1,019 karma · joined July 26, 2024
There is no bad cutting or good cutting, there's only politics. Meaning, an argument for what make some problems more important or worthwhile is not an apolitical argument, it's an argument of how other people should live.
As in, you don't get to conclude that you're obviously right and they're obviously wrong.
> Yes yes. What's "normative" is now "political" (for some reason), and my standard is also "political" and therefore is not relevant.
You're hand-waving. Your stance is political but not irrelevant. Your stance is philosophical (resting on chosen assumptions) and not empirically irrefutable.
Not acknowledging that is why you fail to convince.
You've made this argument about you and your ability to "catch" people. You have no argument that stands on its own construction.
A problem of authorization can be solved with delegated authority. I'm saying your use of it is as evidence for your reasoning is weak. Those are two different problems.
> your point was that there's no normative standard for evaluating science. You said it like, three times.
Yes, but you equated me saying "no normative standard" to "no standards at all." You setup a false dichotomy. My larger point was that what's normative is political. And you saying your standard should be the norm is also political.
> You like the research, therefore I don't know what I'm talking about, and who am I for having an opinion anyway. And then I tell you that I actually do have some relevant knowledge, and you dismiss the knowledge as "normative". Convenient!
You're placing words in my mouth. I didn't say I like the research, I'm saying I don't like your grounds for dismissing it. I don't dismiss your expertise but I reject it as sufficient evidence for your argument.
> Doing the 150,000th derivative observational study finding that poor people are sicker than rich people is not an example. Poor people are sicker than rich people. Let's move on.
If you cannot see the hubris here, if you cannot see how unscientific it is to conclude (reductively) the results of an experiment before the experiment, then we are at an end. Let's move on.
That's a laughable argument based on a claim of authority. Unfortunately, advanced training is not unique to you, and so, you don't get a singular say on what's good or bad.
> so let's not have any standards at all.
Do not misrepresent my point. My point was: if people care, even marginal reduction of uncertainty is worthwhile.
> Go get a grant from the National Endowment for the Arts or a left-wing think tank or something.
And there's your actual point. You hate that science is so affected by the politics of those who control the funding.
But that's always been the case. Wars have done more for physics than curiosity.
> repeating things we already know
Not a terribly scientific stance.
> while legitimate researchers have to scrimp and wheedle to do anything novel
There isn't a normative standard for good research beyond doing good research. Some fields have an easier time setting up and controlling experiments, but no research can predict how useful it'll turn out to be. You're conflating control convenience for utility.
> randomize grants who meet a basic competency threshold
You ignore the political and economic system within which the scientific system sits.
> if it leads to a more focused funding of actual, legitimate science, I'm largely in favor
Again, your normative standard for what is legitimate.
> simply because it's done by a consortium of big names, in trendy areas.
They're trendy for a reason. Science is, at it's core, questioning things because someone cares about it.
Detail-oriented work with lots of output that can cover up the noisy bits of thoughtless garbage? Sure, great.
Analysis-oriented work where decisions have consequences over large amounts of resources? Only an idiot would use these tools for that.
Maybe as a conversational note-taker, but anything more and you don't know what you're doing.
And I'm here trying to get something to make a $1000 per month. What a world.
What were some of the downsides? Illuminating the tradeoffs would elevate this post from good to great.
Whenever I use it, I don't feel like I'm doing anything new when compared to all the other issue trackers and Kanban boards I've used before.
By that same measure: correctness, maintainability, and efficiency are not that useful.
I wasn't saying usefulness is not important, I'm saying this post conveniently crammed the hardest problem of writing software into a fuzzy adjective.
What kind of 'useful'? Normative? Empirical? Prescriptive? Pragmatic?
'Useful' is a very subjective north star.
Made me think this was for companies working on self-driving.
I use Cursor and Codex but the agents keep making regressions and breaking rules. They'll even take shortcuts sometimes, by doing things to make tests pass but with code that would be dangerous in prod.
Now, I use them file by file but it feels more like a typing assistant than something much more.
Your points do not address my post.
You are no moral arbiter.
You engage in personal attack without grace.
We're done.
Unsolicited commercial email is legal in many jurisdictions. There are some good reasons for it being so.
Your moral stance is fine. But it is not final.
Engineers have known that bad engineering cultures cause tiny errors and imprecise code to compound into systemic problems.
Why is this suddenly not a problem? If your AI agent is wrong 5% of the time, what do you think happens when you run it a 1000 times?
Why? On what basis is this claim made? They're trained to probabilistically complete patterns. Where does the confidence in this ability come from?
Modelling a problem is what I'm concerned about. And I'm currently better than any AI agent at doing that, given enough time.
But it can write working code much faster than I can.
And in a lot of cases, unfortunately, faster beats better.
Is the conclusion that growing anti-AI sentiment is a good opportunity for political organization?
One question I had:
Does every sandbox change end (when ready for production) in a pull request? If marketing sends me a pull request and I hate the code, what's the flow like for me to fix it?
I understand the sentiment but I don't think it's useful to take a directly antagonistic stance, especially when it's a losing battle.
For those who feel this way, our best hope is to keep searching for how we can have a world that values human effort and care, even after AI does everything it's proclaimed to do.
We can't declare the world a lost cause and relegate ourselves to only hating. We need to do what we've always done: roll with it.
How can an engineering team say they have a good process if they don't measure the results of the process?
Sure, you can make a lot of logical leaps, but what does it actually produce?
How does this compare to hiring by randomly picking a sufficiently good-looking resume?