Tonight on GeekNights, we consider the ramifications of a ridiculous Texas law being argued before the Supreme Court, and its ramifications for the imperfect but valuable Section 230. In the news, A lot of Redditors hate the Reddit IPO, Reddit is selling your data to Google for AI training, and Reddit admitted that your moderator protests are working. Also Rym has some notes on his new workstation.
Also I tell the story of my ski binding literally exploding at Hunter this weekend.
No livestream tonight because I’ve only just gotten the GeekNights pipeline migrated to the new workstation, and I haven’t build a new streaming pipeline yet. It’s due for an upgrade anyway 
Spam filtering seems like an obvious casualty if the court decision goes the wrong way.
Gmail - spam is whatever our users say it is. Why not do the same on youtube, facebook, etc? Rym even hinted at something crowdsourcey like this in the episode.
If we end up in the world where centralized systems end up just allowing everything, then there will be a rush to build and sell systems that allow users to control their own filtering on the client side. A lot of money could be made by whoever makes the best ones.
This could be somewhat good because it gives users control over what they see and don’t see.
It will mostly be terrible. Unsophisticated users will bee ill equipped to filter well, and their experience will be ruined. A great deal of computing and energy resources will be wasted if the same spam message has to be detected and filtered out thousands of times instead of one time.
Spammers and other evildoers will actually get their messages through more often. If they have a few thousand tiny battles, they’re going to win more than 0 of them. If they lose a battle with a big centralized force, like GMail, they get no messages through.