, particularly when they satisfy niche markets -- as does Apple.
C//
True, part of the reason that win95 died so much wasn't MSes fault. It was bad software downloaded from the highspeed access, corrupting the registry and whatnot.
You're not exactly winning any friends by mentioning the horrifying registry. It's an abortion, and it is all Microsoft's fault. They invented the thing.
C//
However, remaining silent doesn't solve the problem. Somebody has to speak up.
Public name-calling (e.g., "Asshole" was the exact word) is a sign of immaturity. But then again, not that suprising. As one poster aptly put it, "kernel developers have egos the size of small countries."
C//