I think we're about to go chasing after rabbits. We're still on A1, which I'll restate just so I don't lose focus.
A1: Programs are formal (syntactic)
You are privileging minds to have some special version of meaning which a simple circuit can't have.
This really wasn't my intent. Though I can see how my example would lead you to this interpretation; that's my fault.
My problem was, I think, playing fast and loose with the word "meaning", which has connotations that can cause problems. When I say that a symbol has meaning, I mean that there is a relationship between the symbol and what it denotes.
Had I bothered to define that in my last post, I could have saved us both a lot of writing.
The circuit describes the meaning of the switch as on or off, by the nature of it's operation.
This may be closer to the point at which we disagree. In this case, you're still talking about semantics which are externally attributed (what we interpret the function of the switch to be) not about some *intrinsic* meaning.
We already agree that the meaning we attribute to a program (extrinsic meaning) has no effect. The point at which we disagree (hopefully not for long) and that I've pitifully tried to demonstrate, is that programs lack intrinsic meaning.
I don't want to chase the lamp thing too far -- It was, quite possibly, the worst example I could have used. (It confuses "it has" with "it gives", if that makes any sense, and takes the focus away from programs and syntax.)
Anyhow, we can agree (I think) that programs are defined syntactically, composed of formal symbol manipulations.
Formal symbol manipulations *by themselves* have no semantics. That is to say they have no *intrinsic* meaning. Any semantic content is extrinsic.
Some of the confusion, I think, stems from the semantic preserving properties that we're used to in formal systems (they would hardly be useful otherwise!). The semantics, of course, remain extrinsic.
I should point out that this post doesn't address A3, only A1. I've taken great care to avoid it. (When we get to A3, we won't care about the distinction between intrinsic and extrinsic semantics.)
I don't know that I've done the best job here, but I'm hoping that at least my definition helps to clarifies things.
If the appears to the user to exhibit intelligence, it is in fact intelligence.
The Turing test is not only highly specific, it's more than a little contentious.
Your particular interpretation of the TT has lead you to develop this strange method of assessment that lets you ascribe intelligence to virtually anything. Not only to my thermostat, but equally to a teaspoon or a cup of coffee.
Given your bizarre beliefs about what constitutes intelligence, I don't see any way we can come to any agreement or even mutual understanding.
That is my whole point. How long is a string?
One or more bytes for null terminated strings.
AI, like hardware and software development, observes Moore's law
This is completely unsubstantiated nonsense. I think we're done here.
he statement that somehow because the technology uses "lookup tables" or some other mundane algorithms as part of it's logic disqualifies it from being intelligent implies that it has to perform it's task in some "special" way.
Doesn't this (including the unquoted remainder of the paragraph and the paragraph that follows) broaden the definition of AI to include nearly all information processing? I'm not sure where you're drawing the line between what is AI and what is not.
It is entirely possible that the GP simply holds a narrower view, not that he believed intelligence to be magical.
it is still considered intelligent if it's behavior is intelligent.
This is a philosophical statement to which I'm not certain I can agree without a better definition of intelligent behavior. If the behavior of the amoeba in your previous example qualifies I can't agree as it makes the definition of what can be considered intelligent virtually meaningless. (If sensing and responding to an environment qualifies, then my thermostat must be considered intelligent.)
I won't ask you to define intelligence -- it's a very slippery word -- just what qualifies, in your opinion, as intelligent behavior.
I could go on but you get the point. AI is in use everywhere
That wasn't your original claim. The claim was that AI was advancing at an accelerated pace. (I don't know how you'd measure the rate) I don't disagree that what is now being termed AI is in use in many places.
So it's you who keeps changing the definition of what AI is, it's not a lack of forward progress.
It depends on your perspective. Lot's of things that once would not have been considered AI now fall under that umbrella. Expert systems come immediately to mind. From that point of view, it's the field of AI that keeps changing its definition.
What the GP likely considers AI is the original goal of AI -- to create intelligent machines. To that end, the GP is correct, we're not any closer to solving that problem than we were in the 1950's.
On his claim that there is no "significant research" going on, he's WAY off base.
You know what WOULD kill an iPod would be a device that costs the same, has the same ease-of-use (or better, if they make an iTunes killer), has superior audio and is designed with a better UI. Since none of these have come to fruition, the iPod is still #1.
You're under the impression that the iPod is successful because it has the nicest UI, is the easiest to use, and has the best audio.
How wrong you are! Apple hasn't had the best or easiest to use player in years. What apple has, is the *myth* of being both the best and the easiest to use.
Just an example, when my wife was in college she and her roommate both bought mp3 players. She got a $50 RCA Lyra and her room mate bought a 3G iPod Classic.
My wife had her music collection on the player in about 10 minutes. It took her roommate three hours and a call to me to get anything on to her iPod. Now, I can't say that the Lyra was "better" -- only that it was much easier to use. (The roommate ultimately ended up returning the iPod and picking up a Lyra.)
Other players, like the Creative Zen series, are (and were) known for having well-designed UI's -- often beating apple on both features and easy of use. Of course, it's hard to get the message out when "everyone knows" that apple's products are easiest to use.
Apple myths have an amazing immunity. Early iPods were plagued with HD reliability problems yet "everyone knows" that apple builds lasting products of the highest quality. There have been several class-action lawsuits resulting from the shoddy iPod line (bad batteries, insta-scratch displays), yet "everyone knows" that you should pay extra for Apple because you're paying for quality and reliability.
Apple maintains it's market share in the portable digital music player market by virtue of the myths created around it -- Not because their products are easier to use, of higher quality, more innovative, or nicest-looking.
I'm really struggling with putting the concepts in writing in such a way that is clear and unambiguous
You and me both:)
The input and output is NOT a useful definition of a program. The output depends on too many things besides the input. Your example was a definition of a black box, not a program.
Can you think of a better way to define a program other than by it's I/O? It seems perfectly reasonable to me, as the implementation details aren't really relevant -- as I mentioned before, it doesn't matter if it's 6502 assembly or a mechanical contrivance. (I'd rather not spend too much time on this, as it's completely beside the point.)
Moving on, it seems the point where we differ is on the meaning of "meaning". You write:
The action a turing machine takes in response to a specific input is the meaning a turing machine ascribes to that input.
This may be a result of the loose way we use language when we talk about computers. We often make statements that anthropomorphize the computer like "this command tells the computer too..." or "the computer sees that as..." But statements like that don't have any reality to them. We don't actually mean that the computer "understands" in the way you or I understand.
Part of the reason I chose such a simple example is to illustrate that even at the lowest level, meaning isn't relevant. By showing meaning to be irrelevant to function, I was hoping to show that the only meaning involved is that which is externally attributed -- there is no internal attribution of meaning.
Let's start with a simple circuit just a battery, switch, and lamp arranged in such a way that moving the switch to the on position causes the circuit to complete and the lamp to light.
It's easy to see in this example, that meaning is irrelevant to the function of the circuit. The circuit doesn't have any meaning assigned to the "on" or "off" position of it's switch, even though we might be tempted to make statements like "flipping the switch tells the circuit to turn the light on" we don't intend to imply any understanding on the part of the circuit.
Let's add a second switch to our circuit. We've now built an AND gate (or not, see the first example). Still, you can see that meaning is not relevant to the circuit.
We can expand our circuit further, adding electro-mechanical relays in such a way as to build a half-adder, with our switches setting the input. Still, the circuit requires no meaning be attributed to any of its parts.
Now let's go nuts and build an adder and subtracter. Let's create two "registers" (just a set of switches) which serve as input to them both, and a third 1-bit register (a switch and a few relays) which selects which circuit gets its output displayed on our lamps (i.e. which instruction our "computer" will execute).
We can make this more interesting, by using some paper tape to set the switches we were manually setting before. (imagine each manual switch replaced with a brush-contact -- paper in the way breaks the circuit, equivalent to turning the switch off, a hole allows contact, equivalent to turning the switch on)
With a hand-crank serving as our clock, winding the tape through our machine will execute whatever add and subtract instructions that happen to be recorded on the tape. In effect, the tape will contain a (very uninteresting) program composed of add and subtract instructions with their associated operands.
Building this from the bottom up, we can see that in no way must the machine attribute any 'meaning' to its instructions.
It should be clear that further increasing the complexity of our machine would not change this.
You write:
only the meanings the computer ascribes to them
I hope I've made obvious that the computer at no point need ascribe meaning to it's instructions any more than our first circuit need ascribe meaning to the position of it's switch.
I hope that clears that bit up. Do you have any remaining objections to the axioms?
Thanks for being the first person to actually list and explain how they've made use of their tablet. It's very helpful.
That said, if your use is representative of the common case, it's easy to see why others think the device is "useless" -- at least at the price it's set at.
Just a few reasons why an iPad wouldn't suit our household from the examples you give:
The Train: I don't take one with any sort of regularity. I could, perhaps, see it for the wifi on some flights, but I've only been on a few that offered it and I've never felt compelled to connect to it.
Recipes: My wife would't dream of bringing a computer (tablet or otherwise) into the kitchen for fear that it would suffer from accidental spills. She tends to writes recipes down on note cards, so that she has a record and so that she doesn't have to look them up again later. (she discards or trades the ones she doesn't care for.) I don't deny that you find it useful, I just can't see it working for us.
Movies in bed: Would get me kicked out of the bed!
The only tablet that I think I could get any use out of is RIM's Playbook -- but only because it can work as an extension of my phone (which, with the larger display, should make working with documents and spreadsheets easier). Still, I doubt that the use I'll get out of it will be worth the price.
Not really (it's actually an area I'm familiar with). Though I still want to know what application you're talking about that requires a CD under windows but is also available on the iPad.
I can't let this bit go, however:
the superior apple user experience
This is, undoubtedly, a myth. Neither Apple's UI or overall user experience seems to be superior to equivalent products from other vendors. It's really quite inferior in a number of their products -- I'm sure you can think of a few examples (I dare not upset the apple-cart by listing a few of my own!)
Apple has a long history of myths. You might remember hearing that "Macs are better for graphics" over the past 20 years or so -- even though there was absolutely no reason (at any point) to believe that was the case.
With 70%+ market share after a decade, I think most people would disagree,,,,
Ummm... Being popular does not mean that it's easier to use! Also, your 70% figure is not iPod touch -- which is the one with graphics and apps, it's all of apples music player products.
It's also only 70% with respect to apple, microsoft, sandisk, and creative (possibly only in the US, the article mentioned best buy, but otherwise didn't specify) not the whole market
You're also assuming that people NOT buying apple products don't care about sound quality. You not only have no basis for that assumption, you forget the continuous complaints about apples horrid headphones (admittedly, it's slowed down in the past few years).
An informed customer who cared about sound quality wouldn't look to apple.
But go ahead and enjoy your toys. I'll go enjoy to the hundreds of extra dollars in my pocket while I listen to music in my inexpensive mp3 player with upgradeable storage, high-end sound, and more Music Related features.
The things he listed don't make a cheap iPod knock off better by adding a bunch of half-assed feature sets to the device.
First off, MP3 players existed before the iPod -- to call other MP3 players "knock offs" is ridiculous.
Second, the features he listed are, with one exception, essential to the operation of an mp3 player. Physical buttons for operation is *better* because it means the device can be operated without looking at the display. Not needing a separate application to put music on the device makes putting music on the device easier (iTunes is NOT known for being the paragon of ease-of-use -- quite the opposite in fact). Sound quality -- that's pretty important! IIRC, lots of people complained about quality of sound the headphones that came with the iPod provided when it first came out.
Removable storage is equally nice, not a useless feature at all. If you have a 1gb iPod, that's it, you're done. if it had removable storage, you wouldn't have to upgrade the player to add more storage (and hence more music). Begin able to swap out cards allows for innumerable additional benefits which I need not begin to list here.
If you want to talk about half-assed feature set -- or useless features as you're clock-radio toaster implies -- the iPod is the perfect example of that. Touch screen display, wifi, applications -- it's an MP3 player! In what conceivable way do those features make it a better music player? They have Absolutely Nothing to do with Playing Music!
By your own criteria, the iPod has a bunch of useless half-assed features!
Better, as far as the parent is concerned, means 'easier to use' and 'more extensible'. If you think 'harder to use' and 'less extensible' is better, well, I can't help you.
I'd like to point out, first, that A1 is not contentious at all. It's well established in mathematics and computer science circles. I did some looking, and couldn't find anywhere in the literature where A1 is critically addressed. I won't ask you to take my word for it, hence my example. I'll see what I can do to clear things up for you.
The example shows, quite clearly, that the function of the program is irrelevant to the semantic meaning attributed to it. Its function is the same regardless of the 1) the meaning of the symbols it receives/produces 2) the meaning you attribute to it
On the point "The input and output is not the program." You're right, but it is an excellent/definition/ of the program.
On "You only gave the input and output of the program, and guessed what the program itself would be." I did no guessing! The program didn't change. It produced the same outputs for the same inputs. The only thing that changed was the semantic meaning attributed to it. That was the entire point. The program operates completely independently of meaning (semantics are irrelevant).
On "The program would contain either an AND or NOR or perhaps the symbol DOWHATEVER." It doesn't matter how the program is implemented -- 6502 assembly or some mechanical contrivance.
Further, whatever instructions the program is composed of are equally devoid of semantic meaning as the program itself (one of the reasons I made it so primitive was to imply this.)
You argue that
For the program to actually do something, you have to have something which ascribes meaning to the instructions contained within it... If you have a machine which interprets + as subtraction and - as addition, vastly different things result
You're assuming that the machine requires semantics to function -- As I've demonstrated, quite clearly, it does not.
Remember, whatever operations the machine performs on some operands when it encounters the symbol + is only addition as far as YOU attribute it to addition.
Imagine a Turing machine that given a tape similar to "111001111b" upon halting leaves the tape "111111100b" I could call that addition (I could say that the TM performs addition) -- and I'd be right -- but calling it addition didn't have any influence over the machine while in operation. I could just as easily say that it sorts a b terminated list 1's and 0's The meaning which you attribute doesn't have any influence over the machines operation.
The machine requires no semantic context for the the operations performed or the symbols it manipulates.
I don't know how I can make this any clearer. Let's move on
Regarding axiom 2: Is a mind the hardware? He doesn't define a mind, unless it constitutes anything with understanding.
For our purposes, a general understanding of what is meant by mind is sufficient. If you need a better definition, Searle has written extensively about it.
If it's just a definition of the mind, I would suggest that the chineese room, containing both a person and a program, is a mind which happens to contain a different mind, not that no understanding exists within the room.
That has absolutely nothing to do with A2.
A2, as you should already be aware, is "minds have mental contents".
I can see how someone without the proper background could have difficulty understanding A1, but we can't even discuss A2 without assuming its truth!
A2 is (if you'll pardon the pun) self-evident.:)
Really, I'm surprised that you've spent so much time on A1 and A2 -- and they're not contentious at all. I hope that I was able to clarify this for you.
A3 is the contentious one -- though only on very rare occasions is it addressed directly. Usually it's the CR illustration (which is all about A3) that's attacked. I suspect it's because it's easier to misrepresent or pick-on insigni
Sorry, my bullshit detector can't handle much more of this.
Name this mysterious "Special Education" application that requires the CD to run on a windows machine, but that apparently is also available via the app store for the iPad.
Because reading an e-book while sitting on a train, riding to work, is a neck-breaking BITCH with a laptop but not an unpleasant experience with a tablet.
But hell on the eyes -- A dedicated e-ink e-book reader is a cheaper and better solution for that use case.
And with the included stand, I can watch a movie without having to hold the iPad.
On a stand, it's just a notebook without a keyboard -- If it's a "neck-breaking bitch" for reading an ebook wouldn't it be equally horrid to watch a video? Or are you no longer on the train in that scenario?
It's a MP3 player -- who cares about graphics and apps?
See, most people by an MP3 player to -- get this -- play music!
The features the parent listed are actually relevant to the primary function:
iTunes-less upload (much easier to get music on to the player)
removable flash (easy to swap out/expand storage)
actual buttons that you can feel and click (No need to look at the screen to pause/play/skip/adjust the volume, etc.)
better sound quality (subjective, but still relevant criteria)
Better, in the sense the parent is talking about (and made quite clear), means better at performing it's primary function.
For that, there are many far less expensive and *better* options than the iPod touch.
Is this really that difficult for you to understand?
they are intended to be a better fit for certain user tasks....
it doesn't do everything a desktop computer does, but then the people who buy it don't want it to do that. It's a different device for a different set of tasks.
Not to sound snarky, but what are those tasks, exactly? I don't know that I've ever seen anyone list them out -- what do you use it for? Is it better suited for that purpose than other devices? What are the real "killer" applications for tablets?
Don't get me wrong -- I want one! I just can't think of anything I would want to do with one.
That's my problem too. The only tablet that I think I could get any use out of would be RIM's PlayBook -- but only because it would work as a companion to my phone -- the larger display would make it easier to read and edit documents and spreadsheets on the go.
I doubt that the asking price will match what I'm willing to pay, given use I expect to get out of it.
I suspect that that tablet market will dry up once the novelty wears off.
there are MP3 players with better functionnality (iTunes-less upload, removable flash, actual buttons that you can feel and click...) and better sound quality that are much cheaper
Really? I can buy an MP3 player with 32GB of storage, 960x480 display, Wireless-N. equivalent graphic performance, quality 3rd party apps for less than the iPod Touch?
That's not what he said at all. He said that there are MP3 players with "better" functionality which are less expensive than an iPod Touch.
He's absolutely correct. You can buy MP3 players for MUCH less than an iPod Touch which have the features he listed.
His posited separation of programs and mind is proved given those axioms, but that separation is necessary for his axioms to make sense in the first place.
I think this is the flaw in your reasoning. The conclusion isn't that programs and minds are separate -- the conclusion (form the first three axioms) is that programs are not constitutive of nor sufficient for minds
Axiom 1 is trivial -- programs are syntactic. Any meaning you can attribute is irrelevant to the function of the program.
This is true for even trivial operations. Consider a program which accepts as input two 1-bit numbers (0,0) (0,1) (1,0) (1,1) and outputs (0) (0) (0) (1) respectively. If we attribute the meaning of 0 to be false and 1 to be true, this program is equivalent to a logical AND operation. If we assume that 0 is true and 1 is false our program no longer performs a logical AND, but a logical NOR.
The important point here is that the meaning that assigned is completely irrelevant to the function of the program. It didn't operate any differently than it did before, only your interpretation of it changed. It should be clear that the semantics of the symbols (which you have to attribute) have absolutely no influence over the program. That is, the program is purely syntactic.
Axiom 2 is equally trivial -- minds do have mental contents. I don't understand why you'd say that minds don't have semantics as this should be obvious -- just look at the role that "you" play in the example for A1. Given your qualifier (minds without programs), I'm willing to bet that this is related to the flaw listed earlier. If I've misunderstood you, please feel free to correct me.
Anyhow, the argument isn't circular as none of the axioms require the conclusion be true to be true themselves.
Whatever, feel free to enjoy your self-referential nattering.
You're the one who claims that Searle's argument is circular. It's not my fault that you can't show this to be the case. It's your absurd assertion -- Prove it.
Oh, you can't? I'm not surprised. The fact that the argument is NOT circular makes proving otherwise quite difficult.
I'm not talking about Serle, I'm talking about emergent phenomenon
Then why did you write:
Actually the Chinese Room thought experiment says nothing about the possibility of artificial sentience, only that we will have a hard time effectively defining and measuring it.
Because the Chinese Room is Searle's argument -- It addresses specifically the claims of Strong AI -- it says nothing about defining "artificial sentience". The only thing related to measuring it is Searle's objection to the Turing test as sufficient for it's purpose.
Slashdot Mirror
I think we're about to go chasing after rabbits. We're still on A1, which I'll restate just so I don't lose focus.
A1: Programs are formal (syntactic)
You are privileging minds to have some special version of meaning which a simple circuit can't have.
This really wasn't my intent. Though I can see how my example would lead you to this interpretation; that's my fault.
My problem was, I think, playing fast and loose with the word "meaning", which has connotations that can cause problems. When I say that a symbol has meaning, I mean that there is a relationship between the symbol and what it denotes.
Had I bothered to define that in my last post, I could have saved us both a lot of writing.
The circuit describes the meaning of the switch as on or off, by the nature of it's operation.
This may be closer to the point at which we disagree. In this case, you're still talking about semantics which are externally attributed (what we interpret the function of the switch to be) not about some *intrinsic* meaning.
We already agree that the meaning we attribute to a program (extrinsic meaning) has no effect. The point at which we disagree (hopefully not for long) and that I've pitifully tried to demonstrate, is that programs lack intrinsic meaning.
I don't want to chase the lamp thing too far -- It was, quite possibly, the worst example I could have used. (It confuses "it has" with "it gives", if that makes any sense, and takes the focus away from programs and syntax.)
Anyhow, we can agree (I think) that programs are defined syntactically, composed of formal symbol manipulations.
Formal symbol manipulations *by themselves* have no semantics. That is to say they have no *intrinsic* meaning. Any semantic content is extrinsic.
Some of the confusion, I think, stems from the semantic preserving properties that we're used to in formal systems (they would hardly be useful otherwise!). The semantics, of course, remain extrinsic.
I should point out that this post doesn't address A3, only A1. I've taken great care to avoid it. (When we get to A3, we won't care about the distinction between intrinsic and extrinsic semantics.)
I don't know that I've done the best job here, but I'm hoping that at least my definition helps to clarifies things.
If the appears to the user to exhibit intelligence, it is in fact intelligence.
The Turing test is not only highly specific, it's more than a little contentious.
Your particular interpretation of the TT has lead you to develop this strange method of assessment that lets you ascribe intelligence to virtually anything. Not only to my thermostat, but equally to a teaspoon or a cup of coffee.
Given your bizarre beliefs about what constitutes intelligence, I don't see any way we can come to any agreement or even mutual understanding.
That is my whole point. How long is a string?
One or more bytes for null terminated strings.
AI, like hardware and software development, observes Moore's law
This is completely unsubstantiated nonsense. I think we're done here.
he statement that somehow because the technology uses "lookup tables" or some other mundane algorithms as part of it's logic disqualifies it from being intelligent implies that it has to perform it's task in some "special" way.
Doesn't this (including the unquoted remainder of the paragraph and the paragraph that follows) broaden the definition of AI to include nearly all information processing? I'm not sure where you're drawing the line between what is AI and what is not.
It is entirely possible that the GP simply holds a narrower view, not that he believed intelligence to be magical.
it is still considered intelligent if it's behavior is intelligent.
This is a philosophical statement to which I'm not certain I can agree without a better definition of intelligent behavior. If the behavior of the amoeba in your previous example qualifies I can't agree as it makes the definition of what can be considered intelligent virtually meaningless. (If sensing and responding to an environment qualifies, then my thermostat must be considered intelligent.)
I won't ask you to define intelligence -- it's a very slippery word -- just what qualifies, in your opinion, as intelligent behavior.
I could go on but you get the point. AI is in use everywhere
That wasn't your original claim. The claim was that AI was advancing at an accelerated pace. (I don't know how you'd measure the rate) I don't disagree that what is now being termed AI is in use in many places.
You are under the commonly held impression that intelligence requires some special magical ingredient.
I don't see how you can come to this conclusion given what the GP has written.
AI is far more advanced than you realize and advancing at an accelerating pace
For this, a citation is necessary.
So it's you who keeps changing the definition of what AI is, it's not a lack of forward progress.
It depends on your perspective. Lot's of things that once would not have been considered AI now fall under that umbrella. Expert systems come immediately to mind. From that point of view, it's the field of AI that keeps changing its definition.
What the GP likely considers AI is the original goal of AI -- to create intelligent machines. To that end, the GP is correct, we're not any closer to solving that problem than we were in the 1950's.
On his claim that there is no "significant research" going on, he's WAY off base.
You know what WOULD kill an iPod would be a device that costs the same, has the same ease-of-use (or better, if they make an iTunes killer), has superior audio and is designed with a better UI. Since none of these have come to fruition, the iPod is still #1.
You're under the impression that the iPod is successful because it has the nicest UI, is the easiest to use, and has the best audio.
How wrong you are! Apple hasn't had the best or easiest to use player in years. What apple has, is the *myth* of being both the best and the easiest to use.
Just an example, when my wife was in college she and her roommate both bought mp3 players. She got a $50 RCA Lyra and her room mate bought a 3G iPod Classic.
My wife had her music collection on the player in about 10 minutes. It took her roommate three hours and a call to me to get anything on to her iPod. Now, I can't say that the Lyra was "better" -- only that it was much easier to use. (The roommate ultimately ended up returning the iPod and picking up a Lyra.)
Other players, like the Creative Zen series, are (and were) known for having well-designed UI's -- often beating apple on both features and easy of use. Of course, it's hard to get the message out when "everyone knows" that apple's products are easiest to use.
Apple myths have an amazing immunity. Early iPods were plagued with HD reliability problems yet "everyone knows" that apple builds lasting products of the highest quality. There have been several class-action lawsuits resulting from the shoddy iPod line (bad batteries, insta-scratch displays), yet "everyone knows" that you should pay extra for Apple because you're paying for quality and reliability.
Apple maintains it's market share in the portable digital music player market by virtue of the myths created around it -- Not because their products are easier to use, of higher quality, more innovative, or nicest-looking.
I'm really struggling with putting the concepts in writing in such a way that is clear and unambiguous
You and me both :)
The input and output is NOT a useful definition of a program. The output depends on too many things besides the input. Your example was a definition of a black box, not a program.
Can you think of a better way to define a program other than by it's I/O? It seems perfectly reasonable to me, as the implementation details aren't really relevant -- as I mentioned before, it doesn't matter if it's 6502 assembly or a mechanical contrivance. (I'd rather not spend too much time on this, as it's completely beside the point.)
Moving on, it seems the point where we differ is on the meaning of "meaning". You write:
The action a turing machine takes in response to a specific input is the meaning a turing machine ascribes to that input.
This may be a result of the loose way we use language when we talk about computers. We often make statements that anthropomorphize the computer like "this command tells the computer too..." or "the computer sees that as ..." But statements like that don't have any reality to them. We don't actually mean that the computer "understands" in the way you or I understand.
Part of the reason I chose such a simple example is to illustrate that even at the lowest level, meaning isn't relevant. By showing meaning to be irrelevant to function, I was hoping to show that the only meaning involved is that which is externally attributed -- there is no internal attribution of meaning.
Let's start with a simple circuit just a battery, switch, and lamp arranged in such a way that moving the switch to the on position causes the circuit to complete and the lamp to light.
It's easy to see in this example, that meaning is irrelevant to the function of the circuit. The circuit doesn't have any meaning assigned to the "on" or "off" position of it's switch, even though we might be tempted to make statements like "flipping the switch tells the circuit to turn the light on" we don't intend to imply any understanding on the part of the circuit.
Let's add a second switch to our circuit. We've now built an AND gate (or not, see the first example). Still, you can see that meaning is not relevant to the circuit.
We can expand our circuit further, adding electro-mechanical relays in such a way as to build a half-adder, with our switches setting the input. Still, the circuit requires no meaning be attributed to any of its parts.
Now let's go nuts and build an adder and subtracter. Let's create two "registers" (just a set of switches) which serve as input to them both, and a third 1-bit register (a switch and a few relays) which selects which circuit gets its output displayed on our lamps (i.e. which instruction our "computer" will execute).
We can make this more interesting, by using some paper tape to set the switches we were manually setting before. (imagine each manual switch replaced with a brush-contact -- paper in the way breaks the circuit, equivalent to turning the switch off, a hole allows contact, equivalent to turning the switch on)
With a hand-crank serving as our clock, winding the tape through our machine will execute whatever add and subtract instructions that happen to be recorded on the tape. In effect, the tape will contain a (very uninteresting) program composed of add and subtract instructions with their associated operands.
Building this from the bottom up, we can see that in no way must the machine attribute any 'meaning' to its instructions.
It should be clear that further increasing the complexity of our machine would not change this.
You write:
only the meanings the computer ascribes to them
I hope I've made obvious that the computer at no point need ascribe meaning to it's instructions any more than our first circuit need ascribe meaning to the position of it's switch.
I hope that clears that bit up. Do you have any remaining objections to the axioms?
Thanks for being the first person to actually list and explain how they've made use of their tablet. It's very helpful.
That said, if your use is representative of the common case, it's easy to see why others think the device is "useless" -- at least at the price it's set at.
Just a few reasons why an iPad wouldn't suit our household from the examples you give:
The Train: I don't take one with any sort of regularity. I could, perhaps, see it for the wifi on some flights, but I've only been on a few that offered it and I've never felt compelled to connect to it.
Recipes: My wife would't dream of bringing a computer (tablet or otherwise) into the kitchen for fear that it would suffer from accidental spills. She tends to writes recipes down on note cards, so that she has a record and so that she doesn't have to look them up again later. (she discards or trades the ones she doesn't care for.) I don't deny that you find it useful, I just can't see it working for us.
Movies in bed: Would get me kicked out of the bed!
The only tablet that I think I could get any use out of is RIM's Playbook -- but only because it can work as an extension of my phone (which, with the larger display, should make working with documents and spreadsheets easier). Still, I doubt that the use I'll get out of it will be worth the price.
Yes, you'd be surprised whats out there.
Not really (it's actually an area I'm familiar with). Though I still want to know what application you're talking about that requires a CD under windows but is also available on the iPad.
I can't let this bit go, however:
the superior apple user experience
This is, undoubtedly, a myth. Neither Apple's UI or overall user experience seems to be superior to equivalent products from other vendors. It's really quite inferior in a number of their products -- I'm sure you can think of a few examples (I dare not upset the apple-cart by listing a few of my own!)
Apple has a long history of myths. You might remember hearing that "Macs are better for graphics" over the past 20 years or so -- even though there was absolutely no reason (at any point) to believe that was the case.
Of course all his conclusions follow from his premises: he's assuming the very thing he's trying to prove!
You keep saying that, but provide no argument in support of it. Name the premise that is dependent on the the conclusion!
(You won't be able to, as none of his premises are dependent on the conclusion!)
I see that you tried, but inexplicably managed to fail. All you need to say is A1, A2, A3, or A4 -- is that so hard? This is as close as you got:
The one that is also the conclusion: that it takes special "causative powers" to cause a consciousness
And which one is that again? Oh, right, it doesn't exist. Big surprise.
Anyhow, the closest premise to the "causative power" bit is A4 -- which is NOT a conclusion:
A4: Brains cause minds.
Is this the one you object to? Well, okay. Does that mean you agree with A1, A2, A3 and C1?
If we can agree to that point, we can discuss A4 and C2, C3, and C4.
With 70%+ market share after a decade, I think most people would disagree,,,,
Ummm... Being popular does not mean that it's easier to use! Also, your 70% figure is not iPod touch -- which is the one with graphics and apps, it's all of apples music player products.
It's also only 70% with respect to apple, microsoft, sandisk, and creative (possibly only in the US, the article mentioned best buy, but otherwise didn't specify) not the whole market
You're also assuming that people NOT buying apple products don't care about sound quality. You not only have no basis for that assumption, you forget the continuous complaints about apples horrid headphones (admittedly, it's slowed down in the past few years).
An informed customer who cared about sound quality wouldn't look to apple.
But go ahead and enjoy your toys. I'll go enjoy to the hundreds of extra dollars in my pocket while I listen to music in my inexpensive mp3 player with upgradeable storage, high-end sound, and more Music Related features.
The things he listed don't make a cheap iPod knock off better by adding a bunch of half-assed feature sets to the device.
First off, MP3 players existed before the iPod -- to call other MP3 players "knock offs" is ridiculous.
Second, the features he listed are, with one exception, essential to the operation of an mp3 player. Physical buttons for operation is *better* because it means the device can be operated without looking at the display. Not needing a separate application to put music on the device makes putting music on the device easier (iTunes is NOT known for being the paragon of ease-of-use -- quite the opposite in fact). Sound quality -- that's pretty important! IIRC, lots of people complained about quality of sound the headphones that came with the iPod provided when it first came out.
Removable storage is equally nice, not a useless feature at all. If you have a 1gb iPod, that's it, you're done. if it had removable storage, you wouldn't have to upgrade the player to add more storage (and hence more music). Begin able to swap out cards allows for innumerable additional benefits which I need not begin to list here.
If you want to talk about half-assed feature set -- or useless features as you're clock-radio toaster implies -- the iPod is the perfect example of that. Touch screen display, wifi, applications -- it's an MP3 player! In what conceivable way do those features make it a better music player? They have Absolutely Nothing to do with Playing Music!
By your own criteria, the iPod has a bunch of useless half-assed features!
Better, as far as the parent is concerned, means 'easier to use' and 'more extensible'. If you think 'harder to use' and 'less extensible' is better, well, I can't help you.
I'd like to point out, first, that A1 is not contentious at all. It's well established in mathematics and computer science circles. I did some looking, and couldn't find anywhere in the literature where A1 is critically addressed. I won't ask you to take my word for it, hence my example. I'll see what I can do to clear things up for you.
The example shows, quite clearly, that the function of the program is irrelevant to the semantic meaning attributed to it. Its function is the same regardless of the 1) the meaning of the symbols it receives/produces 2) the meaning you attribute to it
On the point "The input and output is not the program." You're right, but it is an excellent /definition/ of the program.
On "You only gave the input and output of the program, and guessed what the program itself would be." I did no guessing! The program didn't change. It produced the same outputs for the same inputs. The only thing that changed was the semantic meaning attributed to it. That was the entire point. The program operates completely independently of meaning (semantics are irrelevant).
On "The program would contain either an AND or NOR or perhaps the symbol DOWHATEVER." It doesn't matter how the program is implemented -- 6502 assembly or some mechanical contrivance.
Further, whatever instructions the program is composed of are equally devoid of semantic meaning as the program itself (one of the reasons I made it so primitive was to imply this.)
You argue that
For the program to actually do something, you have to have something which ascribes meaning to the instructions contained within it ...
If you have a machine which interprets + as subtraction and - as addition, vastly different things result
You're assuming that the machine requires semantics to function -- As I've demonstrated, quite clearly, it does not.
Remember, whatever operations the machine performs on some operands when it encounters the symbol + is only addition as far as YOU attribute it to addition.
Imagine a Turing machine that given a tape similar to "111001111b" upon halting leaves the tape "111111100b" I could call that addition (I could say that the TM performs addition) -- and I'd be right -- but calling it addition didn't have any influence over the machine while in operation. I could just as easily say that it sorts a b terminated list 1's and 0's The meaning which you attribute doesn't have any influence over the machines operation.
The machine requires no semantic context for the the operations performed or the symbols it manipulates.
I don't know how I can make this any clearer. Let's move on
Regarding axiom 2: Is a mind the hardware? He doesn't define a mind, unless it constitutes anything with understanding.
For our purposes, a general understanding of what is meant by mind is sufficient. If you need a better definition, Searle has written extensively about it.
If it's just a definition of the mind, I would suggest that the chineese room, containing both a person and a program, is a mind which happens to contain a different mind, not that no understanding exists within the room.
That has absolutely nothing to do with A2.
A2, as you should already be aware, is "minds have mental contents".
I can see how someone without the proper background could have difficulty understanding A1, but we can't even discuss A2 without assuming its truth!
A2 is (if you'll pardon the pun) self-evident. :)
Really, I'm surprised that you've spent so much time on A1 and A2 -- and they're not contentious at all. I hope that I was able to clarify this for you.
A3 is the contentious one -- though only on very rare occasions is it addressed directly. Usually it's the CR illustration (which is all about A3) that's attacked. I suspect it's because it's easier to misrepresent or pick-on insigni
Sorry, my bullshit detector can't handle much more of this.
Name this mysterious "Special Education" application that requires the CD to run on a windows machine, but that apparently is also available via the app store for the iPad.
Because reading an e-book while sitting on a train, riding to work, is a neck-breaking BITCH with a laptop but not an unpleasant experience with a tablet.
But hell on the eyes -- A dedicated e-ink e-book reader is a cheaper and better solution for that use case.
And with the included stand, I can watch a movie without having to hold the iPad.
On a stand, it's just a notebook without a keyboard -- If it's a "neck-breaking bitch" for reading an ebook wouldn't it be equally horrid to watch a video? Or are you no longer on the train in that scenario?
See, most people by an MP3 player to -- get this -- play music!
The features the parent listed are actually relevant to the primary function:
Better, in the sense the parent is talking about (and made quite clear), means better at performing it's primary function. For that, there are many far less expensive and *better* options than the iPod touch. Is this really that difficult for you to understand?
and does your $800 laptop have a touch-screen?
Why would you want a touch-screen on a laptop? (or on anything else, for that matter)
A touch screen would just encourage people to poke the display with their nasty cheeto-stained fingers. No thanks.
they are intended to be a better fit for certain user tasks. ...
it doesn't do everything a desktop computer does, but then the people who buy it don't want it to do that. It's a different device for a different set of tasks.
Not to sound snarky, but what are those tasks, exactly? I don't know that I've ever seen anyone list them out -- what do you use it for? Is it better suited for that purpose than other devices? What are the real "killer" applications for tablets?
Special Education stuff
That happens to be in the app store?
Don't get me wrong -- I want one! I just can't think of anything I would want to do with one.
That's my problem too. The only tablet that I think I could get any use out of would be RIM's PlayBook -- but only because it would work as a companion to my phone -- the larger display would make it easier to read and edit documents and spreadsheets on the go.
I doubt that the asking price will match what I'm willing to pay, given use I expect to get out of it.
I suspect that that tablet market will dry up once the novelty wears off.
there are MP3 players with better functionnality (iTunes-less upload, removable flash, actual buttons that you can feel and click...) and better sound quality that are much cheaper
Really? I can buy an MP3 player with 32GB of storage, 960x480 display, Wireless-N. equivalent graphic performance, quality 3rd party apps for less than the iPod Touch?
That's not what he said at all. He said that there are MP3 players with "better" functionality which are less expensive than an iPod Touch.
He's absolutely correct. You can buy MP3 players for MUCH less than an iPod Touch which have the features he listed.
Learn to read!
His posited separation of programs and mind is proved given those axioms, but that separation is necessary for his axioms to make sense in the first place.
I think this is the flaw in your reasoning. The conclusion isn't that programs and minds are separate -- the conclusion (form the first three axioms) is that programs are not constitutive of nor sufficient for minds
Axiom 1 is trivial -- programs are syntactic. Any meaning you can attribute is irrelevant to the function of the program.
This is true for even trivial operations. Consider a program which accepts as input two 1-bit numbers (0,0) (0,1) (1,0) (1,1) and outputs (0) (0) (0) (1) respectively. If we attribute the meaning of 0 to be false and 1 to be true, this program is equivalent to a logical AND operation. If we assume that 0 is true and 1 is false our program no longer performs a logical AND, but a logical NOR.
The important point here is that the meaning that assigned is completely irrelevant to the function of the program. It didn't operate any differently than it did before, only your interpretation of it changed. It should be clear that the semantics of the symbols (which you have to attribute) have absolutely no influence over the program. That is, the program is purely syntactic.
Axiom 2 is equally trivial -- minds do have mental contents. I don't understand why you'd say that minds don't have semantics as this should be obvious -- just look at the role that "you" play in the example for A1. Given your qualifier (minds without programs), I'm willing to bet that this is related to the flaw listed earlier. If I've misunderstood you, please feel free to correct me.
Anyhow, the argument isn't circular as none of the axioms require the conclusion be true to be true themselves.
Whatever, feel free to enjoy your self-referential nattering.
You're the one who claims that Searle's argument is circular. It's not my fault that you can't show this to be the case. It's your absurd assertion -- Prove it.
Oh, you can't? I'm not surprised. The fact that the argument is NOT circular makes proving otherwise quite difficult.
You're pathetic.
I'm not talking about Serle, I'm talking about emergent phenomenon
Then why did you write:
Actually the Chinese Room thought experiment says nothing about the possibility of artificial sentience, only that we will have a hard time effectively defining and measuring it.
Because the Chinese Room is Searle's argument -- It addresses specifically the claims of Strong AI -- it says nothing about defining "artificial sentience". The only thing related to measuring it is Searle's objection to the Turing test as sufficient for it's purpose.
The "Chinese Room" argument to which you linked doesn't prove that computer mind is impossible
His conclusions follow for his premises. Which one of Searle's axioms do you object to?
In short, Searle is an idiot, his "Chinese Room" argument is nonsensical
Many have tried, yet none have been successful in toppling Searle.
So, I'll ask again: Which one of Searle's axioms do you object to?
All of his conclusions logically follow from his premises -- you've no alternative but to object to one of his four axioms. Which one is it?