RUNNING FOR COVER. The cable companies' adoption of new pricing strategies has less to do with stopping piracy than with economics and business models. At an average monthly cost of $45, broadband is still perceived as too expensive by many consumers, and in recent months, prices have been rising, rather than dropping. That's slowing subscriber growth. According to market-research firm ARS, the rate of new signups for broadband in the first quarter of 2002 slid to 12%, the worst quarter on record.
Disappointing demand has left cable operators scrambling to cover the $60 billion they spent building and upgrading their networks over the past decade. At the same time, they've tired of seeing a small group of heavy users tax their networks while paying the same flat rate as everybody else. AT&T Broadband says on its system, 1% percent of users account for 16% of bandwidth consumption.
Um, as soon as I said that I went on to refute the notion myself.
The main problem with your assertion that Microsoft, Sun, Apple, etc paid for that research and therefore should be able to use it is that others have also paid for it, but the research is of particular use to existing software firms. That they have the possibility to use subsidized research (whether they've helped subsidize it and in what proportion is not relevant, what's relevant is that they weren't the only financial backers) to create a profit is a concern.
However, it's a concern that is easily answered. Suppose research project Z is produced and put into the public domain. As Z, it costs the users of the world $0 to get the utility of Z (UZ). If a company comes along and works with the free code available for Z and produces a product called Z+a, then presumably the utility of Z+a is UZ+ua. Perhaps the price is raised to $0+$a. So what? If I am content to get utility of UZ, I can pay $0. If I want utility UZ+ua, then I must pay the additional $a. In any case, the company cannot charge more than the market will bear for utility ua itself, because that is the only value they have added to the package and consumers will presumably know they can get marginally less utility at no cost.
Besides, as I said before, there is nothing to stop a Free Software developer from taking the same public domain package and adding to it, then relicensing his or her resulting work (U+f) with the GPL. The developer has therefore added utility (uf) so that the overall utility is UZ+uf. But instead of changing the price to $0+$f, the developer simply changes the license to ensure that anyone who wants to obtain utility uf must abide by the GPL. Which means that the company cannot make (UF+uz)+ua and keep ua proprietary.
Either way, the public has only paid for U and has received utility ua. If U were only available under terms such that a company making U+a had to essentially give away the utility ua, then except as a charitable act, no company would ever use resources to develop U+a.
Well, that's the point, I guess. None of the above were terribly difficult to accomplish and even less careful people could do it and get a lot of protection from it. Even so I still worry constantly that someone will find an exploit for the firewall. And I'd write a cron job to do 'apt-get update' on the firewall, but what if someone compromises the Debian server? I just update to corrupt binaries? At least if I manually do it, I have to pay some attention to what's being loaded and only do updates as a result of known security notices.
This is plain nonsense. Give me five examples of government funded software that are licensed as GPL directly upon their release (note that there is nothing to stop *me* from taking public domain/BSD-licensed software, altering and releasing the new version under the GPL-- same as a corporation can take that public domain/BSD-licensed software and incorporate it into a proprietary product).
Furthermore you have not remotely refuted the assertion that the government *should* GPL all their software. Not that I agree, since the public domain/BSD license still only allows companies to charge for adding incremental value to software. That is, if a company takes a piece of public software and adds to it, the price they can charge is the value of their additions, since the non-value-added version is available to me for free.
You su to root, and the keystrokes get mailed somewhere... with your box IP number and username. Well, the attacker just got your root password. Worried yet?
No. I'm not sure the IP address of 192.168.1.[1..255] is going to do them much good. Especially since one should normally configure SSH not to allow root login, and the firewall shouldn't allow any remote logins at all. And of course root on the general-use machines has a different password than root on the firewall. Have I missed anything?
Actually your biggest concern should be how your neighbors feel about this, whether or not you can persuade them that your camera is harmless (assuming it is not showing the interior of their home), and how the courts in your area would rule should they choose to file a lawsuit (assuming you want to let a webcam start a fued of such proportions).
Now, you could take the advice of many below and fashion cardboard cutouts to block the view of the nieghbor's house (this would probably be simple and seems likely to appease the neighbors). You could tell your neighbor to take a flying leap (I recommend consulting a lawyer in your area first, though... if your webcam is illegal in your area-- and different cities/states may well have different laws-- you could be liable for damages and court costs if you lose your case). You could move the cameras so they don't show the neighbor's house at all. You could get a different hobby altogether.
My own personal opinion as a citizen is that if I don't want someone to see inside my house I pull the shades. I have no right to prevent them from viewing or photographing the outside of the house unless they can be shown to be violating a copyright in doing so.
If I have a webcam that shows me my yard (which I have done so that I could monitor it while I am at work-- damn kids wreck my gardens, it would be nice to catch one in the act), but also happens to show me the house across the street or the one next door then tough bounce-- I'm not going to subvert my own right to monitor my property because someone else thinks they have some sort of privacy rights. Of course, I do consider it a form of harrassment (and would not condone) to be setting up a camera in such a way that it can see parts of my neighbors property that I could not normally see (i.e. putting a camera up on a pole to see over a privacy fence--- which would be very different from having an obvious 2nd story window out of which I could easily see over the fence).
You dont have a problem with criminals getting caught, do you?
I do. Doesn't China have some laws that violate what we might consider human rights? Why would I support catching criminals whose crimes consist of practicing or promoting religion or advocating general elections or simply writing articles critical of the government?
Agreed. Minneapolis would be great since I could save hotel money, but I'd consider (in order of preference) Detroit, Chicago, Dallas, and Salt Lake City as well. And no, I wouldn't buy tickets today, but if there were some cool speakers (Mitnick, that cool Senator we all like but whose name escapes me, Stallman, Linus, Bruce Perens, Bill Gates, Steve Jobs-- just some suggestions), maybe a giant LAN gaming session, some programming/techie special interest meetings, etc, I'd seriously consider going.
This makes the librarian's job harder, I admit, but we are supposed to be a provider of helpful information that helps people... not junk that polutes children's minds...
A big part of the problem is that there is absolutely no conclusive evidence that this "junk" actually pollutes the minds of children at all. This whole discussion isn't even really about "protecting" children from "harm". It's a way for people with certain moral leanings to enforce their view of what others should and should not be looking at onto others, regardless of age.
And while I would understand their concerns if the library suddenly built up a large collection of smut (thus diverting funds from the purchase of supposedly more worthwhile books/magazines/movies), there is no added expense to the library to allow people to view whatever they want on the internet (except that this does pose some risk of liability in lawsuits as you mention-- I know in my city, librarians themselves have filed some sort of sexual harrassment suit based on the purported hostile environment that library patrons viewing erotica creates).
I think it is sane for any library, however, to maintain a policy that no internet user will display porn in the library, where others might see. However, it can get extremely tricky to tell what images that might actually be. I suppose I would say that the test is whether or not you can find a book in the library with a similar image in it and sit at a table in view of the public to look at it. If I go to the figure drawing section and pull out a book full of unclothed women, how is that different from reading playboy.com online?
Don't worry. Most of "your" money (as if you really pay enough in taxes that your one opinion counts for diddley-squat) is used to fund social-security/Medicare (33%), paying off the national debt (principal payments 12%, interest 11%), the military-industrial complex (17%), non-Social-Security social programs (food stamps, etc, @ 16%). Libraries fall into community development at 9%, which also includes schools, agriculture, transportation, science, etc.
As to your lame contention that your money is being used to "fund" pr0n freaks. It's not. It's being used to fund computers and internet access. The incremental expense associated with viewing pr0n is $0 (unless you want to contend that this raises demand for the service-- which means to me that you simply put a time limit on usage). There is, however, an incremental expense associated with filtering. Filtering that doesn't even work. Now, that's wasting tax dollars.
[tax information found on page 23 of http://www.irs.gov/pub/irs-pdf/i1040ez.pdf]
Well, *you* think it ought to be illegal for me to request infringing material, but that puts the burden of proof on the consumer. Of course, if it is clearly stated that these movies are unauthorized copies, then my complicity is plain. But under your rules, I'd have to verify the legality of every movie I rent from BlockBuster, buy a copy of at Best Buy, or watch on pay-per-view. Personally I think the law works fine the way it is. Any amendments will have to be worded very carefully.
Oh please. How much code is written correctly in the first place? The problem is that Microsoft's testing procedures and release policies didn't catch or fix this problem at any point in the last several years (assuming it *is* older code that is at fault here). The further problem is that without public embarrassment and probably a benign exploit or two, MS won't even fix the problem, preferring instead their securityThruObscurity algorithm.
What if YOUR dollar went to buy a ticket for a terrorist.
You mean the way all that US tax money given to Afghanistan over the years may have gone to terrrorists? (see http://www.house.gov/paul/tst/tst2001/tst110501.ht m for more info) Or some other way?
Re:Does this make you a Criminal ? Yes of course.
on
Live from Iran, Film88
·
· Score: 4, Insightful
Actually, you're wrong. But that's okay. You're also a coward.
Put aside your lame "intellectual property" bias for two seconds, and you'll realize that the viewer in this case committed no crime. Copyright law restricts the transmission of works, not the reception thereof. You might argue, in this case, that the viewer is making an unauthorized duplication, but if the bits are streamed then no duplication is made... this is essentially a broadcast. If I set up a radio station and play only infringing materials over the air, the listeners are not guilty of a crime.
Re:Perl's had it's day - It's become like COBOL
on
Apocalypse 5 Released
·
· Score: 1
(Although I'm sure you've heard of it...) If you like Perl but dislike it's OO, check out ruby. Ruby's OO is so clean you can see your reflection in it. Every time I hear Larry Wall talking about Huffman encoding, I look at a ruby script to see it in action. While Perl6 has a fantastic opportunity to way outshine ruby, ruby is here today and quite easy to learn/use. Unfortunately, the Ruby Application Archive is no match for CPAN, but some might consider that an opportunity for fame and fortune rather than a failing.;)
No kidding. A bigger screen at $300 less. This may not eat into old iMac sales so much as the flat iMac sales. Because with the exception of drive offerings, the eMac is essentially the same computer with a different monitor.
Question for anyone with the information: what kind of fans do the various current iMac/eMac models offer? Aren't the old G3 iMacs fanless? What about the flat-screen iMac? Any idea on the eMac?
I agree that the statement was totally unfair-- no code is going to be perfect the first time out. But how is the Office v.X better than Office 2000 without MS Access included? And why do you say OS X has no offie suite? It *comes* with AppleWorks! What's worse, the MS Office for OS X costs more than the same package does on Windows... what a rip!
I'm much more worried he'll be reviewing the upcoming Tom Cruise film 'Minority Report'.
Re:Williams not fitting the role?
on
Review: Insomnia
·
· Score: 1
Don't forget 'World According to Garp' from much earlier in his career. Also, his role in 'Birdcage' was both serious and funny-- certainly not the usual Williams madcap stuff. And of course, 'Good Morning Vietnam' did include a lot of great Williams humor, but was also a very dramatic role.
But seriously. What is a review of this movie doing here, even if it is Jon Katz writing? How is this at all news for nerds? I mean, yeah, I saw and loved 'Insomnia' but it hardly topped 'Attack of the Clones'. Sheesh, Katz. Save your not-so-subtle slams on Star Wars for once okay?
Because square inches doesn't add any information and would require a massive change of methodology and consumer re-education. Given an aspect ratio (TV or cinema or whatever) and a diagonal size, we have a lot of information about a screen's capacity. Given square inches we have not gotten more information.
Now given actual vertical/horizontal dimensions we would have even better information, except that the hypoteneuse of a right triangle is always longer than either of the sides, so going from a 17" screen to one that is 13" x 9" is going to take some explaining as to why a screen that sounds a lot smaller actually isn't.
actually a cube might not be stable enough to support the monitor and its arm, the half-bowling-ball approach at least keeps the base center of gravity fairly low. A pyramid shape would've been even better.
What kind of programmer are you, then? All that information sitting around in analog form is decaying so fast I can hear it from here.
The best part is, probably 95% of it is stuff you printed from digital documents that had you kept online in a project repository would not only be instantly updatable, but also under version control. This would mean that changes could be verified, and the documents easily shared.
Let's not even discuss how difficult it is to back up a shelf full of paper, in case of disaster. Then there's the whole question of when you need to relocate to a different place for some reason (which as far as I can tell is a frequent occurrence).
I haven't read CatB-- and if ESR agrees with me, then it's a coincidence. I'm basing this on personal experience. I have several software developers sitting within 25 yards of me right at this moment... none of whose work will be of any use outside this company without major refactoring (of course the code is not GPL, frankly the license is mostly irrelevant). And while I might use a broad definition of "software developer" to include more than just full-time programmers, I would not include sysadmins unless they wrote a lot of maintenance scripts, web designers unless they developed a lot of code, data analysts unless they write custom analytical applications, university students unless they actually produce useful code (remember: Torvalds was a university student), researchers unless they produce applications others can use (good example would be the festival speech engine). I'm not going to include or disinclude people based on a job title. I am going to include people who spend a significant portion of their time at work coding, or those whose code is valuable to others.
I think even if you took a fairly hardline approach to quantifying this number, you'd find no more than half of all programmers working on shrink-wrapped general-use applications. Take a look at the Fortune 500. Yes, a few of those corporations sell software. But at all the others there are enormous systems upon which those businesses are dependent. And while the software may not be entirely unique, large portions of the code base are certainly going to have been customized for that corporation. Look at the Department of Defense and other major governmental organizations. How much of their systems code is shrink-wrapped versus customized? Even certain applications, like Oracle, pretty much require software development type people to operate since off the shelf it doesn't do a whole lot on its own.
Slashdot Mirror
RUNNING FOR COVER. The cable companies' adoption of new pricing strategies has less to do with stopping piracy than with economics and business models. At an average monthly cost of $45, broadband is still perceived as too expensive by many consumers, and in recent months, prices have been rising, rather than dropping. That's slowing subscriber growth. According to market-research firm ARS, the rate of new signups for broadband in the first quarter of 2002 slid to 12%, the worst quarter on record.
Disappointing demand has left cable operators scrambling to cover the $60 billion they spent building and upgrading their networks over the past decade. At the same time, they've tired of seeing a small group of heavy users tax their networks while paying the same flat rate as everybody else. AT&T Broadband says on its system, 1% percent of users account for 16% of bandwidth consumption.
The main problem with your assertion that Microsoft, Sun, Apple, etc paid for that research and therefore should be able to use it is that others have also paid for it, but the research is of particular use to existing software firms. That they have the possibility to use subsidized research (whether they've helped subsidize it and in what proportion is not relevant, what's relevant is that they weren't the only financial backers) to create a profit is a concern.
However, it's a concern that is easily answered. Suppose research project Z is produced and put into the public domain. As Z, it costs the users of the world $0 to get the utility of Z (UZ). If a company comes along and works with the free code available for Z and produces a product called Z+a, then presumably the utility of Z+a is UZ+ua. Perhaps the price is raised to $0+$a. So what? If I am content to get utility of UZ, I can pay $0. If I want utility UZ+ua, then I must pay the additional $a. In any case, the company cannot charge more than the market will bear for utility ua itself, because that is the only value they have added to the package and consumers will presumably know they can get marginally less utility at no cost.
Besides, as I said before, there is nothing to stop a Free Software developer from taking the same public domain package and adding to it, then relicensing his or her resulting work (U+f) with the GPL. The developer has therefore added utility (uf) so that the overall utility is UZ+uf. But instead of changing the price to $0+$f, the developer simply changes the license to ensure that anyone who wants to obtain utility uf must abide by the GPL. Which means that the company cannot make (UF+uz)+ua and keep ua proprietary.
Either way, the public has only paid for U and has received utility ua. If U were only available under terms such that a company making U+a had to essentially give away the utility ua, then except as a charitable act, no company would ever use resources to develop U+a.
Well, that's the point, I guess. None of the above were terribly difficult to accomplish and even less careful people could do it and get a lot of protection from it. Even so I still worry constantly that someone will find an exploit for the firewall. And I'd write a cron job to do 'apt-get update' on the firewall, but what if someone compromises the Debian server? I just update to corrupt binaries? At least if I manually do it, I have to pay some attention to what's being loaded and only do updates as a result of known security notices.
Furthermore you have not remotely refuted the assertion that the government *should* GPL all their software. Not that I agree, since the public domain/BSD license still only allows companies to charge for adding incremental value to software. That is, if a company takes a piece of public software and adds to it, the price they can charge is the value of their additions, since the non-value-added version is available to me for free.
You su to root, and the keystrokes get mailed somewhere... with your box IP number and username. Well, the attacker just got your root password. Worried yet?
No. I'm not sure the IP address of 192.168.1.[1..255] is going to do them much good. Especially since one should normally configure SSH not to allow root login, and the firewall shouldn't allow any remote logins at all. And of course root on the general-use machines has a different password than root on the firewall. Have I missed anything?
Now, you could take the advice of many below and fashion cardboard cutouts to block the view of the nieghbor's house (this would probably be simple and seems likely to appease the neighbors). You could tell your neighbor to take a flying leap (I recommend consulting a lawyer in your area first, though... if your webcam is illegal in your area-- and different cities/states may well have different laws-- you could be liable for damages and court costs if you lose your case). You could move the cameras so they don't show the neighbor's house at all. You could get a different hobby altogether.
My own personal opinion as a citizen is that if I don't want someone to see inside my house I pull the shades. I have no right to prevent them from viewing or photographing the outside of the house unless they can be shown to be violating a copyright in doing so.
If I have a webcam that shows me my yard (which I have done so that I could monitor it while I am at work-- damn kids wreck my gardens, it would be nice to catch one in the act), but also happens to show me the house across the street or the one next door then tough bounce-- I'm not going to subvert my own right to monitor my property because someone else thinks they have some sort of privacy rights. Of course, I do consider it a form of harrassment (and would not condone) to be setting up a camera in such a way that it can see parts of my neighbors property that I could not normally see (i.e. putting a camera up on a pole to see over a privacy fence--- which would be very different from having an obvious 2nd story window out of which I could easily see over the fence).
You dont have a problem with criminals getting caught, do you?
I do. Doesn't China have some laws that violate what we might consider human rights? Why would I support catching criminals whose crimes consist of practicing or promoting religion or advocating general elections or simply writing articles critical of the government?
Agreed. Minneapolis would be great since I could save hotel money, but I'd consider (in order of preference) Detroit, Chicago, Dallas, and Salt Lake City as well. And no, I wouldn't buy tickets today, but if there were some cool speakers (Mitnick, that cool Senator we all like but whose name escapes me, Stallman, Linus, Bruce Perens, Bill Gates, Steve Jobs-- just some suggestions), maybe a giant LAN gaming session, some programming/techie special interest meetings, etc, I'd seriously consider going.
h. Cowboy Neal
This makes the librarian's job harder, I admit, but we are supposed to be a provider of helpful information that helps people... not junk that polutes children's minds...
A big part of the problem is that there is absolutely no conclusive evidence that this "junk" actually pollutes the minds of children at all. This whole discussion isn't even really about "protecting" children from "harm". It's a way for people with certain moral leanings to enforce their view of what others should and should not be looking at onto others, regardless of age.
And while I would understand their concerns if the library suddenly built up a large collection of smut (thus diverting funds from the purchase of supposedly more worthwhile books/magazines/movies), there is no added expense to the library to allow people to view whatever they want on the internet (except that this does pose some risk of liability in lawsuits as you mention-- I know in my city, librarians themselves have filed some sort of sexual harrassment suit based on the purported hostile environment that library patrons viewing erotica creates).
I think it is sane for any library, however, to maintain a policy that no internet user will display porn in the library, where others might see. However, it can get extremely tricky to tell what images that might actually be. I suppose I would say that the test is whether or not you can find a book in the library with a similar image in it and sit at a table in view of the public to look at it. If I go to the figure drawing section and pull out a book full of unclothed women, how is that different from reading playboy.com online?
Don't worry. Most of "your" money (as if you really pay enough in taxes that your one opinion counts for diddley-squat) is used to fund social-security/Medicare (33%), paying off the national debt (principal payments 12%, interest 11%), the military-industrial complex (17%), non-Social-Security social programs (food stamps, etc, @ 16%). Libraries fall into community development at 9%, which also includes schools, agriculture, transportation, science, etc.
As to your lame contention that your money is being used to "fund" pr0n freaks. It's not. It's being used to fund computers and internet access. The incremental expense associated with viewing pr0n is $0 (unless you want to contend that this raises demand for the service-- which means to me that you simply put a time limit on usage). There is, however, an incremental expense associated with filtering. Filtering that doesn't even work. Now, that's wasting tax dollars.
[tax information found on page 23 of http://www.irs.gov/pub/irs-pdf/i1040ez.pdf]
Well, *you* think it ought to be illegal for me to request infringing material, but that puts the burden of proof on the consumer. Of course, if it is clearly stated that these movies are unauthorized copies, then my complicity is plain. But under your rules, I'd have to verify the legality of every movie I rent from BlockBuster, buy a copy of at Best Buy, or watch on pay-per-view. Personally I think the law works fine the way it is. Any amendments will have to be worded very carefully.
Oh please. How much code is written correctly in the first place? The problem is that Microsoft's testing procedures and release policies didn't catch or fix this problem at any point in the last several years (assuming it *is* older code that is at fault here). The further problem is that without public embarrassment and probably a benign exploit or two, MS won't even fix the problem, preferring instead their securityThruObscurity algorithm.
What if YOUR dollar went to buy a ticket for a terrorist.
t m for more info) Or some other way?
You mean the way all that US tax money given to Afghanistan over the years may have gone to terrrorists? (see http://www.house.gov/paul/tst/tst2001/tst110501.h
Actually, you're wrong. But that's okay. You're also a coward.
Put aside your lame "intellectual property" bias for two seconds, and you'll realize that the viewer in this case committed no crime. Copyright law restricts the transmission of works, not the reception thereof. You might argue, in this case, that the viewer is making an unauthorized duplication, but if the bits are streamed then no duplication is made... this is essentially a broadcast. If I set up a radio station and play only infringing materials over the air, the listeners are not guilty of a crime.
(Although I'm sure you've heard of it...) If you like Perl but dislike it's OO, check out ruby. Ruby's OO is so clean you can see your reflection in it. Every time I hear Larry Wall talking about Huffman encoding, I look at a ruby script to see it in action. While Perl6 has a fantastic opportunity to way outshine ruby, ruby is here today and quite easy to learn/use. Unfortunately, the Ruby Application Archive is no match for CPAN, but some might consider that an opportunity for fame and fortune rather than a failing. ;)
No kidding. A bigger screen at $300 less. This may not eat into old iMac sales so much as the flat iMac sales. Because with the exception of drive offerings, the eMac is essentially the same computer with a different monitor.
Question for anyone with the information: what kind of fans do the various current iMac/eMac models offer? Aren't the old G3 iMacs fanless? What about the flat-screen iMac? Any idea on the eMac?
I agree that the statement was totally unfair-- no code is going to be perfect the first time out. But how is the Office v.X better than Office 2000 without MS Access included? And why do you say OS X has no offie suite? It *comes* with AppleWorks! What's worse, the MS Office for OS X costs more than the same package does on Windows... what a rip!
I'm much more worried he'll be reviewing the upcoming Tom Cruise film 'Minority Report'.
Don't forget 'World According to Garp' from much earlier in his career. Also, his role in 'Birdcage' was both serious and funny-- certainly not the usual Williams madcap stuff. And of course, 'Good Morning Vietnam' did include a lot of great Williams humor, but was also a very dramatic role.
But seriously. What is a review of this movie doing here, even if it is Jon Katz writing? How is this at all news for nerds? I mean, yeah, I saw and loved 'Insomnia' but it hardly topped 'Attack of the Clones'. Sheesh, Katz. Save your not-so-subtle slams on Star Wars for once okay?
Because square inches doesn't add any information and would require a massive change of methodology and consumer re-education. Given an aspect ratio (TV or cinema or whatever) and a diagonal size, we have a lot of information about a screen's capacity. Given square inches we have not gotten more information.
Now given actual vertical/horizontal dimensions we would have even better information, except that the hypoteneuse of a right triangle is always longer than either of the sides, so going from a 17" screen to one that is 13" x 9" is going to take some explaining as to why a screen that sounds a lot smaller actually isn't.
actually a cube might not be stable enough to support the monitor and its arm, the half-bowling-ball approach at least keeps the base center of gravity fairly low. A pyramid shape would've been even better.
What kind of programmer are you, then? All that information sitting around in analog form is decaying so fast I can hear it from here.
The best part is, probably 95% of it is stuff you printed from digital documents that had you kept online in a project repository would not only be instantly updatable, but also under version control. This would mean that changes could be verified, and the documents easily shared.
Let's not even discuss how difficult it is to back up a shelf full of paper, in case of disaster. Then there's the whole question of when you need to relocate to a different place for some reason (which as far as I can tell is a frequent occurrence).
I haven't read CatB-- and if ESR agrees with me, then it's a coincidence. I'm basing this on personal experience. I have several software developers sitting within 25 yards of me right at this moment... none of whose work will be of any use outside this company without major refactoring (of course the code is not GPL, frankly the license is mostly irrelevant). And while I might use a broad definition of "software developer" to include more than just full-time programmers, I would not include sysadmins unless they wrote a lot of maintenance scripts, web designers unless they developed a lot of code, data analysts unless they write custom analytical applications, university students unless they actually produce useful code (remember: Torvalds was a university student), researchers unless they produce applications others can use (good example would be the festival speech engine). I'm not going to include or disinclude people based on a job title. I am going to include people who spend a significant portion of their time at work coding, or those whose code is valuable to others.
I think even if you took a fairly hardline approach to quantifying this number, you'd find no more than half of all programmers working on shrink-wrapped general-use applications. Take a look at the Fortune 500. Yes, a few of those corporations sell software. But at all the others there are enormous systems upon which those businesses are dependent. And while the software may not be entirely unique, large portions of the code base are certainly going to have been customized for that corporation. Look at the Department of Defense and other major governmental organizations. How much of their systems code is shrink-wrapped versus customized? Even certain applications, like Oracle, pretty much require software development type people to operate since off the shelf it doesn't do a whole lot on its own.
Thank you for the off-topic post that started at +1. If you don't like the way people are moderating: go to metamod. Thanks.