It sounds like a pretty silly of Zynot to me. Porting code to lower-level languages is a big net loss in most cases, and I can certainly see why other developers wouldn't be excited about the idea. On desktop systems there would be zero gain, and the program would become calcified by the poor implementation language, and painful to maintain for the (usually volunteer) developers. Creating painful-to-maintain code is a deathwish for a free software project.
But maybe that's just a sign that a fork is the proper response -- Zynot's embedded perspective just doesn't fit with Gentoo. It has requirements that just don't make sense for the rest of the community. It sounds like Zachary wanted to professionalize Gentoo, but that's just not interesting to a volunteer developer base.
I really dislike layers, and not because of the CPU overhead. It creates a conceptual overhead.
In the original article, and in some of the defenses of layers that follow in this thread, there's a notion the problem is how much time a program takes to develop. Writing it the first time isn't the hardest part, in my experience, it's maintaining it.
Layers can work well if each layer is robust, and its role clearly delineated and described. When that is not the case, the system is conceptually heavy and pieces cannot be understood or modified in isolation. One common problem is that a bug in a lower level will propogate to a higher level, but in a way that is mysterious and difficult to debug. Or different layers, when first put together, will interact in unpredictable ways. Everything should be predictable -- that's the most important quality I see in a library.
Some things work great. I certainly don't mind layering my network code on a socket library. Of course, everyone has been over that for a couple decades -- as time goes on the foundation expands. The safety of layering improves, but you have to identify the underlying layers that can really be trusted.
Re:but it's more humane!
on
Chicken Run
·
· Score: 1
That final destination is, after all, their raison d'etre, for the individuals as well as the species.
Anyway, I respect PETA for advocating incrementally better treatment even if many of them would prefer that the animals not be eaten at all.
While I agree with the above statement, there will ALWAYS be those who will refuse to believe or even claim that the discoveries were false. "Oh, some scientist must have forged the data" or "They just want to destroy religion" or "There was contamination".
To be fair, not all religions feel threatened by extraterrestial life. After all, the Catholic church is funding a (telescope?) project in conjunction with SETI -- so they can find aliens and then try to convert them to Catholocism. Terribly optimistic of those Catholics... a bizarre thought to think about them succeding.
Anyway, science and religion don't have to be at odds. In fact, they shouldn't be at odds -- religion and technology may often have a beef with each other, but science should just be seen as exploring God's creation.
Spam naturally seems to come in several categories - porn, penis enlargements, mortgages etc. However, it's unlikely that any one spam will simultaneously advertise porn and mortgages. Simply having a "spam" and a "not" category will not take advantage of distinctions such as that.
Why does it matter what category? To the user they don't care what kind of spam, merely that it's spam. And this isn't just a UI issue -- the filter is not meant to indicate authoritatively what is spam and what is not. Instead it learns what the particular user considers spam. You're only going to introduce inaccuracy if you create more categories, because the user is sometimes going to miscategorize spams (e.g., porn in penis enlargement). The user is not invested in the result of that subcategorization, so it's not a good goal for training.
Certainly there are other categorizations that are useful, e.g., work vs. private mail. Bayesian techniques can be used for further categorization, but they should only be used to categorize as far as the user cares to have their mail categorized.
Bayesian techniques for non-spam wouldn't be that useful, anyway, because non-statistical rules generally work well for everything but spam -- it's only because spammers are specifically trying to defeat non-statistical rules that we need statistical analysis. The only other place for Bayesian techniques, IMHO, is where the user can't articulate the basis of the categorization they desire (but that's probably quite common).
Most employers that do this are either jerks, greedy, or under extreme financial pressure and you have to ask yourself, "do you really want to work for them"?
The financial pressure is really important -- financial pressure turns otherwise petty and ethically loose businessmen into real assholes. It's a point when they are trying to anger as few people as possible, while also passing blame away from themselves. SCO is an obvious instance, but it happens on a small scale all the time. Especially when it comes to legal issues, this is when people get dangerous. This shitty part is it's hard to predict, and certainly control, what happens to your employer.
I estimate that eMusic is a great deal for someone who currently purchases 30 or more CD's a year. I say this because I'm assuming that for someone with varied musical taste, only about half of the albums they want will be on eMusic (though this could get better over time)
It's a lot worse than that. I'm betting it's something more like 5% of the time you'll find it on eMusic, or even less (though it depends a lot on your musical tastes). I'm not saying eMusic is bad, they have some cool stuff, but it's not a place you get the music you want. It's the place you find music you didn't know you wanted.
The exception might be that they have a lot of old stuff, probably stuff the label has decided doesn't have any other sale value. I understand people who love older jazz can find a lot of stuff they'll like on eMusic. The other stuff is lots of indy music, mostly from kind of small labels, or bands that are kind of niche (like They Might Be Giants -- but even TMBG is very popular compared to most eMusic artists).
It's cool, but you have to invest a lot of time to get something out of it. Not like time on P2P, trying to find something, but time to listen to new music and explore, so you can find new things you like.
Hmm... how do you install a new operating system on a computer without a CD-ROM/floppy? Can you boot off USB/Firewire CD-ROM drive? But I don't know anyone who has one of these, and I wouldn't want to buy such a drive just to install a new OS. Network install works well, but only after you've got the new installation started and networked.
Re:Lots of good info here...
on
I, Spammer
·
· Score: 1
If you create hazardous waste, you are liable for that waste even after it leaves your hands. If your waste shows up on the side of the road or thrown into a stream, you don't have to be the one who threw it.
This is of course so some guy with a pickup truck and nothing to lose can't come to your business and offer you a good deal on disposal.
Similar things happen with spam. The businesses don't sent out spam, they just hire someone to do unspecified marketing, or pay for referrals. I think in the case of junk faxing, businesses can be liable. The same thing should be true of spamming. The nice thing is that ultimately you can track down every business, since they have to do something to receive your money. The intermediaries may be harder to find, but that's okay (and once you start prosecuting the businesses they'll quickly fess up).
Now, with waste you can protect yourself by giving your business to somebody who is properly accredited, implying they follow all the regulations. Perhaps similar regulation could be applied to commercial email to someone with whom you do not have a previous business relationship. Regulations would involve proper labeling, no-email lists, accurate return addresses, etc. Probably not a very useful service, since the regulations would make spam easy to block, but eh... at that point it's not a free speech issue, since you are not making the speech illegal, you are just enabling email readers.
Of course, if SCO can say that they cannot be bound to the GPL, because they released their own code only due to fraud or misrepresentation, that seems to protect most of the distributions as well. If SCO is not under any copyright/licensing obligations due to their release of the code, no one else is either. Those distributions all have to stop releasing the code, and presumably everyone has to stop using the code, but they shouldn't be faulted for mistakenly using it in the past.
Whoever misrepresented that code and first violated the copyright on that code may be at fault -- IBM perhaps -- but everyone else should be fine. Any case SCO can bring against, say, Redhat, is the exact same case Linux developers can bring against SCO.
Of course, anyone who uses the code after the copyright violation has been revealed could be at fault. But I don't think anyone worries that the code will be hard to replace, even if they need to use clean room techniques. Potentially users who don't update their system could be at fault, but that's hardly a fruitful path for litigation.
It just depends on the area, there's no national standard for police phone numbers (beyond 911 itself). If the local police tell you to call 911 to contact them, then I would be likely to believe them and just call 911.
Of course the Russians have a money problem -- Russia is a third world country. I don't believe that Europe, Russia, the Soviet Union, or China have put anywhere close to as much money into space exploration as the United States.
Now, NASA arguably has accomplished much more than those other countries, so it's reasonable they'd need more money. But after thinking about it -- what has NASA accomplished? The shuttle, the space station, the Apollo missions... I don't think these accomplished anything. Doing things to prove you can is stupid, because there's too many useful things that need to be done.
That said, NASA has also done useful things -- Voyager, etc. But the manned program is a failure by design -- when a project has has no useful goal, it cannot succede. So how can you judge the efficiency?
The problem with putting tracks along highways is that highways are only part of what you need to replace. No one just drives on the highway, they drive on normal roads, then the highway, then back on normal roads. The highway area itself tends to be unpleasant and without many interesting destinations.
The only practical use for a train in the highway is when it gets off the highway, or goes to a very concentrated destination (like downtown), and when other end has parking. This describes the Blue Line in Chicago, but that's still pretty limited -- traffic to and from downtown is a rather small portion of traffic in the city, and park-and-ride is only a solution for commuters, not something that changes communities or significantly reduces our dependency on cars.
I just don't see much of a future in that sort of rail. BTW, The El' is considered "heavy" rail, like most subway systems, while light rail is something somewhat different. Strangely heavy rail involves lighter cars, so really the term "light rail" is just dishonest marketing.
I agree that cars scale well -- certainly they scale in ways that mass transit doesn't, like scaling to a diversity of destinations, scaling between high-capacity expensive infrastructure (highways), to low-capacity cheap infrastructure (down to the dirt road), scaling trip distances (5 minute drive to crosscountry). And the infrastructure even scales to a large capacity -- traditional rail can't compare.
In certain ways cars don't scale, but we already know those. It's bad for especially dense areas, like a downtown, where there's simply not room for all the cars, when they are driving or when they are parked. But I think there's a point that we're reaching where traffic in general isn't scaling, where increasing highway size doesn't pay off enough, and where the rest of the supporting road infrastructure can't keep up with the the highways anyway.
And while the roads are providing diminishing returns, they are also eating up the thing they are supposed to provide: space. Roads take a significant amount of space (I vaguely remember 25%, according to SimCity:) -- if you just keep building more roads, they are going to start taking a larger percentage of land. Then things have to be even more spread out just to get the same amount of land... that's the sort of scaling issue I'm concerned with.
And the outer suburbs just waste space for no reason. Why is there a strip of grass between the gas station and the Burger King? Does anyone enjoy that space? Why are the residential roads so wide? They don't lead anywhere so there's no point to driving quickly, no one parks on them... why not just two lanes? If people used their space thoughtfully, they could happily live in much closer quarters with no decrease in living standards. Alleys and sidewalks, those are space well spent, but in the developments most of the time they are too lazy to put in a proper curb, so the couple feet of land next to the road are a vague limbo. So much land is lost to a limbo -- useless, empty space. Old suburbs aren't so bad, but the new ones... ick.
In a lot of places the car system is really falling apart. Here in Chicago concerns with traffic keep me from taking car trips at certain times of the day, as even short trips (like 10 miles one way) can have an hour added because of traffic. I like to do my grocery shopping at midnight as a result (someplace where I can't use mass transit). Traffic is also crippling the bus system, which seems to consistently take twice as long as a car, and suffers from traffic just like any car driver does. Rush hour traffic, somewhat oddly, is equally as bad going to the suburbs as coming to the city. You can't win! And I won't even start on parking... (and this is ignoring energy consumption, aesthetics, pollution, high asthma rates, lung cancer from rubber in the air...)
It's not an abstract -- people realize the amount of time and money they are wasting, as well as the stress. Mass transit just isn't stepping up to meet the problem -- busses are a fundamentally crappy means of transportation, and the train system only works for a small set of trips, as both ends have to be close to a station, and transfers mean going out of your way and losing a lot of time. Plus it's expensive -- as an individual mass transit is fine, but in groups of three or more it quickly becomes impractical.
Chicago is a lot worse than most places -- in New York everyone is used to it, but we beat out most other cities. But still, everyone is feeling the pinch, it's getting rapidly worse everywhere. More roads aren't solving the problem -- there's a serious scaling issue going on here. To summarize: I don't think it would be hard to sell people on an alternative to cars.
The alternate system needs to be orthogonal to traffic -- light rail just compounds traffic with at-grade crossings, not to mention having a tiny capacity. Buses suffer from traffic, so they can't attract riders.
PRT seems by far to be the best option to me. But if not PRT, then at least monorail or other elevated systems. It's a big investment, but we were able to invest in these systems 50 or 100 years ago, we certainly ought to be able to afford them today. Nobody else is offering a real alternative.
My third post on the subject, so I'm probably being tedious, but PRT answers these scaling issues:
Small (3-person) elevated train cars mean small tracks, not a lot more expensive than road -- being elevated means you don't need much right-of-way.
Off-line loading means relatively high capacity, again like a road (and unlike light rail or other mass transit) -- moving lots of low-capacity vehicles can carry more people than a small number of high-capacity vehicles, and unlike on a subway you only stop for your destination, not other people's.
Cars that go directly to your destination means no time waiting to make transfers.
Automatic interchanges (that look like a highway cloverleaf, as opposed to traffic signals) mean that dense and decentralized networks won't slow the system down. One-way tracks that make relatively tight loops mean simpler tracks and interchanges, while making it possible to cover an area more densely.
And again because entering and exiting the vehicle occurs offline (i.e., not on the main track), frequent stations won't slow the system down (as they would with mass transit).
Roads stop scaling well at a certain point. Yes, highways are a very efficient way to move people around. But the more lanes you add, the less payoff you get, and that's where a lot of cities are right now. Adding another lane to some highway might only buy them another year at the current growth rate, but it takes two years to build... it's a unsustainable situation.
Highways also require feeders -- large capacity requires a large system of feeder roads (the first and last few miles your trip). These roads have become monstrous, and sources of congestion in their own right.
So, the alternative is to build more highways instead of expanding the current ones. However, that is incredibly expensive, especially because those highways go through developed land, creating social and economic upheaval.
The car system is falling apart. And it's people who want their big back yards that are at the source of it -- people who are too stupid to realize that those distant suburbs are a bad deal. The houses are pieces of crap, built for people who value size above quality. The yards are grassy emptiness, ignored by the developers, and much of the time ignored even by the homeowners (plant some freakin' trees already!) The communities are socially barren. The schools are crappy and overcrowded, often made worse by the cheap landowners who sacrifice their children's education so they can keep their taxes down to buy a bigger TV. The irony of this all is that you people are fucking yourselves more than the rest of us! The suburbs are built for self-centered people... I don't know why people won't realize that being self-centered doesn't pay off in the long run.
P.S., the rest of New Hampshire hates you people:)
PRT! PRT could actually appeal to current car users, unlike most mass transit options that are suggested. Most mass transit ideas are horribly failed -- light rail, for instance, has failed over and over to be a practical option, and yet people keep proposing it. PRT, because it is an individual, decentralized form of transit, can actually adapt to the cities that we currently have (which were built around cars), inside of using transit as a way to punish people into creating the cities we fantasize about.
Of course, transit is only part of the problem (though it's a really large part of the problem). But PRT still offers lots of advantages -- safer streets (less cars), more intimate neighborhoods (because it takes less space), and it would get people to at least walk short distances, which is still a whole lot better than attached garages.
PRT is a starting point that doesn't depend on idealism -- there are real practical benefits, so the mass of people who don't follow their ideals (sad, but true) can still come on board.
Without the cars, my little sister will be much safer as she walks that 3/4 mile in the dark from the subway to her house, instead of driving directly to her door.
I think she very well may be more safe. Cars, and the general use of public space that they encourage (i.e., no pedestrians) encourage crime. Your sister will be much safer in an environment where the streets are not abandoned, occupied only by cars and fearful people trying to get home.
Having people in our public spaces makes those spaces much more safe. It's idiotic how people have lobbied to have public phones and benches removed, because they encourage loitering and make the community less safe. That's bull. Loitering makes a community more safe. It's things like cars that take people out of public spaces and make a community less safe.
I think you underestimate the render extensions. Why generate Postscript, which is interpreted and in turn writes to a canvas, when you can write to the canvas directly?
Resolution independence is certainly a worthy goal. But I don't see why Postscript (or PDF) has to be part of that. There are already higher-level drawing interfaces (in GTK, Qt, GNUStep, etc), so you wouldn't be talking Postscript directly anyway, even with a DPS backend. This is a much more solid abstraction layer, IMHO, than the heavy layers that DPS implies.
X11 doesn't just send pixels (unless there's really just pixels to be send -- as with a bitmap image). Applications speak Xlib (well, usually toolkits speak Xlib), and it implements low-level drawing (lines and such).
The result isn't widgets on the client side -- though systems like that exist (e.g., NeWS, Berlin, PicoGUI). But it's higher-level than pixels (unlike, say, VNC).
Interestingly, it seems like GNUStep has given up on DPS (well, Display Ghostscript), even though they'd long wanted to use it (to better clone NeXT). It was a hold-up for a long time, and someone wrote an Xlib backend and I think it just worked. I'm sure they could take advantage of the RENDER extension, but I don't really see the advantage of a whole Postscript interpreter under there.
Good article. Particularly disturbing is the fact that these machines are being produced by strongly partisan companies. When they say "trust us", we should at least ask who are we trusting? and are they impartial?.
Some of these manufacturers are clearly suspicious. They need to prove themselves. That they refuse to do so should make us assume the worst.
I think it should be added that Linux is particularly suited to the desktop in an institutional setting (as would be found in a government office). In that situation there's typically some support staff, the systems and installations are handled by that staff (so they don't need to be accessible by novices). The required functions are also very specific -- if Linux can satisfy the specific requirements, it doesn't matter if you can't play games, use MS Office, or whatever.
It's also a situation where training can occur specific to their system, and because decisions often come from above, you can overcome some of the inertia of people's comfort with Windows.
Slashdot Mirror
But maybe that's just a sign that a fork is the proper response -- Zynot's embedded perspective just doesn't fit with Gentoo. It has requirements that just don't make sense for the rest of the community. It sounds like Zachary wanted to professionalize Gentoo, but that's just not interesting to a volunteer developer base.
In the original article, and in some of the defenses of layers that follow in this thread, there's a notion the problem is how much time a program takes to develop. Writing it the first time isn't the hardest part, in my experience, it's maintaining it.
Layers can work well if each layer is robust, and its role clearly delineated and described. When that is not the case, the system is conceptually heavy and pieces cannot be understood or modified in isolation. One common problem is that a bug in a lower level will propogate to a higher level, but in a way that is mysterious and difficult to debug. Or different layers, when first put together, will interact in unpredictable ways. Everything should be predictable -- that's the most important quality I see in a library.
Some things work great. I certainly don't mind layering my network code on a socket library. Of course, everyone has been over that for a couple decades -- as time goes on the foundation expands. The safety of layering improves, but you have to identify the underlying layers that can really be trusted.
Anyway, I respect PETA for advocating incrementally better treatment even if many of them would prefer that the animals not be eaten at all.
Anyway, science and religion don't have to be at odds. In fact, they shouldn't be at odds -- religion and technology may often have a beef with each other, but science should just be seen as exploring God's creation.
Certainly there are other categorizations that are useful, e.g., work vs. private mail. Bayesian techniques can be used for further categorization, but they should only be used to categorize as far as the user cares to have their mail categorized.
Bayesian techniques for non-spam wouldn't be that useful, anyway, because non-statistical rules generally work well for everything but spam -- it's only because spammers are specifically trying to defeat non-statistical rules that we need statistical analysis. The only other place for Bayesian techniques, IMHO, is where the user can't articulate the basis of the categorization they desire (but that's probably quite common).
The exception might be that they have a lot of old stuff, probably stuff the label has decided doesn't have any other sale value. I understand people who love older jazz can find a lot of stuff they'll like on eMusic. The other stuff is lots of indy music, mostly from kind of small labels, or bands that are kind of niche (like They Might Be Giants -- but even TMBG is very popular compared to most eMusic artists).
It's cool, but you have to invest a lot of time to get something out of it. Not like time on P2P, trying to find something, but time to listen to new music and explore, so you can find new things you like.
Hmm... how do you install a new operating system on a computer without a CD-ROM/floppy? Can you boot off USB/Firewire CD-ROM drive? But I don't know anyone who has one of these, and I wouldn't want to buy such a drive just to install a new OS. Network install works well, but only after you've got the new installation started and networked.
This is of course so some guy with a pickup truck and nothing to lose can't come to your business and offer you a good deal on disposal.
Similar things happen with spam. The businesses don't sent out spam, they just hire someone to do unspecified marketing, or pay for referrals. I think in the case of junk faxing, businesses can be liable. The same thing should be true of spamming. The nice thing is that ultimately you can track down every business, since they have to do something to receive your money. The intermediaries may be harder to find, but that's okay (and once you start prosecuting the businesses they'll quickly fess up).
Now, with waste you can protect yourself by giving your business to somebody who is properly accredited, implying they follow all the regulations. Perhaps similar regulation could be applied to commercial email to someone with whom you do not have a previous business relationship. Regulations would involve proper labeling, no-email lists, accurate return addresses, etc. Probably not a very useful service, since the regulations would make spam easy to block, but eh... at that point it's not a free speech issue, since you are not making the speech illegal, you are just enabling email readers.
Whoever misrepresented that code and first violated the copyright on that code may be at fault -- IBM perhaps -- but everyone else should be fine. Any case SCO can bring against, say, Redhat, is the exact same case Linux developers can bring against SCO.
Of course, anyone who uses the code after the copyright violation has been revealed could be at fault. But I don't think anyone worries that the code will be hard to replace, even if they need to use clean room techniques. Potentially users who don't update their system could be at fault, but that's hardly a fruitful path for litigation.
It just depends on the area, there's no national standard for police phone numbers (beyond 911 itself). If the local police tell you to call 911 to contact them, then I would be likely to believe them and just call 911.
Taxi 2000 is another implementation of PRT in the US. They've been making some real progress recently.
Now, NASA arguably has accomplished much more than those other countries, so it's reasonable they'd need more money. But after thinking about it -- what has NASA accomplished? The shuttle, the space station, the Apollo missions... I don't think these accomplished anything. Doing things to prove you can is stupid, because there's too many useful things that need to be done.
That said, NASA has also done useful things -- Voyager, etc. But the manned program is a failure by design -- when a project has has no useful goal, it cannot succede. So how can you judge the efficiency?
The only practical use for a train in the highway is when it gets off the highway, or goes to a very concentrated destination (like downtown), and when other end has parking. This describes the Blue Line in Chicago, but that's still pretty limited -- traffic to and from downtown is a rather small portion of traffic in the city, and park-and-ride is only a solution for commuters, not something that changes communities or significantly reduces our dependency on cars.
I just don't see much of a future in that sort of rail. BTW, The El' is considered "heavy" rail, like most subway systems, while light rail is something somewhat different. Strangely heavy rail involves lighter cars, so really the term "light rail" is just dishonest marketing.
In certain ways cars don't scale, but we already know those. It's bad for especially dense areas, like a downtown, where there's simply not room for all the cars, when they are driving or when they are parked. But I think there's a point that we're reaching where traffic in general isn't scaling, where increasing highway size doesn't pay off enough, and where the rest of the supporting road infrastructure can't keep up with the the highways anyway.
And while the roads are providing diminishing returns, they are also eating up the thing they are supposed to provide: space. Roads take a significant amount of space (I vaguely remember 25%, according to SimCity :) -- if you just keep building more roads, they are going to start taking a larger percentage of land. Then things have to be even more spread out just to get the same amount of land... that's the sort of scaling issue I'm concerned with.
And the outer suburbs just waste space for no reason. Why is there a strip of grass between the gas station and the Burger King? Does anyone enjoy that space? Why are the residential roads so wide? They don't lead anywhere so there's no point to driving quickly, no one parks on them... why not just two lanes? If people used their space thoughtfully, they could happily live in much closer quarters with no decrease in living standards. Alleys and sidewalks, those are space well spent, but in the developments most of the time they are too lazy to put in a proper curb, so the couple feet of land next to the road are a vague limbo. So much land is lost to a limbo -- useless, empty space. Old suburbs aren't so bad, but the new ones... ick.
It's not an abstract -- people realize the amount of time and money they are wasting, as well as the stress. Mass transit just isn't stepping up to meet the problem -- busses are a fundamentally crappy means of transportation, and the train system only works for a small set of trips, as both ends have to be close to a station, and transfers mean going out of your way and losing a lot of time. Plus it's expensive -- as an individual mass transit is fine, but in groups of three or more it quickly becomes impractical.
Chicago is a lot worse than most places -- in New York everyone is used to it, but we beat out most other cities. But still, everyone is feeling the pinch, it's getting rapidly worse everywhere. More roads aren't solving the problem -- there's a serious scaling issue going on here. To summarize: I don't think it would be hard to sell people on an alternative to cars.
The alternate system needs to be orthogonal to traffic -- light rail just compounds traffic with at-grade crossings, not to mention having a tiny capacity. Buses suffer from traffic, so they can't attract riders.
PRT seems by far to be the best option to me. But if not PRT, then at least monorail or other elevated systems. It's a big investment, but we were able to invest in these systems 50 or 100 years ago, we certainly ought to be able to afford them today. Nobody else is offering a real alternative.
Highways also require feeders -- large capacity requires a large system of feeder roads (the first and last few miles your trip). These roads have become monstrous, and sources of congestion in their own right.
So, the alternative is to build more highways instead of expanding the current ones. However, that is incredibly expensive, especially because those highways go through developed land, creating social and economic upheaval.
The car system is falling apart. And it's people who want their big back yards that are at the source of it -- people who are too stupid to realize that those distant suburbs are a bad deal. The houses are pieces of crap, built for people who value size above quality. The yards are grassy emptiness, ignored by the developers, and much of the time ignored even by the homeowners (plant some freakin' trees already!) The communities are socially barren. The schools are crappy and overcrowded, often made worse by the cheap landowners who sacrifice their children's education so they can keep their taxes down to buy a bigger TV. The irony of this all is that you people are fucking yourselves more than the rest of us! The suburbs are built for self-centered people... I don't know why people won't realize that being self-centered doesn't pay off in the long run.
P.S., the rest of New Hampshire hates you people :)
Of course, transit is only part of the problem (though it's a really large part of the problem). But PRT still offers lots of advantages -- safer streets (less cars), more intimate neighborhoods (because it takes less space), and it would get people to at least walk short distances, which is still a whole lot better than attached garages.
PRT is a starting point that doesn't depend on idealism -- there are real practical benefits, so the mass of people who don't follow their ideals (sad, but true) can still come on board.
Having people in our public spaces makes those spaces much more safe. It's idiotic how people have lobbied to have public phones and benches removed, because they encourage loitering and make the community less safe. That's bull. Loitering makes a community more safe. It's things like cars that take people out of public spaces and make a community less safe.
Resolution independence is certainly a worthy goal. But I don't see why Postscript (or PDF) has to be part of that. There are already higher-level drawing interfaces (in GTK, Qt, GNUStep, etc), so you wouldn't be talking Postscript directly anyway, even with a DPS backend. This is a much more solid abstraction layer, IMHO, than the heavy layers that DPS implies.
The result isn't widgets on the client side -- though systems like that exist (e.g., NeWS, Berlin, PicoGUI). But it's higher-level than pixels (unlike, say, VNC).
Interestingly, it seems like GNUStep has given up on DPS (well, Display Ghostscript), even though they'd long wanted to use it (to better clone NeXT). It was a hold-up for a long time, and someone wrote an Xlib backend and I think it just worked. I'm sure they could take advantage of the RENDER extension, but I don't really see the advantage of a whole Postscript interpreter under there.
Some of these manufacturers are clearly suspicious. They need to prove themselves. That they refuse to do so should make us assume the worst.
It's also a situation where training can occur specific to their system, and because decisions often come from above, you can overcome some of the inertia of people's comfort with Windows.