Imagine this for a whole neighborhood! Maybe it'd be more efficient for the transport to come at regular intervals though. And while we're at it, let's pick up other people along the way, you'll need a bigger vehicle though, perhaps bus-sized...

Half-jokes aside, if you don't own it, you'll end up paying more to the robotaxi company than you would have paid to own the car. This is all but guaranteed based on all SaaS services so far.

This only works in neighborhoods that are veritable city blocks, with buildings several stories tall standing close by. Not something like northern Houston, TX; it barely works for places like Palo Alto, CA. You cannot run buses on every lane, at a reasonable distance from every house.

The point of a car is takes you door to door. There's no expectation to walk three blocks from a stop; many US places are not intended for waking anyway. Consider heavy bags from grocery shopping, or similar.

Public transit works in proper cities, those that became cities before the advent of the car, and were not kept in the shape of large suburban sprawls by zoning. Most US cities only qualify in their downtowns.

Elsewhere, rented / hailed self-driving cars would be best. First of all, fewer of them would be needed.

Self-driving municipal busses would be fantastic.

Also, a real nightmare for the municipal trade unions. (Do you know why every NYC subway train needs to have not one but two operators, even though it could run automatically just fine?)

Why?

Because the Transport Workers Union fought tooth and nail for it. Laying off hundreds of operators would be a politically very dangerous move.

Huh. I wonder if that makes any sense. It doesn't seem to make sense to keep employing people if you no longer need them. It sucks to be layed off, but that's just how it works.

It also shows a lack of imagination. If you have to provide a union with a job bank, why not re-deploy employees to other roles? With one person per train, re-deploy people to run more trains therefore decreasing the interval between trains. Stations used to have medics but this was cut. How about re-train people to be those medics? The subway could use a signaling upgrade and positive train control. Installing platform screen doors to greatly reduce the incidence of people falling onto the tracks is going to need a lot of labor.

Why would you need buses?

Mass transit is a capacity multiplier. If 35 people are headed in the same direction compare that with the infrastructure needed to handle 35 cars. Road capacity, parking capacity, car dealerships, gas stations, repair shops, insurance, car loans.

Believe it or not, in some cities that have near grid-lock rush-hour traffic - there's between 50-100%+ as many people traveling by bus as by car.

If all of those people switch to cars, you end up with it taking an hour to travel 1 mile by car.

It's almost as if they have busses for a reason.

First, these cities should be fixed by removing the traffic magnets. It's far past the point where we used the old obsolete ideology of trying to supply as much traffic capacity as possible.

But anyway, your statement is actually not true anywhere in the US except NYC. Even in Chicago, removing ALL the local transit and switching to 6-seater minivans will eliminate all the traffic issues.

> First, these cities should be fixed by removing the traffic magnets.

If you remove the jobs and housing, traffic does get a lot better. But it's not much of a city without jobs and housing.

Indeed. And people live better lives, with better job accessibility and variety. Once you remove dense office cores.

Car traffic magnets like highways inside urban cores? Or people traffic magnets like office buildings, colleges, sports stadiums, performing arts venues, shopping malls?

Office buildings. Everything else is just noise.

Large stadium arenas are a special case, but they don't create sustained traffic, and their usage periods typically do not overlap with the regular rush hour.

6-seater self-driving municipal minivans would be fantastic, too. (I would still call that a "bus", but I don't care what we call it.)

> if you don't own it, you'll end up paying more to the robotaxi company than you would have paid to own the car

Maybe for you, I already don't own it and have not found that to be true. I pretty much order an uber whenever I don't feel like riding my bike or the bus, and that costs <$300 most months. Less than the average used car payment in the US before you even consider insurance, fuel, storage, maintenance, etc.

I also rent a car now and then for weekend trips, that also is a few hundred bucks at most.

I would be surprised if robotaxis were more expensive long term.

> Maybe it'd be more efficient for the transport to come at regular intervals though

Efficient for who, is the problem

Cars and personal computers have advantages over shared resources that often make them worth the cost. If you want your transport/compute in busy times you may find limitations. (ever got on the train and had to stand because there are no seats? Every had to wait for your compute job to start because they are all busy? Both of these have happened to me).

This exists in a way -- I'd wager every city has a commercial service that will shuttle you to, say, the airport. They're not cheap, however.

Yep. And it's indeed a good model for this mode of transportation. And they ARE cheap.

For example, in Seattle I can get a shared airport shuttle for $40 with the pick-up/drop-off at my front door. And this is a fully private ADA-compliant commercial service, with a healthy profit margin, not a rideshare that offloads vehicle costs onto the driver. And a self-driving van can be even cheaper than that, since it doesn't need a driver.

Meanwhile, transit also costs around $40 per trip and takes at least 1 hour more. And before you tell me: "no way, the transit ticket is only $2.5", the TRUE cost of a transit ride in Seattle is more than $20. It's just that we're subsidizing most of it.

So you can see why transit unions are worrying about self-driving. It'll kill transit completely.

you made too many false assumptions if you came up with those routes. Experts have run real numbers including looking at what happens in the real world. https://humantransit.org/category/microtransit - (as I write this you need to scroll to the second article to find the useful rebuttal of your idea)

Yeah, yeah: "Major US Public Transit Union Questions “Microtransit”" Read it. Go on. It's pure bullshit.

The _only_ issue with the old "microtransit" is the _driver_. Each van ends up needing on average MORE drivers than it moves passengers. It does solve the problem of throughput, though.

But once the driver is removed, this problem flips on its head. Each regular bus needs around 4 drivers for decent coverage. It's OK-ish only when the average bus load is at least 15-20 people. It's still much more expensive and polluting than cars, but not crazily so.

Scroll down to the other articles as I said in the first place.

self driving changes some things, but there are a lot of other points in the many article linked from there that don't change.

This article is just a bunch of propaganda. You can tell that by the picture with people in the shape of a bus next to the line of cars. Every time you see it, you can immediately blacklist the author and ignore whatever they are saying about cars.

Can you guess why?

Hint: think about the intervals between buses and how you should represent them to stay truthful. And that buses necessarily move slower than cars. And that passengers will waste some time due to non-optimal routes and transfers. And that passengers will waste some time because they need to walk to the station.

So back to my point, can you tell me EXACTLY what I should read in that article? Point out the paragraph, please.

> But why make them significantly cheaper when you can match the price and extract more profits?

Even better — charge 10% less and corner the market! As long as nobody charges 10% less than you…

> Cars are mostly idle and could be cheaper if shared. But why make them significantly cheaper when you can match the price and extract more profits?

Yeah, this would rely on robust competition.

Nah, I don't want to share my car with anyone. It's my own personal space where I can keep some of my stuff and set it up exactly the way I want.

That's how some people feel about airplanes. Presumably you're not one of them. For some people, the inconvenience of being responsible for a car would outweigh the benefit of setting up their stuff inside of one.

It's not even an inconvenience. I like my cars. Dealing with ride hailing services (autonomous or not) is certainly far more inconvenient than owning a car (unless maybe you're stuck living somewhere without convenient parking).

> Let's say, just for the sake of argument, that AI can eventually serve to level the playing field for everything. It outputs novels, paintings, screenplays - whatever you ask it for - of such high quality that they can't be discerned from the best human-created works.

This requires the machine to understand a whole bunch of things. You're talking about AGI, at that point there will be blood in the streets and screenplays will be the least of our problems.

I'm not sure you need AGI to clear that bar; I'm not sure you need more technology than currently exists beyond iterative improvements to things like how expensive it is to train a model.

But let's say it's free-ish to train a model, so you decide that that's how you're going to write the next Marvel movie. You train an LLM specifically on screeplay writing, teaching it to cross reference literary techniques with audience reaction, you teach it the sum total of Marvel canon, you teach it the sum total of the American cinema canon, you train it on all the social media reaction to either, and so on. You teach it to specifically engineer screenplays for Marvel movies that Marvel audiences will do maximum Marvel fanboy shit about. Do you genuinely believe such a dedicated model couldn't output a Marvel movie that everyone would love as much as Endgame?

Obviously, in the economy of 2026, it is cheaper for Disney to hire flesh-and-blood writers instead of doing this madness. But one day it won't be - and this is hardly even the tip of the iceberg. The ability to finely hone models quickly and on-demand (potentially on a per-prompt basis) would unlock another tier of accuracy and performance from LLMs, and for some/most artistic tasks, I think that gets you to "indistinguishable from mass market media."

Marvel movies is the worst example, it's a roller coaster ride, not a movie. I agree any braindead idiot or machine could write one. But they couldn't know to write The Pianist or how the subject could be approached or why it's time to write it.

AI can make slop yes, but it can't make the kind of art people don't get tired of. It's the difference between wisdom and knowledge.

Truly a "why say many words" title!

Of all the packaging systems to do this with, npm is certainly one of them.

It seems fine?

After all, Claude, codex and Gemini-cli are all distributed via npm.

I recently witnessed one such potential fuckup. The AI had written functioning code, except one of the business rules was misinterpreted. It would have broken in a few months time and caused a massive outage. I imagine many such time bombs are being deployed in many companies as we speak.

Yeah; I saw a 29,000 line pull request across seventy files recently. I think that realistically 29,000 lines of new code all at once is beyond what a human could understand within the timeframe typically allotted for a code review.

Prior to generative AI I was (correctly) criticized once for making a 2,000 line PR, and I was told to break it up, which I did, but I think thousand-line PRs are going to be the new normal soon enough.

That’s the fault of the human who used the LLM to write the code and didn’t test it properly.

Exhaustive testing is hard, to be fair, especially if you don’t actually understand the code you’re writing. Tools like TLA+ and static analyzers exist precisely for this reason.

An example I use to talk about hidden edge cases:

Imagine we have this (pseudo)code

  fn doSomething(num : int) {
    if num % 2 == 0 {
      return  Math.sqrt(num)
    } else {
       return Math.pow(num, 2)
    }

  }

Someone might see this function, and unit test it based on the if statement like:

    assert(doSomething(4) == 2)
    assert(doSomething(3) == 9)

These tests pass, it’s merged.

Except there’s a bug in this; what if you pass in a negative even number?

Depending on the language, you will either get an exception or maybe a complex answer (which not usually something you want). The solution in this particular case would be to add a conditional, or more simply just make the type an unsigned integer.

Obviously this is just a dumb example, and most people here could pick this up pretty quick, but my point is that sometimes bugs can hide even when you do (what feels like) thorough testing.

So UBI + universal healthcare? Sign me up!

As it was foretold since the beginning, IA use is breaking security wantonly.

You might say the problem CloudFlare is causing is lesser than the ones it's solving, but you can't say they're not causing a new, separate problem.

That they're trying counts for brownie points, it's not an excuse to be satisfied with something that still bothers a lot of people. Do better, CloudFlare.

Do better, how?

If you have any ideas on how to protect against bad actors in a way that is just as effective but easier for users, please share it.

Because as far as I can tell, this isn't a question of effort. It's a question of fundamental technological limitations.

This is pretty interesting to me, as I do use Grafana in my current role. But none of their other products, and not their helm chart (we're on the Bitnami chart if that's a thing).

So far it's pretty good. We're at least one major version behind, but hey everything still works.

I cannot imagine other products support as many data sources (though I'm starting to think they all suck, I just dump what I can in InfluxDB).

I agree. I think OP has made the mistake of using more than just Grafana for dashboards and perhaps user queries.

I operate a fairly large custom VictoriaMetrics-based Observability platform and have learned early on to only use Grafana as opposed to other Grafana products. Part of the stack used to use Mimir's frontend as caching layer but even that died with Mimir v3.0, now that it can't talk to generic Prometheus APIs anymore (vanilla Prom, VictoriaMetrics, promxy etc.). I went back to Cortex for caching.

Such a custom stack is obviously not for everyone and takes much more time, knowledge and effort to deploy than some helm chart but overall I'd say it did save me some headache. At least when compared to the Google-like deprecation culture Grafana seems to have.

The room is not doing anything of the sort.