It is fascinating. Particularly the "characters" problem you mentioned already falls under the realm of natural language processing, and a very specific problem at that (context recognition).

I concede that defining "logical leap" is mostly a semantic argument -- I personally consider a syllogism (A->B, B->C therefore A->C) as a "logical leap", but it is one that can be perfectly captured by inference engines. Furthermore, one may argue that some aspects of "human intuition" might actually be... illogical. :)

And to answer your hard-coding concern: it ISN'T necessarily needed to hard-code more rules -- that is just a matter of making the system perform better given similar questions in the future. Like I said earlier, modern inference engines do have conflict-resolution mechanisms, as it is very possible in large knowledge bases that many rules are applicable to a particular case. One of the main research areas, I think, is for conflict-resolution mechanisms that can correctly recognize context as opposed to some "simplistic" or "hard-coded" statistical formula. (But what is the difference, really? One may argue that "context" is just more parameters to the formula.) Sure, humans are still quite ahead of A.I. in this regard, but the Mycin system is actually notable in that it was able to diagnose diseases MUCH better than humans can -- and that's in the 1970s.

If by a "highly-specified" domain you mean a domain with its knowledge base and rule base completely specified, yes, inference engines still rely on those. But that IS the general case. Not even a human would be able to solve your example without knowing what movies, actors, novels, pages, characters, magazines and singers are, and what the linking concepts are (i.e., facts encoded in the knowledge base such as "element(actors, movie)", "element(characters, novel)", "contains(novel, pages)"). The real problem is the modeling -- if you say "contains(actors, movie)" instead, the A.I. will backfire and give you "novel : pages", and as knowledge bases become more complex, the modeling might actually make sense for some other cases (other IQ test questions) and then it's up to the inference engine to realize the conflicts and make intelligent guesses using some principled rule set (e.g., Mycin's certainty factors model). On the other hand you can solve the problem at the modeling side -- the whole field of knowledge engineering is devoted to that.

In any case, I'm merely addressing the incorrect statement of the other poster: that there has "NEVER been, in the whole field of AI, a logical leap to join two abstract concepts". It is just strikingly obvious that the poster had never heard of inference engines, or for that matter, other AI concepts outside of genetic algorithms and brute-force matching. :)

Correct. Grandparent post is symptomatic of someone who has not actually studied A.I. or one of its subfields and assumes that Kinect is the "nearest thing we had recently to intelligence." The problem of joining two abstract concepts is long ago solved by inference engines.

Never bought anything in the Android app store. Not even the $1 apps. I did get a lot of ad-supported apps, but I'm not anymore willing to pay for ad-free versions of those since I already got what I need and the ads don't bother me.

On the other hand, I got a lot of old GameBoy games on my 3DS, each purchased at the eShop for at least $2.99. All of those are games I played in my childhood, but couldn't (re-)get in any other way nowadays. Funnily enough, I never bought any of the cheaper games in the eShop, just because I haven't tried them and I don't know how they fare against the classics.

Bottom line, I'm willing to pay more for things when I know exactly what I'm going to get, and I'm sure I can't get them anywhere else for cheaper/free. I'm sure this is in line with the latte analogy. (Btw, I don't go for Starbucks, I just visit the nearest McCafe for my hot chocolate fix since I'm satisfied with the quality and pay much less.)

Just don't complain if you can't do your grocery shopping that whole week.

Exactly. Which is why we need to give the progressive theologians, which John Haught is a representative of, have their say. It's not a full attempt at declaring all science as part of his religion, but is certainly getting there much faster than the laughable ad hominem attempts of Coyne. (really, I saw what he has to say, and it's all very unconstructive at best, and he's a pitifully and hopelessly biased representative of true atheism at worst)

From what I gather from the debate, no attempt was made by Haught to dictate what science really is. All it is is an attempt to unify current understanding of science with religion, which you are in favor of. Coyne took a LOT of his statements out of context, somehow put them together to paint a caricature of Haught as an ultra-conservative that deserves flogging for even attempting to unify his beliefs, then appealed to the Slashdot crowd by misrepresenting his blatant attack as 'the scientific way'.

Pardon me for being blunt, but YOU were the one who changed the subject. My point was originally (and as evidenced by this thread itself):

Then you replied by saying:

Which was basically an attack on JUST a sub-argument -- but do note that in fact I conceded FROM THE VERY START that there are objections, and THEN there are objections to the objections (read my first post again if you dare). Hence I linked you to defenses in the form of revisions to the wager, which proves that it's NOT dead yet. But then now you're arguing (and I quote):

So now you want me to stick to the original Pascal wager alone, despite the fact that I already gave you the on-going studies that continue the argument by extending it? (And conveniently so, because it wins you a pissing match by attacking the straw man?)

Yes it is. Remember the original claim I refuted: "Honestly, if you attempt to justify religion on anything resembling logic, you lose." In this case, showing that there is a debate on this subject with no declared "losers" is a valid refutal, and showing that the debate is indeed going on in scholarly journals is proof to support that refutal. All you did is to show a non-peer-reviewed personal vlog on the matter and then claim that religion has irrevocably lost, while it has not addressed the said revisions to the wager.

If you can't argue with my original point (hint: not the wager itself, if it is still not so absolutely clear to your shortcut-taking mind), say so and stop wasting my time.

Uh, tell you you're an obvious troll god? No, really, that god already failed the principle of sufficient reason (Leibniz) from the outset, so why would I even go to Pascalian logic for that? (Or are you trying to attack just the Pascalian view, which is completely missing the point?)

I'm not saying there aren't any problems with Pascal's wager. I'm saying that it PROVES that you can think logically about religion.

If you are interested, all your other arguments are actually addressed by proponents of the wager too. There's an ecumenical interpretation of the wager (see here) where believing in any OTHER god is acceptable as long as it has the essential characteristics of Pascal's god, and the original argumentation of Pascal's wager still holds. (Specifically, your what-if about a jealous God does not actually disprove the wager, it is only dismissed as a non-genuine option. That is, if there IS a god who really punishes people for worshiping the competition, that god is not a meritorious god either, and therefore isn't Pascal's god.)

That said, even if you believe it's a glaringly gratuitous mechanism for self-justification, it's STILL a more valid justification than other "childish" self-justifications I've seen from some (not all) atheists, e.g., "if your (judeo-christian) god is merciful, why does he punish you and countless others for not believing?" And anyway, I started believing in it after going through an atheistic "science is the only valid philosophy" phase myself, so I'm not so sure it's valid to say it "only works for people who have already made up their minds".

That's even cuter. How did YOU personally rebut MY argument by posting some random guy on youtube as opposed to peer-reviewed journal articles? You know, the one where I'm just saying that the debate is alive and well?

The mere fact that you are arguing with me over this proves there IS a debate. Even though you're prone to dismiss my arguments via your own misreading of the third article (which I gave you so that you can avoid having to cross the paywall). But sure, attack that summary article by equating:

with:

(Oooh. Complete disregard of the truth? The truth being? Oh, your own argument. Flimsily supported by 2 youtube videos by the same guy, attacking sub-articles of Pascal's wager which are not actually concerned with the validity of the main argument but with the implications of the argument to organized religions as opposed to plain faith. Good job begging the question there, by the way.)

But never mind that. All I'm saying is that actual scholarly debate exists (and is very much alive) on this subject. Arguing for religion logically is most certainly NOT a dead end. Now put down that elitism hat and stop attacking me with your kettle.

No, it hasn't.

Refutal: Pascal's wager exists. It has not, by any practical standard, "lost" the debate. (Yes, there are claims that it is a fallacy, but there are also claims that THOSE claims are fallacious. Not much time to explain the subtleties, but the point remains that the wager IS justifiable and that worthwhile debate whether "faith can be a logical construct" can stem out of it.)

So, is your claim -- justifying religion is a dead end -- borne from actual logic, or from plain dismissive elitism on your part?

The 'worse is better' philosophy is more an argument about simplicity rather than price ("worse" functionality correlates to "better" practicality). Some of the best patents are actually for simple inventions used to do something novel. The novelty in UNIX and C isn't price (i.e., cheap/free), but portability (they're VERY simple designs yet powerful enough to write a self-compiler) -- and that made it better than the alternatives such as Algol. Not just marginally, it really WAS much better because hardware was developing so fast at the time (birth of personal computing, remember?) and Algol simply couldn't keep up.

Ritchie definitely could have made a large profit from the whole shebang if he wanted to. He didn't.

I thought he just said hello to it... :)

RIP

What's ambiguous? He's posting on Slashdot, so it's obviously the girlfriend.