Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Gettiers in software engineering (2019)

jsomers.net

Post reply on HN
Print view View original

181–190 of 222 posts

Re: Gettiers in software engineering (2019)

#181

Post by EasyMark »

ndndjdjdn wrote: post #130

The problem is the word "know" being overloaded. First person know is belief. To some extent: this is just faith! Yes we have faith that the laws of physics wont change tomorrow, or we remember yesterday happened etc. Science tries to push that faith close to fact by verifying the fuck out of everthing. But we will never know why anything... The other "know" is some kind of concept of absolute truth and a coincidence…

Maybe, but it’s still a useful overload. No one wants to be pedantic every time they say they know something. It’s more of a spectrum which is really my own overloaded expression. My other one is “it’s a probabilistic matter, there’s no absolute here”

Re: Gettiers in software engineering (2019)

#182

Post by osullivj »

tucnak wrote: post #82
codeulike wrote: post #9

“I am sitting with a philosopher in the garden; he says again and again 'I know that that’s a tree', pointing to a tree that is near us. Someone else arrives and hears this, and I tell him: 'This fellow isn’t insane. We are only doing philosophy.” ― Ludwig Wittgenstein

I often wonder if LLMs would have made Wittgenstein cry...

Suspect not as later Wittgenstein tells us "the meaning is the use"; don't look at the dictionary definition, look at many examples. And that's what LLMs do.

Re: Gettiers in software engineering (2019)

#183

Post by n4r9 »

PaulDavisThe1st wrote: post #51
bee_rider wrote: post #45

Earlier quoted context omitted.

What exactly does it mean to know something then? As distinct from believing it. Just the justification, and then, I guess it doesn't have to be a very good justification if it can be wrong?

I think I would say that knowing means that your belief can resist challenges (to some degree) and that it is capable of driving behavior that changes others' beliefs. The strength of the justification is, I would suggest, largely subjective.

My issue with this definition is that it includes deluded charlatans, can be applied to unfalsifiable (unknowable, even) propositions, and depends on the gullibility and cognitive biases of the general populace. So for example, Jesus "knew" that he was the son of God, even though a more rational interpretation is that he was mistaken in his own belief but charismatic enough to convince many others. (Please replace Jesus for another religion's prophet if you are Christian!)

Also I don't think this definition fits with people's intuition. At least, certainly not my own. There are times where I realise I'm wrong about something I thought I knew. When I look back, I don't say "I knew this, and I was wrong". I say "I thought I knew this, but I didn't actually know it".

Re: Gettiers in software engineering (2019)

#184

Post by bumby »

mjburgess wrote: post #177
bumby wrote: post #171

Earlier quoted context omitted.

Is the extension of your position that we are born with a theory of the body, irrespective of experience? How does that relate to the psychological literature where babies seem to lack a coherent sense of self? I.e., they can't differentiate what is "body" and what is "not body"? If it's an ability that later develops independent of experience with the exterior world, it seems untestable. I.e., how can you test the t…

It might be that its vastly more minimal than it appears I'm stating. I already agree with the high adaptability of the motor system -- indeed, that's a core part of my point, since its this system which does the heavy lifting of thinking. Eg., it might be that the kind of "theory" which exists is un/pre-conscious. So that it takes a long time, comparatively, for the baby to become aware of it. Until the baby has a s…

I think that makes sense as a philosophical thought, but do you think it's testable to actually tell us anything about the human condition?

E.g., If motor movement and causal inference are coupled, would you expect a baby born with locked in syndrome to have a limited notion of self?

Re: Gettiers in software engineering (2019)

#185

Post by efitz »

bonoboTP wrote: post #146
CamperBob2 wrote: post #53

Earlier quoted context omitted.

Hmm. That seems like a better example of the problem than either of the examples at https://en.wikipedia.org/wiki/Gettier_problem . The cases cited in the article don't seem to raise any interesting issues at all, in fact. The observer who sees the dark cloud and 'knows' there is a fire is simply wrong, because the cloud can serve as evidence of either insects or a fire and he lacks the additional evidence needed to…

One should distinguish between one instance and a mechanism/process for producing them. We could take randomness and entropy as an analogy: Shannon entropy quantifies randomness of a sequence generator, not the randomness/complexity of individual instances (which would be more akin to Kolmogorov complexity). Similarly, the real interesting stuff regards the reliability and predictive power of knowledge-producing mech…

Like arguing which level of the OSI model a particular function of a network stack operates at. I’d love to have those hours back from 20’s me.

Re: Gettiers in software engineering (2019)

#186

Post by feoren »

merryocha wrote: post #143

I was a philosophy major in college and semantic quibbling over Gettier problems was popular while I was there. I have always believed that Gettier's popularity was due to the fact that the paper was only three pages, and therefore it was the only paper that the academics actually read to the end. I never thought there was anything particularly deep or noteworthy about the problem at all - it is fundamentally a debat…

The vast majority of philosophical arguments are actually arguments about definitions of words. You can't actually be "wrong" in philosophy -- they never prove ideas wrong and reject them (if they did, we'd just call it "science"), so it's just an ever-accumulating body of "he said, she said". If you ask a philosophical question, that's the answer you get: "well, Aristotle said this, and Kant said that, and Descartes said this, and Searle said that." "... so, what's the answer?" "I just told you." So if you want to actually argue about something, you argue about definitions.

Re: Gettiers in software engineering (2019)

#187

Post by LemonyOne »

pjc50 wrote: post #116
tucnak wrote: post #82

Earlier quoted context omitted.

I often wonder if LLMs would have made Wittgenstein cry...

It's remarkable how LLMs have skipped any kind of philosophical grounding for "how do we know that this output is valid?" and just gone straight to "looks good to me". Very postmodernist. Also why LLMs are going to turn us into a low-trust society. A tool for filling the fields with papier-mache cows.

Because reality is imperfect, and our perception of reality is even less perfect than that. And reality is full of "good enough" things, so if nature is "ok" with "good enough" things, why not artificial things?

Re: Gettiers in software engineering (2019)

#188

Post by csours »

Puttiers: When a junior engineer fixes something, but a different error is returned, so they cannot tell if progress was made or not.

***

justified: in the sense of deriving from evidence

true: because it doesn't make sense to "know" a falsehoood

belief: i.e., a proposition in your head

***

Justified: there is an error message

true: there is an error condition

belief: the engineer observes the message and condition

---

Where's my cow?

Are you my cow? [0]

0: https://www.amazon.com/Wheres-My-Cow-Terry-Pratchett/dp/0060...

Re: Gettiers in software engineering (2019)

#189

Post by goatlover »

visarga wrote: post #163

Knowledge and truth are "centralized" concepts. I prefer "models" which don't have such issues, all models are imperfect and provisional, and there are many models for the same process. Knowledge and truth have a way to lead to endless debates, while models are better understood in their limitations and nobody claims they can reach perfection. In programming we call them abstractions and we know they are always leaky…

But search is a functional term, while consciousness is experiential. Deep philosophical problems exist because there are very puzzling things about existence. Changing terms doesn't make that puzzlement go away.

Re: Gettiers in software engineering (2019)

#190

Post by goatlover »

feoren wrote: post #186
merryocha wrote: post #143

I was a philosophy major in college and semantic quibbling over Gettier problems was popular while I was there. I have always believed that Gettier's popularity was due to the fact that the paper was only three pages, and therefore it was the only paper that the academics actually read to the end. I never thought there was anything particularly deep or noteworthy about the problem at all - it is fundamentally a debat…

The vast majority of philosophical arguments are actually arguments about definitions of words. You can't actually be "wrong" in philosophy -- they never prove ideas wrong and reject them (if they did, we'd just call it "science"), so it's just an ever-accumulating body of "he said, she said". If you ask a philosophical question, that's the answer you get: "well, Aristotle said this, and Kant said that, and Descartes…

Science doesn't prove things, it provides empirical support for or against theories. Philosophical ideas can be shown to be wrong if their reasoning is shown to be invalid. Words have meaning, and philosophical arguments are over the meaning of those words. The problem is there is a "loose fit between mind and world", as one contemporary philosopher put it. We naively thinks words describe the word as is, but they really don't. There's all sorts of problems with the meaning of our words when examined closely.

For example, it feels like we have free will to many people, but the meaning is hard to pin down, and there are all sorts of arguments for and against that experience of being able to freely choose. And what that implies for things like punishment and responsibility. It's not simply an argument over words, it's an argument over something important to the human experience.

Post reply on HN