There’s no z in the Valve equation, though, because they’re not locking you in to their ecosystem on the device.

You’re right about Sony: They sell a console for x (cost of console) - y (subsidy) + z (profit from Playstation games). They’ve calculated that z is going to be significantly more than y over the life of the console, on average. Nearly every game anyone ever plays on a Playstation will involve some amount of money going to Sony. You can’t really do anything else with it.

But with Valve, if they sell a console for x (cost of console) - y (game voucher), there’s no guarantee of any z (profit from Steam games). A ton of people could (and probably will) buy games from third-party stores, or only use the Steam Machine for retro gaming, or pirated games, or as a media console, or have already-existing Steam libraries. A lot of people would use that voucher and never buy anything else on Steam, or only buy a few games on Steam sales. Only a fraction of games ever bought will go to Valve because it’s not a locked-down console.

Maybe if they gave percent-off coupons, but otherwise it’s definitely not. The big console companies subsidize consoles because they know people are going to buy games for that system, probably on their digital storefront, and thus they’re going to earn that subsidy back in the profits from those sales. Even if people buy physical copies, the licensing fees come back to the console manufacturers.

But with Steam credits, that’d just be loss on the other end.

If Wikipedia was truly defunct, the idiot right wouldn’t be trying to destroy or commandeer it.

So, going through your claims: two cases in which the system worked as intended, one case in which the outcome is uncertain because the facts are uncertain, and one secondhand case about a minor celebrity in which the facts may or may not be certain (if there was actually a legal case here, it would be pretty easy to prove and get it taken down). Not only are these not major, it seems like everything is working correctly!

Edit: also, you clearly misunderstood what I said about the former president. I was saying it was a long time ago.

“defunct” 😂

Where have you seen this happen? I think the last time I saw any type of significant misinformation last on Wikipedia for longer than a few days at most, particularly without a disclaimer being added, George W. Bush was still the President of the United States.

Sounds like you need to read up on what Wikipedia actually allows as a reliable source before you complain about what Wikipedia actually allows as a reliable source.

Gonna need a source for that, boss–

1. Wikipedia’s intense scrutiny of sources and requirements for reliable citations are actually one of the reasons that Sanger started his malformed crusade.

2. If there’s actual, provable lies about a notable person in the encyclopedia, then there should be actual, provable truths to combat it; and any Wikipedia editor can update the article in question to correct the record. If an edit war emerges, a community discussion can take place wherein the person in question can have their say. Wikipedia isn’t the wild west, and any reasonable argument that it is died twenty years ago.

–but even if those two things weren’t true–it’s literally impossible to remove misinformation from AI models. I’m not saying that to be dramatic or overstate the problem. When a model is trained with misinformation, the misinformation becomes a part of the model; the entire corpus of everything it was trained on is baked into the neural network on a fundamental level, and humans can’t manipulate it manually. Which means you can’t remove any datapoint from the model without excising it from the training data and then retraining a whole new model.

So now not only are you drinking your dog’s urine, you’re claiming that the tap water is too yellow. Even if your assertion’s true, your alternative is demonstrably worse.

Not sure if troll or…

…ok, I’ll bite. This is maybe the stupidest thing I’ve ever seen. You know that Wikipedia comprises one of the largest publicly-accessible sets of meticulously-curated natural-language data on the planet, right? And so when you’re training up a new model, you’re naturally going to start with that massive, freely-available repository?

Every major AI model has been trained, at least partially, on Wikipedia. This insane viewpoint is essentially saying that you don’t think Wikipedia is reliable, but if some linear algebra chews it up for a few minutes, then it’s ok. You’re turning up your nose at tap water, but drinking your inside dog’s urine.

Wikipedia also has a well-known and well-exercised rule called the Snowball Clause. Basically, if consensus is clear early, don’t waste others’ time or let further harm occur while you’re waiting for a procedure to reach a foregone conclusion.

In the ban discussion, you can see a clear consensus developing very early that Sanger was, by his own admission on external platforms, NOTHERE to build an encyclopedia. He further doubled down on his position in the discussion thread for his own banning, and even intimated that he was rejoining Wikipedia with the express purpose of gathering enough meatpuppets to change Wikipedia’s policies. The Snowball Clause is even mentioned in that discussion; basically, had Sanger and his meatpuppets been allowed to continue to edit for those three days, the damage to the encyclopedia could have been significant.

Oh, interesting! The last time I tried it was long, long ago, so I didn’t realize that they had expanded that configuration. Or maybe I just don’t remember seeing it.

I don’t use sponsorblock myself (though I don’t begrudge anyone who does). For a couple of reasons:

1. I don’t trust SponsorBlock necessarily, since it’s community-noted. It’s become clear that I have different opinions than others about what constitutes a sponsor spot. uBlock can be more certain about what constitutes an ad or not, since it comes from a different domain than the content, but with sponsor spots, they’re part of the same content stream.

2. Sometimes host-read ad spots are actually clever, or integral to the video in some way.

3. I have a lot more sympathy for the individual creator who gets all of that money than I do for the trillion-dollar multinational conglomerate.

In any case, I am typically pretty good at skipping ahead. And if the sponsor segments get too onerous, I tend to just stop watching that channel.

EDIT TO ADD: I’ve been informed that SponsorBlock now does a good job of solving the first problem by categorizing sponsor spots. I’ll have to try it out again.

Yep, I block the embedded ads from YouTube, too. I don’t use SponsorBlock to automatically skip the ad spots that the creator put in themselves, though. Sometimes they’re actually clever, but more to the point I have a lot more sympathy for the individual creator who gets all of that money than I do for the trillion-dollar multinational conglomerate.

You’d think so, but no! My only real ad surfaces anymore are YouTube (sponsored spots), podcasts, and billboards. I am very good at skipping podcast ads and sponsor spots on YouTube, but when I don’t I mostly just fume about how I can’t for whatever reason (usually when I’m washing dishes and my hands are wet). Billboards are easy to ignore most of the time, too, because on my regular routes I know where they are and have apparently trained myself that there’s not anything of interest there.

I know why Discord isn’t offering that, but there are at least three FOSS options (Element, Stoat, and Spacebar) that could really use a good way to get content off of Discord and into an open format (sure, Matrix is a solid option). Any data that can be accessed publicly can be scraped, even if it’d take a while. I think one of those teams should be considering this.

I just want one of these Discord alternatives to code up a “migrate-from-Discord” script that admins can use to scrape all of their messages and media, save out their channel structure and access rules, and pull their user list for later mapping, and box it all up into some sort of standards-compliant package.