-ish. I often keep md files around and after a successful task. I ask Codex to write the important bits down. Then, when I come around to a similar task in the future, I have it start at the md file. It's like context that grows and is very localized. It helps when I'm going through multiple repos at multiple levels.

I’m also doing similar with fairly decent results. AGENTS.md grows after each session that resulted in worthwhile knowledge that future sessions can take advantage of. At some point I assume it will be too big, then it’s back to the Stone Age for the new agents, in order to release some context for the actual work.

As a devout Baptist minister, this is likely about one of two things, avoiding the appearance of evil (gambling, 1 Thess 5:22 - Abstain from every form of evil), and giving up something for the sake of others (gambling addictions within the church, Rom 4:21 - or do anything that causes your brother to stumble).

The reality is that most churches recognize that they were too legalistic in the past, and so now address things like gambling more directly, and are perfectly ok with playing cards. FWIW YMMV :-)

I was under the impression that the injunction against playing cards was because of their proximity to tarot/occult practices. Mormons had the same injunction against playing cards until the 80s, when the teaching was no longer promulgated. Speaking as a former Mormon...

Here in Sweden, where we also have free churches such as Baptists, Laestadians etc., the concern was definitely about gambling.

I think that's not wrong. Same principle, different sin... it looks like gambling, or the occult, or...

Sure, but let's say you do EKS, you set it up once and then it's mostly done, including security, etc. You set up your own, then you upgrade every 6 months manually.... this is a cascading cost.

This takes time and effort, thus, lost opportunity cost. The thing that makes these providers worth it, is that it lets the business focus on their core competencies and just add-on as they scale without worrying about complexity. A business owner who hyper-optimizes for every contract is unlikely to be focusing on growing their business, even if their business is more efficient on paper.

> This takes time and effort, thus, lost opportunity cost.

Why should we assume this for every type of business.

> The thing that makes these providers worth it, is that it lets the business focus on their core competencies and just add-on as they scale without worrying about complexity.

Since when? Mastering the complexity and implementation of infrastructure from US cloud providers is a skill that takes time in itself. Personally I don’t see how Scaleway does not provide the same for example.

At some point we have to question are we choosing AWS, GCP, or Azure out of brand name, convenience, and marketability. Our if they actually enable faster business execution, higher availability, security, and regional compliance that alternatives don’t…

It would make me very uneasy to have my company be 100% dependent on another company. It sure is easy and convenient to just go to AWS/Azure/GCP pick all the components I want and plug them together but I'd say leading a company is not always choosing the easiest but sometimes the most sensible option.

Let’s say there’s a balance between the two, and maybe optimising a bit more is currently a good idea for various reasons…

Missed a link?

I found this pretty interesting, but would love to either see external inputs (news feed), or perhaps give the models more space. It seems like they might start collaborating more fully if they only had more space.

I'm curious if running them at slightly lower voltage would fix it or if it's a software thing.

or (3) get a battery

My first thought was why would someone halt their socials? Too much holiday time? Then I realized this was social media :-P. Socials to me are precisely NOT social media. Is this a common language usage now?

Yes. “Socials” has been functionally interchangeable, especially in written form, for probably 2+ years now, in my experience.

Not where I live if you're older than middle school age

this might be a generational thing or a geographic thing, my nieces and nephews use the word "socials" while I always use "social media".

Yes

Large software projects are an interesting use case because once you get large is precisely when the framework becomes valuable.

A large enterprise project will need security, testing, auth, (AI now too). I'd hate to implement SAML without a library, that would be torture, and likely incompatible with most systems.

While I've often written small self projects from scratch, I wouldn't dream of building a large one that way unless you are sure to have an army of engineers and QA.

As an aside, this is where AI code fails as well. Speed of dev is easy, stability over time and compatibility is hard.