Hacker News Reader

Comment by stevage | Hacker News Reader

I would say that OpenStreetMap is also a pretty good thing on the internet. Maybe not quite as impactful, and I'd say less well run, and rougher around the edges, but it's largely resisted being taken over by evil corporations or heavily influenced in bad directions.

View on HN

The difference is that there are corporate alternatives to OSM, and they're popular. The closest alternative to Wikipedia is asking an LLM to read Wikipedia for you.

>The difference is that there are corporate alternatives to OSM, and they're popular.

I actually think that's probably a good thing for OSM. If OSM was the only game in town, there would be a lot more contention, fighting over how businesses are represented etc etc.

Torrents are also a good thing on the internet.

Me getting free movies and music, piracy good

Meta getting free books to train an LLM, piracy bad

Powerful entities getting a pass, bad.

Individuals getting a pass, good.

See, it just depends on how you slice the Venn diagram. With a bit of imagination you'll be able to start connecting the dots by yourself in no time.

I've already connected the dots of the HN zeitgeist. Besides this was done by a few individuals at Meta, or are you thinking they had a board meeting and shareholder vote on it?

even if we completely ignore organizational structure, do you truly believe not holding companies accountable for a few rogue employees is a good call? Is it too difficult for higher-ups to blame the rank-and-file and arrange scapegoats in the opaque black box that is a corporation? and even still, we can ignore this potential precedent and focus on motivation only: if an employee uses illegal means as a tool to reach their work goals, isn't an investigation into said work goals and culture warranted?

This is all already happening, that's why we know about it. But there's also nuance. Was piracy Meta corporate strategy, as implied ad nauseam on here, or was it some guy taking a shortcut?

Is it actually bad that Meta trained their AI on books? No, court already decided that it's substantially transformative and doesn't harm the publishers. Should Meta employees have stolen the books? No, obviously not. The middle men need their cut.

"Move fast and break things" A guy taking a shortcut is the ethos of Meta, it's the DNA.

For the courts, it seems it's the opposite.. Meta, Anthropic and others seem to be getting away with terabytes of piracy, on a much larger scale than the typical consumer

Linux is also good thing on the internet.

Definitely!

So I guess we need a go-to "good things on the internet" list :)

I submit libgen / annas-archive

Basically. Things that don’t rely either ads or data collection.

I think that’s the key. When users pay, the company works to serve them better. When advertisers pay, the incentives are to attract users by any means (blinking lights, big emotions, fraud, etc.) and keep them on the website longer.

The latest example of this behavior for me is recipe websites that do SEO to get the top spot on Google and then serve you a 30-page long webpage full of ads with the recipe at the bottom.

I think a huge part of the world’s current problems come from “news” sites that are funded by ads.

Add musicbrainz to the list. It was weirdly comforting to rediscover it 15 years later with nothing really having changed (for better and worse)

Reply on news.ycombinator.com