How do you actually incorporate these into your prompt? Because I've tried and I've seen many others try and report that it does not work well, especially over long sessions.
I also don't know how much to trust the model, but I've had the model tell me specifically that certain aspects of ASD-STE100 are unactionable and will just create more noise.
The OP's own skill even leads with something in a very similar vein:
> These rules apply to every response for the rest of the session, not only this one. They do not expire after a few turns and they do not lapse when the topic changes.
My understanding is that phrases like this are at best a _very_ weak signal to the model. It's simply contradictory to how the model works at a level that can't be overridden by injecting tokens.
I actually wish the model would go completely in the opposite direction. Except in the rare circumstances where the model has actually measured something, it is hilariously deficient in its concept of time. It will often output phrases like "this relates to <thing> that you did weeks ago", referring to something that happened in the session just a few turns (and hours or a couple of days) ago. Likewise for estimating how long a coding tasks takes, it is hilariously inept. It honestly feels like it rolls two completely independent dice to select a number and a value from (hours | days | weeks) when it needs to attach an estimate to something. I'd much rather read "a bit" than be distracted by these utterly nonsensical times.
The full skill gives this example:
>Bad: "This will take some work." Good: "About 15 minutes if tests already cover this. An afternoon if not."
My experience is that it is very likely that whatever task this is describing takes anywhere from 1 to 30 minutes, consistently. Maybe I just work way faster than the average person.
I think I'd be a strong fit for some of your roles, but the benefits advertised read to me like the absolute bare minimum to meet industry norms. Is there anything you would say is a real advantage for Apex?
I really hate that I'm even typing these words right now, but wouldn't this be one of the few fields where modern LLMs would potentially provide a real capability that was previously lacking? I'm imagining an LLM might be able to provide rough first pass translations and provide hints about which specific tablets might be most interesting for a human that can't possibly look at them all.
I don't know how broad the trend is, but I am a contractor for the US government. I'm told that agencies are actually required by law to reduce square footage. So when they tore down my old building (which absolutely needed to go), they simply could not as a matter of law build a new building with the same square footage even if they had the budget.
The United States population has more than doubled since 1950. There is no "before" that we could actually go back to. The massive, sustained population growth is built on the massive efficiency growth.
Everyone has different opinions about this stuff, but I for one found _working_ in the office to absolutely be the worst part. And since Covid, my workplace has torn down old buildings and reduced square footage per employee, so if I went back to the office it would be even worse.
You're just lucky that your preferred spelling happens to align with Claude's. It is categorically impossible to get any Anthropic model to consistently use American spelling in the last few releases.
I think we desperately need to start differentiating between "is creating" and "has created". I have a couple of "am creating" projects too, but their proximity to "have created" is directly proportional to how much effort and expertise _I_ am bringing, not so much related to the AI's contribution.
The deli is using it to make advertising flyers. Every consumer is using it as a replacement for google search as far as I can tell. I do a fair bit of work with people in the 18-25 range and almost all of them love to rail against AI, but none of them seem to be capable of completing even simple tasks without using it. Even my 63 year old mother uses it at work to generate reports. I would love to see some hard numbers, but I would be truly shocked if "most people still don't really know what AI is" is in any way justifiable across all adults in OECD countries.
I don't think the style of multiple revisions to walk back the most problematic parts really helps the case. It very much leaves me with the impression that the original version is the one that the author really meant, with the revisions being just a weak attempt to deflect criticism.
Not GP, but I don't think a specific other community was being specified. It's more like some hypothetical other community whose primary characteristic is "not Zig".
Meanwhile, as one of those engineers, they ran fiber down the highway a mile from my house circa 2021, but they did not do any upgrades at all to the last mile infrastructure so I still only have a ~10Mbps DSL option for wired internet at that house, which is a big step up from literally no wired option before, but still vastly inferior to Starlink. (The terrain makes terrestrial wireless a nonstarter in the area). I've since moved back to civilization, but I still own the house. As far as I know, there are no plans at all to improve the last mile infrastructure.
Separately, from SpaceX's own prospectus, Starlink is only a tiny fraction of the overall conglomerate that went public recently. It "only" needs to support double digit billions of valuation to pull its weight inside of the company.
I would describe it as entirely normal. My experience working in a research organization where the majority of my colleagues hold Phds is that education level has a strong inverse correlation with ability/willingness to care about such mundane chores as spelling, grammar and arithmetic.
As someone who works daily with export-control-adjacent hardware and software, my experience is that people tend to aggressively self-censor to a far higher standard than export control regulations actually require. The perceived headache of drawing the ire of whoever it is the enforces this stuff (which as I type this comment I'm just realizing I don't know who specifically is responsible for that) is so scary that people don't want to take any risk at all of being targeted.
If there was ever a game where the secret sauce doesn't have anything at all to do with the code, Eve has got to be it. They could probably release every single thing including all of the assets, complete buildable client and server code, etc and I doubt it would hurt the Eve at all.
In addition to the other comments, I just want to point out that we _do_ do a tremendous amount of CFD. The Pleiades supercomputer [0] sits in a building just down the street from the large wind tunnel at Ames Research Center, is generally ranked somewhere in the top 150 supercomputers in the world, and is largely used for CFD work to complement the wind tunnel work.
There are lots of these hyper-specific "reserved keywords" in the military. Another one from the US Army is "repeat", which is the command for an artillery battery to fire again with the same parameters as their previous barrage. Therefore on the radio we only ever used "say again" to ask someone to repeat their last transmission. Even if no one on the radio had ever or would ever be involved in artillery operations, I imagine it's easier to just train the entire force on a single uniform standard.
What is this, the 60s? Modern gas cars are so computer controlled that the concept of a "tune-up" effectively no longer exists, and they go 10,000 miles between oil changes so most people don't even average a single oil change every 6 months. EVs are even lower maintenance, but the difference isn't nearly as big as you're implying.
First of all, yes, I would object although I've debated whether I could get over it for myself. But it's a moot point because my partner would absolutely _never_ put up with it. And that's assuming that the human cleaner doesn't show up with a go-pro and an ill defined policy about where the video from the go-pro is going and what it will be used for
The company that just IPOed is already overwhelmingly "X AI" financially, regardless of the fact that it says "Space X" in the marketing. Whether SpaceX also buys Tesla is hardly even going to move the needle.
One major reason is probably that most of the liability portion is going to be covering medical bills in the US. I only did a quick skim of the Wikipedia summary but it looks like Finland, like (almost?) all of Europe wouldn't have that particular liability?
Another reason might be that insurance costs vary widely in the US. I recently had a reason to get an insurance quote in Utah, and it was literally one third the price that I currently pay in California.
All that said, you do seem to enjoy remarkably cheap insurance over there from my perspective. I hardly think those two factors are enough to cover such a large difference.