I realise that this is unpopular. But personally while I disagree with the decision to charge (exorbitantly) for the api and appalled at the slander hurled at the dev, I think that is an business choice and one more item that I have to disagree and live with.
But I am very excited about the rise of the fediverse. I know that a company will eventually make a decision that I feel very passionately about, but I will be stuck making a difficult choice. With the fediverse, it provides the users with the opportunity to have control. This power of course often comes with various other costs (lack of a dedicated sre or moderation teams, etc). But I expect that over time this will evolve into options where paid offerings will come up that allows for higher QoS where required.
The api changes really were about protecting their gold mine of data from ai data models scraping for data. Reddit wants to use that data to create its own models and then replace moderators with those models. The ultimate goal here is to turn the existing dataset into an automoderator on steroids that they could sell anywhere. Trouble is someone else is going to beat them to it.
There was a reason these changes lined up so nicely with Google doing the same thing. Everyone’s realizing they’ve been spouting their gold from firehoses for any machine to pick up, and they’re being reactionary and turning them off asap instead of just like, accepting it as a facet of having a public social network.
I realise that this is unpopular. But personally while I disagree with the decision to charge (exorbitantly) for the api and appalled at the slander hurled at the dev, I think that is an business choice and one more item that I have to disagree and live with.
But I am very excited about the rise of the fediverse. I know that a company will eventually make a decision that I feel very passionately about, but I will be stuck making a difficult choice. With the fediverse, it provides the users with the opportunity to have control. This power of course often comes with various other costs (lack of a dedicated sre or moderation teams, etc). But I expect that over time this will evolve into options where paid offerings will come up that allows for higher QoS where required.
Honestly, if spez hadn’t already sold the site to white supremacists, I’d be a lot quicker to defend this.
Who are the white supremacists he sold to?
It was the Chinese that he sold out to. Not the white supremacists.
The api changes really were about protecting their gold mine of data from ai data models scraping for data. Reddit wants to use that data to create its own models and then replace moderators with those models. The ultimate goal here is to turn the existing dataset into an automoderator on steroids that they could sell anywhere. Trouble is someone else is going to beat them to it.
There was a reason these changes lined up so nicely with Google doing the same thing. Everyone’s realizing they’ve been spouting their gold from firehoses for any machine to pick up, and they’re being reactionary and turning them off asap instead of just like, accepting it as a facet of having a public social network.