michaelbuckbee

Generally speaking it has a lot of information from things like OP's blog post on how best to structure the file and prompt itself and you can also (from within Claude Code) ask it to look at posts or Anthropic prompting best practices and adopt those to your own file.

What I struggle more with the things like Grammarly, where it's a mix of fixing very nitpicky grammar spelling structure issues that push things from casual writing with my own voice into more of a professional tone.

I agree with you. All I can think is that it's affecting some central mental or physical aspect that has many different outward benefits.

All I could think looking at the (amazing) dioramas was how much time they could save with something like the EufyMake Printer that could print out all the textures they're making by hand (or 3d printing).

Adam Savage review: https://www.youtube.com/watch?v=SD8r1UBwLL0

Something I often find are "categorical" bugs where it's really 3 or 4 different bugs in a trench coat all presenting as a single issue.

Speed is considered one of many ranking factors. I think less because Google explicitly says "we will rank you higher if your site is faster" and more "the user behaviors we track will improve if your site is faster".

This is things like going from the SERP -> Site -> Back to the SERP.

That's implicitly telling Google that the site didn't satisfy their search intent and something they for sure track.

Maybe the user clicked back b/c the page didn't load fast enough, maybe the content was bad but either way it's a search quality signal.

In my current personal development workflow with Claude Code, I've switched entirely to using CLI tools and scripts over MCP as the experience is much more deterministic and flexible.

A great example is Github, it's a significantly better dev experience having CC call out to the gh cli for actions than trying to invoke the MCP.

Something missed in the continual tracking/privacy discussion is the primary motivations for doing so, with the assumption that it's about advertisers trying to sell you more stuff.

I'm 100% in the need for personal privacy camp, but mention this only because without addressing the underlying issues, it's hard to come up with larger solutions.

And the big issues really come down to fraud and cyber attacks:

- Years ago, the NYTimes was found to be doing some kind of homegrown fingerprinting with canvas. They have plenty of ways of doing analytics and tracking subscribers, they were trying to root out ad fraud.

- When Spotify has people every day putting up AI-generated streams and attempting to "listen" to them with bot networks, they start to look at things like fingerprinting.

- Massive credential stuffing attacks on sites are often thwarted at the technical level by fingerprinting.

- Bot traffic (and in particular AI indexing bots not respecting robots.txt) has shot up dramatically in the last year, and fingerprinting is one of the strongest ways of bot identification.

Again, want the personal privacy, but think we need to fix the professional problems to get there.

The poll mentioned in the article has 56% of respondents saying that Brexit was a bad idea, so roughly half seem ok with it.

Different scientist and different set of experiments, but a much more fun and educational video of bees + time:

https://www.youtube.com/shorts/xlGuBT5GT10

This feels the most like Cloudflare branching out to something closer to a traditional hosting model and further from their networking efforts (though things like workers and R2 certainly have blurred that line).

Notably, Amazon has already invested $8 billion in Anthropic/Claude, so I'm hoping this is actually something wildly different and with a different approach.

There's that maniac who is building a quad-copter skateboard contraption who got in trouble with the FAA who successfully reported that he was flying, but got fined for landing at a stoplight.

Considering they have a pure service in selling Geforce Now (game streaming), that doesn't seem in any way far fetched.

For years, both the financial and sports news sides of things have generated increasingly templated "articles", this just feels like the latest iteration.

I've heard experts comment on this from the other side, that they'll give a quick layperson's soundbite about their subject of expertise that doesn't defensibly lay out all the possible exceptions and edge cases and weirdness for reasons of time and audience interest and then they'll be inundated with comments calling them a liar and accused of falsifying things or not actually understanding the subject.

A better term might be antagonism. X seemed to switch to a system of rewarding views as a method of engagement far above all else, which led to people (generally and deliberately) ramping up the extremeness of their hot takes in a bid to get as much attention as possible.

A parallel term is "hate click", where there's a headline that's so stupid or off that you click it just to see what the hell they were talking about.

An example of this vile genre was someone tweeting about how:

"Teachers make plenty of money, and I think they should provide school supplies to their students out of their own pocket instead of making hard-working parents pay for them."

It was a message _designed_ to get people to yell at them, and for all of that, it wasn't any of the really hot-button stuff around politics, race, or any of the other divisive things that drive antagonistic engagement.

Twitter could have (and previously did) reward all sorts of other types of engagement, but the shift to rewarding divisiveness was just at another level.

To your point, AT&T's "You Will" commercials started airing in 1993 and present both an optimistic and fairly accurate view of what the future would look like.

https://www.youtube.com/watch?v=RvZ-667CEdo

You'd need to do the "/clear" or other context manipulations.

Don't sleep on using Claude Code to improve your Claude Code config. Switch to plan mode and try the following prompt:

read the document at https://blog.sshh.io/p/how-i-use-every-claude-code-feature and tell me how to improve my Claude code setup

The prerequisite thought here is that you're using CC to invoke CLI tools.

So now you need to get CC to understand _how_ to do that for various tools in a way that's context efficient, because otherwise you're relying on either potentially outdated knowledge that Claude has built in (leading to errors b/c CC doesn't know about recent versions) or chucking the entirety of a man page into your default context (inefficent).

What the Skill files do is then separate the when from the how.

Consider the git cli.

The skill file has a couple of sentences on when to use the git cli and then a much longer section on how it's supposed to be used, and the "how" section isn't loaded until you actually need it.

I've got skills for stuff like invoking the native screenshot CLI tool on the Mac, for calling a custom shell script that uses the github API to download and pull in screenshots from issues (b/c the cli doesn't know how to do this), for accessing separate APIs for data, etc.

I'm not actually sure about that (that turning it into an MCP would help). I've seen more momentum building around having better cli tool integration with ClaudeCode than MCP reliance.

He's already been cast as (and I can't believe I'm about to type this) Skeletor in the Masters of the Universe reboot.

I think the point was more that even if you're on a coast a lot of other problems are more likely to impact you first: higher energy and more frequent hurricanes, insurance, costs, etc.

In John Scalzi's Old Man's War, there is a discussion of how the more advanced society that they're interacting with deliberately put a space elevator on Earth, not because it was the easiest or cheapest solution, but as a sort of constant reminder of just how much more technologically sophisticated they were.

I dug down trying to get the actual paper and it appears it's a pre-print with only the abstract available (but maybe I wasn't looking in the right spot).

That being said, the quotes from the author were more to the point of it being a milestone that in the UK that Solar+Battery systems were now less expensive than gas/coal.

To my understanding, this is a milestone vs "raw" production numbers, which you're correct in saying have had solar as the cheapest option for years.

The author is describing the typical pattern of these things (SIM farms, email spam hosts, Proxy Networks, etc.) which don't typically get VC money or large investments, but grow organically.

The stats are "regularly revised" (sometimes up, sometimes down).

Though honestly, I wish the terminology were changed to "forecasted" and "actual" to be clearer.

This feels like a very knee-jerk reaction to this and not what's actually the case: a new system with weird bugs.

It just seems very similar to the sort of articles that came out when online ordering or touchscreen ordering first appeared.

Like one of the big knocks on the Taco Bell AI ordering was that it let people ask for a 1000 waters on their order, which yeah is dumb, but it's the kind of thing the humans actually making the food are going to catch.

Anthropomorphizing big industries is always fraught, but looking at their actions, I'd concede it means something that the American Gas Association is doing things like hiring influencers to promote gas cooking over induction ranges.

https://www.treehugger.com/the-marketing-of-gas-stoves-never...