This will probably be the best primer on BTC+blockchain for generations to come, enjoy: https://www.youtube.com/watch?...

After 2011 you pretty much had to join a public pool to produce any reward. There was a tiny window of about a month when the very first asics came out where my $150 Butterfly 5GH/s was doing almost 0.1BTC a day in Slushpool (early 2012 iirc). That was pretty high because of the low difficulty, then difficulty quickly skyrocketed as the asic (and gpu) floodgates opened. But adding your hash to a pool was still the only realistic way to get any reward, solo was dead at that point for the little guy.

Have I ever told you about that new programming language Apple created a few years back?

Think that's just referring to the Home assistant addon. Pihole is still going strong. It's overkill but I have a VM instance for Pihole running on my basement server under Proxmox providing a dedicated IP for serving local DNS. Then, as an backup, an old RPi under a secondary IP for the rare instance when the server gets rebooted (or the small downtime when nightly VM backups run). Simply point your network(s)' DNS on the router to those primary and secondary IPs and everyone in the house is covered. To keep things simple start with a RPi or spare machine then graduate to redundancy like above.

But these days the easiest way to do blocking is to use one of those ad-blocking DNS providers (e.g. NextDNS). Just point your router's DNS to them and done. But even with a paid subscription, like all the scummy VPN providers out there, I'd be wary of what browsing data any of those companies are selling for added revenue. NextDNS appears to be above board but caveat emptor.

True, but Community Notes has been working pretty well in the fight against mis/disinformation, imo. The ability for a volunteer group to quickly assemble an explanation and set of source links that directly rebut the offending tweet's content has been pretty powerful. Then it's up to the reader "to decide". That would have been unthinkable under Twitter's old regime of dealing with the problem exclusively through employees' banhammers+censorship.

Doesn't mean that community notes couldn't devolve into politically correct groupthink again but time will tell. Even with its flaws, the moderation concept has worked decently here for 20+ years.

Same, been fighting Honda's collision ghosts for a year now. It appears there are just too many corner cases for it to be reliable, mainly in the implementation of the detection algorithms, but also by adding in hardware variables (e.g. misaligned or dirty sensors, etc.). Albeit infrequent, it's been dangerous in some cases.

Can't speak to the math side, but he's a very good writer imho. For example, his ChatGPT post/book is an enjoyable primer on LLMs. Helps fill the void between articles that are too fluffy and those that assume too much.

Case in point, Mozilla's embarrassing lack of due diligence with this lowlife:

https://krebsonsecurity.com/20...

Doesn't help for Safari but for uBlock Origin users in general, adding this to your Settings->My Filters kills it:

||accounts.google.com/gsi/*$xhr,script,3p

In addition, the Context/Community Notes feature has been a useful tool against disinformation campaigns. It's almost always based on high-quality information sources with a clear description and links. For those unfamiliar with it, here's one example of many that popped up in my feed earlier:

https://x.com/sandylocks/statu...

Just checked and I'm currently at 98.6 degrees F. Albeit n=1, your assertion does appear to be true.

Heh, fair enough. And to be honest, as someone naturally skeptical of such situations, it still surprises me each time it works out well.

Absolutely, and you do need to use common sense to make it all work. Like capping the risk to a comfortable price level combined with the possible lack of warranty (mine has worked out to under $200 items). And the type of item. Bought a save with used SSD for $50 last week that was $100 new. The thought being that it was listed as "good" so even if it was opened, there's no way someone wrote 500TB on it before returning. Sure enough it came factory sealed, someone just decided to return it unopened. Then Amazon graded it "good" instead of "like new" which would have reduced the discount.

Stuff like that.

I've shifted almost all of my Amazon orders over to buying "saved with used" when available. Haven't kept track but the money saved over the last few years instead of buying new is probably well over $2000. Even the stuff marked "good" is usually in sealed packaging and, depending on the item, can be up to 50% off. Never had a dud so Amazon must be doing a really good job grading the returned stuff that is offered at a discount.

I don't know how the economics work out, especially with the double transport costs and packaging. But it does mean that a perfectly good item does have a path away from the landfill. As in nature where the vultures and maggots digest the fallen corpses, there are guys like me doing our part in ecommerce.