bmcahren

It's easy with hindsight to believe you could have capped expense at 200% medicare but getting what we got passed was nearly impossible at the time. Before Affordable Care Act, insurers had every tool available to deny care, maximize profits, and skim more than 20% off the top. It's great we're getting closer to the point that it feels to you like incompetence that these things aren't fixed today but your anger with the medical lobby is clearly misplaced here.

Every major piece of legislation needs revisions to chase circumvention and we're well past due on updates but no legitimate bills have been presented that cover these topics and that's not a one-party issue.

Those with a bit more experience understand faster is not always better. Databases thought to be battle-tested encounter incredibly complex and near impossible to predict failures of the most absurd kind. You can go back and look at some crazy behavior hundreds of people have worked to resolve regarding TTL contracts within Redis.

The ease of "appearance of value" today is the uncanny valley for software. The repo looks professionally organized, you can PAY for it, the preliminary benchmarks are looking good. Overlooked are the testing, validation, backup, failure recovery, practical behaviors, and most importantly: honesty.

These projects would get more love if it was declared up front that they were heavily AI generated projects and shouldn't be used in production since it has the air of practical utility.

It's probably a great drop-in replacement for Redis for a raspberry pi project that has low stakes. The smaller 1MB disk footprint and the performance difference could be impactful. Personally, I wouldn't be using this in production for at least a few years after hobbyists have their go at revealing its hidden near-guaranteed flaws.

At least I can broach TTL issues and gather reasonable insight on Redis vs Elasticache nuance based on the thousands who have encountered the issues.

LLMs do not encode nor encrypt their training data. The fact they can recite training data is a defect not a default. You can understand this more simply by calculating the model size as an inverse of a fantasy compression algorithm that is 50% better than SOTA. You'll find you'd still be missing 80-90% of the training data even if it were as much of a stochastic parrot as you may be implying. The outputs of AI are not derivative just because they saw training data including the original library.

Then onto prompting: 'He fed only the API and (his) test suite to Claude'

This is Google v Oracle all over again - are APIs copyrightable?

MongoDB Atlas was around 500% more expensive than in-house every time I evaluated it (at almost every scale they offer as well).

They also leaned too heavily on sharding as a universal solution to scaling as opposed to leveraging the minimal cost of terabytes of RAM. The p99 latency increase, risk of major re-sharding downtime, increased restore times, and increased operational complexity weren't worth it for ~1 TB datasets.

A huge benefit of single-database operations at scale is point-in-time recovery for the entire system thereby not having to coordinate recovery points between data stores. Alternatively, you can treat your queue as volatile depending on the purpose.

Missed a huge opportunity to play the sound of a monstrous wooden door sound when the lid closes. Looking forward to the update!

I think the other important step is to reject code your engineers submit that they can't explain for a large enterprise saas with millions of lines of code. I myself reject I'd say 30% of the code the LLMs generate but the power is in being able to stay focused on larger problems while rapidly implementing smaller accessory functions that enable that continued work without stopping to add another engineer to the task.

I've definitely 2-4X'd depending on the task. For small tasks I've definitely 20X'd myself for some features or bugfixes.

So there's this thing called "Setup Scripts" but they don't explicitly say these are equivalent to AWS Metadata and configured inside of Codex web interface - not a setup.sh or a package.json preinstall declaration. I wasted several hours (and lots of compute where Codex was as confused as I was) trying to figure out how to convince codex to pnpm install.

Counter-point A: AI coding assistance tools are rapidly advancing at a clip that is inarguably faster than humans.

Counter-point B: AI does not get tired, does not need space, does not need catering to their experience. AI is fine being interrupted and redirected. AI is fine spending two days on something that gets overwritten and thrown away (no morale loss).

I'm pro-accessibility and have contributed privately to blind developer initiatives. Unfortunately Ubisoft insists on implement user-hostile accessibility that screams at the user using voice-to-text when they open their games and is quite difficult to get through even as an abled user.

How about Ubisoft work with Sony/Microsoft/Valve and get vision and hearing disability implemented at the device level rather than harassing abled users every new game which I'm sure through this frustration is contributing in some small way to these anti-intellectual movements against accessibility.

From day one. We would have had LLMs years before if Google wasn't holding back. They knew the risk - google search would be dead as soon as the internet were flooded with AI content that google could not distinguish from real content.

Then you could look at how the first "public preview" models they released were so neutered by their own inhibitions they were useless (to me). Things like over-active refusals in response to "killing child processes".

I am going to cite you in a decade. Already today ChatGPT is _far_ better than Google. Instead of finding a keyword optimized page for "frob language", I can get the objectively best sources for frob language and even find the best communities related to it. Zero frob ads, zero frob-optimized pages that are designed to trick google, etc.

Traditional search is dead, semantic search through AI is alive and well.

I can't yet count once AI misunderstood the meaning of my search while Google loves to make assumptions, rewrite my search query, and deliver the results that pay it the best which have the best ads (in my opinion as a lifetime user).

Lets not even mention how they willingly accept misleading ads atop the results which trick the majority of common users into downloading malware and adware on the regular.

This is actually perfect for AI robots to blend in waiting in public. Just like bartenders polishing glasses, you can't have them just staring making people uncomfortable.