Because there's now a class of programmers who are very anti AI when it comes to coding because they think anybody who relies on it are degenerate vibe coders who have no idea what they are doing. You can see this in pretty much every single HN post w.r.t AI and coding.
The first one has four important phrases: “negative correlation,” “mediated by increased cognitive offloading,” “higher educational attainment was associated with better critical thinking skills, regardless of AI usage,” and “potential costs.”
The second paper has two: “students using GenAI tools score on average 6.71 (out of 100) points lower than non-users,” and “suggesting an effect whereby GenAI tool usage hinders learning.”
I ask you, sir, where exactly do you get “AI over-reliance will make us worse…because it’s true” from TWO studies that go out of their way to make it clear there is no causative link, only correlation, point out significant mediations of the effect, identify only potentiality, and also show only half a letter grade difference, which when you’re dealing with students could be down to all sorts of things. Not to mention we’re dealing with one preprint and some truly confounding study design.
If you don’t understand research methods, please stop presenting papers as if they are empirical authorities on truth.
This is a somewhat idealistic/naive view. If your project is even mildly successful and you get motivated to keep contributing to it more and more and form a strong community soon the leechers arrive. They start demanding more and more for literally nothing in return under the pretence that it's in the best interest of the project to add the features they want and to accelerate development.
It soon becomes exhausting to deal with them on a regular basis while still feeling good about contributing. People are faster to demand rather than appreciate and you start wondering what the point of all this is.
So yes, if you end up painting something that's like a high school project, sure, it's easy to leave it on the wall and not care. But if your painting starts getting displayed in galleries and there's a little "demand" for it, everything becomes a headache.
> in return under the pretence that it's in the best interest of the project to add the features they want and to accelerate development.
Which is why you shouldn't work under false pretenses. You willing to turn that project into a day job? No? Then tell them no. You owe them nothing. You've already done more than your "fair share" giving to the community by publishing open source code, no matter how good (or not) it is, or how many "galleries" it hangs in (or not). They have a problem? Well, by definition it's their problem, not yours. Let them own it, fork it, and solve it, if it's such a big deal.
"No" is how you set expectations and maintain integrity. You only have so many resources in a day. Don't become a load bearing internet person for free.
I have to admit, it's one of my guilty pleasures to tell people "no" when they ask for unreasonable things from me. Haven't had the luck of managing a large open source project, but the prospect of saying "no" to an overwhelming stream of demands excites me.
- Crawlee has out-of-the-box support for headless browser crawling (Playwright). You don't have to install any plugin or set up the middleware.
- Crawlee has a minimalistic & elegant interface - Set up your scraper with fewer than 10 lines of code. You don't have to care about what middleware, settings, and anything are or need to be changed, on the top that we also have templates which makes the learning curve much smaller.
- Complete type hint coverage. Which is something Scrapy hasn't completed yet.
- Based on standard Asyncio. Integrating Scrapy into a classic asyncio app requires integration of Twisted and asyncio. Which is possible, but not easy, and can result in troubles.
> You don't have to install any plugin or set up the middleware.
That cuts both ways, in true 80/20 fashion: it also means that anyone who isn't on the happy path of the way that crawlee was designed is going to have to edit your python files (`pip install -e` type business) to achieve their goals
I've been working on a crawler recently and honestly you need the flexibility middleware gives you. You can only get so far with reasonable defaults, crawling isn't a one-size fits all kinda thing.
Crawlee isn’t any less configurable than Scrapy. It just uses different, in my personal opinion more approachable, patterns. It makes it easier to start with, but you can tweak whatever you want. Btw, you can add middleware in Crawlee Router.
> Crawlee isn’t any less configurable than Scrapy.
Oh, then I have obviously overlooked how one would be able to determined if a proxy has been blocked and evict it from the pool <https://github.com/rejoiceinhope/scrapy-proxy-pool/blob/b833...> . Or how to use an HTTP cache independent of the "browser" cache (e.g. to allow short-circuiting the actual request if I can prove it is not stale for my needs, which enables recrawls to fix logic bugs or even downloading the actual request-response payloads for making better tests) https://docs.scrapy.org/en/2.11/topics/downloader-middleware...
Unless you meant what I said about "pip install -e && echo glhf" in which case, yes, it's a "simple matter of programming" into a framework that was not designed to be extended
Cache management is also what I had in mind. Ive been using golang+colly and the default caching behavior is just different enough from what I need. I haven't written a custom cache middleware, but I'm getting to that point.
That’s odd. I was just talking about how Spotify falls into the same cul du sacs that all similarity based recommendations fall in. Like it might show you some decent stuff for a few days if you give it a hard kick by finding something outside of Spotify first, but soon it’s the same three bands and two albums.
That has definitely not been my experience. The Made For Your section shows me tons of artists I've never heard from. I suppose it could be limited if you listen to a very specific niche where they don't have a large dataset on. I love finding new artists and then suddenly diving into a rabbit hole of discovering their older stuff.
Weird. As an anecdote my "Daily Mix 1" has 50 songs, and only two bands I didn't recognize. 62% of the songs are by artists with more than one song in the list. The most songs by an artist? 7 songs over 3 albums. Of the 50 songs, only 4 were put out this decade, only two bands did I not recognize.
"Daily Mix 2" is better with 30 bands, with the top band having 4 songs across 2 albums. Coincidently 62% of songs are also by artists with more than one song in the last, but the most common artist only has 4 songs across 2 albums. 19 unrecognized bands, so that's something.
> Did the NYT not do the most basic research on Elon, which would have shown his consistent history of this kind of behavior they are now criticizing him for.
Twitter wasn't as big back then as it is now. Earlier all his thoughts were kept mostly to himself and social media wasn't as big. Twitter amplifies his inner thoughts. People don't seem to like it.
Love the humblebrag. This "investing" method works until once day it doesn't. Don't get me wrong, HN is wrong a lot of times. Those two specific examples are about companies that are monopolies for all intents and purposes. Google and Meta pretty much control their primary markets. Tesla does *not* control their primary market. In fact it looks like their competitors are closing in on them very fast. No one was closing in on Google Search or Instagram/Whatsapp. The numbers never indicated that at any point in time. For tesla on the other hand...
India has rampant corruption. Almost all the "winners" in bids are pre selected. There is no real competition, hence the fat profit margin. To be fair, a significant part of the "profits" gets sent up the food chain to ensure more contracts keep coming.
