I don’t know what to tell you folks, with regards to the missing Edit button. I’m wondering if someone has misbehaved, and lost everyone it’s use, and we just weren’t notified. Dunno. Again, I’ll ask Erwin when I get a chance.

Jan Bogaerts - Mar 20, 2011:

(missing the edit button)
But Hans also has a point that there is a mismatch. I think it’s an ‘and’ story instead of ‘or’.

I do agree that the ‘mismatch’ is not the all encompassing answer, and other issues are clouding the debate as well

Hans Peter Willems - Mar 20, 2011:

C R Hunt - Mar 20, 2011:

I have yet to see one actual journal paper get referenced

Me too, by anyone here (including yourself). This is a discussion forum, not a research paper. We all post links to stuff, not posting exact references. Get real.

What? I just did not a few posts up with regard to g-loading.

Hans Peter Willems - Mar 20, 2011:

BTW, I have pointed in several discussions that relate to this issue, to David Chalmers (just about the most referenced opponent of the Chinese Room argument), his work on ‘Qualia’, the related scientific discussion on the ‘symbol grounding problem’ and the discussion around the ‘easy and hard problems in AI’.

Yes, but you do see the difference between citing an entire research area (‘symbol grounding problem’) and a specific research endeavor that favors your view, right?

Hans Peter Willems - Mar 20, 2011:

So let’s just agree to disagree. I will write my upcoming research paper on the Chinese room (with all proper references in place of course), and you will…. well, we have to see.

Agreed. Out of curiosity, what journal are you aiming for your paper to find a home in? And who is your collaborator?

C R Hunt - Mar 20, 2011:

Out of curiosity, what journal are you aiming for your paper to find a home in? And who is your collaborator?

No idea yet to which paper(s) I’m going to submit it. Having it published in a journal might give ‘some’ credibility, but we now have many other (Internet-) ways to get such a paper reviewed by other researchers in the field (as that is what I want to happen of course). It will be available on-line for those interested, but I will also send it to certain renowned scientists in the field.

As for collaboration; I can’t name him (for reasons I won’t mention now) but I’m working with someone who holds a doctorate degree and has authored/co-authored several papers that where published by IEEE Computer Society, John Wiley & Sons, Kluwer Academic Publishers, Elsevier Science Inc, to name a few. He is now working as Lecturer at the University of applied sciences here in my region. After I gave him a short presentation on my current research, he requested to join my project. I will write the paper on ‘the Chinese room’ by myself, but the upcoming paper on ‘Core-concepts’ will probably be done in collaboration with him. He brings specific knowledge to the project, like ‘invariant representations’ and ‘analogies’.

I think it would be in your interest to aim for a peer-reviewed journal. Definitely the best avenue to have your research gain exposure, and be considered seriously. (I’m coming from a natural science background here, so the “etiquette” so-to-speak may be different in computer science, but this is my experience.) There are certainly many to choose from.

I find it highly strange that you can’t name the guy you’re working with. And that you can’t even divulge the reason. But I’ll reserve judgment on the oddness of this until the reasons are made clear later. What’s the time scale on this secrecy?

C R Hunt - Mar 21, 2011:

I find it highly strange that you can’t name the guy you’re working with.

It’s not strange at all; he asked me to, because of his ‘current situation’, and because of that I can’t say any more on the subject. As soon as we publish a joint paper, his name will be on it.

Besides that, it’s not really that important. He is working with me, on my project, not the other way around. I didn’t agree for him to join, as to gain some kind of credibility for my project, I did because he brings some very valuable knowledge, insights and experience to the project.

Credibility has to come from the quality of the content I’m generating, not because someone’s name is on it. And let’s be honest, IF I’m right (which has to be proven yet of course, I still might be completely wrong) then the AI-research field can hardly afford to ignore that.

Hans Peter Willems - Mar 21, 2011:

Credibility has to come from the quality of the content I’m generating, not because someone’s name is on it. And let’s be honest, IF I’m right (which has to be proven yet of course, I still might be completely wrong) then the AI-research field can hardly afford to ignore that.

You would think so, eh? Except if nobody established in the field is associated with the work, it doesn’t tend to see the light of day (get published). And then nobody reads it. And then nobody knows what they’re missing, so to speak. I’m not talking about how things should be, I talking about the way they are. Nobody said life was the f-word.

(Fair, that is. )

C R Hunt - Mar 22, 2011:

You would think so, eh? Except if nobody established in the field is associated with the work, it doesn’t tend to see the light of day (get published). And then nobody reads it.

I think this is a very limited view on the world. First of all, lots of books get published (and bought, and being read) that are NOT written by, or associated to, academics. Second, the Internet has given us ‘viral marketing’; stuff can get into the spotlights in totally different ways then being published in traditional ways.

Take a look at this forum: we don’t link only to academic works, we link to anything and everything that might seem useful in light of the discussion at hand.

If only I put my stuff on-line on my own website and start throwing links around on a few forums, it will get at least some exposure. Take a look at Arthur’s work; It’s not academically published and you may agree or disagree with him, but many people are in fact aware of his work being in existence at least.

However, how I’m going to publish my work, who I work with, and the validity of my ideas and views in that regard, are not the scope of this discussion.

I think this is a very limited view on the world. First of all, lots of books get published (and bought, and being read) that are NOT written by, or associated to, academics. Second, the Internet has given us ‘viral marketing’; stuff can get into the spotlights in totally different ways then being published in traditional ways.

Though I would tend to agree on the ‘limited view’, I’d direct it to the community in general. I think CR was referring to publishing in an academic journal with respect to the collaborator. Sad as it is, it’s a simple fact that, if you are new to a field, you need to know someone to get published.
From the other side of the story, of course, there’s this thing of ‘mass production’. If I had to review all those papers, I’d also put in some sort of criteria as an initial filter.

Jan Bogaerts - Mar 22, 2011:

Sad as it is, it’s a simple fact that, if you are new to a field, you need to know someone to get published.

Jan, I do agree with this, contacts make the world go round. But those contacts are not necessarily academic, as many books on AI have been published without being written or co-written by academics. The AI-research field is not ‘only’ an academic space.

Also, I see no problem in writing a paper, put it online ‘somewhere’, have it picked up by ‘someone’ (might be an academic) and then being asked to submit it for publishing. What I mean by that is this; having someone to endorse your paper for publication can very well happen AFTER you write it. It has to be REALLY good of course but that stands without reason.

I have always had a problem with the Chinese Room argument. I have mulled on it a lot, and although someone else has probably brought up similar concerns, I couldn’t easily find it in any of the discussions so let me discuss it here. Since a lot of AI discussion revolves around this thought exercise, here is my concern;

It boils down to: “The problem is the problem.”

I believe the Chinese Room is a False Dilemma. By answering the question, “Does the entity in the room understand Chinese?” you buy into the fact that it is possible for this test to answer that question.
I think the question actually has 4 possible results:
The entity in the room understands Chinese. (positive)
The entity in the room does not understand Chinese. (negative)
The entity in the room understands Chinese, but responds incorrectly. (false negative)
The entity in the room does not understand Chinese but responds correctly. (false positive)

Searle (1999) summarized the Chinese Room argument concisely:

Imagine a native English speaker who knows no Chinese locked in a room full of boxes of Chinese symbols (a data base) together with a book of instructions for manipulating the symbols (the program). Imagine that people outside the room send in other Chinese symbols which, unknown to the person in the room, are questions in Chinese (the input). And imagine that by following the instructions in the program the man in the room is able to pass out Chinese symbols which are correct answers to the questions (the output). The program enables the person in the room to pass the Turing Test for understanding Chinese but he does not understand a word of Chinese.

Let me rephrase the problem and see if you would say the entity in the room “understands”.

Imagine in the room is a native English speaker. Imagine that people outside the room send in questions on any topic. And imagine that all the answers to the questions the native English speaker returns are wrong (the output).

The native English speaker in the room fails the test. Does he understand English? Yes, but he has failed the test.

Imagine a native English speaker who knows no binary locked in a room with a list of ASCII binary symbols together with a book of instructions for manipulating the symbols. Imagine that a computer outside the room send in other binary symbols which, unknown to the person in the room, are questions in binary (the input). And imagine that the man in the room is able to pass out binary symbols which are correct answers to the questions (the output).

The instructions enables the person in the room to pass the Turing Test for understanding binary but he does not understand a word of binary. Or does he? Is he able to understand binary? What if he did not need to look up the symbols? Does he understand binary then? If no one can understand binary, how did we invent computers?

If the test is not valid then: “The problem, is the problem.”

If the argument states the answers in the output are correct, then the output could not be incorrect.  If the argument says the instructions in the program determine the output, then the agent of execution of the instructions is effectively removed from the “results” other than how effectively/efficiently the agent processes the instructions. If required to make judgement on the agent depending upon the output, only the quality of computation is available.

The conclusion sought is whether the agent is the program, that is, is what the program does the same as what the agent does?  The misrepresentation is the agent having the ability without the program (the person knows Chinese.)  The other misrepresentation is the program having the ability without the agent (the system knows Chinese.)  Neither is right.  This is the nature of the mind/body dilemma.  Where are thoughts, in the brain?  Or does it require language (to reason?)

The whole thing is worse now that we run programs on computers. For some, machines can’t be intelligent, but programs can. Processing data into information is more than the processing because the program knows what to do.

Searle asks, how can the program know anything?

If you seek to find the agent, you must not fall into the pit of the program or the processing dilemma.  You need to remove the constraints of the task and explore the quality of the output.  Only then can you find that there might be something more to speaking Chinese that the processing of a program.  An agent who speaks Chinese may not always answer the question it is given or may answer in such a way that the true nature of the agent is revealed.