Besides, getting a culture started requires more than intelligence, it requires hands.
While I was with you for most of what you said, that sentence seems a bit off. Tool-making requires hands (or other appendage(s) capable of fine manipulation), but culture requires some level of communication. Exactly how much is an open scientific question; if someone manages to nail that down, they'll become very famous indeed.
As evidence, an important part of human culture for thousands of years was bardic tradition, the reciting of poems and songs created by former generations. This all dated from well before anyone knew how to write — Homer, author of the Iliad and the Odyssey, lived in a non-literate era, yet it was assuredly an age of culture, and it forms part of our culture today. Were hands required for its creation and maintenance? It's hard to see how that can be; the voice though, that's vital. Animals definitely can make noises, and we just don't know yet how semantically-sophisticated they can be.
So do some animals have some sort of culture? Probably. Is it as sophisticated as ours? Probably not, but that might be just a stupid parochialism on my part; we won't know until we figure out how to look properly.:-)
That's assuming you've overcome the software limitation of slow processing that plagues most touch keyboards.
Having seen the speed at which some teens use touch-screen phones, I'd guess that that limitation is at least somewhat overcome in some configurations. Whether that would be good enough for a full keyboard-equivalent, I don't know. I doubt it would be great for an FPS though...
Maybe I'm missing something, but why does PHP have its own version of strtod()? It's a standard C99 function, so you'll find it in libc or equivalent in any C99-compliant platform (including Windows) and more effort has probably gone into optimising that version than the PHP version, although if you're converting from strings to floating point values anywhere performance critical then you're probably Doing It Wrong.
A number of languages have their own versions of strtod(), and it's because the system version is often deeply sub-optimal. (Going the other direction is even worse, especially if you want minimal representations of numbers.) Typically, system library authors stop once they've got something that works well enough, whereas other clients of that sort of functionality are much more exposed to it and so care much more about it being optimal. (After all, system library authors tend to think that people don't convert back and forth a lot. Mostly they don't, but since when did that cover 100% of cases?)
Which isn't to say that if you're is reimplementing a system library function, you can get out of testing it very thoroughly. Oh no, not at all...
1 single point of failure, outside of your control.
You can host your own OpenID server, if you wish, and use that with any site that allows the use of OpenID for logging in (assuming they aren't 100% lame and restrict the server URLs; if they do that, don't use the site). If that doesn't constitute control, I don't know what does. Since the authentication point is a (particular kind of) website, you can use the usual methods for replicating it.
Of course, for these things you'll pay (whether in money, time or effort) but I'm assuming that you believe that it is worth it. It's your choice.
nahh, the problem is a misunderstanding of statistics (thinking that post-hoc analysis with this fishing for statistical significance) is as valid as proper hypothesis testing. The proper way is where the hypothesis is fully pre-formed and then tested. The numbers and statistics apply ONLY TO THE HYPOTHESIS being tested, so you cannot hunt for a statistical significance just somewhere in the data and then re-formulate your hypothesis.
The problem is that there are a lot of fields (e.g., astronomy, economics) where it is not possible to conduct proper lab experiments. That means you've got to just collect all the data that you can and try to work with it. The best way to do that is to partition the data and use part of it to search for candidate hypotheses, and the rest (possibly with additional partitions) to check those hypotheses, and yet it's never entirely certain that enough data is present in either set for correct conclusions to be drawn. It's challenging statistically (and part of why I prefer to write programs instead).
Further, then we have to find another mechanism to distinguish between immediate commands and commands to store to create a program or else explain the REPL concept early.
I don't know that you have to explain it. What is important is giving them something which will do what they tell it to immediately, i.e., a REPL or equivalent. A few obvious pitfalls that they'll have to navigate their way around aren't a bad thing either, so long as there's no lasting damage from getting it wrong. It's also good to minimize the gap between what they type immediately and what they do in a saved program.
Linux distributions solve this by using a central package manager.
FTFY. (The kernel doesn't care; the kernel developers don't care much more than the kernel. Sensible folks!)
For an application "vendor" (using the term loosely to mean everyone who is upstream of multiple distribution vendors) working with the distribution vendors is usually a PITA. For example, they all have different policies for how to get updates in, and they all have unhelpful models for reporting issues upstream. (The exceptions tend to be certain very large OSS systems, but that's usually because some vendor employs someone to handle the liaison as part of their job in those cases.) Not to put too fine a point on it, relying on the central package manager for something where you're doing a complicated networked application (e.g., Skype) where protocol versions need to be properly synched across many platforms, well... it's rather less than ideal.
This will just lead to more state parties moving to closed primaries. This means independents, most Americans, will have even less say in who our leaders are.
A political party is a private association of people who agree to nominate people (typically members of their association) to particular public positions that are filled by public vote. Why should anyone not a member of a private association have any right to control the decision making process of that association (as opposed to setting the general framework - the legal system - within which that association operates, of course)? While it might be advisable for the party to listen to the opinion of the ordinary man in the street when selecting a candidate, it's still a private organization that's ultimately doing the nominating; if their general principles lead them to nominate the unelectable, they have that right. It's wrong to do otherwise.
If you're living somewhere where a private organization like that has such a total grasp on power that you feel that it is reasonable for the public to have such a say in their decision making, you're living in an effective one party state. Dress it up how you like, but that's one of your big problems right there.
This feature will flop in florida et al. No engine, no a/c.
Bet whether the engine is going will depend on the load on the overall system. If the A/C is on full, the load will be enough to justify the engine being on (well, at least a lot of the time; that'll depend on a bunch of other factors such as how well insulated the passenger compartment is and what size of battery is fitted).
Sorry, but that has never been a "rule of science".
Theories stand or fall on their own merit. Science does not care about sources.
That's a nice "theory" (in a different sense to how you've used the word) but as it's done in practice, that's not at all how science works. Hypotheses from people inside the field tend to get totally different profiles of scrutiny to those from people outside. In particular, proposals from outsiders tend to be checked for howlers — things that have been proved false ages ago — first, whereas proposals from insiders get checked for subtle things first as it's assumed that their proponents know to avoid the obvious errors. An insider's proto-theory will much more rapidly attract the attention of some lowly PhD student for reproducing the experiments and calculations that lead to the conclusion, although before a proposal from any source can truly be an Accepted Theory, they've got to be fully checked anyway, which is the truth of the pudding. But the lifeblood of science is the process of turning "that's odd" into an accepted theory (together with checking if the current accepted theories are sufficiently consistent with reality).
Note that I'm saying "outsider" and "insider" deliberately, and I wish to distinguish these terms from "pro" and "amateur". Professionals can still be outsiders, and amateurs can be insiders. The usual mark of an insider is that they know what they're doing, what other insiders are doing, and what people have done before. Outsiders are much more likely to skip straight to the part where they propose a perpetual motion machine or something else equally idiotic.
All modern supercomputers are massively clustered, using various shared memory architectures.
Actually, they do very little memory sharing because it doesn't scale at all. Shared memory systems top out at on the order of 1k cores, after which the memory backbone becomes just too damn expensive, even by supercomputer standards. Instead, supercomputers use message passing (especially various MPI implementations) over what is still very fast dedicated interconnect. Algorithms have to be very carefully written to take good advantage of that sort of system. (Some will actually have a mix of technologies, being essentially a cluster of smaller shared-memory machines, with the programming model being mainly OpenMP within each smaller unit, and MPI between them. But that's getting a bit esoteric and difficult to develop for.)
To my mind, it makes much sense (where possible) to split the program up into smaller pieces that can run pretty much independently on commodity hardware. That fits an awful lot of scientific and engineering problems, and scaling up to full datasets is just a matter of using more nodes. I've been doing such things successfully (using both clusters and the cloud) for some time now. The systems we use don't enter into the top500 but we've got access to lots of them, and can process full parameter sweeps and datasets quite fast enough to be really useful.
An optional nofollow would be allowed, but no other attributes.
How do you know all the attributes which will make sense in the future? I know I don't, and that's why I think your rule is foolish and over-prescriptive. Instead, the determining factor should be one of intention; is the linker being scummy by making the link? That's not a mechanical decision, but rather a human one, and it's the kind of thing that courts can handle. Remember, the law is primarily about people and their intentions and actions; it's not a programming language or computer. Whether a link is good or bad is primarily about a human distinction; computers don't make such value judgements (though they can help enforce them).
Your single-line-queue logic requires all transaction times to be roughly the same to work, I believe.
Queueing theory addressed this donkey's years ago. The single-line multiple-servers approach works particularly well when there is a substantial amount of variance in the time to service a particular entry in the queue. It can gum up if every server is loaded with something slow, but that's always a possibility, and by having a single queue, once any slow entry is done, the whole queue will move forward fast again. That's a big win. (Also, if this is happening frequently then there's not enough servers given the nature of the workload.)
If everyone takes about the same amount of time to serve though, there's much less of a need to have a single queue; the gains from doing so are minimal. Multiple queues with load leveling on entry to the queues will work just as well. It's also a relatively rare situation; it doesn't happen in most stores (though there the determining factor is often not to do with queue length; there's often not enough room for a single queue) where there's multiple servers in the first place.
Ultimately, I think stores don't want customers to leave quickly.
That depends on how long the queue is. Customers really don't like queueing, and while they'll put up with a bit of it, if it gets too much (which is largely determined by the time they're waiting in relation to the reward they feel they receive for it) then they'll start to avoid the store in future if they can. The store does not want to lose future custom, and so has to invest in increasing the number of servers (since the minimal immediate-cost option to the store is a single server).
This is an area that's been studied extensively, both in queueing theory and in retail management.
Governments prefer big corporations. One or two big corporations are much easier to control than a lot of small companies (some of which the government might not even be aware of).
On the other hand, small corporations can't cause nearly so much trouble for governments as large ones so are preferable to government in other respects. What changes isn't what the effects of scale actually are, but what value is attached to particular parts of those effects; that's what drives the cycles.
It's still ongoing right now, albeit intermittently. I'm seeing drop-outs on the distribution of my skype status (despite my local 'net connection being fine).
Plain X forwarding is fine on a LAN, but it's not really usable over ADSL.
Only because you insist on having masses of eye candy. Forego that, and you can run just fine with ADSL. Heck, I've done real work with remote X running over a 14k4 modem (ick!) and that worked (by "real work" I mean Framemaker and xfig; everything else that I needed at the time was just fine locally or in a plain terminal). It's only for eye candy (or photo editing, though I've never needed that professionally) that you really want to have a high-bandwidth low-latency connection between application and graphics display.
Ah, so you're one of those "there are people in California, and people in a couple of cities going down the eastern coastline, and nothing else counts" sorts, huh?
In population terms, yes. There's no excuse for urban populations having crap broadband, and there's lots of people in cities and towns in the US. If you're out in the boonies, it's going to impact on your speed (or costs) but that's true all over the world. But more to the point, just look at where the majority of people are, in urban and suburban areas. Is there any reason why it's impossible for such a large fraction of them to get broadband? (Well, yes there is, and it's got to do with lack of real competition between providers. Regulatory fail.)
His statement is pretty spot on -- there are some pretty wide swaths in this country where you've either got low population density or geographical problems making it difficult. Look at Appalachia as a whole, for example -- a good chunk of it is "difficult" geographically, and having a significant percentage of the populace nestled in mountain hollows doesn't help.
Because cables can't go down into mountain hollows... (Or did you think that the rest of the world does broadband always by wireless telecoms?)
Mass Effect 2 was so awesome I actually went out and bought a genuine copy of it after I beat it. I think I've got at least 200 hours logged on that game as many times as I've replayed it.
That's a game that specializes in making you feel like James Bond in space, where you are totally the Hero out to save the galaxy even if they don't think they need it. That sort of experience can only be delivered as a single-person game; you just can't have 100k players all simultaneously being the most awesome person in the universe.
What this does emphasize is that the big difference between single and multi-player games will be between story-driven and interaction-driven gaming. There's plenty of room for both.
Or expensive. Some niches are defined by having extremely high quality requirements on entry, which make it hard for non-commercial software to get started precisely because none of the users will consider any option that isn't very good indeed. Those tend to be pretty conservative niches, but for some things (e.g., anything safety-critical) that's good; it's reasonable to spend on having things dependable.
Incidentally, there's a fine Russian word for a hierarchical system of representation, in which smaller governmental bodies choose representatives to the national government: "Soviet." Yeah, that sure helped protect the liberties of the people and the long-term interests of the republic, didn't it?
Null argument. The official name of East Germany was (after translation) The German Democratic Republic. Does that mean that there's a problem with democracy? Or republics? Or just that names chosen for propaganda reasons are bunk and what matters is what happens on the ground? Hmm...
Slashdot Mirror
Rabbits and hamsters 'have distinct personalities' and 'can think about the future'. But that doesn't mean I'd give them the vote and the car keys.
I think they'd have trouble pressing the gas pedal anyway.
Besides, getting a culture started requires more than intelligence, it requires hands.
While I was with you for most of what you said, that sentence seems a bit off. Tool-making requires hands (or other appendage(s) capable of fine manipulation), but culture requires some level of communication. Exactly how much is an open scientific question; if someone manages to nail that down, they'll become very famous indeed.
As evidence, an important part of human culture for thousands of years was bardic tradition, the reciting of poems and songs created by former generations. This all dated from well before anyone knew how to write — Homer, author of the Iliad and the Odyssey, lived in a non-literate era, yet it was assuredly an age of culture, and it forms part of our culture today. Were hands required for its creation and maintenance? It's hard to see how that can be; the voice though, that's vital. Animals definitely can make noises, and we just don't know yet how semantically-sophisticated they can be.
So do some animals have some sort of culture? Probably. Is it as sophisticated as ours? Probably not, but that might be just a stupid parochialism on my part; we won't know until we figure out how to look properly. :-)
That's assuming you've overcome the software limitation of slow processing that plagues most touch keyboards.
Having seen the speed at which some teens use touch-screen phones, I'd guess that that limitation is at least somewhat overcome in some configurations. Whether that would be good enough for a full keyboard-equivalent, I don't know. I doubt it would be great for an FPS though...
And I will only let go of it when they pry it from my cold dead hands!
As a bonus, you're also prepared for the zombie apocalypse.
Maybe I'm missing something, but why does PHP have its own version of strtod()? It's a standard C99 function, so you'll find it in libc or equivalent in any C99-compliant platform (including Windows) and more effort has probably gone into optimising that version than the PHP version, although if you're converting from strings to floating point values anywhere performance critical then you're probably Doing It Wrong.
A number of languages have their own versions of strtod(), and it's because the system version is often deeply sub-optimal. (Going the other direction is even worse, especially if you want minimal representations of numbers.) Typically, system library authors stop once they've got something that works well enough, whereas other clients of that sort of functionality are much more exposed to it and so care much more about it being optimal. (After all, system library authors tend to think that people don't convert back and forth a lot. Mostly they don't, but since when did that cover 100% of cases?)
Which isn't to say that if you're is reimplementing a system library function, you can get out of testing it very thoroughly. Oh no, not at all...
1 single point of failure, outside of your control.
You can host your own OpenID server, if you wish, and use that with any site that allows the use of OpenID for logging in (assuming they aren't 100% lame and restrict the server URLs; if they do that, don't use the site). If that doesn't constitute control, I don't know what does. Since the authentication point is a (particular kind of) website, you can use the usual methods for replicating it.
Of course, for these things you'll pay (whether in money, time or effort) but I'm assuming that you believe that it is worth it. It's your choice.
nahh, the problem is a misunderstanding of statistics (thinking that post-hoc analysis with this fishing for statistical significance) is as valid as proper hypothesis testing. The proper way is where the hypothesis is fully pre-formed and then tested. The numbers and statistics apply ONLY TO THE HYPOTHESIS being tested, so you cannot hunt for a statistical significance just somewhere in the data and then re-formulate your hypothesis.
The problem is that there are a lot of fields (e.g., astronomy, economics) where it is not possible to conduct proper lab experiments. That means you've got to just collect all the data that you can and try to work with it. The best way to do that is to partition the data and use part of it to search for candidate hypotheses, and the rest (possibly with additional partitions) to check those hypotheses, and yet it's never entirely certain that enough data is present in either set for correct conclusions to be drawn. It's challenging statistically (and part of why I prefer to write programs instead).
Further, then we have to find another mechanism to distinguish between immediate commands and commands to store to create a program or else explain the REPL concept early.
I don't know that you have to explain it. What is important is giving them something which will do what they tell it to immediately, i.e., a REPL or equivalent. A few obvious pitfalls that they'll have to navigate their way around aren't a bad thing either, so long as there's no lasting damage from getting it wrong. It's also good to minimize the gap between what they type immediately and what they do in a saved program.
Linux distributions solve this by using a central package manager.
FTFY. (The kernel doesn't care; the kernel developers don't care much more than the kernel. Sensible folks!)
For an application "vendor" (using the term loosely to mean everyone who is upstream of multiple distribution vendors) working with the distribution vendors is usually a PITA. For example, they all have different policies for how to get updates in, and they all have unhelpful models for reporting issues upstream. (The exceptions tend to be certain very large OSS systems, but that's usually because some vendor employs someone to handle the liaison as part of their job in those cases.) Not to put too fine a point on it, relying on the central package manager for something where you're doing a complicated networked application (e.g., Skype) where protocol versions need to be properly synched across many platforms, well... it's rather less than ideal.
This will just lead to more state parties moving to closed primaries. This means independents, most Americans, will have even less say in who our leaders are.
A political party is a private association of people who agree to nominate people (typically members of their association) to particular public positions that are filled by public vote. Why should anyone not a member of a private association have any right to control the decision making process of that association (as opposed to setting the general framework - the legal system - within which that association operates, of course)? While it might be advisable for the party to listen to the opinion of the ordinary man in the street when selecting a candidate, it's still a private organization that's ultimately doing the nominating; if their general principles lead them to nominate the unelectable, they have that right. It's wrong to do otherwise.
If you're living somewhere where a private organization like that has such a total grasp on power that you feel that it is reasonable for the public to have such a say in their decision making, you're living in an effective one party state. Dress it up how you like, but that's one of your big problems right there.
6) Stop getting treatment
7) ???
8) Die.
The "???" is "Decline slowly in chronic pain and poverty while mean assholes say it's your fault for not being a millionaire."
This feature will flop in florida et al. No engine, no a/c.
Bet whether the engine is going will depend on the load on the overall system. If the A/C is on full, the load will be enough to justify the engine being on (well, at least a lot of the time; that'll depend on a bunch of other factors such as how well insulated the passenger compartment is and what size of battery is fitted).
Sorry, but that has never been a "rule of science".
Theories stand or fall on their own merit. Science does not care about sources.
That's a nice "theory" (in a different sense to how you've used the word) but as it's done in practice, that's not at all how science works. Hypotheses from people inside the field tend to get totally different profiles of scrutiny to those from people outside. In particular, proposals from outsiders tend to be checked for howlers — things that have been proved false ages ago — first, whereas proposals from insiders get checked for subtle things first as it's assumed that their proponents know to avoid the obvious errors. An insider's proto-theory will much more rapidly attract the attention of some lowly PhD student for reproducing the experiments and calculations that lead to the conclusion, although before a proposal from any source can truly be an Accepted Theory, they've got to be fully checked anyway, which is the truth of the pudding. But the lifeblood of science is the process of turning "that's odd" into an accepted theory (together with checking if the current accepted theories are sufficiently consistent with reality).
Note that I'm saying "outsider" and "insider" deliberately, and I wish to distinguish these terms from "pro" and "amateur". Professionals can still be outsiders, and amateurs can be insiders. The usual mark of an insider is that they know what they're doing, what other insiders are doing, and what people have done before. Outsiders are much more likely to skip straight to the part where they propose a perpetual motion machine or something else equally idiotic.
All modern supercomputers are massively clustered, using various shared memory architectures.
Actually, they do very little memory sharing because it doesn't scale at all. Shared memory systems top out at on the order of 1k cores, after which the memory backbone becomes just too damn expensive, even by supercomputer standards. Instead, supercomputers use message passing (especially various MPI implementations) over what is still very fast dedicated interconnect. Algorithms have to be very carefully written to take good advantage of that sort of system. (Some will actually have a mix of technologies, being essentially a cluster of smaller shared-memory machines, with the programming model being mainly OpenMP within each smaller unit, and MPI between them. But that's getting a bit esoteric and difficult to develop for.)
To my mind, it makes much sense (where possible) to split the program up into smaller pieces that can run pretty much independently on commodity hardware. That fits an awful lot of scientific and engineering problems, and scaling up to full datasets is just a matter of using more nodes. I've been doing such things successfully (using both clusters and the cloud) for some time now. The systems we use don't enter into the top500 but we've got access to lots of them, and can process full parameter sweeps and datasets quite fast enough to be really useful.
An optional nofollow would be allowed, but no other attributes.
How do you know all the attributes which will make sense in the future? I know I don't, and that's why I think your rule is foolish and over-prescriptive. Instead, the determining factor should be one of intention; is the linker being scummy by making the link? That's not a mechanical decision, but rather a human one, and it's the kind of thing that courts can handle. Remember, the law is primarily about people and their intentions and actions; it's not a programming language or computer. Whether a link is good or bad is primarily about a human distinction; computers don't make such value judgements (though they can help enforce them).
Your single-line-queue logic requires all transaction times to be roughly the same to work, I believe.
Queueing theory addressed this donkey's years ago. The single-line multiple-servers approach works particularly well when there is a substantial amount of variance in the time to service a particular entry in the queue. It can gum up if every server is loaded with something slow, but that's always a possibility, and by having a single queue, once any slow entry is done, the whole queue will move forward fast again. That's a big win. (Also, if this is happening frequently then there's not enough servers given the nature of the workload.)
If everyone takes about the same amount of time to serve though, there's much less of a need to have a single queue; the gains from doing so are minimal. Multiple queues with load leveling on entry to the queues will work just as well. It's also a relatively rare situation; it doesn't happen in most stores (though there the determining factor is often not to do with queue length; there's often not enough room for a single queue) where there's multiple servers in the first place.
Ultimately, I think stores don't want customers to leave quickly.
That depends on how long the queue is. Customers really don't like queueing, and while they'll put up with a bit of it, if it gets too much (which is largely determined by the time they're waiting in relation to the reward they feel they receive for it) then they'll start to avoid the store in future if they can. The store does not want to lose future custom, and so has to invest in increasing the number of servers (since the minimal immediate-cost option to the store is a single server).
This is an area that's been studied extensively, both in queueing theory and in retail management.
Governments prefer big corporations. One or two big corporations are much easier to control than a lot of small companies (some of which the government might not even be aware of).
On the other hand, small corporations can't cause nearly so much trouble for governments as large ones so are preferable to government in other respects. What changes isn't what the effects of scale actually are, but what value is attached to particular parts of those effects; that's what drives the cycles.
It's still ongoing right now, albeit intermittently. I'm seeing drop-outs on the distribution of my skype status (despite my local 'net connection being fine).
Plain X forwarding is fine on a LAN, but it's not really usable over ADSL.
Only because you insist on having masses of eye candy. Forego that, and you can run just fine with ADSL. Heck, I've done real work with remote X running over a 14k4 modem (ick!) and that worked (by "real work" I mean Framemaker and xfig; everything else that I needed at the time was just fine locally or in a plain terminal). It's only for eye candy (or photo editing, though I've never needed that professionally) that you really want to have a high-bandwidth low-latency connection between application and graphics display.
Ah, so you're one of those "there are people in California, and people in a couple of cities going down the eastern coastline, and nothing else counts" sorts, huh?
In population terms, yes. There's no excuse for urban populations having crap broadband, and there's lots of people in cities and towns in the US. If you're out in the boonies, it's going to impact on your speed (or costs) but that's true all over the world. But more to the point, just look at where the majority of people are, in urban and suburban areas. Is there any reason why it's impossible for such a large fraction of them to get broadband? (Well, yes there is, and it's got to do with lack of real competition between providers. Regulatory fail.)
His statement is pretty spot on -- there are some pretty wide swaths in this country where you've either got low population density or geographical problems making it difficult. Look at Appalachia as a whole, for example -- a good chunk of it is "difficult" geographically, and having a significant percentage of the populace nestled in mountain hollows doesn't help.
Because cables can't go down into mountain hollows... (Or did you think that the rest of the world does broadband always by wireless telecoms?)
Mass Effect 2 was so awesome I actually went out and bought a genuine copy of it after I beat it. I think I've got at least 200 hours logged on that game as many times as I've replayed it.
That's a game that specializes in making you feel like James Bond in space, where you are totally the Hero out to save the galaxy even if they don't think they need it. That sort of experience can only be delivered as a single-person game; you just can't have 100k players all simultaneously being the most awesome person in the universe.
What this does emphasize is that the big difference between single and multi-player games will be between story-driven and interaction-driven gaming. There's plenty of room for both.
Niche products also tend to be rather buggy.
Or expensive. Some niches are defined by having extremely high quality requirements on entry, which make it hard for non-commercial software to get started precisely because none of the users will consider any option that isn't very good indeed. Those tend to be pretty conservative niches, but for some things (e.g., anything safety-critical) that's good; it's reasonable to spend on having things dependable.
Incidentally, there's a fine Russian word for a hierarchical system of representation, in which smaller governmental bodies choose representatives to the national government: "Soviet." Yeah, that sure helped protect the liberties of the people and the long-term interests of the republic, didn't it?
Null argument. The official name of East Germany was (after translation) The German Democratic Republic. Does that mean that there's a problem with democracy? Or republics? Or just that names chosen for propaganda reasons are bunk and what matters is what happens on the ground? Hmm...
out-of-state websites would be billed at a higher rate (except for nights and weekends).
What makes you think that they'd lower their rates for any reason?