wpietri

22 ms·

301.
by wpietri 2y ago
Thanks. Good answers. For what it's worth, the people I've had the most success working with are people I've already spent some time working with. But yes to those other things too.
302.
by wpietri 2y ago
Saying that it "is expected" is a rhetorical trick, projecting your personal notions onto the structure of the universe. You may expect that, but I don't. To me, software is a community effort. Communities are made of peopl
303.
by wpietri 2y ago
If you're still reading this, could you say more about how you found your co-founder? And do you have thoughts on how next time you'd go about picking one differently?
304.
by wpietri 2y ago
That seems like a false dichotomy to me. Aren't there more ways to run things than a binary "user narrow numeric KPIs" vs "people doing absolutely anything"?
305.
by wpietri 2y ago
> Paying more will get more and better submissions, and it doesn't seem to be particularly expensive to me. They might get more, but will they get better? The most passionate people I know are pretty insensitive to pay rate. Whereas
306.
by wpietri 2y ago
That strikes me as likely to do more harm than good, and research is a good example. We already have major problems with research quality because publication quantity has risen as important metric. I can't think of a metric for art qua
307.
by wpietri 2y ago
I think the point is that it has the same sort of latency tradeoff that disk swap did: it's awful, but sometimes better than nothing.
308.
by wpietri 2y ago
I appreciate you demonstrating so clearly that you're not engaging at all.
309.
by wpietri 2y ago
I have no reason to think you even know what my argument is based on, because all I've seen are nitpicky side-tracks.
310.
by wpietri 2y ago
I never claimed that I was reporting on the whole arc of publishing law over hundreds of years. I was laying the foundation for the next sentence, which is pointing at why this was so important to put in the US Constitution. The relevant te
311.
by wpietri 2y ago
There's a question in a (reasonably) dead reply that I want to resurrect: > Are we to go out and memorize everyone’s personal framework/religion (cause I can tell I’m not going to) or do you have something else in mind except a
312.
by wpietri 2y ago
As long as we're nit-picking, we're talking about the Apache license. But my point isn't about the specific license. It's about one set of people very generously doing lots of labor and publishing the result in hopes of
313.
by wpietri 2y ago
You are confusing legality and morality. Legality can be an indicator of morality, in that laws tend to flow out of our moral senses. But legality is a lagging indicator, and new circumstances tend to create gaps where things are legal but
314.
by wpietri 2y ago
It might not to you, but it does to me. I think it's bad not just from a social perspective (because exploitation acts to diminish the amount of open code) but from an investor one. Forking a repo, committing IP crimes, and then doing
315.
by wpietri 2y ago
And I think those conversations are worth listening to. Do we want a society where people release a lot of open-source code? Or do we want one where people get tired of doing free labor for greedy jerks, and so stop releasing things openly?
316.
by wpietri 2y ago
Is your standard for ethics really, "Well it's not a criminal violation of the law?"
317.
by wpietri 2y ago
Do you think those two are somehow unrelated?
318.
by wpietri 2y ago
Got it. I am talking about a function, and particularly the expected reward function for a behavior. If some factor has no influence on the score, the function does not depend on that factor. If it does, it does. Even in your sense of the t
319.
by wpietri 2y ago
You're using a meaning of "depends on" which I'm unfamiliar with. Where are you getting it from?
320.
by wpietri 2y ago
They explicitly said, "If others see it, great. If not, no big deal." So I agree they get some reward from the writing. (Like all writers do, and like we can all do just as well with a journal.) But the part I just quoted is clear
321.
by wpietri 2y ago
Hey, I'm glad to hear that. Some years back I came across a great online test, one created by autistic people. I can't find it now, but I strongly remember the graph, which showed a bimodal distribution and marked my place on it.
322.
by wpietri 2y ago
Very nice to see this point being made. One way I explain it to people: Imagine a corporation that only has a PR department. Extremely good at generating press releases and answering reporter questions. But without the rest of the company,
323.
by wpietri 2y ago
I just described two different ways of thinking about self-worth, so I am not suggesting it "can only be defined" as anything.
324.
by wpietri 2y ago
I thought maybe I was using the phrase wrong, but the first two definitions I find are "the internal sense of being good enough and worthy of love and belonging from others" and "a feeling that you are a good person who deser
325.
by wpietri 2y ago
Because upvotes control visibility, you are still describing a reward function that depends on the upvote.
326.
by wpietri 2y ago
This is a great point. In one of Nassim Nicholas Taleb's books, he talks about the emotional impact of looking at one's portfolio performance. If you do it rarely, like quarterly or annually, it'll generally be a positive exp
327.
by wpietri 2y ago
I think your notion of "the truth" is heavily influenced by exactly the context she's pointing at. Google and Meta's algorithmic rankings aren't "the truth", and "the truth" doesn't change o
328.
by wpietri 2y ago
This is an odd thing to say on a site that depends so heavily on the upvote.
329.
by wpietri 2y ago
I think there are two ways you can look at it. One is what I think of as the neurotypical way. They spend a lot of energy on social modeling, on fitting people into hierarchies of privilege. (E.g., "respecting your elders", older
330.
by wpietri 2y ago
I don't think that's what they're saying at all. They're talking not about qualia in the human sense, but specifically about "the qualia of their own training". That is, the corpus that LLMs "learn" f

More ›