BCS: BCS_HST_2024-06-19/otter

comparison BCS_HST_2024-06-19/otter_ai.txt @ 6:abb1b1e2f6fc

trying alternative sources of free speech-to-text

author	Henry S Thompson <ht@inf.ed.ac.uk>
date	Wed, 21 Aug 2024 19:34:07 +0100
parents
children

comparison

equal deleted inserted replaced

-:f3b043032519
+:abb1b1e2f6fc
+BCS  0:00
+Record says recording here,
+HST  0:03
+yep, it just I clicked it as you spoke, or just before or something like that, right? How are you doing? You? Do you Well, I said this last time, and you disagreed with me, but you look okay, yes, oh, I [mixed up]
+BCS  0:17
+actually think I am okay this time.
+BCS  0:22
+I I'm a little
+BCS  0:32
+compromised in various ways which I'm going to tell you about,
+BCS  0:35
+sure, well,
+Speaker 2  0:38
+one of them being that I haven't done my homework for a reason. I want to try to explain, actually, but,
+HST  0:45
+well, I mean, it was a short it was short notice, but I figure we do this, oh, I don't know. It's like going, this is a comparison I use too often in too many ways. It's like we used to do with the kids, which was that we would go to the west coast of Scotland for the Maybank holiday weekend every year, and without paying any attention to what the weather forecast was, because you needed to book in advance to get a cheap place and so on. And sometimes that meant famously. And you know, family history is a good thing eating our sandwiches in a phone booth on the ferry pier between sky and Razi because it was raining too hard, didn't want to sit in the car to have our picnic. But sometimes it meant, you know, swimming off white sand beaches in arise in 20 degree weather, and it looked and felt like the Caribbean. So you win some and you lose some. Then if this is not, well, they're not as well prepared as you'd like, then we'll talk anyway. We'll talk
+Speaker 2  1:51
+anyway. And I have a question about substance. So here's the problem.
+BCS  2:00
+I have to get [should continue as BCS]
+Speaker 2  2:02
+the final draft of the reflections book to the press by July 8, right[HST] which, which deadline I'm not going to make, but I need to make it enough that my good standing with the press remains such that I can get an extension, and I think even the uncertainty about my lifespan, to say nothing of maybe just efficiency overall, I just need to do that. So this morning, I kind of thought, Look, am I going to spend the morning reading old versions of God, approximately which I would like to do? And I slapped myself on the other wrist.
+BCS  2:54
+Is that a well founded instruction?
+BCS  2:56
+Probably not. But anyway, I mean,
+BCS  3:04
+and have been working on it.
+HST  3:08
+That's that, I mean, you, you are the only person who can correctly set your priorities
+BCS  3:13
+right. So I think I have to do that now. July 8 is not very far away. [HST]No, it's not[HST]. So that might mean delaying our project by rather short amount of time. But realism, the aforementioned realism, means it'll probably mean deferring it for longer than that. But
+HST  3:34
+understood, but we can, we can reduce, at the very least, reduce the frequency, but I may try to keep it ticking over one way or
+BCS  3:41
+another. Yeah? Sure, sure. Well,
+Speaker 2  3:42
+so here's a question, if I can just plunge in. Maybe there are other of
+BCS  3:46
+course, yeah, go. So
+BCS  3:51
+I was struck when I wrote the
+BCS  3:56
+postscript note to our last meeting
+BCS  4:00
+by how I was framing everything
+BCS  4:04
+in terms of,
+BCS  4:09
+well, actually, I don't even remember the last note.
+Speaker 2  4:12
+Hang on a second. Maybe I should take a look at it. I should was it email? Probably, I
+HST  4:18
+believe. Well, I'm sorry if it wasn't emailed, then I don't have it. But that doesn't mean that it's not worth looking at. So
+BCS  4:31
+I'm desperately
+HST  4:41
+waiting I gather from Jim that some progress has been made on the map project,
+BCS  4:52
+on which project the
+HST  4:53
+Save Brian's Mac project?
+Speaker 2  4:57
+Oh, yes. I. Not enough to have the Mac saved.
+HST  5:05
+Well, he was hopeful of his next meeting with you, but maybe it didn't happen that way. So when did I Okay? Here we are. Call this week. No, that was quick. Thought it says, Oh,
+BCS  5:23
+that's it, okay, right,
+BCS  5:26
+all right, sorry, yeah, I've got it. I
+BCS  5:43
+uh, right?
+BCS  5:46
+So, as in the first paragraph, I say
+Speaker 2  5:50
+call these two historical and metaphysical approaches, right? And what I have not done is read any so what you think you have, or what you know that you have, is something like version 11. Is that
+HST  6:08
+right? That's That's correct. 2009 version 11, which I would say in terms of this dichotomy, is entirely the historical approach, okay? And I think that's consistent with the note at the top, which says, In previous versions of this, I tried to produce a metaphysics which would underpin what I'm talking about, but didn't get far enough to make it worth reproducing, or something like that.
+BCS  6:38
+And I did say in previous versions of this,
+HST  6:41
+I believe, so let me just get the fact of the matter in front of me, which it nearly is.
+BCS  6:55
+Right. Wait a minute, I'm just Looking at the wrong place. I
+HST  7:37
+Okay, I think it's this one. Yes, it is.
+HST  7:46
+A number of manuscripts have been circulated under this title over the last 15 years. Right? This one lacks any sketch of a worldview exhibiting the characteristics described, I presume that means described below, as it were, somewhat in response to the first version, which tried to provide such a view without explanation of what was interesting or mattered about it. If it seems worthwhile, I may someday incorporate all the various versions into a single, long, it says short monograph,
+BCS  8:27
+stereograph,
+BCS  8:28
+yes, something like that,
+Speaker 2  8:33
+right? Okay, well, that's very helpful, actually, to me, Bob, thank you for finding that. Yes, I think that longer monograph yet to be produced, longer monograph is what I feel as if we're aiming it.
+BCS  8:52
+And I don't actually know
+BCS  8:56
+whether I
+Speaker 2  8:58
+made any attempt to say that these lead to the same view.
+BCS  9:10
+I have actually thought about that.
+Speaker 2  9:17
+So let me actually recite from memory four or five sentences, and tell me if they ring a bell. If you were Have you ever read them
+BCS  9:29
+go something like this? Start at the beginning.
+BCS  9:33
+That is,
+Speaker 2  9:36
+start at what those who'd like to start at the beginning. Start with
+BCS  9:46
+bosons, fermions,
+BCS  9:49
+quarks,
+BCS  9:53
+assemblages,
+BCS  9:57
+pressed into atoms and molecules and I.
+BCS  10:01
+DNA and so on, as it were. And
+BCS  10:04
+then the second paragraph, saying,
+BCS  10:13
+of course, something like that's i
+BCS  10:23
+is not a beginning. Many will argue, whatever,
+BCS  10:34
+and and
+Speaker 2  10:35
+then something like but actually, it doesn't matter where we start, we'll end up in the same place.
+BCS  10:46
+So in the media there, there would be something like
+BCS  10:51
+other people would say, start with stories
+BCS  10:56
+or something like that. Anyway,
+HST  10:58
+I see what you're saying. Okay. I mean, I think it's important that you well, it changes the where you go next to have something like the stories line, because otherwise it's all just about where you cut the physics. And that, I think, is, is is not enough. That's just what I think of as I had this version of this conversation last week with my redder, my regular Quaker interlocutor, right there. There are these two questions, which I believe I which I tend to attribute to Kant, but I may get wrong. Why is there something rather than nothing, and how would I live my life? Most you know, and if you talk to Dominicans, for instance, they will happily talk about one or the other, but usually find it challenging to see what the relationship is between likely answers to the first and likely answers to the second, right? That's another way of saying what the what it is you're trying to bring together. I think, right? I
+BCS  12:12
+think so. Yeah, I think so. And
+Speaker 2  12:19
+I think what I put in the note after the historical approach
+BCS  12:26
+is sort of a story about
+BCS  12:32
+how Our understanding of
+BCS  12:38
+framions and bosons as it were, I has
+Speaker 2  12:40
+been
+BCS  12:49
+pressed into
+BCS  12:50
+service as a grounds for normativity and
+BCS  12:57
+maybe objectivity and so on and so forth.
+BCS  13:01
+I don't think successfully, but
+HST  13:06
+there is that's, that's, that's really the, that's the, the first large paragraph in the email right,
+BCS  13:15
+which I've now buried under lots of windows. Well, I deep,
+HST  13:23
+the pure mechanism of classical science, then rationality, with reference to frigate logic, then normativity, and the current paradigm of deriving it from evolutionary field,
+BCS  13:31
+etc, right?
+BCS  13:42
+Yeah. So then
+BCS  13:52
+the argument would go something like this, That
+BCS  14:02
+the only tenable version of
+BCS  14:18
+the well, either
+BCS  14:26
+the only tenable version of
+BCS  14:28
+the metaphysical approach,
+BCS  14:36
+well, sorry,
+BCS  14:38
+The only tenable version of both approaches
+BCS  14:42
+ends up being
+BCS  14:45
+indistinguishable from the tenable version of the other.
+BCS  14:56
+And
+BCS  14:58
+one is.
+BCS  15:01
+A crucial
+BCS  15:06
+factor in that, I believe, is that
+BCS  15:13
+both stories
+BCS  15:16
+have to do justice to our being here.
+HST  15:22
+Yeah, I mean, I've been thinking, you know, the phrase, the thing, which, I think is very bizarrely labeled, the anthropic principle, right? Which says, which attempts to dissolve the first of the Kantian questions by saying, because if there weren't something, we wouldn't be here to ask the question, get over it.
+Speaker 2  15:45
+Yes, but I think that the instruction is misapplied radically because they try to understand what the world needs to be like in order to support
+BCS  16:01
+life or inquiry or something like that.
+HST  16:05
+Yeah. I mean, yeah, certainly. The the what little I remember of the time I heard somebody talk about this at length was Planck's constant is what it is. And the fact that if you varied it by not very much in either direction, nothing would work. Isn't something that needs explanation, because it evidently is the case, right? And I mean, yes, exactly you, you, you know it is at least a minor theoretical interest to establish what the what the bounding box is, right, in which we would still be here to ask that question. But having done that, there's nothing more to be said, right? But I think you're so, I mean, I don't think that just that that changes the the availability of of both projects, essentially, I
+Speaker 2  17:06
+think that's right. And I I actually think, you know, this is, well, I'm going to have to agree to the long rather than short, um, assuming if I go down this pathway, but, um, I actually think the fact well as I've, as I put it, which is transparent to nobody, the ontological warrant for the epistemic fact that we Use differential equations
+BCS  17:36
+to express physical laws
+BCS  17:39
+is actually, I mean,
+Speaker 2  17:43
+I don't know if I said this in the objects, but, but anyway, underlies the Diocese of the world, which I think is fundamental to
+BCS  17:50
+consciousness and self and various things like
+HST  17:55
+that. But because of the uncertainty. No, not the uncertainty because, I mean, is this what I remember from the objects book, which I've already apologized for, is very little is about the importance of sloth.
+BCS  18:09
+Yeah, no, that's a different thing.
+HST  18:11
+That is a different thing. Okay, never mind then, wrestleman, what's the
+Speaker 2  18:24
+press on regardless? Yeah, I'm not sure I should accept it regardless just now, but,
+BCS  18:37
+yeah, the Dyches stuff is, I think, important to
+BCS  18:41
+to self
+Speaker 2  18:44
+and something else that's interesting. This is going to sound a little bit like a non sequitur, but,
+BCS  18:52
+but I think it's not for obvious reasons.
+BCS  18:54
+The
+BCS  18:57
+fact that
+BCS  19:00
+llms And
+BCS  19:05
+the fact that llms are based on language
+BCS  19:10
+is, I think,
+Speaker 2  19:13
+possibly consequential, but possibly not, the reason for their success. Because, I think the power of them stems from the fact that the
+Speaker 2  19:34
+relationality that they encode
+BCS  19:39
+is so stupefyingly huge that
+Speaker 2  19:45
+all the content of the state of the network is bizarrely non conceptual in the sense of that,
+HST  19:58
+absolutely I. Mean they got somewhere by not being representational. Well, not sorry, but not being explicitly representational that no amount of additional funding to Doug Leonard and company would ever have gotten to Right, right. Exactly
+BCS  20:21
+how to say that? Well, is not trivial, but, but I completely agree,
+HST  20:26
+yeah. I mean, it was, it was, you know, I just it would be useful in the, in the indefinitely, in foreseeable future, to have a conversation involving Fernando Pereira about this. Because,
+BCS  20:41
+have you ever met Fernando? Not clear. Oh, yeah,
+BCS  20:44
+I knew him.
+BCS  20:48
+God knows if he was a student. But anyway, 100 years ago,
+HST  20:51
+no, he was our student. So because I did his PhD oral, no, but I think he was, he was in California at the time of the oral so it's possible it doesn't matter. Anyway, he was here six months ago for a guest talk during our 60th anniversary celebrations. And the talk was interesting, but not great and not recorded, but lunch beforehand, which was just me and him and one other person, was hugely more valuable because he was expanding to a to an audience that could hear of the two of us on his anger about the fact about the impact of his own company's work. You know, indirectly in terms of open AI, but, but, you know, in chat, GPT and so on, because he's, he's recently changed within Google being responsible for the the natural language work to being responsible for the sort of theory practice interface within Google, and he's very angry about
+BCS  22:09
+the way in which
+HST  22:12
+people are treating the natural language problem as having now been solved and or being soluble by only by the technologies of llms and but what he did for us in that conversation, and I wish I had recorded it, was give me a much clearer sense of the scale of the the base model, and also the scale of the priming that it gets in order to make it a question. Answer, yeah. What's
+BCS  22:51
+that? Called
+HST  22:53
+the prompt? It's not the prompt, but it's something prompt engineering, yeah, prompted, yeah, the prompt engineering is, there are three aspects of this. I think there's, there is the base model,
+BCS  23:05
+right, which is, which is something like
+BCS  23:11
+100 billion gigabytes or something,
+HST  23:12
+yeah, it's, well, it's, it's certainly that many dimensions. And I don't know, you know, there's this whole business about projecting to lower dimensionalities for use
+BCS  23:24
+that I don't understand, but
+HST  23:27
+there's the base model. There is the make this a question answer. You know, make a question answer from this base model, right? And there's what do. What do we add to the to the to the conjunction of those two, from your from your question, from their question, right?
+BCS  23:48
+And is it a third of those, what's called proptening?
+HST  23:51
+I think so. Okay, but I could be wrong.
+BCS  23:58
+Doesn't matter anyway, no
+BCS  24:00
+more than, yeah, anyway,
+BCS  24:02
+and you know, the,
+HST  24:04
+even though, the the the it's the interesting part, in a way, is, in a sense, from from the performance point of view, is not the base model, but it's the the thing you make a question answer out of it with, right, right? Um, because that's what you that's what the people who don't have any money scrimp on, skimp on, right? And why you then get things which lie and fabulate and contradict themselves, and in general, or, you know, start imitating witty ticky Ray rather than a human being, or whatever it might be. Because actually, there's another kind of farm, rather than the GPU farm, that you need to build some. Thing like, as successful as it is, that's chat GPT, which is a huge farm of ordinary human beings asking questions and feeding back to the engineers the wrong answers and saying, you know, this is you've got to, you've got to stop this kind of answer. All right.
+BCS  25:21
+Yeah, that's not a trivial and
+HST  25:23
+that's an open, ended and in principle, impossible task,
+BCS  25:28
+right? Interesting.
+Speaker 2  25:32
+Anyway, that was all a total footnote. You could have expressed your thought at the beginning of your
+BCS  25:40
+what you just said that
+BCS  25:42
+that's what people who scrimp skimp on, yes,
+HST  25:48
+something like that, anyway, but, but so I think, from your perspective, it's really, it's really GPT three that you're interested in, which is the base model. It's now GPT four, and they won't tell you anything about that. The only thing we have any information about is GPT three, right? Well, that's the only thing I've seen published information about from Google Anyway,
+Speaker 2  26:15
+yes, I mean, I think that's open. AI, sorry, yeah, from open. Ai, yeah. I think that's what I was just talking about. I mean, it doesn't prove that I'm not interested in the other ones.
+HST  26:27
+But, I mean, it's there, for instance, that we come back to the thing that you said, which I think is why I think Dyches is certainly in there is not only do they not know that there's a world that not only does that, does that 100 million gigabytes, whatever it is, million gigabytes, what it doesn't have is any obligation to the World about which right that is some kind of representation, right?
+BCS  27:05
+But that
+HST  27:08
+you know that there are that that that responsibility can be
+BCS  27:19
+decomposed in any particular instance,
+BCS  27:23
+to being only about a certain small part of
+HST  27:30
+which amounts, I guess in many cases, to some kind of story about Reference and dices. And it does. It does. I am tempted to bring Jonathan back into this again, Jonathan Reese, because of his, what, you know, what he's been spending the last two or three years on, is trying to articulate a story about reference which is simply defined in terms of
+BCS  28:01
+of of propositions
+HST  28:05
+that include this are vulnerable to changes in that, that is that include this referring expression are vulnerable to changes in that bit of the world as a way of talking about, what does that referring expression refer to? Well, because he's a radical empiricist. Basically, he wants right anyway. Sorry, that is taking us away now.
+BCS  28:35
+No, no, not entirely, because the I
+BCS  28:47
+there was a title of a talk I was thinking of putting together, sort of
+BCS  28:53
+something like the nonverbal meaning of words,
+BCS  29:00
+if we talk about,
+BCS  29:07
+not only about Sussman, but let's say
+BCS  29:12
+and what he meant by
+BCS  29:20
+empirical or something.
+BCS  29:23
+But just we talk about,
+BCS  29:32
+well, the things we're talking about, the three, the three parts,
+Speaker 2  29:38
+the base model, the delta that turns it into a question answering machine and the prompt engineering that turns a particular prompt into a particular prompt, basically particular question into a particular prompt. Say it's.
+Transcribed by https://otter.ai

Mercurial > hg > BCS

comparison BCS_HST_2024-06-19/otter_ai.txt @ 6:abb1b1e2f6fc