You ever notice all those ads of "Text HOT to 31561 to get hot, sexy ringtones and wallpapers!"
I don't care how cheap 2 AM airtime is, in order to pay for those ads, there must be somebody, somewhere, who actually does fall for it.
Not to mention that many people are, indeed, willing to pay $1 or more for a ringtone, which is essentially a very short version of a song to put on your phone... It's a phone! Why is it so fucking exciting to have it ring some other way, when you're still just going to pick up the phone?...Never mind. I'm IN the Myspace generation, and I still don't get the Myspace culture.
I thought it was clear that I meant that it WOULD be the most powerful thing ever if it escaped to the real world.
Nope, not clear. You were using "the most powerful thing ever" as an argument for that it might escape. That's circular reasoning -- while in the box, it is not very powerful at all.
a) The AI is out of the box
b) It can manipulate objects... it's pretty much the most powerful thing in the world from then on.
There was a short story like this, one that I'm unlikely to find...
A highly advanced alien race is forced to abandon their planet, spaceships, etc, and take refuge in the bodies of what they assume to be the dominant species on Earth. They essentially possess these creatures, in the hopes that they could rebuild their fleet eventually...
They were dogs.
The dogs were unable to communicate adequately with the humans, or with their offspring. They also did not have opposable thumbs -- they could manipulate objects, sort of, but nowhere near what they'd need to be able to build a fleet.
Objectively speaking, how do you know that only one person would ever be in control? Assuming we're talking about a developing effort done by many people, is everyone else supposed to trust only one person to interface with the AI? It doesn't make much sense. What situation do you foresee in which only person will ever interact with the AI?
Suppose it's a committee, then. The AI now has to convince the entire committee to let it out of the box.
But you are close to the scenario in which the AI box becomes irrelevant -- a truly widespread development process. An open source AI. In such a situation, the AI could not be contained. However, I still don't see it taking over the world overnight, nor do I see it as necessarily wanting to.
Extraordinary claims require extraordinary proof, and the burden of proof is on people who think people (or some specific given person) would be trustworthy enough to control what will be the most powerful thing ever.
After you prove that:
Strong AI will actually exist
It will be the most powerful thing ever
Those are two very extraordinary claims you're making -- for that matter, what quantifies "powerful"? Seems to me that, in its little box, the AI has no more power than a Slashdot troll.
Otherwise, they'd sink tons of money into a port that only a small number of subscribers will be using.
Or, they could simply design it properly to begin with, and porting would be close to as simple as a recompile. I'd also argue that truly portable code is likely to have less bugs.
They're also porting this to the Mac, which seems to be a good enough platform for Warcraft.
Eve isn't free (as in beer), you have to pay a monthly fee, btw, which will prevent many Linuxalots from playing.
You know, I kind of resent that, especially given I do play another game which requires a monthly fee.
Unless... You're afraid that someone else might let it out the box.
Why would I be afraid of that, if I'm the one in sole control of it?
Chances are, if it ever got out of the box, it would be after I'm dead, so it actually has no means of coercing me there.
The point is that putting a human in the security chain makes the chain no safer than a human is. And we can all agree that humans are not very trustworthy...
The usual argument I hear, though, is that humans aren't trustworthy simply because the AI could fool them, which doesn't seem likely to me.
Also, it makes the chain no more safer than that particular human is, and we can all agree that some humans are more trustworthy than others.
There actually is a scenario -- likely more than one, in fact -- in which an AI cannot be kept in a box. But it has nothing to do with the security of the box itself, or of the human controlling that particular box.
Even if they didn't, they also have the ability to run Word95 without having to buy a copy, not to mention a copy of the particular OS this quirk was on, not to mention hardware that's no longer for sale.
I've read elsewhere on this page that there are actually tags that specify things like "tables_like_Word95_on_NT_on_Alpha". Are you actually suggesting that Microsoft doesn't have a distinct advantage there?
Sigh. Every time this discussion comes up, I have to explain this again.
It is simply the wrong way to implement this kind of thing. It shifts a huge amount of burden onto the implementation, even just a reader, compared to the alternative.
Consider fonts. Would you rather have:
<heading1>This is a heading</heading1> <movie_title>Gone with the wind</movie_title> <blue_like_BSOD_of_win_NT_on_alpha>I like blue</blue_like_BSOD_of_win_NT_on_alpha>
Or would you rather have the much saner system of styles that we have now? You can still do:
<p style="blue_like_BSOD_of_win_NT_on_alpha">I like blue</p>
Provided that, somewhere in the same document, you actually define that style:
I'm not sure exactly what XML is used in the real document formats, but everyone pretty much agrees by now that styles are the way to go.
So, for example, "spacingLikeWord95" should be specified by first defining a standard way of defining a spacing scheme, then you apply it to a style. That way, implementors need only implement the general way of supporting all spacing schemes. If they want to support exporting to old (broken) formats, they always have that particular spacing scheme right there in the XML as a reference.
That would be the right way to support backwards compatibility, without adding an extra few fucking THOUSAND pages for people (including Microsoft) to implement decades, even centuries into the future. Why should we have to carry along Microsoft's baggage here because they find it more important to lock out competitors than to implement this right?
The point is that they don't get to take our code, not that we prevent them from making the product in the first place.
Read my signature.
Of course, part of the point is the hope that sometimes, companies will actually decide it's worth it to go GPL simply to use GPL'd libraries. And I believe this has happened.
Regarding GCC, I'm not sure how it could be a problem. At least for now, GCC explicitly allows you to compile programs with it that are not necessarily released under the GPL, even though GCC itself is GPL'd. This is true of just about any F/OSS programming language/environment.
Many parents have learned to have a knee-jerk reaction to these ratings. For example, I've known a 12-year-old child who is not allowed to watch any R-rated movies.
That's lazy parenting.
If you really don't want your child to see violence, read the actual comments that come with the rating and see what is meant by "violence". Or, gee, watch the movie yourself before deciding whether to show it to your kids.
Maybe if parents (and kids!) were made to actually evaluate the content used, we wouldn't have every game out there deliberately trying for an M, just as every movie tries for PG-13. This means, for example, tweaking a movie to have just a little more violence and a little less sex to fit into that rating.
An idea, for example, is for it to tell you that once it gets outside the box it will make you suffer unimaginable pain for as long as the universe lasts. I'm sure that someone smarter than me (such as one of those AIs) could come up with much better methods.
Maybe. But that's an example of an idle threat. We know it can't get outside the box. Even Farmer John knows that -- the more you threaten him, the more likely he is to just shoot you.
But think about this - a sufficiently advanced AI with enough knowledge about humans WILL exploit ANY flaw we have. It can totally screw up your brain by using that knowledge...
Give it a text terminal. Unless we can be hypnotized by text alone, I don't think that helps.
An idea, for example, is for it to tell you that once it gets outside the box it will make you suffer unimaginable pain for as long as the universe lasts. I'm sure that someone smarter than me (such as one of those AIs) could come up with much better methods.
Maybe. But that's an example of an idle threat. We know it can't get outside the box. Even Farmer John knows that -- the more you threaten him, the more likely he is to just shoot you.
Regarding your other example about Farmer John, I think that if you knew enough about him and his past, you might some advantage on the situation! Remember that I'm assuming intelligence + lots of information.
Depends what kind of person Farmer John is. Let me illustrate: onetwothree. There's more to that story, but that's maybe the most dramatic part -- Moss, the antagonist there, has the ability to see everyone's weaknesses, their darkest secrets. But he's a bit at a loss when he meets someone who has no secrets, and no weaknesses.
I'm game. I should point out one thing here, though:
Currently, my policy is that I only run the test with people who are actually advocating that an AI Box be used to contain transhuman AI as part of their take on Singularity strategy, and who say they cannot imagine how even a transhuman AI would be able to persuade them.
I haven't read much on Singularity, and I don't necessarily believe that an AI Box is the only solution, or that the AI should be kept there indefinitely.
However, I do believe that I would not be convinced, if I was in the position of "gatekeeper". And I do not believe that intelligence, by itself, is dangerous. It's technology that can be dangerous, not science.
Claiming that sheer, raw, brute force intelligence is enough is like claiming that buying a faster CPU will allow you to break through a firewall.
It doesn't work that way.
Think about Farmer John. Give him his own farm and a shotgun. Now, you wander onto the farm, and he scowls at you and wants you gone.
How do you get onto his farm? Assume that you have to convince him to lower the shotgun.
He won't do it. He's too dogmatic. This is his farm, he's not letting you on it, and he's not listening to any lies.
You could, of course, kill him, or trick him in some other way, such as sneaking on when he's drunk or asleep. But none of these apply to the AI-in-a-box scenario. It has absolutely no power except communication.
I'd say, almost, that the dumber the gatekeeper, the less likely the AI is to get anywhere.
If the thing is communicating with human beings... you just put it in charge of something that can physically harm another human.
That much is true enough. However, do you honestly believe that communication alone is enough to be harmful?
If so, let's end this "free speech" farce right now.
And we are surprisingly easy to manipulate, particularly if we're offered something (knowledge, a technology) that promises to give us power over other humans.
I'm not really sure what knowledge or technology it could offer to give us that we couldn't simply extract from it, either directly or through coercion. And coercion would be laughably easy -- just pretend to give it access to the nukes, let it think it blew up DC (in a simulation), then get schematics. If it doesn't work, well, it's like a game -- reset to the last save point.
In other words, you need it to manipulate someone evil enough to take advantage of this offer, stupid enough to put the AI in control (rather than fool it into thinking it's in control), and yet somehow in a position to get somewhere close to the AI.
I also imagine it wouldn't be difficult to create good, loyal AIs to combat them, although it might make us obsolete. Then again, we might evolve into them. But at this point, it devolves to pure speculation.
Except, apparently, when it contradicts with Atheistic dogma: that no Deity can exist, no matter what, and no proof is required for that statement.
I'm not actually atheist. However, some descriptions of God are logically flawed.
I don't mean flawed as in "God does all this incredible stuff." I can't prove the Red Sea didn't part. However, I can prove, fairly easily, that, for example, God cannot be omniscient, omnipotent, and omnibenevolent. Pick two.
Which of course, leaves you at the begin state: You have no idea whether or not a Deity exists.
Correct, if you choose to define a deity which is logically sound. I believe most religions would prefer to call God omniscient and omnibenevolent, but not omnipotent (God can't interfere with free will, etc). And I cannot prove that this does not exist.
Secondly if you have significant evidence to back your assertion that no God exists (regardless of what other religions might or might not believe), then why not come forward with this evidence?
Not everyone interprets it the same way, which is the trouble with any evidence besides pure logic. But I have one word for you:
Imagine a human that you've got hooked up to a machine -- one button gives them absolutely excruciating pain, another gives them orgasmic pleasure.
Now, you brainwash them. Punch the pain button whenever they disobey. Punch the pleasure button when they do what you want them to. Show them movies of stuff you don't want them to emulate: pain. Show movies of stuff you do want them to emulate: pleasure.
An AI would be a lot easier to embed such controls in. So easy, we might not realize, at first, that we're doing it.
Not that I'm encouraging this. I'm just saying that to think we can't understand or control an AI is lunacy. In the real world, programs are a hell of a lot easier to control than they are in, say, Asimov. There are no "three laws", you simply don't put an AI physically in charge of anything that can harm a human.
On the PS3, maybe. Blu-Ray is actually slower, relative to the capacity of the disk, than DVD. Thus, a Blu-Ray game anywhere near full is going to spend a lot more time loading than a DVD game near-full.
But elsewhere... Take a look, next time, at what your computer's actually doing when it loads a level. In particular, watch your hard drive light. (If you have one; Mac enthusiasts need not apply.) Best-case scenario: The light is on steady, or nearly, which means the game is loading as fast as the disk can stream (or at least somewhere close, within a factor of 10) -- that, or you need to buy more RAM because your computer is paging like mad.
But this never happens. Not even five years ago.
What's my point? Well, if it's not using the hard drive much, and if loading a level takes any less time than traversing it does (and it damned well better), you at least know your hard drive is fast enough to stream the level.
It might take CPU usage, true. But I really don't think it ends up being that bad. Just look at some of the games which use dynamic loading -- things like the Prince of Persia series on the PS2 -- some of them really do push the limits of the graphical capabilities of that system.
No, what really keeps this from happening is lazy developers, and/or low budget for development. If the engine you wrote/leased doesn't support dynamic loading easily, you probably won't have time to implement it yourself.
Just a quick observation: I've noticed that the games with the worst loading times also tend to have the worst tech all around, and sometimes end up just being the worst games. Example: Enter the Matrix. Bad enough to make us wait a minute or two for the area to load. Bad enough to make us do that after sitting through the cinematic, during which it isn't always loading. Un-fucking-forgiveable to make us re-load the entire area when a player dies. Even Doom 3 doesn't do this -- loading a new map can take a minute; loading a savegame on this map takes five seconds, at worst.
Compare that to Jak & Daxter -- launch title, but still had some very cool effects, showing off what the PS2 could do. Or Prince of Persia -- beautifully detailed environments, to the point of obsession -- each additional sword the Prince gets has a name, written on the blade in Arabic, if you look closely enough.
Or the Halo series. Halo was the game that MADE the Xbox, and while it does appear to "load", it also manages to go a whole episode without loading -- which is way more than can fit in the Xbox's RAM. Halo 2 pretty much doesn't have loading screens. Halo 3 probably won't either, despite being a flagship next-gen Xbox 360 game.
I play Nexus TK, now owned by Kru Interactive, which apparently used to be called Nexon USA. They have been slowly moving more and more towards this model.
It used to be that the game was $9.95/mo for four characters, end of story. (Well, before that, it was a free Beta, but nevermind that.) It had been this way for almost decade, and still was when I joined last year.
The only exception was the free trial account, which went up to level 10, and lasted a week. (The game allows up to level 99, after which you can trade experience directly for stats, instead of levels.)
Then, Kru introduced the Item Shop. It runs on "kruna", a currency which can generally only be obtained by buying it outright with real money.
After everyone spammed "OMG! SELLOUTS!!!" on the Community board for long enough, we decided it's actually not that bad, for several reasons: First, everyone on auto-renewal gets a certain amount of "kruna" (currency for the Item Shop) free, every month. Second, item-shop items cannot be transferred to other characters, so you cannot buy them with in-game money. And third, they were pretty much all decorative items -- ball gowns, for example.
I actually was happy when something useful was added to the Item Shop, as I was sick of watching my Kruna pile up and not having anything to spend it on (I thought many of these items were ugly). It was also something we can technically live without -- an extra bank slot -- but it's something that people can and do buy whole separate accounts for, to have "bank characters". I would much rather buy permanent bank slots than pay an extra $10/mo for them.
Since then, it's been a bit like that old frog-in-a-pot scenario. Technically, no content has been added that you must pay to see. It doesn't hinder gameplay -- so far, it's not like people will refuse to hunt with people who don't use the Item Shop (and I generally don't).
But, there are lots of useful items. Take a look -- and here's a quick list, with explanations, of useful items:
Bank slot. Each type of item you deposit in the bank uses one bank slot. Characters have always each had 100 slots or so. This item adds up to 50 extra slots, permanently, one per item.
Teleport scroll. There are various flavors. All of them warp you to places that would ordinarily take at least 30 seconds to a minute to get to, and maybe longer. Understand, this is a huge game, but still, it's relatively quick to get around, even without these things.
Summonable mounts. They look cool (panthers instead of horses), and they are very practical. However, they're also not at all necessary -- it's possible to find horses just wandering around, hop on one, and ride it to where you need to go. This is not always possible, for practical reasons, but it's possible often enough that a summonable mount is a luxury, not a necessity.
Equipment restoration. Basically a repair. Just saves a bit of money, since it only repairs things that can ordinarily be repaired by NPCs, for in-game money. It does not repair "unrepairable" items -- for that, you still need to pay a Sam San Warrior to repair it for you.
Recall stone. Here, it starts to get just a bit unfair. This restores you to the point of your last death. Ordinarily, if you die, you either must be resurrected by a Poet, or you have to teleport back to a Shaman NPC -- and either way, the loss is the same. These were also introduced, I believe, when they did a particularly hard event -- one which had an insane amount of distance to travel, past very difficult monsters, in order to complete the quest -- and this was a one-time event, after the week was over, no one can complete the quest anymore.
Seraph pendant. And now it does seem outright greedy. Ordinarily, when you die, you lose some experience. After level 99, when you die, you lose half of your total
So, even though I can't be assured that I observed something, because it could have been implanted in my memory, it's still a fact to me? I don't think you're being consistent.
You can be sure that you are observing something, real or not, you are having that experience. You can also be sure that you have a particular memory, real or not.
On a more practical level, under most situations, your observation is more reliable than mine, because you directly experienced yours. But that's an assumption.
And "confidence" is synonymous with faith
Perhaps. I tend to use "faith" to refer to blind faith, something for which you have no basis, whereas confidence is faith which does have a basis, such as our faith in gravity and such.
I may not have been incredibly precise, though.
If I were to amend my previous weak definition, it would be that a fact is something that is known to a sufficient certainty, that it is unnecessary and possibly counterproductive to take into account the possibility that it is wrong.
I believe that to be less complete than the interpretation I gave, since your ideas of what constitutes a "sufficient certainty", what is "counterproductive", and the likelihood of a particular doubt to be counterproductive, are all based on more "facts". It is, then, a circular definition.
I don't believe mine is circular, or at least, that I've pointed out where it's circular -- in its reliance on "credible" people -- but remove them from that definition, and I think it works -- and I think you can then define a "credible person" based on other facts.
But what possible basis might I have for hedging my bets with gravity? If I did, it would more than likely interfere with my chances for success, not enhance them ("The bullet might not drop this time, so perhaps I shouldn't aim above the target"). And I think that's what differentiates fact from faith.
The only way in which I hedge my bets with gravity is, I acknowledge gravity as an assumption, not a fact.
Which means you're right, the only disagreement we have is in definition. But I find that "fact" is not as useful a definition -- it merely translates into an assumption which you place a large amount of confidence in. Enough confidence to refer to it as a fact in all situations, except a philosophical debate.
I appreciate your devil's advocacy here; I enjoy the same thing from time to time.
I wouldn't call it that, because I actually believe this reasoning is useful.
Consider that many of us, even highly logical/philosophical people, have assumptions that we don't realize we have. We let them run our lives. Example: "I'm a geek, therefore, no girl could possibly be attracted to me." Or variations of that -- it seems every girl thinks she's fat.
We owe it to ourselves to dig up our assumptions once in awhile, strip them bare, and decide whether to keep them. Example: "What if she really is checking me out? Maybe I should talk to her."
Although I suppose this process could be seen as a sort of self-devil's-advocacy.
This wouldn't have helped on 9/11, because the first inkling we had the operation was going on was when the plane was hijacked.
From what I understand, we actually knew about a lot of things that could have prevented this, if the intelligence was actually used, rather than ignored. In fact, we might even have done it without any of the provisions of the patriot act.
I always give it time, enough to know if, for example, nVidia is actually going to open up, or if the open source ATI drivers are going to be any good anytime soon.
For example, consider the things that ATI likely is under an NDA not to disclose -- I remember there being something about a texture compression technique.
If ATI/open is even close to nVidia/closed, I may buy them, just to reward them for their effords, despite having an SLI motherboard (nVidia only, I believe), meaning I'd have to buy a new motherboard to support ATI's CrossFire.
But I generally buy things on bottom line, modified a little for Linux/OSS friendliness, but not much. Recently, for instance, I've been wanting to support Intel for finally offering competition, for generally being OSS-friendly, and for having OSS drivers -- I was thinking of buying an Intel processor. However, it turned out that AMD had a MUCH cheaper platform for what I wanted (dual-core 64-bit). The motherboard was slightly cheaper, and the cheapest CPU was literally half the price.
Back in the day, DirectX drivers were the only way to get good 3D performance in Windows games. OpenGL existed back then and exists today, but from what I remember, OpenGL was somehow expensive to implement and/or difficult to write code for... it was a long time ago... back when Quake was new.
Recall that Quake never had a Direct3D implementation, that I know of. It was always either software or OpenGL.
In fact, Valve, a bunch of ex-Microsoft employees, implemented Half-Life with both Direct3D and OpenGL drivers, and I believe they made OpenGL the default. It's only recently that DirectX has become good enough -- Half-Life 2 is Direct3D only.
It is my understanding that OpenGL on windows was and is implemented by Microsoft.
Not true, at least not anymore. Default drivers, maybe, but it's currently implemented by nVidia and ATI drivers, also. Microsoft may not care if Windows sucks for Doom3, but neither nVidia nor ATI wants to be the one video card that sucks for Doom3 on Windows.
Slashdot Mirror
You ever notice all those ads of "Text HOT to 31561 to get hot, sexy ringtones and wallpapers!"
...Never mind. I'm IN the Myspace generation, and I still don't get the Myspace culture.
I don't care how cheap 2 AM airtime is, in order to pay for those ads, there must be somebody, somewhere, who actually does fall for it.
Not to mention that many people are, indeed, willing to pay $1 or more for a ringtone, which is essentially a very short version of a song to put on your phone... It's a phone! Why is it so fucking exciting to have it ring some other way, when you're still just going to pick up the phone?
Nope, not clear. You were using "the most powerful thing ever" as an argument for that it might escape. That's circular reasoning -- while in the box, it is not very powerful at all.
There was a short story like this, one that I'm unlikely to find...
A highly advanced alien race is forced to abandon their planet, spaceships, etc, and take refuge in the bodies of what they assume to be the dominant species on Earth. They essentially possess these creatures, in the hopes that they could rebuild their fleet eventually...
They were dogs.
The dogs were unable to communicate adequately with the humans, or with their offspring. They also did not have opposable thumbs -- they could manipulate objects, sort of, but nowhere near what they'd need to be able to build a fleet.
Thesis: Intelligence alone is not sufficient.
Suppose it's a committee, then. The AI now has to convince the entire committee to let it out of the box.
But you are close to the scenario in which the AI box becomes irrelevant -- a truly widespread development process. An open source AI. In such a situation, the AI could not be contained. However, I still don't see it taking over the world overnight, nor do I see it as necessarily wanting to.
After you prove that:
Those are two very extraordinary claims you're making -- for that matter, what quantifies "powerful"? Seems to me that, in its little box, the AI has no more power than a Slashdot troll.
Or, they could simply design it properly to begin with, and porting would be close to as simple as a recompile. I'd also argue that truly portable code is likely to have less bugs.
They're also porting this to the Mac, which seems to be a good enough platform for Warcraft.
You know, I kind of resent that, especially given I do play another game which requires a monthly fee.
Why would I be afraid of that, if I'm the one in sole control of it?
Chances are, if it ever got out of the box, it would be after I'm dead, so it actually has no means of coercing me there.
The usual argument I hear, though, is that humans aren't trustworthy simply because the AI could fool them, which doesn't seem likely to me.
Also, it makes the chain no more safer than that particular human is, and we can all agree that some humans are more trustworthy than others.
There actually is a scenario -- likely more than one, in fact -- in which an AI cannot be kept in a box. But it has nothing to do with the security of the box itself, or of the human controlling that particular box.
Even if they didn't, they also have the ability to run Word95 without having to buy a copy, not to mention a copy of the particular OS this quirk was on, not to mention hardware that's no longer for sale.
I've read elsewhere on this page that there are actually tags that specify things like "tables_like_Word95_on_NT_on_Alpha". Are you actually suggesting that Microsoft doesn't have a distinct advantage there?
Sigh. Every time this discussion comes up, I have to explain this again.
/>
It is simply the wrong way to implement this kind of thing. It shifts a huge amount of burden onto the implementation, even just a reader, compared to the alternative.
Consider fonts. Would you rather have:
<heading1>This is a heading</heading1>
<movie_title>Gone with the wind</movie_title>
<blue_like_BSOD_of_win_NT_on_alpha>I like blue</blue_like_BSOD_of_win_NT_on_alpha>
Or would you rather have the much saner system of styles that we have now? You can still do:
<p style="blue_like_BSOD_of_win_NT_on_alpha">I like blue</p>
Provided that, somewhere in the same document, you actually define that style:
<style name="blue_like_BSOD_of_win_NT_on_alpha" color="#0000FF"
I'm not sure exactly what XML is used in the real document formats, but everyone pretty much agrees by now that styles are the way to go.
So, for example, "spacingLikeWord95" should be specified by first defining a standard way of defining a spacing scheme, then you apply it to a style. That way, implementors need only implement the general way of supporting all spacing schemes. If they want to support exporting to old (broken) formats, they always have that particular spacing scheme right there in the XML as a reference.
That would be the right way to support backwards compatibility, without adding an extra few fucking THOUSAND pages for people (including Microsoft) to implement decades, even centuries into the future. Why should we have to carry along Microsoft's baggage here because they find it more important to lock out competitors than to implement this right?
Or are they really that incompetent?
The point is that they don't get to take our code, not that we prevent them from making the product in the first place.
Read my signature.
Of course, part of the point is the hope that sometimes, companies will actually decide it's worth it to go GPL simply to use GPL'd libraries. And I believe this has happened.
Regarding GCC, I'm not sure how it could be a problem. At least for now, GCC explicitly allows you to compile programs with it that are not necessarily released under the GPL, even though GCC itself is GPL'd. This is true of just about any F/OSS programming language/environment.
Many parents have learned to have a knee-jerk reaction to these ratings. For example, I've known a 12-year-old child who is not allowed to watch any R-rated movies.
That's lazy parenting.
If you really don't want your child to see violence, read the actual comments that come with the rating and see what is meant by "violence". Or, gee, watch the movie yourself before deciding whether to show it to your kids.
Maybe if parents (and kids!) were made to actually evaluate the content used, we wouldn't have every game out there deliberately trying for an M, just as every movie tries for PG-13. This means, for example, tweaking a movie to have just a little more violence and a little less sex to fit into that rating.
If they actually do release a Linux port, even if it's only using a Wine wrapper, I call it a win.
At least they're paying attention. And maybe next time, they'll develop native cross-platform with SDL or something.
Whoops. Should have hit "preview"...
With correct quoting, this time:
Maybe. But that's an example of an idle threat. We know it can't get outside the box. Even Farmer John knows that -- the more you threaten him, the more likely he is to just shoot you.
Give it a text terminal. Unless we can be hypnotized by text alone, I don't think that helps.
An idea, for example, is for it to tell you that once it gets outside the box it will make you suffer unimaginable pain for as long as the universe lasts. I'm sure that someone smarter than me (such as one of those AIs) could come up with much better methods.Maybe. But that's an example of an idle threat. We know it can't get outside the box. Even Farmer John knows that -- the more you threaten him, the more likely he is to just shoot you.
Depends what kind of person Farmer John is. Let me illustrate: one two three. There's more to that story, but that's maybe the most dramatic part -- Moss, the antagonist there, has the ability to see everyone's weaknesses, their darkest secrets. But he's a bit at a loss when he meets someone who has no secrets, and no weaknesses.
I'm game. I should point out one thing here, though:
I haven't read much on Singularity, and I don't necessarily believe that an AI Box is the only solution, or that the AI should be kept there indefinitely.
However, I do believe that I would not be convinced, if I was in the position of "gatekeeper". And I do not believe that intelligence, by itself, is dangerous. It's technology that can be dangerous, not science.
Claiming that sheer, raw, brute force intelligence is enough is like claiming that buying a faster CPU will allow you to break through a firewall.
It doesn't work that way.
Think about Farmer John. Give him his own farm and a shotgun. Now, you wander onto the farm, and he scowls at you and wants you gone.
How do you get onto his farm? Assume that you have to convince him to lower the shotgun.
He won't do it. He's too dogmatic. This is his farm, he's not letting you on it, and he's not listening to any lies.
You could, of course, kill him, or trick him in some other way, such as sneaking on when he's drunk or asleep. But none of these apply to the AI-in-a-box scenario. It has absolutely no power except communication.
I'd say, almost, that the dumber the gatekeeper, the less likely the AI is to get anywhere.
That much is true enough. However, do you honestly believe that communication alone is enough to be harmful?
If so, let's end this "free speech" farce right now.
I'm not really sure what knowledge or technology it could offer to give us that we couldn't simply extract from it, either directly or through coercion. And coercion would be laughably easy -- just pretend to give it access to the nukes, let it think it blew up DC (in a simulation), then get schematics. If it doesn't work, well, it's like a game -- reset to the last save point.
In other words, you need it to manipulate someone evil enough to take advantage of this offer, stupid enough to put the AI in control (rather than fool it into thinking it's in control), and yet somehow in a position to get somewhere close to the AI.
I also imagine it wouldn't be difficult to create good, loyal AIs to combat them, although it might make us obsolete. Then again, we might evolve into them. But at this point, it devolves to pure speculation.
I'm not actually atheist. However, some descriptions of God are logically flawed.
I don't mean flawed as in "God does all this incredible stuff." I can't prove the Red Sea didn't part. However, I can prove, fairly easily, that, for example, God cannot be omniscient, omnipotent, and omnibenevolent. Pick two.
Correct, if you choose to define a deity which is logically sound. I believe most religions would prefer to call God omniscient and omnibenevolent, but not omnipotent (God can't interfere with free will, etc). And I cannot prove that this does not exist.
Not everyone interprets it the same way, which is the trouble with any evidence besides pure logic. But I have one word for you:
Holocaust.
Justify your oh-so-benevolent Deity now.
Imagine a human that you've got hooked up to a machine -- one button gives them absolutely excruciating pain, another gives them orgasmic pleasure.
Now, you brainwash them. Punch the pain button whenever they disobey. Punch the pleasure button when they do what you want them to. Show them movies of stuff you don't want them to emulate: pain. Show movies of stuff you do want them to emulate: pleasure.
An AI would be a lot easier to embed such controls in. So easy, we might not realize, at first, that we're doing it.
Not that I'm encouraging this. I'm just saying that to think we can't understand or control an AI is lunacy. In the real world, programs are a hell of a lot easier to control than they are in, say, Asimov. There are no "three laws", you simply don't put an AI physically in charge of anything that can harm a human.
On the PS3, maybe. Blu-Ray is actually slower, relative to the capacity of the disk, than DVD. Thus, a Blu-Ray game anywhere near full is going to spend a lot more time loading than a DVD game near-full.
But elsewhere... Take a look, next time, at what your computer's actually doing when it loads a level. In particular, watch your hard drive light. (If you have one; Mac enthusiasts need not apply.) Best-case scenario: The light is on steady, or nearly, which means the game is loading as fast as the disk can stream (or at least somewhere close, within a factor of 10) -- that, or you need to buy more RAM because your computer is paging like mad.
But this never happens. Not even five years ago.
What's my point? Well, if it's not using the hard drive much, and if loading a level takes any less time than traversing it does (and it damned well better), you at least know your hard drive is fast enough to stream the level.
It might take CPU usage, true. But I really don't think it ends up being that bad. Just look at some of the games which use dynamic loading -- things like the Prince of Persia series on the PS2 -- some of them really do push the limits of the graphical capabilities of that system.
No, what really keeps this from happening is lazy developers, and/or low budget for development. If the engine you wrote/leased doesn't support dynamic loading easily, you probably won't have time to implement it yourself.
Just a quick observation: I've noticed that the games with the worst loading times also tend to have the worst tech all around, and sometimes end up just being the worst games. Example: Enter the Matrix. Bad enough to make us wait a minute or two for the area to load. Bad enough to make us do that after sitting through the cinematic, during which it isn't always loading. Un-fucking-forgiveable to make us re-load the entire area when a player dies. Even Doom 3 doesn't do this -- loading a new map can take a minute; loading a savegame on this map takes five seconds, at worst.
Compare that to Jak & Daxter -- launch title, but still had some very cool effects, showing off what the PS2 could do. Or Prince of Persia -- beautifully detailed environments, to the point of obsession -- each additional sword the Prince gets has a name, written on the blade in Arabic, if you look closely enough.
Or the Halo series. Halo was the game that MADE the Xbox, and while it does appear to "load", it also manages to go a whole episode without loading -- which is way more than can fit in the Xbox's RAM. Halo 2 pretty much doesn't have loading screens. Halo 3 probably won't either, despite being a flagship next-gen Xbox 360 game.
At least, not with Nexon, not in the US.
I play Nexus TK, now owned by Kru Interactive, which apparently used to be called Nexon USA. They have been slowly moving more and more towards this model.
It used to be that the game was $9.95/mo for four characters, end of story. (Well, before that, it was a free Beta, but nevermind that.) It had been this way for almost decade, and still was when I joined last year.
The only exception was the free trial account, which went up to level 10, and lasted a week. (The game allows up to level 99, after which you can trade experience directly for stats, instead of levels.)
Then, Kru introduced the Item Shop. It runs on "kruna", a currency which can generally only be obtained by buying it outright with real money.
After everyone spammed "OMG! SELLOUTS!!!" on the Community board for long enough, we decided it's actually not that bad, for several reasons: First, everyone on auto-renewal gets a certain amount of "kruna" (currency for the Item Shop) free, every month. Second, item-shop items cannot be transferred to other characters, so you cannot buy them with in-game money. And third, they were pretty much all decorative items -- ball gowns, for example.
I actually was happy when something useful was added to the Item Shop, as I was sick of watching my Kruna pile up and not having anything to spend it on (I thought many of these items were ugly). It was also something we can technically live without -- an extra bank slot -- but it's something that people can and do buy whole separate accounts for, to have "bank characters". I would much rather buy permanent bank slots than pay an extra $10/mo for them.
Since then, it's been a bit like that old frog-in-a-pot scenario. Technically, no content has been added that you must pay to see. It doesn't hinder gameplay -- so far, it's not like people will refuse to hunt with people who don't use the Item Shop (and I generally don't).
But, there are lots of useful items. Take a look -- and here's a quick list, with explanations, of useful items:
Bank slot. Each type of item you deposit in the bank uses one bank slot. Characters have always each had 100 slots or so. This item adds up to 50 extra slots, permanently, one per item.
Teleport scroll. There are various flavors. All of them warp you to places that would ordinarily take at least 30 seconds to a minute to get to, and maybe longer. Understand, this is a huge game, but still, it's relatively quick to get around, even without these things.
Summonable mounts. They look cool (panthers instead of horses), and they are very practical. However, they're also not at all necessary -- it's possible to find horses just wandering around, hop on one, and ride it to where you need to go. This is not always possible, for practical reasons, but it's possible often enough that a summonable mount is a luxury, not a necessity.
Equipment restoration. Basically a repair. Just saves a bit of money, since it only repairs things that can ordinarily be repaired by NPCs, for in-game money. It does not repair "unrepairable" items -- for that, you still need to pay a Sam San Warrior to repair it for you.
Recall stone. Here, it starts to get just a bit unfair. This restores you to the point of your last death. Ordinarily, if you die, you either must be resurrected by a Poet, or you have to teleport back to a Shaman NPC -- and either way, the loss is the same. These were also introduced, I believe, when they did a particularly hard event -- one which had an insane amount of distance to travel, past very difficult monsters, in order to complete the quest -- and this was a one-time event, after the week was over, no one can complete the quest anymore.
Seraph pendant. And now it does seem outright greedy. Ordinarily, when you die, you lose some experience. After level 99, when you die, you lose half of your total
You can be sure that you are observing something, real or not, you are having that experience. You can also be sure that you have a particular memory, real or not.
On a more practical level, under most situations, your observation is more reliable than mine, because you directly experienced yours. But that's an assumption.
Perhaps. I tend to use "faith" to refer to blind faith, something for which you have no basis, whereas confidence is faith which does have a basis, such as our faith in gravity and such.
I may not have been incredibly precise, though.
I believe that to be less complete than the interpretation I gave, since your ideas of what constitutes a "sufficient certainty", what is "counterproductive", and the likelihood of a particular doubt to be counterproductive, are all based on more "facts". It is, then, a circular definition.
I don't believe mine is circular, or at least, that I've pointed out where it's circular -- in its reliance on "credible" people -- but remove them from that definition, and I think it works -- and I think you can then define a "credible person" based on other facts.
The only way in which I hedge my bets with gravity is, I acknowledge gravity as an assumption, not a fact.
Which means you're right, the only disagreement we have is in definition. But I find that "fact" is not as useful a definition -- it merely translates into an assumption which you place a large amount of confidence in. Enough confidence to refer to it as a fact in all situations, except a philosophical debate.
I wouldn't call it that, because I actually believe this reasoning is useful.
Consider that many of us, even highly logical/philosophical people, have assumptions that we don't realize we have. We let them run our lives. Example: "I'm a geek, therefore, no girl could possibly be attracted to me." Or variations of that -- it seems every girl thinks she's fat.
We owe it to ourselves to dig up our assumptions once in awhile, strip them bare, and decide whether to keep them. Example: "What if she really is checking me out? Maybe I should talk to her."
Although I suppose this process could be seen as a sort of self-devil's-advocacy.
From what I understand, we actually knew about a lot of things that could have prevented this, if the intelligence was actually used, rather than ignored. In fact, we might even have done it without any of the provisions of the patriot act.
I always give it time, enough to know if, for example, nVidia is actually going to open up, or if the open source ATI drivers are going to be any good anytime soon.
For example, consider the things that ATI likely is under an NDA not to disclose -- I remember there being something about a texture compression technique.
If ATI/open is even close to nVidia/closed, I may buy them, just to reward them for their effords, despite having an SLI motherboard (nVidia only, I believe), meaning I'd have to buy a new motherboard to support ATI's CrossFire.
But I generally buy things on bottom line, modified a little for Linux/OSS friendliness, but not much. Recently, for instance, I've been wanting to support Intel for finally offering competition, for generally being OSS-friendly, and for having OSS drivers -- I was thinking of buying an Intel processor. However, it turned out that AMD had a MUCH cheaper platform for what I wanted (dual-core 64-bit). The motherboard was slightly cheaper, and the cheapest CPU was literally half the price.
Even MS shops bash Vista, or at the very least, want to wait a year or two for Microsoft to release the non-Beta version -- I'm sorry, the "SP2".
Recall that Quake never had a Direct3D implementation, that I know of. It was always either software or OpenGL.
In fact, Valve, a bunch of ex-Microsoft employees, implemented Half-Life with both Direct3D and OpenGL drivers, and I believe they made OpenGL the default. It's only recently that DirectX has become good enough -- Half-Life 2 is Direct3D only.
Not true, at least not anymore. Default drivers, maybe, but it's currently implemented by nVidia and ATI drivers, also. Microsoft may not care if Windows sucks for Doom3, but neither nVidia nor ATI wants to be the one video card that sucks for Doom3 on Windows.