[vidarh]

Maybe? I don't think it's a very interesting question, because nobody has a particularly strong emotional reason to care whether we call it swimming or something else.

But people do have entire belief systems built around humans having a special position in the world.

[jddj]

I think if you revisited the Dijkstra quote you might find that you dismissed it a bit too easily.

In a way it's saying the same thing that you and I have said, just a bit more generally and eloquently.

In another way it's wisely asking us, "so what?".

[vidarh]

I did not dismiss it. I think he made a good argument when he made it, about computing at the time.

What I disagreed with was not Dijkstra, but applying it to AI today given that whether or not you think that there shouldn't be anything interesting about it even with AI, the social context means that there very clearly is whether or not you think peoples beliefs around it are reasonable.

To rephrase: At the time, computers unambiguously did not in any way get even close to the line, just like a sub gets nowhere close to replicating swimming. That made the question ludicrous and the comparison a good illustration.

Today there is ambiguity with computers, but no more ambiguity with respect to subs, and that ambiguity is such that it matters deeply to a lot of people in a way the question of subs swimming never will even if you close that gap. As such the comparison has lost its utility.

[jddj]

Interesting.

When I read it I assume that it is a retort to very similar hysterical sentiments to the ones we see today. I don't read it as a sub swimming being as ridiculous as a computer thinking, but rather that the question "is that machine swimming or not?", much like in the robot walking example, isn't particularly valuable.

We don't break a sweat when we say that a robot is walking, because we don't care about walking. We haven't internalised it as the final frontier of humanness. I read the quote as saying that whether a computer can think or not should be as pointless a question as whether or not a robot can "walk".

I'm intrigued enough now to try to hunt down the context.

[vidarh]

I don't think any of what you've written is wrong in terms of how to interpret it,

I just think the context has changed enough that it is more reasonable - whether you agree with them or not - for people to care whether computers think or not, whereas not long ago it would take bizarre misconceptions.

I would also agree that it ought to be pointless, incidentally.

I do find it fascinating to poke into the beliefs and assumptions surrounding it, though.