So… @lazide replied with a hahaha. It got flagged to death, understandably.
If it were anybody but @lazide I’d have flagged it too, but… I can’t help but grimace and nod.
To hazard a loosely-informed guess at their subtext… and politely ignoring your opening salvo of calorie-free populism… I’m not speaking from a place of certainty, but my impression is that there seem to be elements of “not even wrong” here. Some contra- lines of argument:
* code, especially in the huge fundamental aspects, tends to flow from the NEC downward, rather than from municipal experiments in playing with literal fire. Not insurmountable, I guess, but it seems like a pretty big chicken-and-egg kind of obstacle to your proposed course of action;
* individual household demand is several orders of magnitude removed from the considerations informing grid design and operations;
* the last thing a grid needs is a meaningful amount of random feed-in that it can’t centrally (and near-instantly) manage;
* the costs of home-scale labor/permits/whatever are only the biggest ones from the perspective of toting up individual homeowners considering uneconomical projects, not from utility-scale (much less whole-polity-scale) scratchwork… would an army of volunteer homeowners’ labor really still outperform specialist labor installing hundreds-of-homes’ worth of capacity per day at an energy facility somewhere?
* the thing holding back the “explosion of products,” consumerist-salvation overtones aside, seems more likely to be {physics vs the existing built environment} than permits—that, plus the lack of a mass market;
* one barrier to “plug and play” is that said plugs are safe only because of an interlocking set of assumptions upstream of said plugs—which definitely do not contemplate feeding back in whole-home current from a random convenience outlet on your front porch;
* Rate tiers are an economic device to influence consumption patterns under the status quo. As the Californians are learning, there’s a hard floor on the price of “having a grid” vs “not having a grid”; as consumption patterns shift, rate schemes can and must shift too;
* building codes care about safety, grids and ISOs and energy providers care about capacity and stability. Players on grids can realize enormous economies of scale with respect to pretty much any generation or storage technology. The bottleneck you describe is…
I dunno, man… the only bottleneck to energy too cheap to meter is fusion, obviously all we have to do is just go crack fusion and we’ll have a fusion pot on every stove!
I love your sense of empowerment; I’d encourage you to bring it to bear at a scale where it matters— much bigger than a home, much smaller than a whole polity… I think there are benefits to decentralization and to independence but I don’t think those are the benefits you’re emphasizing here.
> * code, especially in the huge fundamental aspects, tends to flow from the NEC downward, rather than from municipal experiments in playing with literal fire. Not insurmountable, I guess, but it seems like a pretty big chicken-and-egg kind of obstacle to your proposed course of action;
UL 3700 exists. Local efforts are underway to let anyone use plug-in sola without contractors, permits, utility approval: https://pluginsolarusa.com/
> individual household demand is several orders of magnitude removed from the considerations informing grid design and operations;
The grid connects millions of individual homes. I'm not sure what you are trying to say here?
> the last thing a grid needs is a meaningful amount of random feed-in that it can’t centrally (and near-instantly) manage;
Plug-in solar or batteries do not feed in to the grid. They are for consumption within the home.
> the costs of home-scale labor/permits/whatever are only the biggest ones from the perspective of toting up individual homeowners considering uneconomical projects, not from utility-scale (much less whole-polity-scale) scratchwork… would an army of volunteer homeowners’ labor really still outperform specialist labor installing hundreds-of-homes’ worth of capacity per day at an energy facility somewhere?
There is no work involved other than just plugging it in. There is no volunteer home owners labor. Even if there is, it is their own home. When you turn the lights on, do you add that to your spreadsheet of volunteer home owners labor? Or do you consider plugging a cable into an outlet a big task that should require contractors, project managers, managers, directors, forecasts, estimates, daily standups? If thats your view, thats understandable, most enterprises work that way.
> the thing holding back the “explosion of products,” consumerist-salvation overtones aside, seems more likely to be {physics vs the existing built environment} than permits—that, plus the lack of a mass market;
plug-in battery uses existing built environment. Thats the whole point.
> one barrier to “plug and play” is that said plugs are safe only because of an interlocking set of assumptions upstream of said plugs—which definitely do not contemplate feeding back in whole-home current from a random convenience outlet on your front porch;
> Rate tiers are an economic device to influence consumption patterns under the status quo. As the Californians are learning, there’s a hard floor on the price of “having a grid” vs “not having a grid”; as consumption patterns shift, rate schemes can and must shift too;
Rate tiers are a politically convenient way to hide cost-recovery problems, cross-subsidize favored groups, and extract revenue while pretending the underlying cost structure of the grid is volume-driven when large parts of it are not. California homeowners are buying record numbers of home batteries to pair with solar panels. If the rates are entirely unreasonable, homeowners will go offgrid. California homeowners are smart enough to figure out how to freeze property taxes close to zero, I'm sure they will figure out the right laws when electricity rates become unreasonable.
> building codes care about safety, grids and ISOs and energy providers care about capacity and stability. Players on grids can realize enormous economies of scale with respect to pretty much any generation or storage technology. The bottleneck you describe is…
Yeah, just leave it to the utilities and regulators, they have nothing but consumers best interest at heart.
> I dunno, man… the only bottleneck to energy too cheap to meter is fusion, obviously all we have to do is just go crack fusion and we’ll have a fusion pot on every stove!
plug-in batteries are impossible and require home owners volunteer labor, but fusion, it can manifest from thin air!
Matteo Wong writing on new “Nano Banana” features in Google Earth. He is not a fan, but he did create some rather impressive fake satellite imagery of cataclysm.
Linkbait title as published: “Google Just Ruined One of Its Most Important Tools”
Corresponding Google press release—I’m not sure whether or not we’d consider it a more appropriate/neutral source:
You just know this is going to be 2-3 times the price of an economy seat, which means you're going to buy the economy seat anyway. In flights, price beats everything else.
I think this is an ancillary specifically for economy class passengers—you pay a certain amount to reserve a 4-hour slot on a bunk or something. I have no idea, but my guess is economy + nap slot would still come in well under the cost of a business seat. Plus they can sell those same 6 berths what, 4 or 5 times over in the same 22-hour flight?
If I recall correctly they’re not certified for use during taxi takeoff and landing, I’d imagine due to orderly-evacuation kinds of requirements.
Good enough for what? Certainly good enough to direct people to the correct links for their platform.
And, to me at least, to signal ahead of time that it’s a Claude-flavored game. Personally I’m not much for those (I don’t really snack on games) but I know people who enjoy burning through them. And “Claude-game” does feel like a steadily-more-coherent genre, with its own aesthetic language and high- and low-quality examples.
But of course it all comes back to the same old open-source questions. Is it marketing if you’re not selling anything? Does somebody have to be good at—or care about—writing or marketing in order to share their GPL game with the public for free?
To skip the landing page, the source repo is here:
I really think the way Claude writes now is beyond tolerable. It makes me actually angry. I wish i knew how to articulate it, and maybe its just reaching a tipping point because its been so many months of daily usage but its really awful and I think a lot of people are starting to feel the same.
We have this incredible piece of technology that has consumed much more literature than imaginable but it insists on this corporate-therapist writing style which is obsessed with accountability.
Some of these municipal water plants in the US are independently operated by municipalities of awfully few people and even fewer resources to spare, often serving relatively vast areas…
It’s probably not how you or I would set things up—especially after many years of warnings and slick best practices guides-but I can sympathize with “if it’s not broken…”-style maintenance, especially at the local level.
These things have lifespans measured in decades, and budgetary cycles to match… and for all of CISA’s good work (on limited budgets, and with power that’s more persuasive than fearsome) [0], it seems hard to get all 148,000 [1] system operators to afford to care, much less to afford to fix things—much less to check their work.
I think knowing what you want (and being able to articulate it with adequate specificity, and detecting when processes drift away from it) is a real skill/field of expertise. I just don't think it's specific to interfacing with "AI."
I almost entirely agree with this. However, after coding quite a few large projects with LLM assistance, I can say with certainty that there is some form of "art" at play with regards to structuring requests for LLMs.
However, this "art" is not so much about how to present a given request to the LLM, but rather guestimating what the scope of the next chunk of work should be to balance getting as much out of the model as possible while avoiding the machine going off the rails.
Obviously, this is a moving target and different models perform differently for various chunk/scope of work. I look at my successful sessions with LLMs and I'm not sure I'd be able to articulate a clear set of rules to apply here. You just... gradually build a intuition for how much you can throw at the LLM at once.
That being said, I'm pretty convinced at this point that this is a property of the coding assistants as they exist today, and what "working well with LLM assistance" means will keep on changing.
I've found that one of the most useful things I can do when using LLMs is paying attention to the tool calls so that I can tell when it's giving an explanation that's obviously not based on any evidence. If it explains why a bug happens without reading the relevant files, it's obviously just a guess. If I ask about what things exist that fit certain criteria, and it doesn't search online, at best it's relying on outdated knowledge from training and at worst it's literally inventing things that don't actually exist.
In a lot of scenarios for software engineering, the cost is just wasted time without anything useful as a result, and that's already bad enough. I can't even imagine working as a lawyer and not even taking the time to validate so I don't end up reprimanded by a judge in front of my clients, but there have been so many news stories like this that obviously this is not anywhere close to a universal view...
IME "creating a good spec" (and good acceptance criteria) has always been one of the most valuable skills in... well, everything.
The machine is really good at working the spec on its own now, which is amazing, science fiction shit. But you've still got a garbage in, garbage out problem at the end of the day, which is pretty much the only hope we who work in software have of remaining somehow employed.
I'm curious if there has been any study on the output quality of LLMs after actively antagonizing them. It's easy to forget that at the most basic level these are just auto-complete machines, and humans tend to not respond so well after antagonized. It stands to reason that the same should follow for an auto-complete system trained on human interaction.
Intuitively I'd say lashing out on an LLM overconstraints it and so worsens the output, therefore the skill to stay calm and neutral is useful for guiding a machine as well. Would be useful to have data on that.
I’m familiar with cities that conduct time-bounded surges of policing in reaction to specific outbreaks of criminality.
When inter-faction violence in a neighborhood escalates to the level of a murder, for example, they surge for a couple weeks. Massively increased visible patrol presence—cops on every corner—portable light towers attached to generators, to provide near-daylight-level illumination on the sidewalks and streets... a conspicuous level of “eyes on the street” that discourages retaliation and responds to residents’ heightened fear in the immediate aftermath. While at the same time
being obviously temporary—not permanent measures that signal the neighborhood as structurally crime-ridden.
It’s a “fly the flag at half mast for a bit” kind of signal, not “this neighborhood is full of criminals and deserves the corresponding built environment.”
I wonder if something in that spirit might be a reasonable long-term equilibrium for things like Flock. Garbage bags over the camera poles for ordinary times, almost as a point of civic pride that the place we’ve built and maintained is safe enough not to need to use them. Pull the garbage bags off if there’s a problem—spate of auto thefts, high-profile crime, fugitive on the loose—but we demand an explanation when that happens and a concrete end point where they go back on.
Not all crimes are equal. I'd like very much for my stuff to not be stolen. I'd like it even better if I didn't get murdered. Optimally, no murders would happen.
reply