3 ms·
Yes, Twitter is so “successful” Musk had to write off 50% of his “investment” out of the gate, and had to come up with an AI-hyped merger with SpaceX to compens
by xondono 7d ago
Yes, Twitter is so “successful” Musk had to write off 50% of his “investment” out of the gate, and had to come up with an AI-hyped merger with SpaceX to compensate the banks that went in with him.
The only reason Twitter has more active users is that it removed some of the protections for bots, in particular broad filtering coming from certain country involved in a war.
You don’t think Musk is any close to the Oceangate guy? Let’s review some interesting decisions by his companies then:
- Musk mucking around with the blue checks and the bot filters in Twitter skyrocketed the amount of scams in the platform.
- Tesla uses very strict NDAs and settlement to silence the reporting of accidents
- The infamous electric door handle that locks up Teslas when power goes out was an obvious point of failure, it was ignored, people died, and now several countries have legislation against it
- Reportedly Musk had safety lines and other safety features in the factories removed for serious reasons like “not liking yellow”
- Any understanding of how to parse public insurance data will show you Tesla’s have accident rates that are multiples of other brands
- The Starship, a vessel that is intended for human transport, has no safe abort system, no way to fit one in its current state, and no plans to modify that.
Just a reminder that even a very SpaceX-friendly count (in which flights that did not attempt to land are considered “good”) puts the current reliability rate at <40%.
As a naive boundary, for clearing the 97.5% predicted crew survival, SpaceX would need to fly in the order of 300 good Starship launches.
Those are just of the top of my head.
- ETH_start 6d agoIndependent web traffic analysis shows that the average X user spends 12 to 14 minutes on the site. It's impossible that these are bots.
- xondono 6d agoAre you seriously claiming that as a fact? You really believe 12 minutes is too long for a bot or something? Because that has to be the lamest excuse I’ve ever heard. Measuring bot activity became harder after the API changes (I wonder why!), but there’s been some attempts at finding ways to estimate bot activity. All those measures are compatible with increases in bot activity, and incompatible with bot decline.
- ETH_start 6d agoThis is not what I personally believe. This is what experts in web analystics believe. Bots don't behave like that. They don't simulate human browsing activity by staying on a site for 10 to 12 minutes, entering the site directly (through url or bookmark), and browsing 14 pages during their visit.
- vinay_ys 6d agoBots are perfectly capable of simulating long user sessions. Many large-volume scrapping traffic for example simulate lots of long persistent user sessions.
- xondono 5d agoExperts from 2005 I guess, because that’s the last time I heard about bots not being able to mimic human-like interactions. Everything you’re talking about is trivial to automate with playwright.