Actually I'd agree with a fairly gloomy outlook for the future. But a 5000 times reduction in value, in 15 years? That's either a joke or insanity, no way that would happen.
Yes the US has huge debts. No, they will not be all called due at once because that would disastrous for the world economy, not just the US. A more likely scenario is that they grow steadily more painful and the US economy slowly declines.
In short, I think there are plenty of real, serious problems that need dealing with (global warming and other environmental issue concern me as well), but doomsday is not upon us.
Perhaps I was a bit too hasty in my criticism of the author, who probably intended it as a joke. Instead just aim my criticism at the moron who modded the comment 'Insightful'. Either way, it's a problem. Confusion of comedy and serious information undermines both.
Jighly moderated, idiotic statments like this with no support whatsoever make me glad that football preseason is here and I can spend time reading sports commentary instead of Slashdot.
Seriously, the most worthless sports website message board crap is more intelligent than garbage like this. Karma be damned, if I read much more drivel like this at +4 I won't stick around anyways.
I might agree that incremental updates can be worth a new release, but unfortunately EA's lack of care shows in Madden 2006.
As you might see from my sig, I run a website that features custom designed plays for the Madden series. I started this website shortly after the release of Madden 2003, the first to feature the Create-a-Playbook feature. This feature has gone largely unchanged since this time. In some ways this is good, as all of my older plays mostly work in the newer versions, but this has also meant that long wanted updates, such as the ability to save audibles with a custom playbook (currently they must be set for every game if you want to use them) have not materialized. Other features are lacking, some which would be difficult to implement, but many like the audibles or the ability to reorder formations within a playbook would be trivial.
Well this year EA's incompetence has sunk to a new low. Custom playbooks are not available to be used in franchise mode, only in exhibitions. For many people this will be serious enough to disregard custom playbooks completely. It's probably a mistake and was not done by design, but this kind of step backwards should simply not happen when you are only trying to improve an already stable product.
You make a good point, but keep in mind that $10 billion is just over one-quarter of Microsoft's annual revenues, making it at the least very plausible as an amount a monopoly can overcharge for such an arbitrarily priced good as shrinkwrap software licenses. This is in stark contrast to the RIAA which uses an obviously flawed methodology as counting every illegally copied song as a lost sale.
I'm not positive, but I believe that the system that Orlowski is critical of is the general state of copyright and not the Creative Commons, i.e. Orlowski is likely a proponent of Creative Commons.
If so, it would be more like Interviewing Linus Torvalds to understand the benefits of Linux (to correct your metaphor). Unfortunately Dvorak uses Orlowski's statment of what CC does that copyright doesn't ("It does nothing") as a straw man attack on CC itself, cherry picking the part of the argument that CC does not supercede copyright to argue that it is completely useless.
Well like I said, I don't really know Ruby so I didn't want to talk about it. Remember also that Python is a continually evolving language, so it's possible that it might include higher level lambdas at some point. Additionally I'm not an expert on functional programming style, but I believe I've read an argument that Python can do everything LISP can through lambdas in a slightly less direct manner because the key ability lies in treating functions as first class data types. In this case LISPs syntax would win out, but not by a huge margin. Correct me if I'm mistaken here.
Python was just an example in any case, my point being that even if LISP is the purest language ever created, it is not necessarily the most productive across a wide selection of developers because the syntax may not provide the best mapping of a mental model of computation to the processor for this set of developers.
LISP solved programming about 45 years ago. once you reach a certain level of features (recursion, lambdas, continuations, redefining functions) you've pretty much written a dialect of lisp.
I'd say this is at best a half truth. What you say is true about how LISP and similar languages function under the hood, but largely ignores the actual act of writing the code, something that many people do not appreciate about LISP with its unusual syntax. I think that Python (don't know Ruby personally) might be viewed as an attempt to reconcile some of the core features of LISP with a more friendly coding style.
It's common to see people argue in favor of LISP on the merits of its expressive capabilities while glossing over it's syntax, despite the fact that history has shown that programmers overwhelmingly prefer more procedurally derived, block structured syntaxes. Language purists might argue that syntax is not important, or that LISP's syntax is superior, but in terms of which languages thrive and which lapse into obscurity, it's evident that this type of preference counts quite a bit.
Except that under a BSD license a future employer could require him to make modifications to the original code and require that those modifications be proprietary and stay within the company.
With the GPL a future employer could require the same problem be resolved with completely new proprietary code, but at least they have the option to use a tried and true GPL version if they do not require proprietary code.
Conduct more studies on the subject. No, this isn't a smart ass answer, I'm serious. Any type of study is best confirmed or denied by conducting additional studies, using different samples, varying methodologies, wider scope, etc.
Maybe the correct answer isn't exactly one-third (this is only an estimate after all), but additional studies can help home in on a more precise number.
Hmmm, I and likely many others consider anything on the desktop to be a shortcut. From a true user perspective the desktop may be an actual storage place, but people like you and I realize the the desktop is just a metaphor for a special kind of directory.
I'm glad you're willing to make improvements to your work, I hope this is a point that you clarify, considering the article is intended for a technical audience that might think of all desktop icons as shortcuts.
He's not misinterpreting your article really. You made a big point of only having a few standard desktop icons, made it a point to say that this was all that was needed. The rest seems an afterthought at best and contradictory at worst. I quote:
Everything else should be kept off the desktop. In particular, it is rather important for the system to NOT have desktop shortcuts in order to prevent the common glut of special offers and installers
Maybe your audience would appreciate your work better if you were less condescending and actually treated these widespread misconceptions as potential flaws in your own work?
Sorry, but no. Very few Slashdotters advocate illegal copying of Microsoft software, though most accept it as a simple reality. Most people who advocate illegal copying of any kind do so with the argument that they would not have bought the (license to) the item even if no means of free acquisition were available.
Most specific software copying that is advocated, or at least admitted to, involves things like Adome Photoshop or expensive modeling software like Maya. People get illegal copies of this software, learn it, and often build a career out of it involving legitimate purchases/licenses. Most others who use illegal copies do not do so professionally and would never have bought the software anyways.
Possibly the key factor is whether the creator of a work is deprived of money he would have received if illegal copying was not an option. In the case of Photoshop, this is probably no. In the case of Peter Jackson, he most certainly (assuming the complaint is legit) is being deprived of money that he already has earned. Apples and oranges.
The worst thing you can do in any business is advertise a product too soon.
Definitely not true in many cases. Companies like Apple and Nintendo follow this line of thinking, but Microsoft sure doesn't, and you'd be hard pressed to think up any case where early hype really as hurt them.
One more thing, the biggest beneficiary of such a blacklist would probably corporate sysadmins, who could setup their gateways/firewalls to block known spyware sites, independent of the web browser each desktop runs.
Somebody needs to start scanning Web sites where this crap comes from, report the assholes to the law, and get the lot thrown in jail. NONE of this stuff came in through email because my client uses Web mail exclusively. That means it came from Web sites. So why not set up a Web scanner that visits suspicious Web sites, downloads this crap into a sandbox, logs everything as evidence, then publishes it as a blacklist - a "reverse honeypot"?
I recently thought of something like this when reading one of the periodic articles describing the giant abuses of email blacklists. The problem with email blacklists are that
1) Source IP addresses of spam are easily changed (by moving to new zombies) and often conflict with legitimate users (e.g. cable model or DSL pools)
2) Domain sources are fairly easily concealed by falsified headers
Neither of these issue applies to websites. Websites generally don't work unless a proper domain name is used, and even if a surfer is sucked into a spyware ridden IP address only site, that site isn't going anywhere. There aren't legions of zombified webservers like there are legions of zombified spambots.
So why don't email blacklist operators actually do something useful with their lives and create spyware blacklists. These would actually be effective. ISPs would have the option of using them, though most would probably be reluctant to do so considering that lots of spyware comes from sites people actually like, so web browsers might be the best point of defense. It would be easy to whip up a SpyBlock plugin that works just like e.g. the AdBlock plugin and prevents downloads from know malicious sites, and these sites would have a much more difficult time than spammers of staying ahead of the blacklists.
How is letting a student have free reign scary? There's no rule that says they have to have access to your primary CVS repository, or that you have to actually use what they produce in your main codebase.
As for mentoring time/difficulties, that's what the $500 (the chunk of the total $5000 that goes directly to the mentoring organization) is for.
it shouldn't matter one iota what your competitors are doing. Listen to your customers, not your competitors. Your customers will let you know what they really want that your competitors have, and anything else isn't worth looking at.
Sorry, but this is a really dumb attitude. While it's a poor approach to copy feature willy-nilly, it's worse to ignore expereince, however it is gained. I think you've likely taken a Spolsky idea and warped it into something he himself wouldn't agree with. Ignorance is never a virtue, and should not be sought after.
Customers don't always know what they want, or what is possible (or feasible). A competing product is evidence of what is possible and how well it works, and can provide key insights for new designs. Chances are your competitors have thought of a few good ideas that you haven't. A customer may not be aware or fully informed about the competition; if you truly want to provide something better then you should be keeping abreast of new ideas so that you can share them with a customer.
I use the word customer loosely here, mainly because it is what was used in the above post, but what is really meant is audience, and applies to anything which will be used by others.
I joined the army, put every dime possible into my GI bill (5-1 match at the time) and when I did my time I went to college.
I worked for and earned EVERYTHING I have. NOTHING was given to me ever.
You know, while your story is quite respectable on the whole, I cringe just a bit when someone talks about how they earned everything they ever had, but made it through school and paid for their education using military service and the GI bill. My fiance has taken advantage of some of these same plans through the Air Force National Guard, but she's quite realistic about how sweet a deal it can be, especially when used to pay for school.
No doubt you worked hard, but somethimes I think that military service is basically welfare that you work for. We spend a ridiculous amount of money on the military in the US, more than is necessary IMHO, and when I see people reaping the benifits of that money it smacks of welfare to me. Not because the labor isn't there, but because it's labor that no one (or very few) would be willing to pay for directly. Not only that, but you got out as soon as you could, effectively maximizing the return (wages plus tuition) on your investment (time and effort). You definitely got more out of the deal than your labor would have been worth on any open market.
Now, I don't blame you for taking advantage of this part of the system, especially since you did obviously work hard. That doesn't mean that the government didn't give you a lot more than your labor and service were actually worth though, otherwise you probably would have used the private sector to bootstrap your career. Your hard work story would be more credible in my mind if it was not done using inflated government spending.
I've never done it, but Knoppix offers what is called a poor man's install" which I believe is pretty much a direct copy of the ISO to a hard drive (which I believe is optional, and I may be wrong about this being part of the process), and storage of user files and prefs alone separately on the hard drive or a USB drive.
This allows for the installation to be easily upgraded (just get the new version of Knoppix), while using your computer mostly like normal because you can save and modify files normally.
Your argument is emotionally appealing but ultimately comes down to a logical fallacy. It boils down to: if people are allowed to make decisions about how to (run their lives/raise their children, create a family) they might make a bad or wrong choice.
Everyone has the potential to screw up every time they make a choice. Some choices will ultimately lead to greater happiness and fulfillment in our lives, others won't. Even after the fact we can never be sure we've made the right or best choices because we can never be sure what the alternatives really would have turned out like.
If your parents had been given the choice you describe it sounds like keeping you would have beena good call. But what if you had developed just fine physically, but turned into a child molester, or murderer? Or what if in terminating you they decided to try for another child who ends up curing cancer or performing other great feats in service of humanity? Unlikely, but you see that every choice is a gamble.
We can't make perfect choices, but we have to make choices based on the limited information we have. Agenetic test proides additional information that will hopefully be used to make more informed choices with better chances of "success" (the definition of success will vary from person to person here). Some bad calls will be made, but proper testing could ultimately save thousands or millions of parents from frustration and hopelessness raising severly disabled children, and save their children from lifetimes of potential pain and/or suffering.
You dismiss rear-projection far too quickly. I have a 5 year old Toshiba 40H80 40 inch rear projection HDTV. It cost me $2300 new, which was pretty close to the least you could pay for anything in that class at the time. Rear projection offers some of the best overall value (not too expensive, big but not too heavy, good clarity and response times).
Projection TVs don't have the huge viewing angles that CRTs do, but in practice it's not likely a problem as long as you have a decent place for the set in your room. The display appearance will dim rapidly as you move away from the optimal viewing angle vertically, but horizontally its perfect up to about 45 degrees from center (about the same as a modern LCD), with gradual falloff after that. Not many people would choose to watch from outside this angle anyways, so not really a problem in practice.
As far as the picture sucking up close, how is this different from any other TV? A big screen means big pixels up close, even at HDTV resolutions. I'm not sure how this is different for rear projection than for any other type of set.
If you were going to bring up a knock on projection sets it probably should have been convergence. My TV has 9 different convergences that must be individually adjusted to get the best picture (center, 4 sides, 4 corners). It's a bit tedious, and the crosshairs used to determine convergence don't always give the best indication of convergence (this likely varies between models and manufacturers). Not a big issue, but something that you don't need to worry about with any other type of display.
I checked CUPS and I think I had already tried that. Then I looked at the OO printer manager, which I had never used before. I had never needed to use it because I had automatically found my correct printer when I needed to print once it was setup in CUPS.
Now I've actually setup my printer in OO and set the paper size to Letter in the printer manager, and it looks like the setting has stuck, although my printer now shows up twice in the print dialog (I get Generic Printer, the printer I setup manually, and the one originally detected). I haven't tested it yet but I think it should work.
Slashdot Mirror
Actually I'd agree with a fairly gloomy outlook for the future. But a 5000 times reduction in value, in 15 years? That's either a joke or insanity, no way that would happen.
Yes the US has huge debts. No, they will not be all called due at once because that would disastrous for the world economy, not just the US. A more likely scenario is that they grow steadily more painful and the US economy slowly declines.
In short, I think there are plenty of real, serious problems that need dealing with (global warming and other environmental issue concern me as well), but doomsday is not upon us.
Perhaps I was a bit too hasty in my criticism of the author, who probably intended it as a joke. Instead just aim my criticism at the moron who modded the comment 'Insightful'. Either way, it's a problem. Confusion of comedy and serious information undermines both.
Jighly moderated, idiotic statments like this with no support whatsoever make me glad that football preseason is here and I can spend time reading sports commentary instead of Slashdot.
Seriously, the most worthless sports website message board crap is more intelligent than garbage like this. Karma be damned, if I read much more drivel like this at +4 I won't stick around anyways.
I might agree that incremental updates can be worth a new release, but unfortunately EA's lack of care shows in Madden 2006.
As you might see from my sig, I run a website that features custom designed plays for the Madden series. I started this website shortly after the release of Madden 2003, the first to feature the Create-a-Playbook feature. This feature has gone largely unchanged since this time. In some ways this is good, as all of my older plays mostly work in the newer versions, but this has also meant that long wanted updates, such as the ability to save audibles with a custom playbook (currently they must be set for every game if you want to use them) have not materialized. Other features are lacking, some which would be difficult to implement, but many like the audibles or the ability to reorder formations within a playbook would be trivial.
Well this year EA's incompetence has sunk to a new low. Custom playbooks are not available to be used in franchise mode, only in exhibitions. For many people this will be serious enough to disregard custom playbooks completely. It's probably a mistake and was not done by design, but this kind of step backwards should simply not happen when you are only trying to improve an already stable product.
You make a good point, but keep in mind that $10 billion is just over one-quarter of Microsoft's annual revenues, making it at the least very plausible as an amount a monopoly can overcharge for such an arbitrarily priced good as shrinkwrap software licenses. This is in stark contrast to the RIAA which uses an obviously flawed methodology as counting every illegally copied song as a lost sale.
I'm not positive, but I believe that the system that Orlowski is critical of is the general state of copyright and not the Creative Commons, i.e. Orlowski is likely a proponent of Creative Commons.
If so, it would be more like Interviewing Linus Torvalds to understand the benefits of Linux (to correct your metaphor). Unfortunately Dvorak uses Orlowski's statment of what CC does that copyright doesn't ("It does nothing") as a straw man attack on CC itself, cherry picking the part of the argument that CC does not supercede copyright to argue that it is completely useless.
Well like I said, I don't really know Ruby so I didn't want to talk about it. Remember also that Python is a continually evolving language, so it's possible that it might include higher level lambdas at some point. Additionally I'm not an expert on functional programming style, but I believe I've read an argument that Python can do everything LISP can through lambdas in a slightly less direct manner because the key ability lies in treating functions as first class data types. In this case LISPs syntax would win out, but not by a huge margin. Correct me if I'm mistaken here.
Python was just an example in any case, my point being that even if LISP is the purest language ever created, it is not necessarily the most productive across a wide selection of developers because the syntax may not provide the best mapping of a mental model of computation to the processor for this set of developers.
I'd say this is at best a half truth. What you say is true about how LISP and similar languages function under the hood, but largely ignores the actual act of writing the code, something that many people do not appreciate about LISP with its unusual syntax. I think that Python (don't know Ruby personally) might be viewed as an attempt to reconcile some of the core features of LISP with a more friendly coding style.
It's common to see people argue in favor of LISP on the merits of its expressive capabilities while glossing over it's syntax, despite the fact that history has shown that programmers overwhelmingly prefer more procedurally derived, block structured syntaxes. Language purists might argue that syntax is not important, or that LISP's syntax is superior, but in terms of which languages thrive and which lapse into obscurity, it's evident that this type of preference counts quite a bit.
Except that under a BSD license a future employer could require him to make modifications to the original code and require that those modifications be proprietary and stay within the company.
With the GPL a future employer could require the same problem be resolved with completely new proprietary code, but at least they have the option to use a tried and true GPL version if they do not require proprietary code.
Conduct more studies on the subject. No, this isn't a smart ass answer, I'm serious. Any type of study is best confirmed or denied by conducting additional studies, using different samples, varying methodologies, wider scope, etc.
Maybe the correct answer isn't exactly one-third (this is only an estimate after all), but additional studies can help home in on a more precise number.
Hmmm, I and likely many others consider anything on the desktop to be a shortcut. From a true user perspective the desktop may be an actual storage place, but people like you and I realize the the desktop is just a metaphor for a special kind of directory.
I'm glad you're willing to make improvements to your work, I hope this is a point that you clarify, considering the article is intended for a technical audience that might think of all desktop icons as shortcuts.
Maybe your audience would appreciate your work better if you were less condescending and actually treated these widespread misconceptions as potential flaws in your own work?
Sorry, but no. Very few Slashdotters advocate illegal copying of Microsoft software, though most accept it as a simple reality. Most people who advocate illegal copying of any kind do so with the argument that they would not have bought the (license to) the item even if no means of free acquisition were available.
Most specific software copying that is advocated, or at least admitted to, involves things like Adome Photoshop or expensive modeling software like Maya. People get illegal copies of this software, learn it, and often build a career out of it involving legitimate purchases/licenses. Most others who use illegal copies do not do so professionally and would never have bought the software anyways.
Possibly the key factor is whether the creator of a work is deprived of money he would have received if illegal copying was not an option. In the case of Photoshop, this is probably no. In the case of Peter Jackson, he most certainly (assuming the complaint is legit) is being deprived of money that he already has earned. Apples and oranges.
There is a wee bit of difference between a year and years. Sheesh.
Neither is punctuation. Honestly, quit using ellipses every other sentence, it's annoying.
Definitely not true in many cases. Companies like Apple and Nintendo follow this line of thinking, but Microsoft sure doesn't, and you'd be hard pressed to think up any case where early hype really as hurt them.
One more thing, the biggest beneficiary of such a blacklist would probably corporate sysadmins, who could setup their gateways/firewalls to block known spyware sites, independent of the web browser each desktop runs.
I recently thought of something like this when reading one of the periodic articles describing the giant abuses of email blacklists. The problem with email blacklists are that
1) Source IP addresses of spam are easily changed (by moving to new zombies) and often conflict with legitimate users (e.g. cable model or DSL pools)
2) Domain sources are fairly easily concealed by falsified headers
Neither of these issue applies to websites. Websites generally don't work unless a proper domain name is used, and even if a surfer is sucked into a spyware ridden IP address only site, that site isn't going anywhere. There aren't legions of zombified webservers like there are legions of zombified spambots.
So why don't email blacklist operators actually do something useful with their lives and create spyware blacklists. These would actually be effective. ISPs would have the option of using them, though most would probably be reluctant to do so considering that lots of spyware comes from sites people actually like, so web browsers might be the best point of defense. It would be easy to whip up a SpyBlock plugin that works just like e.g. the AdBlock plugin and prevents downloads from know malicious sites, and these sites would have a much more difficult time than spammers of staying ahead of the blacklists.
How is letting a student have free reign scary? There's no rule that says they have to have access to your primary CVS repository, or that you have to actually use what they produce in your main codebase.
As for mentoring time/difficulties, that's what the $500 (the chunk of the total $5000 that goes directly to the mentoring organization) is for.
Sorry, but this is a really dumb attitude. While it's a poor approach to copy feature willy-nilly, it's worse to ignore expereince, however it is gained. I think you've likely taken a Spolsky idea and warped it into something he himself wouldn't agree with. Ignorance is never a virtue, and should not be sought after.
Customers don't always know what they want, or what is possible (or feasible). A competing product is evidence of what is possible and how well it works, and can provide key insights for new designs. Chances are your competitors have thought of a few good ideas that you haven't. A customer may not be aware or fully informed about the competition; if you truly want to provide something better then you should be keeping abreast of new ideas so that you can share them with a customer.
I use the word customer loosely here, mainly because it is what was used in the above post, but what is really meant is audience, and applies to anything which will be used by others.
You know, while your story is quite respectable on the whole, I cringe just a bit when someone talks about how they earned everything they ever had, but made it through school and paid for their education using military service and the GI bill. My fiance has taken advantage of some of these same plans through the Air Force National Guard, but she's quite realistic about how sweet a deal it can be, especially when used to pay for school.
No doubt you worked hard, but somethimes I think that military service is basically welfare that you work for. We spend a ridiculous amount of money on the military in the US, more than is necessary IMHO, and when I see people reaping the benifits of that money it smacks of welfare to me. Not because the labor isn't there, but because it's labor that no one (or very few) would be willing to pay for directly. Not only that, but you got out as soon as you could, effectively maximizing the return (wages plus tuition) on your investment (time and effort). You definitely got more out of the deal than your labor would have been worth on any open market.
Now, I don't blame you for taking advantage of this part of the system, especially since you did obviously work hard. That doesn't mean that the government didn't give you a lot more than your labor and service were actually worth though, otherwise you probably would have used the private sector to bootstrap your career. Your hard work story would be more credible in my mind if it was not done using inflated government spending.
I've never done it, but Knoppix offers what is called a poor man's install" which I believe is pretty much a direct copy of the ISO to a hard drive (which I believe is optional, and I may be wrong about this being part of the process), and storage of user files and prefs alone separately on the hard drive or a USB drive.
This allows for the installation to be easily upgraded (just get the new version of Knoppix), while using your computer mostly like normal because you can save and modify files normally.
See details on Poor Man's Installs here.
Your argument is emotionally appealing but ultimately comes down to a logical fallacy. It boils down to: if people are allowed to make decisions about how to (run their lives/raise their children, create a family) they might make a bad or wrong choice.
Everyone has the potential to screw up every time they make a choice. Some choices will ultimately lead to greater happiness and fulfillment in our lives, others won't. Even after the fact we can never be sure we've made the right or best choices because we can never be sure what the alternatives really would have turned out like.
If your parents had been given the choice you describe it sounds like keeping you would have beena good call. But what if you had developed just fine physically, but turned into a child molester, or murderer? Or what if in terminating you they decided to try for another child who ends up curing cancer or performing other great feats in service of humanity? Unlikely, but you see that every choice is a gamble.
We can't make perfect choices, but we have to make choices based on the limited information we have. Agenetic test proides additional information that will hopefully be used to make more informed choices with better chances of "success" (the definition of success will vary from person to person here). Some bad calls will be made, but proper testing could ultimately save thousands or millions of parents from frustration and hopelessness raising severly disabled children, and save their children from lifetimes of potential pain and/or suffering.
You dismiss rear-projection far too quickly. I have a 5 year old Toshiba 40H80 40 inch rear projection HDTV. It cost me $2300 new, which was pretty close to the least you could pay for anything in that class at the time. Rear projection offers some of the best overall value (not too expensive, big but not too heavy, good clarity and response times).
Projection TVs don't have the huge viewing angles that CRTs do, but in practice it's not likely a problem as long as you have a decent place for the set in your room. The display appearance will dim rapidly as you move away from the optimal viewing angle vertically, but horizontally its perfect up to about 45 degrees from center (about the same as a modern LCD), with gradual falloff after that. Not many people would choose to watch from outside this angle anyways, so not really a problem in practice.
As far as the picture sucking up close, how is this different from any other TV? A big screen means big pixels up close, even at HDTV resolutions. I'm not sure how this is different for rear projection than for any other type of set.
If you were going to bring up a knock on projection sets it probably should have been convergence. My TV has 9 different convergences that must be individually adjusted to get the best picture (center, 4 sides, 4 corners). It's a bit tedious, and the crosshairs used to determine convergence don't always give the best indication of convergence (this likely varies between models and manufacturers). Not a big issue, but something that you don't need to worry about with any other type of display.
I checked CUPS and I think I had already tried that. Then I looked at the OO printer manager, which I had never used before. I had never needed to use it because I had automatically found my correct printer when I needed to print once it was setup in CUPS.
Now I've actually setup my printer in OO and set the paper size to Letter in the printer manager, and it looks like the setting has stuck, although my printer now shows up twice in the print dialog (I get Generic Printer, the printer I setup manually, and the one originally detected). I haven't tested it yet but I think it should work.