Sometimes you plan for features that aren’t actually there. I found using mailsAI helped me focus on what’s really available, which made managing expectations easier. It’s a simple way to keep things clear.
Adding a feature because ChatGPT incorrectly thinks it exists
441–450 of 451 posts
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#442Earlier quoted context omitted.
> But https://languagelog.ldc.upenn.edu/nll/?p=2922 "The passive in English" says: >> The relevance of participles is that a passive clause always has its verb in a participial form. > So, what are you saying they do think it was meant to apply to, if every passive clause always includes a participle? I'm confused. OK, you're confused. In the general case, an English verb has five forms†: "plain form" [go], "preterit…
It sounds like you're using the word "participle" in a way that conflicts with not just Pullum's explanation but also, for example, https://en.wikipedia.org/wiki/Participle , which draws no distinction between "participle" and "participial form", specifically claiming that the past participle and present participle are participal forms. When you say "We can be even more specific than that: the verb is always in -en f…
> When you say "We can be even more specific than that: the verb is always in -en form," you're also contradicting Pullum when he points out that the present participle is also used to form certain passive clauses. An example of this is also given in the Wikipedia article, "I saw John eating his dinner."
(a) Yes, Geoff Pullum identified a use of the -ing form that is plausibly called passive. I hadn't looked at that before writing my comment.
(b) But you didn't. "I saw John eating his dinner" is not an example of that usage, and is not plausibly called passive in any sense. If you pulled that out of the wikipedia article you just linked, you should have seen the annotation of that very example noting explicitly that the participle is active.
> for example, https://en.wikipedia.org/wiki/Participle, which draws no distinction between "participle" and "participial form"
This isn't even true. Look at the opening sentence:
>> In linguistics, a participle (from Latin participium 'a sharing, partaking'; abbr. PTCP) is a nonfinite verb form that has some of the characteristics and functions of both verbs and adjectives.
Or here:
> The linguistic term, past participle, was coined circa 1798 based on its participial form, whose morphology equates to the regular form of preterite verbs.
We have a distinction between the participle and the form of the participle. The participle is, in this analysis, named after the form, though there is room for debate on that point.
Compare the sentence I've been there before", where been* is in a participial form but is not a participle† and, obviously, displays no characteristics or functions of adjectives.
The second (and final) mention of "participial forms" does confuse the two concepts:
> Some languages have extensive participial systems but English has only two participial forms
This is a conceptual error; for example, Latin has more participial forms (4, or if you really want to stretch it 48) than English, but its participial system is less extensive.
> I think you're mistaken. I'm not sure exactly how, because I'm not sure exactly what you're trying to express, but I'm sure that sufficient study of sources like those will make it clear to you.
You talk a surprisingly big game for someone who doesn't even claim to know what he's talking about.
† There is a school of formalism within linguistics that says that grammatical categories don't exist within a language unless they are reflected in the language's inflectional system. The best-known idea from this school is that English has a past tense but not a future tense. I don't find this plausible; compare https://en.wikipedia.org/wiki/Lexical_aspect , or just the general concept of an isolating or agglutinative language. On this analysis, but only on this analysis, you could say that in the phrase I've been there before, the token been is a participle. You'd have a lot of trouble explaining why it's a participle, though. On any other analysis, it's part of a finite verb, and by definition cannot be a participle.
But by that school of analysis, you would also have to say that English has no passive voice. (Except in the edge case using -ing forms.) And you run into some very awkward problems more or less immediately; when you look at the sentence "I'm going out for the evening", you have to claim that this sentence is fundamentally about being (the finite verb) rather than going out.
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#443Earlier quoted context omitted.
It sounds like you're using the word "participle" in a way that conflicts with not just Pullum's explanation but also, for example, https://en.wikipedia.org/wiki/Participle , which draws no distinction between "participle" and "participial form", specifically claiming that the past participle and present participle are participal forms. When you say "We can be even more specific than that: the verb is always in -en f…
> When you say "We can be even more specific than that: the verb is always in -en form," you're also contradicting Pullum when he points out that the present participle is also used to form certain passive clauses. An example of this is also given in the Wikipedia article, "I saw John eating his dinner." (a) Yes, Geoff Pullum identified a use of the -ing form that is plausibly called passive. I hadn't looked at that…
Thanks for the correction about John! You're indeed correct on that point, and I was wrong, and I feel like ChatGPT.
The rest of it I don't understand.
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#444Earlier quoted context omitted.
Experiences are not materially different from knowledge once they are both encoded as memories. They're both just encoded in neurons as weights in their network of connections. But let's assume there is some ineffable difference between firsthand and secondhand experience, which fundamentally distinguishes the two in the brain in the present. The core question here, then, is why you are so certain that "creativity" r…
Experiences are not materially different from knowledge once they are both encoded as memories. The storage medium is not relevant here, and actual experience has sensorially laden additions which are not currently possible from reading. Memories are laid down differently as well. In terms of what top neuroscientists know or do not know, you pulled such out of your back pocket, and threw them into onto the floor, per…
> It is merely that an LLM is not it, and will never become it.
Okay, I didn't want to put words in your mouth by claiming you said this, but now that you have, I can address it.
You have literally no way of knowing this. You don't understand how cognition actually works, because no one does, and you don't understand how LLMs actually produce a facsimile of intelligence, for very similar reasons. So you can't say that with certainty, and likewise you cannot claim to know what is actually required for cognition (without leaning heavily on human exceptionalism)
Skeptics of LLMs have been claiming that it "cannot possibly X" for the better part of a decade, and time and time again they have been proved wrong. Ironically, I was just reading an article this morning[1] that reiterated this point:
> [W]e’re still having the same debate - whether AI is a “stochastic parrot” that will never be able to go beyond “mere pattern-matching” into the realm of “real understanding”.
> My position has always been that there’s no fundamental difference: you just move from matching shallow patterns to deeper patterns, and when the patterns are as deep as the ones humans can match, we call that “real understanding”. This isn’t quite right - there’s a certain form of mental agency that humans still do much better than AIs - but again, it’s a (large) difference in degree rather than in kind.
> I think this thesis has done well so far. So far, every time people have claimed there’s something an AI can never do without “real understanding”, the AI has accomplished it with better pattern-matching.
While I can't claim to have been quite as prescient as the author, I agree with his position.
It wasn't so long ago that our standard conception of AI was that it could never make anything that could be called "art"[2], and now we have model that churn out images in seconds and poetry that average people rate as better than most humans[3]
You have a whole, well-spoken argument, but it despite claiming otherwise, every point boils down to "in order to have a trait associated with human thought, it needs to be more like a human brain". Your reasoning is circular and all comes down to human exceptionalism.
> But they provide no means for cognition. And creativity requires cognition. Creativity is a conscious process, for it requires imagination, which is an offshoot of a conscious process.
"Consciousness", the philosophical term primarily infamous for the fact that no one understands or agrees how it works - only that humans have it and some other animals may or may not. But somehow you know that "imagination" (a term not otherwise defined or justified) requires it, and that cognition (again undefined except to assert that LLMs don't have it, despite being able to take in information and retain it) requires that, and therefore, LLMs can't be creative unless they are more human-like.
> At that point, we should inject agency. A reason to exist. With current life, the sole primary reason is "reproduce". Everything else has derived from that premise. I spoke of the mating urge, we should recreate this here.
Again, algorithms are infamous for following goals and chasing rewards even more intently than humans, but you only count that as a "purpose" if it's the same one humans have.
And so on.
[1] https://www.astralcodexten.com/p/now-i-really-won-that-ai-be... [2] https://knowyourmeme.com/memes/can-a-robot-write-a-symphony [3] https://www.nature.com/articles/s41598-024-76900-1
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#445Earlier quoted context omitted.
The problem here is societal, not technological. An end state where people do less work than they do today but society is more productive is desirable, and we shouldn't be trying to force companies/governments/etc to employ people to do an unnecessary job. The problem is that people who are laid off often experience significant life disruption. And people who work in a field that is largely or entirely replaced by te…
> However, there's no reason it has to be this way - the fact people having their jobs replace by technology are completely screwed over is a result of the society we have all created together, it's not a rule of nature. I agree. We need a radical change (some version of universal basic income comes to mind) that would allow people to safely change careers if their profession is no longer relevant.
Reminds me of Mondragón, a corporation and federation of worker coops in Spain. It builds new coops to meet the needs of its community, and when a coop ends, workers are given financial support and trained for new jobs in its other coops.
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#446Amazing story.
Had something similar happen to us with our dev-tools saas. Non devs started coming to the product because gpt told them about it. Had to change parts of the onboarding and integration to accommodate it for non-devs who were having a harder time reading the documentation and understanding what to do.
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#447Earlier quoted context omitted.
> If they're not customers it doesn't matter at all what they think That kind of thinking is how you never get new customers and eventually fail as a business.
It is the kind of thinking that almost all businesses have. You have to focus on the actual products and services which you provide and do a good job at it, not chase after any and every person with an opinion. Down voters here on HN seem to live in a egocentric fantasy world, where every human being in the outside world live to serve them. But the reality is that business owners and leaders spend their whole day thi…
> You have to focus on the actual products and services which you provide and do a good job at it, not chase after any and every person with an opinion.
That's a nice straw man you have there.
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#448If there is a strong demand for a feature, regardless of the source of the request - good enough reason to add it.
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#449Wouldn't some GEO tool like AthenaHQ help with this?
Re: Adding a feature because ChatGPT incorrectly thinks it exists
#450I've found this to be one of the most useful ways to use (at least) GPT-4 for programming. Instead of telling it how an API works, I make it guess, maybe starting with some example code to which a feature needs to be added. Sometimes it comes up with a better approach than I had thought of. Then I change the API so that its code works. Conversely, I sometimes present it with some existing code and ask it what it does…
From https://tonsky.me/blog/gaslight-driven-development/ today:
> Any person who has used a computer in the past ten years knows that doing meaningless tasks is just part of the experience. Millions of people create accounts, confirm emails, dismiss notifications, solve captchas, reject cookies, and accept terms and conditions—not because they particularly want to or even need to. They do it because that’s what the computer told them to do. Like it or not, we are already serving the machines. (...)
> You might’ve heard a story of Soundslice [adding a feature because ChatGPT kept telling people it exists](https://www.holovaty.com/writing/chatgpt-fake-feature/). We see the same at Instant: for example, we used `tx.update` for both inserting and updating entities, but LLMs kept writing `tx.create` instead. Guess what: we now have `tx.create`, too.
> Is it good or is it bad? It definitely feels strange. In a sense, it’s helpful: LLMs here have seen millions of other APIs and are suggesting the most obvious thing, something every developer would think of first, too.
> It’s also a unique testing device: if developers use your API wrong, they blame themselves, read the documentation, and fix their code. In the end, you might never learn that they even had the problem. But with ChatGPT, you yourself can experience “newbie’s POV” at any time.