alexjurkiewicz

I think half the article talks about why CL is good because it can resume from an exception, and half the article talks about why DSLs are good.

I think others have pointed out that most modern scripting languages can halt at exceptions without unwinding the stack. Python & Node both support this with core tooling.

As for DSLs, they constrain the LLM which generally helps with code quality. However why implement your DSL in the unconstrained chaos of CL? You can write DSLs in Rust which gives you static typing, a borrow checker, and clippy.

The first time you get an AI email, it's interesting. The hundredth time it won't be.

It would probably look a lot like existing S3 concepts that have explicit hot and cold tiers. For example Intelligent Tiering, or Glacier.

> While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models

This is covering for Tristan saying something like, "Actually, I was using my friend's account for half of this work".

It's four pages below the fold. The fact this is a paid tool is deliberately obscured.

> if (!line.accept('[').isEmpty() ) // [section] header.

Is this really ergonomic?

Because IOT devices are cheap and networks are ad hoc, flaky connectivity is a likely occurrence. Thread must be reliable in these situations.

How many houses in Britain have wiring that is less than 800 W away from starting a fire? How many of those houses have the disposable income to buy a plug in battery?

Yes, more electricity means more risk of fire. But 800W batteries do not seem like a major safety issue

Come on, "not publicly available"? You know how git works right? Each Claude release probably has the HEAD of main compiled as "1.4.0" until there is a real official release.

Paying customers get LTS. Are any paying customers asking for a Zig branch LTS? Or are you expecting open source maintainers to do free work for no particular reason?

AI training uses wildly more compute than most companies, who are generally building domain specific CRUD apps.

Compare AI costs per-engineer-salary-dollar, because more expensive engineers probably need more expensive AI.

It was Gemini Flash, probably an even faster variant optimised for immediate response on search pages.

Showing overseas based workers of Microsoft as another company name on caller ID is a phishing risk.

It might take a large org several years to migrate off core systems like VMWare. If you think the customer is likely to churn within a few years anyway it makes economic sense to hike their fee.

As the article mentions towards the end, AWS EKS, GCP GKE, and other competitors have made k8s setup turnkey. You can deploy a new cluster with all the controllers you mentioned in a single click / Terraform.

AWS ECS and GCP Cloud Run are this. Run a container on abstract compute. But they aren't "without all that complexity" because it turns out all that complexity is required for even simple use-cases. Load balancing with SSl certs, cloud API keys, deployment pipelines, sidecars, etc.

If the real credentials owner was running the agent, why do it from a new GitHub account?

Someone's bug tracker account was hacked.

Firstly, none of this post is the contract people are signing. So it's merely a summary.

Secondly, like all contracts I'm sure there will be exceptions for holding data longer than 30 days with reasonable cause, eg a legal hold.

NPM is getting all the attacks and attention because it is the biggest. But there's nothing language specific to this class of attacks.

Nice. It would have been interesting to dive into the exact differences in components. Are higher wattage PSUs in a series just using higher rated components or does the quality go up too?

I agree that many skills are overblown and unnecessary. But there's a lot of value in giving AI the right process. See how much more effective Claude can be for moderate or large changes when using the superpowers skill.

There can be problems with open source projects run by for-profit companies, but this fork seems a little premature.

And it's unnecessarily rude. Grammarly and Hemingway can identify the same sort of issues without "you are a stupid robot" vibes.

If Finney and Back were working together, why does Finney post a lot as himself through Satoshi's existence while Back does not?

That implies it's just Back who was the Satoshi poster. And if so, you don't have any evidence Finney is a technical co-founder.

Doesn't that break password manager autofill?

"I'm trying to do something illegal and Anthropic are aware. Why do they keep banning me??"

Claude is leaning into the idea of a local "session" being the host where everything connects.

I guess this makes sense for now. You can build integrations leveraging the user's personal access credentials. Later, once Claude takes over the world, they can move sessions to live in their own walled garden.

I wish sudo-rs and all projects being used in production would suck it up and release a major version v1. Seeing v0.xxx on a program used by Ubuntu as a base system utility is ridiculous. You don't have to try and contort the version numbers of a user-facing utility to fit into the semver API spec, but at least bump to v1.0.0 and continue from there!

I enjoyed reading this as a reflection on the 90s/00s MMO era, which has been slowly dying ever since. Game designers do seem to have learned the lessons of this time. Well, at least they are no longer building games where 'multiple fictions' are on offer. The idea of developers supporting player-generated content (remember 'UGC'?!) is also dead.

What remains in MMOs place are massively singleplayer/co-op experiences. ARC Online, Helldivers, etc.

The Web of Trust failed for PGP 30 years ago. Why will it work here?

For a single organisation, a list of vouched users sounds great. GitHub permissions already support this.

My concern is with the "web" part. Once you have orgs trusting the vouch lists of other orgs, you end up with the classic problems of decentralised trust:

1. The level of trust is only as high as the lax-est person in your network 2. Nobody is particularly interested in vetting new users 3. Updating trust rarely happens

There _is_ a problem with AI Slop overrunning public repositories. But WoT has failed once, we don't need to try it again.