I realize I've described my employer's system at a very high level. I don't know to what extent our system has been described outside the company. I do know it has been described outside the company based on comments made by one of the lead patent attorneys that runs the system, since he was relating others' reactions outside the company to what we do. I also don't know the terms under which our scheme was disclosed.
Some additional comments:
The structure of your scheme bears a striking resemblance to ours, which inclines me to believe that this structure is popular and (one hopes) useful. You might want to consider the impact of multiple inventors on a given patent: Do you give each inventor the same award regardless of whether there was 1 inventor or 10? Do you split a fixed award among the inventors (e.g. that $500 gets split 10 ways with 10 inventors)? Do you split the difference at some threshold (e.g. $500/inventor up to $XXX dollars, and $XXX/N above that threshold)?
You might also want to consider the long term impact of the patent. In some sense, giving an equity stake in a company that could grow quickly does tie your incentive to the success of the company and can be the right thing to do. That might make sense in a young startup, but not in an established company whose value on the market can be very starkly disconnected from the activity of any given contributor. Neither one, though, ties incentive to the value of the actual invention. So, perhaps consider having a re-review either at the time the patent issues, or sometime after the patent issues to see how much revenue you can tie to the patent in order to gauge a reward.
I could (and indeed would like to) go into greater depth with the relative merits of the system we use at work, but I don't want to divulge proprietary information. I hope my broader insights are useful.
If you release your invention to the public domain...
I want to be clear on this, since there's a notable difference between copyright and patent here: If you publish the details of your invention without having first filed for a patent on it, then the invention is effectively in the public domain. It's a bit like losing trade secret privilege. Once an idea is known, it can't become unknown.
Now, aspects of how you implement your invention might be copyrightable, such as source code. You could have a completely proprietary application under full copyright protection. If you publish its techniques and protocols without filing patent disclosures first, then those techniques and protocols are no longer patentable. But, the actual code that implements the techniques and protocols is still protected under copyright. So, the invention is effectively in the public domain, but your specific implementation is not.
An example of this might be the SMB protocol that Microsoft uses for sharing files over a network. Portions of the protocol may be protected by patents. Had they instead chosen to document the protocol without filing for protection, then these aspects would become unpatentable, and everyone could implement the protocol without getting granted special exemptions. In either case, though, Microsoft retains the copyright to its Windows implementation.
I wanted to make this clarification, since copyrights and patents so often get confused.
(And then there's chips with their "mask work" protection...)
Don't think that making your product open source will protect you from litigation - quite the contrary.
If you release your invention to the public domain, if someone subsequently tries to patent it, the patent won't be granted (or will be invalidated on a re-review if it slips through) on the basis of prior art. This is, of course, if you do actually invent something before someone else does.
Releasing your invention to the public domain when you weren't actually the first to invent it does not absolve you of liability to another inventor's patent, if they invented it before you. In that case, if their patent is invalid, then you need to find someone else's prior art, or find some other grounds to illustrate that the patent is invalid.
On the other hand having a patent portfolio can be used as a weapon against others threat to sue. "You using our patent so unless you license this patent of yours we'll sue."
Exactly. One strategy is make a competitor back off is to threaten to countersue with your own portfolio. Or, if a patent troll threatens you, you can point to your own patent and say "We're not violating your patent. This is an implementation of our own patent." Then the shoe is on the other foot, and the patent troll must invalidate your patent. Hopefully, the added cost of litigation will make the troll back off.
Pure patent trolls have the annoying property of not actually having any product that might infringe on something in a company's stack of patents. That's what makes them so potent, dangerous and effective. They've got nothing to lose and everything to gain. And since their lawsuits tend to end in settlements, rather than judgments, the validity of the patent doesn't get a chance to be tested in court.
My big problem with the stack-o-patents is that it makes it really hard for the little guy to get into the business. This is true even with pure hardware inventions. Competing Megacorps A and B know to not sue each other without very good reason because they both have huge piles of IP, and who knows where one or the other may have accidentally infringed on the others' patent. Startup C, on the other hand, doesn't have this parity. Megacorp A or Megacorp B could decide to sue Startup C whenever they like, because it has far fewer resources for defending itself, and not much of a portfolio if anything. This is but one way patents stifle innovation.
Personally, I wouldn't mind if patents were modified to be less strictly exclusive. That is, disclose your invention and have exclusive rights for a very short period (say, 3-5 years from the date of filing), and for the remainder of the term open it up for compulsory licensing. That'd poke a hole in the patent troll's bubble (since they seem to collect patents from failing concerns near the end of the patent's lifetime), and it'd lessen the impact of the stack-o-patents mutually assured destruction effect. It would also protect the initial product development time frame and initial time in the market, to provide a little kick to the first-mover advantage, without scuttling long term support of products by patent-blocking interoperability.
I think if you remove the "You Won The 20 Year Exclusive Right To Charge What You Wish Lottery" aspect from patents, then a lot of the frivolous bets people place will disappear, since you just switched out PowerBall for Pick-4. People will still play, but the payout will be much less, and you won't get the rush of people angling for that $300M jackpot that arises from time to time.
At our company, all patent disclosures are reviewed by other technical experts to determine their suitability to file. After all, it isn't free to file patents. Thus, there's a scale: Small bonus for decision to file to encourage disclosures, larger bonus for the patent issuing, and a variable bonus some years later on review as I talk about here.
The technical guy isn't the one that would be flooding the patent office with bogus disclosures in an attempt to write himself an endless stream of bonus checks. There's always at least one layer that could catch the bogons: the patent attorney that drafts the patent and works all the procedural issues. I suggest an additional layer of technical review to sort good from bad and to prioritize among the good so the most important ones get filed first.
Then, the employee feels they are being treated a bit more fairly as apposed to some little bonus. And why not? If a company makes millions of dollars off of an employees creativity why shouldn't they partake.
Where I work, we do regular reviews of our issued patents to determine which were valuable and which were not. The valuable ones win the inventors an incentive award. And, since it's a company full of engineers, yes, we do have fancy schmancy formulas that drive the process, and determine the amount of the award. (There are also fixed awards for "decision to file" and "patent issued.")
Assigning specific patent royalty amounts rarely makes sense. Sometimes you can point to specific licensing actions that generate revenue associated with a specific patent, but most often you can't. Many patents are cross licensed to other companies through bulk cross licensing agreements. Others just form a "mutually assured destruction" sort of bulwark against other companies. (This latter property doesn't work so well against patent trolls, since trolls don't make anything--they just have their patent(s) that they chase people with.) Thus, we tend to look at how much revenue was associated with products that use the patent (whether the revenue was ours or someone else's) and go from there.
As for my background: I'm a member of one of our internal committees that reviews disclosures to determine what to file. At the very least, that acts as a quality filter internally. I've also filed several patent disclosures of my own, and have even had a few patents issue. Some have even gotten incentive awards in those periodic reviews.:-) I'm not a big fan of the patent system, but as long as it's there, I figure I may as well use it and let it die under its own weight.
Indeed. For some reason I thought the Apple/// came out later than that. I remember those things. I used to write papers in/// Easy Pieces, and even played a bit with Pascal on it. I never got to play with Business BASIC, though I did see the demo. (Hastings Manufacturing, the piston ring manufacturer, had bought a bunch of these back in the day, and donated them to the public library and YMCA when they got rid of them. While I was in high school, I consulted for both, setting up the systems and the data disks for them.)
As far as I'm concerned, as long as there's a key in that location that I can use to remove the character to the left of the cursor, I'm happy.
Next up: The never ending confusion between the ASCII backspace (BS, character #8), the ASCII delete (DEL, #127), and the DEC VT-100 "Remove" escape sequence.:-)
For the word processor users of the world, it should just have a "paragraph" symbol on it. After all, in a word processor, you're only supposed to press that key at the end of each paragraph, when you're done entering the paragraph.
And carriage returns? What carriage would this be? If you're editing your documents the old fashioned way--and I still use vim plenty--the key should be labeled "new line" or similar. But wouldn't most Mac users be entering text in a word-processor-like environment that provides line breaks in paragraphs automatically?
Personally, I'm not arguing what the label should be. All the labels suck for one reason or another. (Why doesn't the "return" key return me to the previously visited webpage?;-) ) At least "delete" actually deletes somethings on modern Apples, as opposed to inserting that checkerboard character like it did on the Apple//es. (At least, in Applesoft BASIC. AppleWorks did the right thing of course.)
This is why I recommend people use paid ISPs for their real E-mail accounts, and perhaps use Yahoo, Google, or Rocketmail for registering on spammy websites where they want an E-mail address so they can make their advertisers happy.
When I signed up for DSL service, it was with SBC Yahoo! DSL, you insensitive clod!
Well, once Apple finally got a numeric keypad. On the PC, though, both have been named "enter" from the beginning.
My point, of course, was that complaining about "backspace" being a typewriter function by citing "delete" as a better choice overlooks that "enter" is a better choice than "return" by similar reasoning.
Someone just needs to hit RUN STOP-RESTORE on this whole thread.;-)
No, cmd.exe is the Windows NT shell. WinNT and its offspring are not built on MS-DOS, though they do provide versions of most of its commands. You're thinking of COMMAND.COM.
Murrrh? That's ridiculous. You're not touching my perl. It's sufficient to say:
print "The memory count not be \"read\".";
Or, more likely, it's in a function that prints errors, so the whole thing looks like this (ignoring the extra dots to get "indentation"):
sub memory_error_string($$$)
{
. . return "The instruction at '$_[0]' referenced memory at '$_[1].' The memory could not be '$_[2].' Click OK to terminate program.";
}
So what sorts of outcomes are being measured in outcome based education? To me, "outcome" means "measure of achievement toward a goal" such as "acquiring a certain set of knowledge." I'm all for measuring and grading someone on what they actually managed to learn and achieve by the end of the course, rather than on how many busyworks they handed in on time. The outcome of education should be educated individuals. If individuals demostrate that they're educated, then it doesn't matter so much that they didn't hand in all their homework.
But from how you describe it, "outcome" means "the student's been in our care for a sufficient period of time." (aka. automatic advancement.) Actual level of competence doesn't matter. How is "final level of competence" not part of "outcome"?
(I'm not trolling and this is not flamebait. This is an honest, sincere question. I don't have children and have been out of public schools for over a decade and a half. Gimme a break.)
When I was in high school in Michigan some years back, there was a state-level requirement on attendance. More than 15 days absent for any reason (IIRC) in a marking period would be an automatic failure for reason of lack-of-attendance. (I think the school board could override in special circumstances, such as an extended medical emergency, but otherwise even extended illness still counted toward the 15 days.) For a primary school, this makes a certain amount of sense.
I got the sense while I was there that the net effect was to keep the athletic departments from pulling the athletes out of class too often for away games and so on. That's reasoning can get behind.
Even so, it puts attendance above achievement. Public school is, effectively, a national day care for minors, so I can see attendance being important. In college, I made ample use of flexible attendance policies.
Instead, courses like Biology and Shop (classes I wasn't particularly interested in) weighted my average down.
Before the grammar Nazis get here... It'd read better if I said "the type of classes I wasn't particularly interested in." Also, my word choice would have been better if I said "weighed my average down" instead of "weighted my average down."
I actually had the teacher that taught the programming course (in high school) tell me that he refused to have me in the class, because he didn't see the point of "teaching" a student that knew more than the teacher. (This was for an Applesoft BASIC class back in ~1990.)
You kids with your IT fetish. I spent a summer working at an ISP while getting my engineering degree. I am soooo glad I didn't get sucked into IT.
Now if I just could have tested out of that programming course and had gotten credit for it, that'd be awesome. Courses like that that I would have aced never showed up on my GPA. Instead, courses like Biology and Shop (classes I wasn't particularly interested in) weighted my average down.
Goes to show that GPA is a deeply flawed measure. It only gives you an average score among the classes that you cared to and were allowed to attempt.
Showing my age here... The convention stretches back at least as far as Microsoft BASIC in the early 80s. It wouldn't surprise me that other formats adopted it. Nor would it surprise me to learn it predates BASIC.
Which reinforces my point: IBM doesn't contribute patches to Linux altruistically, and yet IBM contributes quite a large number of patches. They have good reasons for doing so, and it amounts to enlightened self interest. Previously, they focused primarily on improving AIX to ship servers (that may or may not run DB/2). Now they contribute heavily to Linux.
I'm sure they don't mind all the positive PR they get for their contributions, either.
Check your reading comprehension. I was accusing the Iraqis for incompetence, by burying their arms and losing them for 15 years, rather than rushedly burying them to hide them from weapons inspectors in the months just before we invaded so that they could be easily readied to attack us.
We were taken to war on the premise that Saddam could attack the US, Europe or Israel with weapons of mass destruction in short order, and that he had active weapons development programs in order to add nukes to the mix. Neither of these statements were true.
(BTW, who sold him the weapons he used on the Kurds? We did.)
Anyway, yes, the world could end up better off without Saddam, although the way things have gone since we've shown up, that certainly wasn't true in the short run. The real question is, why Iraq? It's not like Saddam had a global monopoly on brutal dictatorship, genocide, and so on.
You realize, of course, that long before FOSS was big, over 80% of software written was never sold. It was developed for internal consumption. That's a huge piece of the pie.
As for software sold to others, have you ever heard of "support contracts"? That's where folks like RedHat make their money. Even Microsoft makes money on support. They make a lot of money off of certifying people to work on their software too.
And then there's sponsored development. This is where the two paragraphs above intersect. Suppose Company X really like some package Y, but it's missing some feature it really needs. It can code it itself (the old internal development model) and spend the money internally, or it can hire someone outside to implement the feature. Not an ounce of altruism there. The FOSS license ensures that the feature is able to become part of the overall product. Company X derives direct benefit, and likely has strong influence over the shape it takes.
IBM doesn't send zillions of patches to Linus out of altruism. They send patches because they want Linux to behave better and have the features they want so they can ship more servers. Freescale doesn't send patches to Linus out of altruism. They do it because they want Linux to run well on their embedded chips so that more people will buy them. And so on.
You've got this vision that this is all a big charity. No, it's enlightened self interest.
Keep efficiency in mind. Remember, once you delete one link in the circular reference, then your reference counting garbage collector will do the rest.
You do raise some very good points, and I guess I hadn't fully considered the long-term maintainership, or the impact of focused Wikis.
Having focused Wikis (such as one I heavily contribute to, IntelliWiki) capitalizes on one of the Internet's main strengths—decentralization.
I guess, ultimately, I want access to as much information as possible, and I like the Wiki concept as one means for organizing it. And as you correctly and cogently point out, that Wiki doesn't need to be Wikipedia.
Slashdot Mirror
My high level inputs here, in case you missed it. (Read both posts--one links to the other.)
I realize I've described my employer's system at a very high level. I don't know to what extent our system has been described outside the company. I do know it has been described outside the company based on comments made by one of the lead patent attorneys that runs the system, since he was relating others' reactions outside the company to what we do. I also don't know the terms under which our scheme was disclosed.
Some additional comments:
The structure of your scheme bears a striking resemblance to ours, which inclines me to believe that this structure is popular and (one hopes) useful. You might want to consider the impact of multiple inventors on a given patent: Do you give each inventor the same award regardless of whether there was 1 inventor or 10? Do you split a fixed award among the inventors (e.g. that $500 gets split 10 ways with 10 inventors)? Do you split the difference at some threshold (e.g. $500/inventor up to $XXX dollars, and $XXX/N above that threshold)?
You might also want to consider the long term impact of the patent. In some sense, giving an equity stake in a company that could grow quickly does tie your incentive to the success of the company and can be the right thing to do. That might make sense in a young startup, but not in an established company whose value on the market can be very starkly disconnected from the activity of any given contributor. Neither one, though, ties incentive to the value of the actual invention. So, perhaps consider having a re-review either at the time the patent issues, or sometime after the patent issues to see how much revenue you can tie to the patent in order to gauge a reward.
I could (and indeed would like to) go into greater depth with the relative merits of the system we use at work, but I don't want to divulge proprietary information. I hope my broader insights are useful.
--Joe
I want to be clear on this, since there's a notable difference between copyright and patent here: If you publish the details of your invention without having first filed for a patent on it, then the invention is effectively in the public domain. It's a bit like losing trade secret privilege. Once an idea is known, it can't become unknown.
Now, aspects of how you implement your invention might be copyrightable, such as source code. You could have a completely proprietary application under full copyright protection. If you publish its techniques and protocols without filing patent disclosures first, then those techniques and protocols are no longer patentable. But, the actual code that implements the techniques and protocols is still protected under copyright. So, the invention is effectively in the public domain, but your specific implementation is not.
An example of this might be the SMB protocol that Microsoft uses for sharing files over a network. Portions of the protocol may be protected by patents. Had they instead chosen to document the protocol without filing for protection, then these aspects would become unpatentable, and everyone could implement the protocol without getting granted special exemptions. In either case, though, Microsoft retains the copyright to its Windows implementation.
I wanted to make this clarification, since copyrights and patents so often get confused.
(And then there's chips with their "mask work" protection...)
--Joe
If you release your invention to the public domain, if someone subsequently tries to patent it, the patent won't be granted (or will be invalidated on a re-review if it slips through) on the basis of prior art. This is, of course, if you do actually invent something before someone else does.
Releasing your invention to the public domain when you weren't actually the first to invent it does not absolve you of liability to another inventor's patent, if they invented it before you. In that case, if their patent is invalid, then you need to find someone else's prior art, or find some other grounds to illustrate that the patent is invalid.
Pure patent trolls have the annoying property of not actually having any product that might infringe on something in a company's stack of patents. That's what makes them so potent, dangerous and effective. They've got nothing to lose and everything to gain. And since their lawsuits tend to end in settlements, rather than judgments, the validity of the patent doesn't get a chance to be tested in court.
My big problem with the stack-o-patents is that it makes it really hard for the little guy to get into the business. This is true even with pure hardware inventions. Competing Megacorps A and B know to not sue each other without very good reason because they both have huge piles of IP, and who knows where one or the other may have accidentally infringed on the others' patent. Startup C, on the other hand, doesn't have this parity. Megacorp A or Megacorp B could decide to sue Startup C whenever they like, because it has far fewer resources for defending itself, and not much of a portfolio if anything. This is but one way patents stifle innovation.
Personally, I wouldn't mind if patents were modified to be less strictly exclusive. That is, disclose your invention and have exclusive rights for a very short period (say, 3-5 years from the date of filing), and for the remainder of the term open it up for compulsory licensing. That'd poke a hole in the patent troll's bubble (since they seem to collect patents from failing concerns near the end of the patent's lifetime), and it'd lessen the impact of the stack-o-patents mutually assured destruction effect. It would also protect the initial product development time frame and initial time in the market, to provide a little kick to the first-mover advantage, without scuttling long term support of products by patent-blocking interoperability.
I think if you remove the "You Won The 20 Year Exclusive Right To Charge What You Wish Lottery" aspect from patents, then a lot of the frivolous bets people place will disappear, since you just switched out PowerBall for Pick-4. People will still play, but the payout will be much less, and you won't get the rush of people angling for that $300M jackpot that arises from time to time.
--Joe
It's a deal!
At our company, all patent disclosures are reviewed by other technical experts to determine their suitability to file. After all, it isn't free to file patents. Thus, there's a scale: Small bonus for decision to file to encourage disclosures, larger bonus for the patent issuing, and a variable bonus some years later on review as I talk about here.
The technical guy isn't the one that would be flooding the patent office with bogus disclosures in an attempt to write himself an endless stream of bonus checks. There's always at least one layer that could catch the bogons: the patent attorney that drafts the patent and works all the procedural issues. I suggest an additional layer of technical review to sort good from bad and to prioritize among the good so the most important ones get filed first.
Where I work, we do regular reviews of our issued patents to determine which were valuable and which were not. The valuable ones win the inventors an incentive award. And, since it's a company full of engineers, yes, we do have fancy schmancy formulas that drive the process, and determine the amount of the award. (There are also fixed awards for "decision to file" and "patent issued.")
Assigning specific patent royalty amounts rarely makes sense. Sometimes you can point to specific licensing actions that generate revenue associated with a specific patent, but most often you can't. Many patents are cross licensed to other companies through bulk cross licensing agreements. Others just form a "mutually assured destruction" sort of bulwark against other companies. (This latter property doesn't work so well against patent trolls, since trolls don't make anything--they just have their patent(s) that they chase people with.) Thus, we tend to look at how much revenue was associated with products that use the patent (whether the revenue was ours or someone else's) and go from there.
As for my background: I'm a member of one of our internal committees that reviews disclosures to determine what to file. At the very least, that acts as a quality filter internally. I've also filed several patent disclosures of my own, and have even had a few patents issue. Some have even gotten incentive awards in those periodic reviews. :-) I'm not a big fan of the patent system, but as long as it's there, I figure I may as well use it and let it die under its own weight.
Indeed. For some reason I thought the Apple /// came out later than that. I remember those things. I used to write papers in /// Easy Pieces, and even played a bit with Pascal on it. I never got to play with Business BASIC, though I did see the demo. (Hastings Manufacturing, the piston ring manufacturer, had bought a bunch of these back in the day, and donated them to the public library and YMCA when they got rid of them. While I was in high school, I consulted for both, setting up the systems and the data disks for them.)
As far as I'm concerned, as long as there's a key in that location that I can use to remove the character to the left of the cursor, I'm happy.
Next up: The never ending confusion between the ASCII backspace (BS, character #8), the ASCII delete (DEL, #127), and the DEC VT-100 "Remove" escape sequence. :-)
For the word processor users of the world, it should just have a "paragraph" symbol on it. After all, in a word processor, you're only supposed to press that key at the end of each paragraph, when you're done entering the paragraph.
And carriage returns? What carriage would this be? If you're editing your documents the old fashioned way--and I still use vim plenty--the key should be labeled "new line" or similar. But wouldn't most Mac users be entering text in a word-processor-like environment that provides line breaks in paragraphs automatically?
Personally, I'm not arguing what the label should be. All the labels suck for one reason or another. (Why doesn't the "return" key return me to the previously visited webpage? ;-) ) At least "delete" actually deletes somethings on modern Apples, as opposed to inserting that checkerboard character like it did on the Apple //es. (At least, in Applesoft BASIC. AppleWorks did the right thing of course.)
When I signed up for DSL service, it was with SBC Yahoo! DSL, you insensitive clod!
Well, once Apple finally got a numeric keypad. On the PC, though, both have been named "enter" from the beginning.
My point, of course, was that complaining about "backspace" being a typewriter function by citing "delete" as a better choice overlooks that "enter" is a better choice than "return" by similar reasoning.
Someone just needs to hit RUN STOP-RESTORE on this whole thread. ;-)
No, cmd.exe is the Windows NT shell. WinNT and its offspring are not built on MS-DOS, though they do provide versions of most of its commands. You're thinking of COMMAND.COM.
If you're going to trot out the Apple // line, you may as well know its history.
Not true. These were added on the Apple //e, which antedates MS-DOS. Take a look at the Apple ][+ as compared to the Apple //e.
The closed-Apple key didn't become Option until the Apple IIgs. (The IIgs unit.) They weren't even on the Apple //e Enhanced. The familiar Macintosh Cmd and Option keys, though debuted with the original model, though there was no control key. But, then, a Mac isn't an Apple //, is it?
So is "return" (as opposed to "enter"). Your point was again? Now get off my lawn.
--Joe
(I grew up with these machines, and I remember their sometimes frustrating differences well.)
Murrrh? That's ridiculous. You're not touching my perl. It's sufficient to say:
print "The memory count not be \"read\".";
Or, more likely, it's in a function that prints errors, so the whole thing looks like this (ignoring the extra dots to get "indentation"):
sub memory_error_string($$$)
{
. . return "The instruction at '$_[0]' referenced memory at '$_[1].' The memory could not be '$_[2].' Click OK to terminate program.";
}
So what sorts of outcomes are being measured in outcome based education? To me, "outcome" means "measure of achievement toward a goal" such as "acquiring a certain set of knowledge." I'm all for measuring and grading someone on what they actually managed to learn and achieve by the end of the course, rather than on how many busyworks they handed in on time. The outcome of education should be educated individuals. If individuals demostrate that they're educated, then it doesn't matter so much that they didn't hand in all their homework.
But from how you describe it, "outcome" means "the student's been in our care for a sufficient period of time." (aka. automatic advancement.) Actual level of competence doesn't matter. How is "final level of competence" not part of "outcome"?
(I'm not trolling and this is not flamebait. This is an honest, sincere question. I don't have children and have been out of public schools for over a decade and a half. Gimme a break.)
--Joe
When I was in high school in Michigan some years back, there was a state-level requirement on attendance. More than 15 days absent for any reason (IIRC) in a marking period would be an automatic failure for reason of lack-of-attendance. (I think the school board could override in special circumstances, such as an extended medical emergency, but otherwise even extended illness still counted toward the 15 days.) For a primary school, this makes a certain amount of sense.
I got the sense while I was there that the net effect was to keep the athletic departments from pulling the athletes out of class too often for away games and so on. That's reasoning can get behind.
Even so, it puts attendance above achievement. Public school is, effectively, a national day care for minors, so I can see attendance being important. In college, I made ample use of flexible attendance policies.
Before the grammar Nazis get here... It'd read better if I said "the type of classes I wasn't particularly interested in." Also, my word choice would have been better if I said "weighed my average down" instead of "weighted my average down."
I actually had the teacher that taught the programming course (in high school) tell me that he refused to have me in the class, because he didn't see the point of "teaching" a student that knew more than the teacher. (This was for an Applesoft BASIC class back in ~1990.)
You kids with your IT fetish. I spent a summer working at an ISP while getting my engineering degree. I am soooo glad I didn't get sucked into IT.
Now if I just could have tested out of that programming course and had gotten credit for it, that'd be awesome. Courses like that that I would have aced never showed up on my GPA. Instead, courses like Biology and Shop (classes I wasn't particularly interested in) weighted my average down.
Goes to show that GPA is a deeply flawed measure. It only gives you an average score among the classes that you cared to and were allowed to attempt.
Showing my age here... The convention stretches back at least as far as Microsoft BASIC in the early 80s. It wouldn't surprise me that other formats adopted it. Nor would it surprise me to learn it predates BASIC.
*points* Hey, it's a BASIC programmer!
In C that'd be "\"read\"".
Which reinforces my point: IBM doesn't contribute patches to Linux altruistically, and yet IBM contributes quite a large number of patches. They have good reasons for doing so, and it amounts to enlightened self interest. Previously, they focused primarily on improving AIX to ship servers (that may or may not run DB/2). Now they contribute heavily to Linux.
I'm sure they don't mind all the positive PR they get for their contributions, either.
Check your reading comprehension. I was accusing the Iraqis for incompetence, by burying their arms and losing them for 15 years, rather than rushedly burying them to hide them from weapons inspectors in the months just before we invaded so that they could be easily readied to attack us.
We were taken to war on the premise that Saddam could attack the US, Europe or Israel with weapons of mass destruction in short order, and that he had active weapons development programs in order to add nukes to the mix. Neither of these statements were true.
(BTW, who sold him the weapons he used on the Kurds? We did.)
Anyway, yes, the world could end up better off without Saddam, although the way things have gone since we've shown up, that certainly wasn't true in the short run. The real question is, why Iraq? It's not like Saddam had a global monopoly on brutal dictatorship, genocide, and so on.
Oh, wait, Iraq has oil.
You realize, of course, that long before FOSS was big, over 80% of software written was never sold. It was developed for internal consumption. That's a huge piece of the pie.
As for software sold to others, have you ever heard of "support contracts"? That's where folks like RedHat make their money. Even Microsoft makes money on support. They make a lot of money off of certifying people to work on their software too.
And then there's sponsored development. This is where the two paragraphs above intersect. Suppose Company X really like some package Y, but it's missing some feature it really needs. It can code it itself (the old internal development model) and spend the money internally, or it can hire someone outside to implement the feature. Not an ounce of altruism there. The FOSS license ensures that the feature is able to become part of the overall product. Company X derives direct benefit, and likely has strong influence over the shape it takes.
IBM doesn't send zillions of patches to Linus out of altruism. They send patches because they want Linux to behave better and have the features they want so they can ship more servers. Freescale doesn't send patches to Linus out of altruism. They do it because they want Linux to run well on their embedded chips so that more people will buy them. And so on.
You've got this vision that this is all a big charity. No, it's enlightened self interest.
Keep efficiency in mind. Remember, once you delete one link in the circular reference, then your reference counting garbage collector will do the rest.
(I wonder what that does to its PageRank.)
You forgot to ask how far right I turn...
You do raise some very good points, and I guess I hadn't fully considered the long-term maintainership, or the impact of focused Wikis.
Having focused Wikis (such as one I heavily contribute to, IntelliWiki) capitalizes on one of the Internet's main strengths—decentralization.
I guess, ultimately, I want access to as much information as possible, and I like the Wiki concept as one means for organizing it. And as you correctly and cogently point out, that Wiki doesn't need to be Wikipedia.