> if you are going to ask me for help, at least make some effort first
It's actually the other way around. You should think what makes you feel they didn't make an effort? Why do you think I am asking - because I think you have a better answer than I can get from Google or AI.
But this is where it's apparently going. We will all talk to AI rather than each other. And we will pat ourselves on the back how self-sufficient and non-lazy we all are. :-)
> You should think what makes you feel they didn't make an effort?
Long experience. There are a lot of people out there in the workforce who ask their boss or a more senior coworker a question the moment they think of it, with no attempt to find the answer via tools at their disposal. Maybe not as many as 80%, as implied by @sdoering below in a sibling thread, but quite a few.
Unfortunately this is true; and if you're not careful with your time, a lot can be wasted by people who realize "I can email so-and-so instead of putting in 5 minutes to finding the issue myself".
They're usually pretty courteous in their interaction, which makes it all the more difficult to be "rude", in my case, by adding an exponential falloff in response times - after I realize what's happening, I tend to take a little longer for each reply so they figure out it's faster to just do the research on their own most times.
> after I realize what's happening, I tend to take a little longer for each reply so they figure out it's faster to just do the research on their own most times.
Agreed, and I do the same. They still get a courteous reply, but they also feel a little "pain" when they don't get a timely answer - an effective teacher.
Indeed - I had a team that called this "remote brain execution" (we were a build team that used Bazel, and often fielded questions about why someone's build broke).
My favorite phrase on that team was "What have you tried so far?"
Ironically, I have to edit out my "what I have tried so far" when asking questions, because I'm more likely to go into a long-winded explanation of the headers that I hacked and the kernel module I installed to fake my way around this or that, when the actual answer tends to be "uh... are you sure you're building the code you think you are? That sounds like you're running from the wrong directory or wrong branch."
Not just the workforce, my parents still barely know how to use a computer because any time they hit the slightest snag, they immediately call me for help.
If someone doesn’t make an effort I don’t care what the excuse is “you’ll know faster, I don’t know what to look up” etc. I won’t enable learned helplessness. At best you’ll get a “maybe read up on X” and that’s about it, if I’m in a good mood.
If I can tell you tried to figure it out via vocabulary and things you tried, I’ll do everything I can get help get you across the finish line.
That's a perfectly valid response for the situation you're describing. But that's not the parent's situation, where the party being asked just silently asks AI (or googles) and feeds the result back without any added expertise.
"I don't know, here's what I would do to find out" is teaching someone; returning an AI response is not.
>> what makes you feel they didn't make an effort?
Because everyone has had that person who you help out, and become their path of least resistance to an answer. They are not looking for the BEST or a GOOD answer, just the least effort. It's completely reasonable to push back with "what have you tried so far?"
I think a lot of people are also missing the value-add of asking a person to Google something for you.
Some large fraction of the time when someone asks me a question, I also end up googling it... and then I use my domain knowledge and experience to weed out bad information and outdated information and identify the right references.
And some fraction of the time when I ask someone else an "easily Google-able question", I've googled the question, found a number of sources, maybe even one I think might be right, but want some confirmation that I'm not going down a rabbit hole.
If you're telling them all that and not wasting their time, that's fine.
But if you just ask them the question and don't tell them what you've found or where you got stuck, you're asking them to stop doing what they're doing and spend all that same time you just spent working on your problem.
IME it usually means they have some good reason to ask, which you are not aware of. For example, people might believe you are an expert or can give a better answer in the context.
Frankly I find it flattering when people asked me certain kinds of questions!
Maybe I’m just more generous with my time than others (or perhaps I don’t value it), but all these responses saying friends and colleagues asking you questions they could’ve theoretically looked up are “wasting your time” are rather perplexing. If somebody’s asking me, like you said, I generally assume they have a reason for asking. Or maybe they’re just tired and don’t want to spend an hour looking it up and verifying it because they know I have a quick answer that takes me little to no effort. I don’t see anything wrong with that. I get asked camera questions all the time, and frankly I just see those as opportunities to stay sharp. No better way to learn/reinforce your knowledge than teaching others
Yes, they might believe you are an expert, but they often ask "experts" trivial questions they should just have Googled. As evidenced by how just using Google to answer them will often make people think you are an expert even at topics you're clueless about.
Most of finance and IT runs on arbitraging complexity. If humans were a rational species, we would have no soldiers and wars. Will AI change it, and reduce the complexity? It could, but I doubt it.
The future is bleak for many, but jobs lost is the least of our problems.
> Most people don't do this, they will get something for free and fiercely defend this right to get things for free.
So what? I think, if you read a good book, learn something or are well-entertained, it's a positive externality, so there is no problem with people doing it for free.
The only real issue with IP piracy is when someone gets money by copying the works. Which were originally the cases copyright tried to prevent.
Maybe you can clarify why you see people doing these things for free a problem, when there is a net benefit to society and also you.
If I didn't have a resource like AA I would likely read less and in the end spend less on books.
When people around me ask about how to "get into reading" I tell them to just find stuff they like online (via AA) or at the library and go from there. If you don't pay initially you don't feel as bad about trying things that may be "bad" or that you aren't interested in.
> The owner takes the risk that there is no (or negative) money left over after paying employees and all other costs.
This is not true, and somewhat confusing, because "takes the risk" means two distinct things - making decisions and living with the consequences of them. Economic production is a collective effort. The management of the company is who usually makes the decisions; these might coincide with owner (especially in small business) but often they're just another employee.
On the other hand, bad decisions made by management affect everyone in the company, not just the owner. The rich enough owner rarely lose their livelihood (we have limited liabilities btw), but the employees might lose the only source of income.
And the system where you have only one person (owner as main manager) making decisions ("take risks") that can negatively impact many people (his employees, customers and what not) is structurally risky, it actually increases the risk of something going wrong (aside from it being a moral hazard). (POTUS is an extreme example of this.) The risk is shared (collectively owned, if you will) and so should be the decision-making.
If they want to "share with society", it's nice, but they can do it after taxes, just like everyone else.
Even if a rich person reinvests everything, the control over large amount of money is what makes it problematic.
Also the idea that welfare doesn't go to investments is wrong. When you buy groceries (or anything really), there is a decision made by the management of the company you buy these things from to reinvest part of it to maintain or build productive capacity.
There is no need for a "capitalist" (owner of the enterprise) to insert themself into the process, they are useless middlemen who get a cut, essentially. (They are not so useless when they do actual managerial work, but then they can be just an employee like everyone else.)
You could always replace him with a "banker" who instead loans you the means of production on credit, but nobody is going to make you a lathe for free out of the goodness of his heart, nor buggies for your grocery, nor the produce for your shelves and meat for your coolers. The banker makes you take the risk because, if it fails, he probably takes your house.
Believe it or not co-ops exist just fine and some do very well. It sounds like what you would like is a co-op and I will be quite happy for you if you start one.
Bankers are even worse than shared ownership due to moral hazard (that's where muslim and christian view of money lending comes from), but yes, as a socialist I am in favor of coops.
> Even if a rich person reinvests everything, the control over large amount of money is what makes it problematic.
What about the control that out-of-touch politicians and bureaucrats have over large amounts of taxpayers' money? Shouldn't we find that far more problematic overall?
> There is no need for a "capitalist" (owner of the enterprise) to insert themself into the process, they are useless middlemen who get a cut, essentially.
Then why do newly created enterprises almost universally seek outside capital investment? Sounds like there is a need after all, otherwise you could just have a partnership structure and take no outside money whatsoever.
> Shouldn't we find that far more problematic overall?
These things are connected and both are problematic. See e.g. Citizens United vs FEC.
> Then why do newly created enterprises almost universally seek outside capital investment?
I agree there is a need to raise capital, because any large scale economic activity is a collective enterprise. Whether this collective enterprise should be owned by a single person, or small subgroup, or by all participants, is an orthogonal question.
What's your solution to the local knowledge problem? Capital markets are like other markets. Capital gets allocated to where it's needed, holding aside market failure (which is what regulations are supposed to fix). Capital owners seek out the best return/risk, and because they are on the ground and diverse in views and skills, that solves for the local knowledge problem.
This is where startup seed funding comes from, capitalists like YC who are good at it rather than some incompetent. It's why bad companies eventually lose the ability to raise, freeing up societal resources.
What appears to be implicit in your comment ("There is no need for a capitalist") is an advocacy for central planning for capital. Although you also say "they can pay taxes" so maybe that's not what you're advocating for.
If you want to know what I think is best, it's possibly a wealth tax applied on global wealth, along with stronger regulations around media concentration, political spending, and a few other things. But to eliminate capital markets and push it all into a central planner is bad.
No, it isn't. You can have democratic control over capital, respecting subsidiarity principle. It's no more centralized or decentralized than under capitalism.
> But to eliminate capital markets and push it all into a central planner is bad.
If you have a high concentration of wealth you have the same centralized control, regardless whether the mechanism is capital markets or gosplan.
Your reading of 1984 is a bit shallow.. It was a warning against censorship in the UK as well. (And honestly I think the authoritarian state of 1984 is just a setting, I think Orwell's main point was we shouldn't give into extremism of any form, otherwise we lose our humanism.)
Orwell has been quoted that Animal Farm was a also a critique of capitalism, in favor of democratic socialism.
You also say GP is naive about China. But China has been actually less oppressive as time goes along. In fact, historically, authoritarian states often become less oppressive without foreign interference (my home country, Czechoslovakia, was on path towards democratic socialism in the 1960s, unfortunately, it was reversed for geopolitical reasons; such has been experience of many American client states as well). (And you also have liberal states becoming more authoritarian on their own, we can see that in the western world, due to concentration of wealth.)
This indicates there is no "natural law" that makes things more (or less) authoritarian. It depends on people pursuing politics, and being informed.
Well, tell that to my management, that it's a pointless endeavor. The structure of (required) social relations in the workplace is a graph; so there will always be some flaw when trying to conform this into a tree.
Compilation is not deterministic, see JITs and GCs. What is deterministic is the resulting program output, but not its performance. So with compilers, we traded away the determinism over performance in exchange for ease of programming.
With LLMs, we are trading away the determinism of the program output as well, in exchange for even more easier programming. Is it a good or bad thing? There are ways to mitigate the problem, just like there are with compilers.
You could argue the determinism of the program output was never really there, because the specification at the high enough level was always unclear. So we are not really losing that much, just accepting more messy reality.
Then the only question remains, can these computer programs (LLMs) do a better job (and where) than a SW developer, who is supposed to translate unclear specifications into a formal language (source code). It happened with compilers - eventually they got better than all of assembler programmers. Same happened to chess players.
> Compilation is not deterministic, see JITs and GCs. What is deterministic is the resulting program output, but not its performance.
Does JIT compiles some other program code instead of the one being run? Does it produce bytecodes for a differenr VM? Does it tries to compile parts of the program that have not been executed or aren’t going to be?
Does GC destroy objects being in use? Does it ignores instances and memory that has been properly released?
JITs and GC are deterministic algorithms, you can predict its behavior by just reading their code. LLM tooling involves an actual random generator for its output.
> Does JIT compiles some other program code instead of the one being run? Does it produce bytecodes for a different VM? Does it tries to compile parts of the program that have not been executed or aren’t going to be?
Sure, but the same is true for LLMs - the lead models no longer make trivial mistakes like answering "What is the capital of France?" wrong.
> JITs and GC are deterministic algorithms, you can predict its behavior by just reading their code.
On large enough systems, you can't, just like it's difficult to predict weather. Determinism has little to do with it. At work, I have just witnessed a bug in JIT (it seems to have been fixed in OpenJDK 25). It inlined a wrong method. We weren't able to reproduce the error conditions without a private customer dataset.
And the fact is, historically, there have been many bugs in compilers, or they have been bad at their job, writing performant programs. The output (resulting program) of a good compiler is difficult to understand (because it is written to be efficient). LLMs (for the programming use case) are different quantitatively, not qualitatively.
It’s really weird how you shift the goalposts and your own definitions.
No one is saying that a compiler can’t have bugs. What we have been saying is that if we take the compiler has a blackbox, we’re reasonably certain given we know the input, what the outputs will be. And the output will stay the same if you keep the input the same.
But you can send the LLM the same prompt, and it will gives you a different answer each time. And it’s not even about the verbiage used.
LLM doesn't have to be non-deterministic, it can work just like any other deterministic algorithm.
But I am not sure why the insistence on the relevance of (non)determinism, rather than on the chaotic relation of the output to the input (which is true for both compilers and LLMs). In practice, inputs to the LLM, as well as to the compiler, change. And the fact is, the output can change radically due to that.
I think nobody really sends the same prompt twice to the LLM, so nobody cares about it being deterministic. I think what you're looking for is something different, some form of stability (as opposed to chaotic behavior). Although it's hard to define exactly, because in case of LLMs theory lacks behind praxis. (And as I said - we already gave up on stability with respect to performance by using compilers. We resolve that issue by doing performance testing.)
(I asked AI what's the opposite of "chaotic", I use "stable", but it seems that people use "deterministic" or "predictable" also in that meaning. So if you're using "deterministic" in that meaning, then you don't really care about sampling and temperature, i.e. determinism in the philosophical sense, but rather whether the output is consistent, albeit expressed differently.)
The whole point of technology is about control and consistency. Even with random parameters, we want their value to an item in a specific sets. When I use a tool, I want it to produce the outcome I want, not any other outcome it wants to produce. If it fails at that, it’s a defective tool.
LLMs can be deterministic as well - same prompt on the same model produces the same input. On the other hand, compilers can be quite undeterministic - you get a new version of compiler, or change compiler options (turn on optimizations) - you might get a very different binary. And JIT compilers (and GC languages) even less deterministic, their compilation can depend on the nature of the inputs.
But I think, in the analogy compiler ~ LLM, the issue is more of a trust than determinism. It took decades to assembler programmers to trust compilers enough not to write code in assembler. The similar will happen with AI - some will embrace it sooner than others.
> LLMs can be deterministic as well - same prompt on the same model produces the same input
> compilers can be quite undeterministic - you get a new version of compiler, or change compiler options (turn on optimizations)
That’s a whole other level pf bad faith argument right here. Flags and options are input too.
> It took decades to assembler programmers to trust compilers enough not to write code in assembler.
You do realize that Cobol, Algol, and Lisp are very old, and they were not assembly. And that Unix were written in C shortly after the language was created.
> That’s a whole other level pf bad faith argument right here.
Not sure where you see the bad faith argument. (Btw I mean "same output", not "same input", it was a typo.)
Take for example JVM. It used to be horribly bad and unpredictable, performance wise, in the 90s. Sun tried to base a desktop environment on it - it didn't work.
> You do realize that Cobol, Algol, and Lisp are very old, and they were not assembly.
Of course! But people have been hand-writing assembler until late 2000s, because compilers were simply not that good.
The same will happen with LLMs - some people will not trust it and won't use it for decades, possibly. Some have already embraced it.
You proof for your argument that a compiler is undeterministic is to change the whole compiler to another version and saying it won’t produce the same output as the old one.
> But people have been hand-writing assembler until late 2000s, because compilers were simply not that good.
And we have software like Unix, enacs, ksh, awk… that’s all written in C. I strongly believe that those people who were writing assembly was optimizing stuff or dealing with constraints (like the 640kb of DOS). Just like today, you may still have to write assembly for microcontrollers or video codecs. Compilers were expensive, but people were paying for them.
> You proof for your argument that a compiler is undeterministic is to change the whole compiler to another version and saying it won’t produce the same output as the old one.
Fair enough. What I meant though was that compilation as a process is not deterministic, because often when you recompile couple years later, you're using a different compiler. (In modern world it can be much shorter time, actually.)
> And we have software like Unix, enacs, ksh, awk… that’s all written in C.
So? IIRC, first compiler was FORTRAN, invented in 1958. OpenAI Codex, first coding LLM, came out August 2021. So we are like in a year 1963. For this comparison, we have ten more years to produce (using a coding LLM) a compiler and operating system just from the textual specification, without an intermediate formal programming language. Funny - we have actually already done that (Claude C Compiler, VibexOS).
> So? IIRC, first compiler was FORTRAN, invented in 1958. OpenAI Codex, first coding LLM, came out August 2021. So we are like in a year 1963. For this comparison, we have ten more years to produce (using a coding LLM) a compiler and operating system just from the textual specification, without an intermediate formal programming language.
Nope, the timeframe would have been three years
In 1961, the MCP was the first OS written exclusively in a high-level language (HLL).[0]
So by 2024, we should all have been able to verify that LLMs are reliable to produce a good enough product. Instead, it’s just slop everywhere, where the one producing it does not even care about its creation.
> if you are going to ask me for help, at least make some effort first
It's actually the other way around. You should think what makes you feel they didn't make an effort? Why do you think I am asking - because I think you have a better answer than I can get from Google or AI.
But this is where it's apparently going. We will all talk to AI rather than each other. And we will pat ourselves on the back how self-sufficient and non-lazy we all are. :-)
reply