750 comments but I can't believe how bad this article is. The paragraph on MAU says similarweb is untrustworthy but then cites "most estimates"...which? Sorry,

750 comments but I can't believe how bad this article is. The paragraph on MAU says similarweb is untrustworthy but then cites "most estimates"...which? Sorry, I can't read this without thinking you're biased against Ed Zitron. The best part is the spreadsheet, which makes him sound like he's just sloppy, but danluu specifically says he's being deliberately misleading elsewhere, like the opposite of hanlon's razor now? So is he being deliberately sloppy in his spreadsheets or is he just incompetent? I suppose both is a better read but you need more proof for the "deliberately misleading" line and the spreadsheet frankly is danluu's best evidence beyond the list of failed predictions which is less striking in my mind. While I think he's probably wrong on the timing, which clearly has been wrong since 2024, it doesn't really address the greater critique, which is that ai companies and nvidia and memory manufacturers need revenues collectively in the trillions to make their investors whole. That is still unsustainable such that it's been called out by other investors and financial journalists.