DI frameworks are tricky beasts. Either they sacrifice flexibility for simplicity (I’ve seen this done in Go and in Scala, where the DI essentially generates basic instantiation and more advanced resolution is left to the app developer) or they can get really complex but do some handy things (.Net 4.x DI frameworks like Castle Windsor provided some neat lifecycle management tools but was internally very complex).

Cycle detection gets a little hairer the more complex a dependency/ class of dependencies gets. The process itself doesn’t change but the internal representation of the graph needs to be sufficiently abstract enough to illustrate a cycle for all possible resolution scenarios.

Based on the commit to fix the particular bug, it looks like the change will address a specific scenario but will probably fail to address similar issues.

All this to say “the problem isn’t too hard to think about but the solution isn’t straight-forward”, also “this is a fine short- term fix but longer-term would involve redefining the internal representation of a dependency graph”, and finally " An LLM-provided solution is at best a band-aid, in the most generous light.’

If there’s one thing that coding LLMs do “well”, it’s expose the need in frameworks for code generation. All of the enterprise applications I have worked on in modernity were by volume mostly boilerplate and glue. If a statistically significant portion of a code base is boilerplate and glue, then the magical statistical machine will mirror that.

LLMs may simulate filling this need in some cases but of course are spitting out statistically mid code.

Unfortunately, committing engineering effort to write code that generates code in a reliable fashion doesn’t really capture the imagination of money or else we would be doing that instead of feeding GPUs shit and waiting for digital God to spring forth.

randomly placed and statistically average, just like real rivers!

I hear they have the biggest bal of tine in Hond.

That’s a bummer of a post but oddly appropriate during the darkest season in the northern hemisphere in a real bummer of a year. Kind of echoes the “Stop talking to each other and start buying things!” post from a few years back though I forget where that one came from.

Came across this gem with the author concluding that a theoretical engineer can replace SaaS offerings at small businesses with some Claude and while there is an actual problem highlighted (SaaS offerings turning into a disjoint union of customer requirements that spiral complexity, SaaS itself as a tool for value extraction) the conclusion is just so wrong-headed.

The before-games-went-woke sector was large enough to bring us gamergate so there’s definitely a sizable available crossover.

I have my eye on the McDonnell Douglas-branded Neo Geo, which will be a value-engineered trijet and a brick of explosive all while only running the version of worm that was on the nokia brick phone.

robots gonna make us Faygo, better stab a data center and by data center I mean land lord who had nothing to do with AI shit.

the real xrisk was the terror clowns we made along the way.

I’m surprised lagering caves, except for servers instead of beer hasn’t come up yet.

This doubly disappoints me because in a professional capacity I strive for being incredibly intentional and accurate while recreationally I aspire to shitposting, and “more accurate than most” satisfies neither. I really need to get my blood boy to write better material.

it’s likely that the user will spend at least 30 minutes to an hour unable to articulate language.

This presumes Johnson was able to articulate language in the first place, which given that his brain has melted at an incredible pace since 2020 may be a bit of a stretch.

will these guys ever get to rainbow-chart levels of galaxy brain or will just be content to fudge some numbers on regular visuals?

This is reminiscent of memestock/buttcoin charts where a new asymptomatic curve is dropped onto the same rather flat graph over and over again.

hey, there’s been some wild shit lately. I can handle reading about a game company CEO doing a dumb with and without the help of a klanker. I didn’t know if scratching the surface would reveal some hidden horrors.

Do I even want to know what a Subnautica or a Krafton is?

they are too high-IQ to understand.

In other news, it appears that Cloudflare has pulled a Crowdstrike

edit: looks like the problem resolved after a few hours

“I’ve made a self-licking lollipop so the fun times will never end and I see no way this could ever go wrong.”

I see I was indeed being presumptuous. Based on replies to my original (and somewhat incorrectly formed statement) it seems that while the parts of the process are understood in abstract, there’s points in an actual running implementation that become, I suppose, unfathomable or incomprehensible. Is that fair to say or am I being wrong in a different direction now?