Its funny how many variations of meaning people assign to agent related terms. Conflating agent with cli and as opposite spectrum of ide is a new one i did not encounter before. I run agents with vscode-server also in a vm and would not give up the ability to have a proper gui anytime i feel like and also being able to switch seamless between more autonomous operation and more interactive seems useful at any level.
The same could be said for color, just to then go into detail that using color codes on too many columns and with unclear associations are the problem. This is just a weird way to frame an article. Colors and icons are indeed essential to quickly parse and use big amounts of tabular data. All the criticisms the author brings forward are specific problems of some implementations or just poor choices. All this is solved by putting a few key icons in their own column next to the text field they belong to, so users can hide and show them as needed depending on the task. I have eg. a basic indicator that is a red exclamation mark or yellow question mark or green checkmark and another one with a colored icon representation of a category field. Its 100 times easier to get a quick impression and find something than without those. if someone has a problem its 2 clicks to hide the columns, thats the universality of a table ui! Don’t let absolutist click bait articles fool you.
As nice as zulips aspirations may be, every time i have to use it for a community i effectively stop interacting with them after a short while just because everything is janky, ugly and feels like a drag to interact with, just tried opening it on my phone to see if it improved but the header ui is just plain broken.
Are you using iOS? Safari 26 has several changes that break the mobile web app layout, and it's proven quite difficult to fix. I'd suggest using the actual mobile app on iOS if you've upgraded to Safari 26.
(My understanding is we are far from the only web app broken by Safari 26, and we're working on it).
I don’t use native apps as general principle, while i happened to be on ios safari when i tried and i am the first to criticize that browser, the bugs i am seeing do not happen in other apps and should be easily fixable in a proper build webapp. Also keep in mind that this was just a coincidence, every other browser and platform i had to use zulip in had a bad experience.
For what it's worth, essentially every main view surface was visually redesigned over the course of the last 2 years. So while I can't promise you'll like the new design, it certainly isn't the same as it was 2 years ago.
One of the other nice features of the new design implementation is there are handy settings for font size and line spacing. It turns out that different people have very different desires for how dense content is in chat apps, and empirically there's a significant portion of users with just about every combination.
I've spent a bit of time last year trying to convey my product instincts to the Zulip team and mostly stopped because I felt like they didn't care enough / weren't moving very fast. The basic problem is that the mobile app is, like it or not, the way most people will use the product, and it needs to be designed by an opinionated person who actually will say no to things.
In my view, the home page should be just like a proper messaging app: show every recent thread ("topic" in Zulip nomenclature) that I'm involved in, across all my channels, with unread ones indicated using a 'dot'. Or, if you really want to be like Slack, just copy Slack more directly. In either case, the other views (Inbox, Combined Feed, DMs, etc) should be under menus, not primary actions.
The other thing is that it's often hard to figure out how to reply to a topic. In the Combined Feed, which is my preferred view for consuming updates, the UX for replying sucks -- first you have to figure out to tap the headers; and even then, you can accidentally tap into a channel instead of a topic. It's extremely non obvious when you've done this and constantly causes people to reply in the wrong topic.
I vibecoded some improved Inbox UX using Claude Code and I think it would be a big step up, but it's hard to know what the steps would be to get it shipped, since I don't have time to spin up properly on the codebase and I doubt my changes are acceptable as-is. If Zulip team wants them I'd happily share though.
there are just so many issues, where do i start? its just apparent no designer or usability person ever used it or was involved in anything for this project. there is a weird search button with uncentered icon, scrolling makes some tooltip flicker and partially scroll on top of the header, the content of the page reappears on the top of the header when scrolling past it. everything just feels like one giant glitch. and when you scroll, there is a focus outline around whatever item you happened to drag the scroll area with. This is what i encountered in 5 seconds testing just opening and scrolling up and down.
I have looked at the rust Zulip forums, which are bulky. But with moderation and rules and having people on the autistic spectrum [citation needed], it perhaps is usable for large organizations. Just kidding.
We are using Zulip for 300+ members in a makerspace, and at 40 members, we were not happy. Scaling to 300 never broke not being happy, since we all hate the UI ever since.
I cannot re-open Zulip threads, which are also issues with an atomic "solved/unresolved" state, unless I have elevated access. It is not a true forum like PHP forums, where we ask people to name threads, and you might just skip reading more than the title, or locate interesting threads by activity and find stickies about important announcements in a pull, not push, way of doing things.
It instead is a chat where a thousand group chats are open, and no once wants to read any of them.
If they wanted to re-invent forums, they should have cloned the "discourse" web app/forum. Still looks like shit on every platform, mobile or desktop, but at least does not break down on mobile.
It instead is a chat where a thousand group chats are open, and no once wants to read any of them.
I really wanted to like Zulip and use it as a personal chat service for a small group and it was exactly that feature that made it basically unsuitable. Forcing everything into titled threads did not make any sense for lots of user to user interactions that are ad-hoc in nature.
I didn't think it was terrible software by any stretch of the imagination - just not really suitable for informal communication.
> It instead is a chat where a thousand group chats are open, and no once wants to read any of them.
Do you mean people are happy to post on a thousand different threads, but no one reads posts from anyone else?
Why do you even have so many different active threads? Why not just let "resolved" threads be, and funnel conversations into fewer threads? (esp if you want ephemerality i.e. conversations to expire with time)
> 300+ members in a makerspace
If you have 300+ people discussing a wide variety of things, how do you ever expect to maintain your sanity with only channels and without threads? Won't every channel be quickly flooded and really hard to resurface anything useful from past discussions?
> I cannot [...] unless I have elevated access
Is your complaint that your Zulip space is not moderated well and that it would be helpful to ad-hoc decentralize some of the maintenance work across more participants?
While i appreciate anthropic making a proof of concept like they did with claude code cli on which they can then do RL to optimise the patterns that work, I expect this to be as unusable as the cli itself. Its a big difference if a model provider internalises something like thinking mode which mainly depends on context and text or if they try to grab a part of the agent loop which has to run on the side of the systems we build and use.
We cannot allow model providers to own the browsers, CLIs, memory, IDEs, extensions and other tooling. Its not just a matter of power but also they just suck at it as i experience every time i have to use claude code instead of amp.
I truly hope we get the pattern of innovation that looks like:
- some dude vibecodes a really cool idea
- model providers build into their reference implementations
- model providers optimize models to work optimally
- startup and/or open source projects step in and build something that is actually usable and opens a new market segment
We saw this play out beautifully with amp, kilo, roo, cline, continue
Another aspect is that we do not want interfaces just made for agents to work in teams, we want software made for humans and agents, that are true platforms for these agent teams to collaborate in.
No product had such a fast transition from novelty to "omg i never want to interact with a human again". I feel about 100% less stressed and happier using a waymo or riding motorbike or bicycle next to a waymo than with human drivers. I hope this next phase will bring availability and prices down. We need this in europe.
Im fortunate to live in an area dense with traditional taxis and Ubers, no Waymo yet.
I rarely take taxis, the exception is when I have to haul my gear to the studio for a jam session. I always take a taxi, because it’s cheaper and faster than using an app to call an uber.
On 80% of the trips, I end up having a nice chat with the driver and learn something new about humanity or myself.
I really enjoy these interactions, but I feel for the drivers, it’s a very tough job where most taxi drivers have to scramble to find places to urinate or do so in an empty bottle between their legs. There is not much dignity in the job. I feel a negligible segment enjoy it as a reliable career.
I wonder what will happen to the drivers if a large representation of the 1 million+ daily trips are displaced by automation?
I used to feel this way. In the early days of "ride sharing," I preferred Lyft and would sit up front so I could have a conversation with the driver, which they encouraged. It was really fun for a while, and I enjoyed meeting people from different walks of life. Over time, though, transportation became much more functional for me, and now when I take non-autonomous rides, it's more irksome than enjoyable when drivers strike up conversations.
Why the change? I think a big part of your experience is the fact that you "rarely take taxis." Once you're doing it daily or near-daily, the amount of smalltalk becomes more tiresome. Also, with kids and a busy life, I'm usually either looking to get things done or enjoy a rare moment to myself as I'm moving from place-to-place. I agree with OP that Waymo is a huge step up on those dimensions. There's no other human in the same space to feel awkward around.
The fact that they drive more safely and smoothly is a huge improvement, as well. Ironically, I thought this was going to be something I would hate about Waymo. "You mean it drives the speed limit and follows all the traffic laws? It will take forever to get anywhere." It took approximately one ride for my perspective to completely flip. It's so much nicer to not feel the stress of a driver who is driving aggressively or jerking to a stop/start at every intersection. It's not like you can tell them to just ease up a bit, either. When we ride with our kids, we feel massively safer in Waymos.
Yes, it will be disruptive, and I don't particularly love the dominance that big tech has in all of our lives, but I do think Waymo is a marvel, and I hugely appreciate it as an option. As soon as they can take kids alone to all their various activities, it will be yet another massive unlock for parents.
Driving to work is the most common way of commuting everywhere in the US except NYC. So in that sense, no, taking a taxi to work daily is not normal, just as walking, biking, and taking public transit aren’t normal.
When I worked in San Francisco I took Caltrain to the city, but I took Waymo from the train station to the office. San Francisco, like almost all US cities, has poor local transit coverage. In my case there was a bus that took a similar route, but it only ran every 20 minutes even during commute hours and wasn’t coordinated with the train, so if everything was running on time it would have been a 17 minute wait (plus an extra 5 minutes walking). I was busy and well paid enough that spending the extra $10 to save ~20 minutes of travel (and the uncertainty of when the bus would arrive, and how strongly it would smell like piss) was well worth it.
According to [1] the median Bay Area big tech worker earns $272k/year - or $130/hour.
According to [2] Uber drivers make $15 to $25 an hour, before expenses like fuel.
So while it's not normal it's certainly plausible that some people take taxis on a daily basis.
More broadly, as levels of wealth inequality rise in a given society, more people end up working in the personal service sector doing things like cleaning, food delivery, taxi driving etc.
As a former Lyft driver in SF I felt kinda weird when saw the bit about urination. Like, that's just not a problem. As a driver you just plan ahead as in any other job out there where you're not allowed to disengage at a whim. Pilots and surgeons don't pee in bottles, why would drivers? It's kinda funny when people try to empathize but come up with these creative scenarios of what's challenging. The parts that are bad are same as any other thing done for a living: money and dealing with other people. The job was shit when people were shit and/or when the money was shit.
I enjoyed it as a job, not a career. But that was in 2015.
Pilots and surgeons surely have easily accessible bathrooms as a part of their workplace, no? They’re also compensated significantly more and (IMHO) given a lot more dignity
In my city public bathrooms are extremely rare and it’s not trivial to find one. I’m sure taxi drivers are a bit more in tune with where they are out of necessity but even then it’s no guarantee they can find convenient parking/be in the right place/etc.
No. Not for some surgeons at least. Once you start cutting you may have to stay until the job is done so get good at holding it. In the The Way I Heard It with Mike Rowe podcast episode Dr. Rahul Seth talks about doing 12 hour surgeries. No breaks, no bathroom, constantly on his feet working.
Commercial pilots flying airliners generally have it a bit easier. As for military pilots flying tactical aircraft, well this song might give you an idea of what they face.
Yep. This is a really weird thread. The no bathroom piss in bottle thing is not a thing I encountered in my IRL XP. Never felt this imaginary problem, never affected my dignity.
Funny enough, I did later work on surgical training tech and went into O.R.'s. And yeah, everyone in the room stays until the work is done, no easy pee pee breaks. Back to back procedures. But then also nobody ever complained about that there either. It's a fun job.
Idk. I'd reiterate a point I was getting at: what makes any job less dignified is dealing with shit people and/or shit pay. Fwiw Bathrooms you can plan for same as you plan for getting hungry by packing a lunch.
> I really enjoy these interactions, but I feel for the drivers, it’s a very tough job where most taxi drivers have to scramble to find places to urinate or do so in an empty bottle between their legs.
Public toilets, their condition and their non-existence are an often-overlooked issue! It's not just highly problematic for taxi drivers, but also for parcel and postal delivery people... and it's not just relevant for workers either, it's also (IMHO) a violation of anti-discrimination laws.
Imagine you're old and don't have much bladder control or volume, or you're a woman who recently has given birth, or you got one of the variety of bowel related diseases, or you've got a child who is still dependent on diapers. Your range of free unimpeded movement is basically limited to where you have easy and fast access to a toilet or at the very least a place to take care of yourself/a child.
>I wonder what will happen to the drivers if a large representation of the 1 million+ daily trips are displaced by automation?
If it happens gradually enough, they will just find other jobs. After the transition, society will be producing more with the same labor force, and thus the aggregate utility will increase.
In the past when automation displaced many jobs, we did things like raise the age kids could stay in school. There used to be huge numbers of e.g. 14 year olds who previously would be expected to go to work that would now have the opportunity to stay in school. Kind of like a mini UBI as in the transition period they would usually get given food, healthcare etc at least minimally. What’s the equivalent now?
try talking to young attractive women on their experiences and you'll maybe appreciate this somewhat forced interaction less. my partner has been literally kidnapped multiple times (refused to take her to her destination and refused to let her out for over an hour), had drivers refuse to unlock doors until she gave them her number at least once every two months, and constantly has drivers take detours and longer routes to force conversation for longer.
the sooner we can stop subjecting people to having to interact with strangers in a semi-private setting just for basic needs like getting around, the better off vulnerable people will be
I think there will still be delivery services where you need someone to go into the restaurant and then up to the customers door. That’s going to stick around unless we get to a point where the restaurant is responsible to load up the Waymo and the customer is responsible for getting it out which probably won’t happen anytime soon. The whole delivery market was also mostly created overnight from something that didn’t exist before.
In Miami, there are several competing companies like Coco Robotics which employ human "pilots" to monitor a small fleet of robot delivery boxes where the restaurant deposits the food in the box and the box unlocks with integration into the app.
Just figured you'd want to know anytime soon was at least a year ago.
I’m aware of those but those only go 2-3 miles so they don’t work for the majority of suburban and rural Americans. Also they don’t have the convenience of delivery to your door unless they can start using elevators.
these things are all over the city i live in, too. absolute menace and an abuse of the commons. i've had them literally run into me more than once and i've started physically moving them out of my way when they stop in the middle of the sidewalk.
That's a pretty dismissive attitude for ~100 million professional drivers worldwide, making a living doing actual useful work on a forum where the vast majority of users do not do any useful work.
There is also a demographic cliff most of the world is currently going off, declining birth rates and labor shortages. Would you rather have a human nurse in your very old age retirement, or a human driver. Because we don’t have enough young people now for both.
So let's poach these people from the third world and...what about the third world? People can't just be made in factories like robots and self driving cars can. It seems inevitable that either we will have really sucky retirements (please die early grandpa, we can't take care of you!) OR (hopefully) automation will come to the rescue despite luddite protests.
Plenty of people from the third world are interested in moving, trying something new. We should all be free to try new things, but of course you he world isn't set up that way. Seems like we could match up dual needs. The western developed world is in the midst of a racist and fascist period, so not the best time to try this. We have competing changes, shortage of workers in many job areas in the West like the trades in the US, also shortage of jobs for young people in the west.
I'm all for immigration, but the world isn't producing enough people to make that a very viable long term solution. Eventually we have to reduce our demand for labor, especially when our civilization is lopsided for awhile with older people and not enough young people (a problem that will fix itself eventually as the old people die off, I guess).
I'm OK with robots driving cars like I'm ok with not needing an elevator operator anymore to use an elevator.
Artificially protecting jobs by holding back technology is terrible form. At best it’s short term before the economics become an order of magnitude cheap and at worst it’s hamstringing your economy so you’re left behind.
Be that as it may, I would argue there's a straight line from "it's okay to destroy this fairly-low-skill-career for the good of the economy" to the overall situation the US finds itself in today
I figure that’s the way of the world. We’ve gone from a majority low skill economy to a much more complex one over the decades. It will probably continue.
Huh? how can one possibly generalize whatever experience they have not only to one country but to “other countries”, i.e. to the world. I’ve taken taxi in many countries, in all continents, and my experience have been that the drivers are generally helpful. There are scams and bad experience, but that’s minority. That applies to any country, the US included
I’m surprised they don’t have opt-in LLM-based “chatty mode” where you can talk to the AI personality of your choice while riding. Obviously shouldn’t be the same AI that’s deciding whether to run over the child or crash into the oncoming train.
why would anyone need this when then can pull out their phone and use their LLM of choice, if they wanted. I expect some large percentage of social users will just facetime chat with their friends during the ride
For me, this is the major selling point to own a car. I may drive a few times a week, and taxis might be much cheaper, but no way I'm going to deal with human taxi drivers if I have a choice.
This seems weird to me, maybe it's a generational thing. Is it really that bad to share a car with someone? You don't have to talk to them the whole time.
As a woman, while 95% of the ridesharing trips I take are perfectly pleasant and sometimes great with conversation the 5% of rides where you are trapped in a car with a creep asking you extremely off putting questions sours the entire concept of ride sharing for me.
Same. Ever been a vulnerable woman stuck in a car with a man who starts ranting that "nobody wants to date men who aren't rich anymore" and it turns out the driver is angry because the women that are trapped as riders won't go out with him?
Yea, I can't imagine being a woman and having to deal with some of these drivers.
This doesn't compare, but as a man I get really put off by the amount of invasive questions (where I work, where my family is from, etc) when I'm just trying to get from point A to point B.
I'm a mid-millenial FWIW, so I very much remember a world of only having old school taxis.
the situations you've described and the fact that our answer as a society is seemingly to throw up our hands at our inability to solve these situations other than by increasing the number of cars on the road in a way that funnels even more wealth to a tiny group of unfathomably wealthy sociopaths who also use ourour personal information to impact our spending habits... very depressing. i really hate it here.
I like humor, however, for the peanut gallery who might not get sarcasm:
Every credible scientific study of women and guns in the last two decades strongly indicates that a firearm in a woman’s home is far more likely to be used against her or her family than to defend against an outside attacker.
More women carrying guns makes them more likely to get shot, and, mostly, not by strangers.
> Is it really that bad to share a car with someone?
Sometimes it is, and you never really know when.
Some of my most unpleasant experiences involved a couple of reckless drivers, even more nutters who insisted on talking about their politics or pet peeves, I fear one of them may have gone beyond mere eccentricity and probably required some medical intervention, but couldn't figure out how to report that without possibly resulting in the driver being punished by the app.
Hah, I had a 2am conversation with a woman from Argentina about Javier Milei which is one of my Uber riding highlights.
But then another time a guy warned me not to open his glove box because his Glock was in there and he sounded deranged and it’s the one time I’ve literally gotten out of the car and cancelled my Uber.
One female Uber driver told me about how she had to go to court because a drunk man threatened to stab her with a knife (that he was brandishing), then he passed out and the police had to haul him out of her car. The .1% ruin it for everyone else.
Personally, I find it odd to have interactions with anyone just based of transactionality. I want to interact with people because I have relationships with them. I've always found it hard to figure out exactly how nice to be with someone you don't know. I don't think this is a maladjustment on my part, I think you probably shouldn't be overly nice to people before you establish trust with them... and that takes time.
The human driver could be nodding off because they didn’t bother sleeping last night, or maybe they just had some food with lots of garlic, or…ya, this has all happened to me before. I’ll take the Waymo over uber.
The Uber Driver who told me all about his Glock in the glove box was pretty off putting.
Also the Jeep that picked me up in August with broken air conditioning, although that was an annoyance vs “what is happening right now am I going to die”.
I’m fine to share a car. I’m less keen on dying in one.
Riding in a car is easily the most dangerous thing I do in my daily life and my subjective impression of how well uber/lyft/taxi drivers drive is not great.
I always (as soon as I could) owned a car, first on independence, but soon that became on price. A car costs between $350 and $500 per month, plus about 2 gas tanks, let's say $600. That's only 10-15 short taxi rides and two long taxi rides at best.
And now I have a family, there's 5 of us. A car is easily less than half the price of public transport for what I need to do (because you pay per person).
I hate traffic, and I don't really like driving, but since a car is easily 30 minutes faster than public transport to drive in to work, sadly 30 minutes of traffic in the morning is still faster than public transport, no matter how annoying it is. Oh and no waiting in the rain/cold is a nice bonus.
A car does not cost $350 and $500 per month. If it does, it’s a status symbol, not merely a tool to get from A to B, and therefore it is unfair to compare it against taxi rides.
There’s nothing wrong with wanting a nicer more luxurious car for yourself. But it’s just ingenious to compare that against taxis with beaten-up and spartan but reliable cars.
A status symbol will easily run you $1000/mo. I currently pay $350/mo (including cost of capital), and I don't know how I would pay less for a car that's not actively falling apart. Chevy Spark, manual transmission, $7k KBB value, averaging 500 miles per month.
There's no shot that number isn't being driven up by people purchasing more car than they need. You can get a used car for $10,000 or less, there's no reason one needs to pay $500/mo.
You sure about that? A $7k car expensed over 18 months plus insurance and road tax is ballpark $450 /mo. That omits maintenance and fuel but conversely also underestimates how long the vehicle will be kept on average. Depending on where you live parking will range from free to potentially more than $100 /mo.
If you manage to stretch $10k cars out to 5 years on average with zero maintenance it's less than $200 /mo but ... no maintenance in 5 years?
I think $300 /mo plus fuel and parking is probably a reasonable estimate for frugal behavior.
6 years ago, I was driving a Subaru BRZ which averaged 32 mpg. My commute was ~30 miles each way, add in a couple miles for weekly errands, and let's just say I was using 10 gallons/week. If gas was $3, that's $30/week, $120/month. Plus $150/month for insurance, it's $270/month.
Still way under your 350-500/month figure, but that's also assuming the car is paid off.
> If it does, it’s a status symbol, not merely a tool to get from A to B, and therefore it is unfair to compare it against taxi rides.
$350-500/month is cheaper than taxi rides. Even with a more reasonable 5-10 mile commute, I'd be spending probably $50/DAY if I took taxis.
You're not including an amortized cost of maintenance, registration fees, etc. Adding in tires alone ($720 for a set, 50k/1200 ~=41 months, ~$17.50/mo) brings it to almost $290/mo. Oil change every 6 months or so, add another $10/mo or so. Now we're at $300/mo and hoping nothing in the car breaks and needs repairs on a car that's already paid off, and we still haven't paid our taxes and registration fees.
Now figure in the fact you've got several thousand dollars in a car instead of even something like a high-yield savings account. At even 4% APY, if you had just $8k tied up in that car that's another ~$27/mo of income you're missing out on.
I'm not making the argument riding a taxi for every trip is cheaper than this. Just pointing out there's a lot of things people don't think about when they think of the cost of car ownership.
Insurance alone can be 100€ a month (and more so for younger drivers). At a very modest 5 liters / 100km and a one way route of 20 km you're at 800km a month / 40 liters of gas => 1.80€ a liter => 72€ in fuel. Your average car then has 20 ct/km for maintenance costs (inspections, spare parts, oil changes, tires, workshop time), so another 160€ a month - and more if it is a run-down junker car.
That are just the fixed running costs you have with pretty much every car, around 330€ a month. We haven't talked about depreciation yet at all. Even if you say you buy a barely road worthy wreck for 3000 € and run it until it's only ripe for the junkyard to fetch maybe 500 € every two years, that's still about 100€ a month you're paying.
And what we also haven't had a single talk about is operating and purchase taxes, highway tolls, city-core tolls, rental spots for parking (including the price you have paid for the garage in your house, it's a lot of real estate), that also can easily add to many hundreds of euros each year.
Cars are expensive once you actually include replacement/depreciation and maintenance costs.
> Cars are expensive once you actually include replacement/depreciation and maintenance costs.
Yep, that describes cars. High up front cost that barely goes up when you need more done (meaning: family of 5? Car beats even the bus fare for a 3km ride to school). In trade for independence, cheap groceries, cheaper travel (at least in opportunity cost), cheap days out with the family, bigger house is realistic, ability to go work in not so well connected places (I'm a consultant), capacity to actually get heavy things, collect people, not waiting/dragging things around in cold/rain/...
Oh and these DON'T add up. Bring the kids to school AND drive to work AND get groceries by car? You don't pay 3 times like you do with any other means of transport, you pay 1.2 times what you pay when doing only one.
With 2 people in the car it easily matches public transport costs if you use it enough. Oh and even by yourself it's like half taxi/uber fares, a third or less of waymo fares (though at least those don't charge per person).
The standard tax deduction for car travel is $0.70 / mile in the US, which accounts for things like insurance, gas, maintenance, and depreciation. So $500 / month is around 700 miles, which probably around 90% of US drivers surpass.
Car insurance has essentially doubled in price over the past few years, from a combination of
- cars becoming more complicated to repair. Marco Arment of Overcast related an incident where his Rivian had a simple fender bender, and his insurance was billed $15,000 in labor and parts to fix it because of the monobody construction where you have to tear apart half the car to fix anything
- inflation in both goods and services means car repair costs are going up
- more reckless and uninsured drivers thanks to general post-covid norm breakdowns
Insurance alone can now be $150-200/month even if you don't have a particularly nice car. Combine that with gas, maintenance, and registration taxes, and I think most people in the US are paying at least $350/month for their car even if amortized costs mean they don't realize it.
Marco Arment of Overcast related an incident where his Rivian had a simple fender bender, and his insurance was billed $15,000 in labor and parts to fix it because of the monobody construction where you have to tear apart half the car to fix anything
Hyundai Ioniq 5, backing into the garage next to the RV, and at "backing into the garage" speed ran into the RV. The fiberglass body of the RV suffered a 3 inch diameter break in the fiberglass that I could have fixed myself. The Hyundai? 17,000 American dollars. The rear quarter panel took a dent, and (IIRC) the bumper might have had some damage. Part of the problem was that there really isn't a "rear quarter panel" anymore. No, as I looked at it, that piece of sheet metal goes all the way from the rear bumper to the front of the passenger compartment. The shop didn't replace that piece, but rather cut the dented piece out and welded in new sheet metal.
Between that, and all the sensors, etc., $17K for backing into a piece of fiberglass at not even a walking pace. Now that the car has some years on it, if I do that again they'll probably total it.
huh? i bought a used, very low-end/utilitarian 10 year old car and paid more than half upfront and my monthly payment was like $300. factor in insurance and gas and i was easily close to 400-450. the days of $1000 beaters that actually run well are gone :(
The cost is a factor -- and something that I think policy makers should very much push to change.
For our family of four, two of us pay for public transport as of now. That adds up to $12 round trip; which is often more expensive than parking in the even in a high density area. Once we have to start paying for the kids too, that would add up to $24 for a round trip, which ends up being more expensive than driving. I get that public transportation is expensive to operate; maybe that alone is the root of the problem here.
Yes, all those things. Except on cost, at least in SF, MUNI is free for children.
We mostly drive wherever we need to go, especially when it's all of us. But if we're going to a Warriors game, we always take Muni, at it's more convenient (and free for adults too if you show your ticket).
Also, it's generally faster and more convenient (and fun) to get to Chase Center via Muni than driving. Getting back is tough both because this is peak Lyft/Waymo demand as well as peak Muni demand.
In most of America there is abundant free parking on private property including homes, stores, and workplaces. That is hardly a subsidy. I understand the argument that dense cities shouldn't have so much free public street parking but there are only a handful of neighborhoods where that even matters.
The "free parking" isn't really free, you just have land that is really cheap devoted to it. And where it isn't...well, American housing prices and rents are increased to pay for them. Street parking matters in almost every neighborhood in Seattle now...since parking on its own is expensive, and you will also have to pay for a few busted windows on your car for the pleasure of free street parking.
The highways are heavily subsidized by general funds these days since raising the gas tax outside of a few states isn't very popular.
I'm American but in the other countries I lived in (Switzerland and China) and the many countries I've visited, private car ownership is always a luxury, not a cheap necessety attainable by everyone.
If you have a driveway. I had to look around hard for a house with a car port that wasn't just a slot in a crowded alley, heck, I saw some beautiful houses that had no effective parking at all (maybe they had sunk garages built in 1920 that were not usable by modern cars).
> And where it isn't...well, American housing prices and rents are increased to pay for them.
The driver of housing cost in US cities is lack of supply. Parking spaces are a drop in the bucket versus what is missing. The root cause is zoning laws; particularly the height restrictions as they currently stand.
That was 2022, $56k is probably about 10% of a one or two bedroom condo price.
> The root cause is zoning laws; particularly the height restrictions as they currently stand.
Tokyo is, as I understand it, the libertarian ideal for a city that doesn't let zoning get in the way of a good time, and parking space prices are still expensive there:
> Monthly rental rates for spots in the 23 Wards range from ¥30,000 to over ¥80,000, which reflects high underlying property value.
I think there's a misunderstanding. I'm not claiming that parking space if charged at the market rate for an unimproved cement room in a high rise is particularly cheap. You can only fit so many cars within the footprint of a typical condo after all.
I'm claiming that removing parking (ie converting the raw sq footage over to living space) would not meaningfully impact housing prices. The existence of parking, free or otherwise, is not a significant contributor to the housing shortage. The issue is one of scale. That's what my "drop in the bucket" comment is referring to.
You specifically said "American housing prices and rents are increased to pay for them". That is technically correct but in context it is blatantly wrong. The price increase as it stands is approximately zero.
The error is failing to differentiate between cost due to construction and maintenance versus cost due to land value. The latter is linked to total supply and thus height restrictions. The former is not the primary component in HCoL cities. You can easily verify this by checking the cost to purchase an apartment building in say San Francisco versus a small town in the midwest. (I refer to the cost to purchase the entire building there, not the cost to rent a single unit.)
Parking garages in HCOLs are expensive, they definitely aren’t free. You can’t build a new multi family without planning for one or two levels of garage underneath. But you are correct that sub-basements, at least in the USA, wouldn’t have been used for living space anyways.
Right, I specifically called out that I agree with you on that. They aren't cheap. But then most places around here charge $50 or even $150 per month per parking space so it's not like the spots are being given away either.
> you are correct that sub-basements, at least in the USA, wouldn’t have been used for living space anyways.
That isn't what I said. I claimed that the amount of space dedicated to parking, if converted to housing, would not meaningfully reduce the housing shortage. It's a simple numbers game. The shortage is far larger than all of the current parking combined. We badly need to build much farther upwards but it is not permitted to anywhere near the extent necessary.
Another way of looking at it is to ask, if every unit of housing in a major city added additional square footage equal to a single car, would that make or break the market? Even at the scale of the entire market it would still be well under 10%, probably under 5%. The typical apartment in the US is definitely larger than a 5x2 grid of parking spaces. Meanwhile most HCoL cities could do with double the housing inventory at absolute minimum. Probably substantially more.
This is the same problem with reducing setbacks. Unscrupulous developers keep lobbying for that (and often getting it). We don't need to reduce buffer space. A few extra feet around the perimeter of a lot is nothing compared to doubling (or 3x, or 5x, or ...) the height.
We are suffocating under our own political dysfunction.
> That isn't what I said. I claimed that the amount of space dedicated to parking, if converted to housing, would not meaningfully reduce the housing shortage. It's a simple numbers game.
In HCOLs places, parking garages, usually basements, are the solution to this problem. If you want to argue that they wouldn't solve the housing problem in SFH neighborhoods...well, SFHs aren't going to solve the housing problem anyway that you look at it, so...
> We badly need to build much farther upwards but it is not permitted to anywhere near the extent necessary.
You are also right. You just need to add your budget of the garage into your housing projects costs, or not, since people of the option to buy condos in buildings that do not mandate you also buy a parking spot (which can pay for the underground garage construction).
> Another way of looking at it is to ask, if every unit of housing in a major city added additional square footage equal to a single car, would that make or break the market?
OMG, yes, if you mean major cities in China. How the heck would you even build enough underground garage space to even think about doing that? The US is nice because our cities are small and not very dense, so we aren't talking about adding parking for every unit in a 40 story...heck, the road infrastructure alone to get that many cars in and out of the garages would bankrupt Beijing or Shanghai.
> This is the same problem with reducing setbacks. Unscrupulous developers keep lobbying for that (and often getting it). We don't need to reduce buffer space. A few extra feet around the perimeter of a lot is nothing compared to doubling (or 3x, or 5x, or ...) the height.
More first world problems and American exceptionalism I guess. No, I disagree, but you should really visit Tokyo.
> If you want to argue that they wouldn't solve the housing problem in SFH neighborhoods
I am arguing that anyone who blames the presence of parking for housing supply issues has failed to understand both the geometry and scale of the problem (or more likely imo is actively attempting to push an anti-personal-car narrative).
> the road infrastructure alone to get that many cars in and out of the garages would bankrupt Beijing or Shanghai.
I wasn't talking about traffic engineering problems. Only raw square footage for living space. You can generalize the question I posed as - in an alternate reality where every housing unit in (say) Beijing were precisely 10% larger, and total stock were reduced proportionally to accommodate that change, but everything else were exactly the same (improbably, I know) would that make or break the housing market in terms of supply and demand? The answer is that it would not. Housing supply problems are not due to a mere 10% shortage.
> More first world problems and American exceptionalism I guess.
I don't follow? What about my objection to reducing buffer space comes across as "American exceptionalism" to you? And why do you disagree?
It's simply a matter of geometry. Expanding the footprint by a few feet versus duplicating the entire structure upwards multiple times. Obviously that doesn't apply to places that already build upwards to the extent physically possible but since approximately nowhere in the US does that it's neither here nor there.
> No product had such a fast transition from novelty to "omg i never want to interact with a human again".
I still recall when taxi services were the only offering, and Uber et al were marketed as ride sharing services instead of ride hailing services. It's hard to put into words the transformative effect that ride hailing services had throughout the world. Overall rides are now far safer and more reliable, to the point where the old days feel like the dark ages.
My anecdote: My wife had to literally have two drinks before here first Waymo ride. Now she doesn't want to use anything else other that Waymo when we can't drive ourselves, and totally agree with her
Having said that, Uber was amazing experience when it started too, now it's on par with cabs.
I'm not against automated driving at all, but in my experience we actually don't have that much use for stuff like this in most (big) European cities, since almost all of them have good public transport options already. I think trams especially fill the hole of "low-friction transport in a city" perfectly. I think having less vehicles on the road is a benefit to us all, but I understand some cities are not as tightly packed for public transport to work that well.
Yeah, this is one of my guilty automation pleasures alongside self-checkout. I hate that I am displacing a human, and I mourn for the handful of really pleasant taxi / Uber experiences I’ve had over the years, but damn is Waymo such a better default experience right now.
I really hope there’s enough viable competition over time to keep costs down or I worry this will evolve into robo-limos rather than a nice cheap default option for areas without good public transit infrastructure. The DUI prevention alone is such a huge win.
There is the matter of surveillance though. I don’t love that I have to take their word on not abusing the cabin recordings, but I guess that’s pretty much all modern vehicles (via onStar and the like) not just robo-taxis. Pretty much every Sci-fi dystopian with urban infrastructure has that scene where the corrupt authorities have someone’s self-driving car pulled over remotely, that seems important as well given the state of things lately.
I don't live in Berlin, but even if: have you ever taken public transport in less mainstream lines? Apart from nothing working and connections taking forever and operations stopping at night, horrible signage that lets you stress even more, you sit next to human excrements, hooligans coming from football games, nazis wanting to beat you up, stink, rude music and beggers. I sometimes miss it for sentimental value, but compared to a world of robots driving us with relaxing music in a clean and safe space i know what future I want.
> have you ever taken public transport in less mainstream lines?
Yes, I have. I never drove a car myself and maybe used a taxi 10 times in the last 30 years.
Will waymo even be available in less mainstream areas? It seems more reasonable for them to go for dense places instead and leave the unprofitable regions for someone else.
> Will waymo even be available in less mainstream areas?
Ever is a long time. It's not reasonable to predict beyond a decade or so. It's easily possible that this becomes huge and in the 2040s people are astonished that "driving yourself" was a thing, the same way it's hard to comprehend now that most people weren't literate. Not "Couldn't write an essay / read a newspaper" but "Couldn't sign their name / read a postal address"
But it's also possible that this goes nowhere, and outside of a few large cities there is never a robot taxi market, it just doesn't exist. Waymo is, among other things, a bet that there is a large market.
Dense places are where it starts, but that was also true for the telephone. Bell didn't provide service to tiny rural settlements, they wired places like Boston and New York, AIUI the general service provision was a government initiative even in the US, it was never strictly profitable enough for huge corporations to spend their own money making it universal.
I mean, I can understand wanting to start in dense places. But those are also the places where public transit is a viable existing solution.
Personal transit just looks incredible inefficient and unscalable if everyone would use it. I could totally see it as a last resort solution for situations where nothing else is available, but that's an unattractive market that isn't going to make anyone rich.
In an urban area the "last resort" cases add up. The last time I was in a taxi it was the middle of the night, and I'd just smashed my head open, so I had concluded that I must not trust my own judgement and should seek immediate medical attention, buses don't run in the middle of the night (on that route)
Ambulances are for emergencies. An ambulance could be dispatched depending on availability, but the dispatch team has more experience with this than I do and so they - like the hospital's initial triage team - would put me in the "injured but not dying" category and maybe I get an ambulance in an hour or two depending on other priorities.
They don't want me to go home and fall asleep, because it is possible that I have a brain injury and will never wake up, but their advice is going to be "Can you get somebody else to drive, or maybe call a taxi?" not "We will Blue Light an ambulance to you ASAP".
Is there really that much poop on Berlin public transit?
Seattle has some of the highest per capital homeless in the US, and a dearth of public toilets, and yet there's not that much poop on our public transit.
I am also skeptical that y'all's violent crime rate is higher than ours.
Granted I haven't taken Berlin public transit in 20 yrs, so I don't know.
It wouldn't take much to have more violent crime in Seattle, according to my gut (yeah, I know, "show me the numbers"). Granted, it's probably gotten worse since we moved here 25 some years ago, but coming from places like my old hometown of Indianapolis, Seattle didn't have any place I wouldn't feel comfortable walking at night. Again, it's changed a lot since (there are some areas I would avoid at 2 a. m. now), but I still feel much safer in Seattle than other large cities.
I’d put Berlin ubahn halfway between nyc and japan in terms of cleanliness and orderly behavior, the bigger problem is that there’s no ac in the summer
Everyone has the "fear" of being near other people, regardless of their affluence. That's why apartments are not built for 20 but got 2-5 people and doors exist. I don't see why it must be a rich people thing when it comes to self driving cars. Could also become super interesting by making remoter areas more serviceable.
I literally got a taxi today in Berlin because the trains were on strike, and the other day because the trains and trams were broken due to ice.
You really can’t think of valid uses for a taxi?
Taxis are also public transport and so their provision in cities is in fact part of the transport fabric. Since there must be taxis, why not improve them?
This isn't about poor people, at least for me, I'd much rather be alone than with Elon fucking Musk. If I want to hang out with people I will choose when and who. The least good bit of being in a taxi is small talk with the driver.
Waymo is not solving driving, it is closer to a sophisticated Disney Parkland ride. It is running inside a tightly constrained Operational Design Domain:
- Geofenced areas
- HD pre-mapped roads
- Curated infrastructure
- Remote ops fallback
This is not general autonomy, it is highend automation inside a controlled distribution. The system degrades exactly where humans do not: construction, unmapped lane shifts, police manually directing traffic, chaotic mixed behavior.
A cop overriding a light is not an “edge case”, it is a semantic and social reasoning problem that current perception stacks still do not robustly solve.
It works because the world is pre modeled, not because the car understands driving.
Scaling that beyond a few mapped US suburbs into Europe is a totally different problem. Dont get fooled by Wall Street stock pumping.
From a consumer perspective, no one cares what Waymo really is. If customers can pay and get from point A to point B reliably and safely then it doesn't matter how the sausage is made. Regardless of technical challenges and limitations they're obviously going to expand coverage to more areas.
If you put human lives on the line, both on the shared public road and inside the Waymo then how the sausage is made totally matters, as directly applies to what the failures modes are. Safe from A to B only holds in ideal conditions and limited zones. The hard problems like rare edge cases, weather, unpredictable humans, are precisely why it cant scale easily.
If the tech was truly solved, Waymo would not be geo fenced or expanding so slowly.
It may not scale easily but it is scaling. Waymo (Alphabet) has access to essentially infinite capital to make that happen. I predict that within 10 years the majority of the US population will have access to their rides.
Im not because the statement you wrote is too generic, « not solving driving » … solving driving could mean many different things. It’s the classic rage bait one liners you see on Reddit from Tesla bag holders. It’s funny because everything you’ve described IS solving « driving » from my perspective. I sit in the back and go read hacker news without having to drive.
Roads are not solving transportation, they are closer to a sophisticated trace track. Roads are a constrained Operational Design Domain:
- Geofenced areas
- pre-build structures
- Curated infrastructure
- fallback to gravel in times of the inevitable event of maintenance.
This is not general transportation, it is a highend infrastructure inside a controlled environment. The system degrades exactly where humans/horses do not: River crossings, Creeks, steep hillsides, marshes, beaches.
A river flooding a road is not and "edge case", it a usual occurrence, and a problem that roads do robustly solve. It works due to extensive maintenance, not because the asphalt can actually deal with water.
Scaling that beyond a few mapped US suburbs into Europe is a totally different problem. Dont get fooled by Wall Street stock pumping.
Asphalt is not marketed as Level 5 intelligence. You can make analogies but that is very different from a rebuttal...that you did not do. The hard part is still unstructured human chaos. Time will prove which one of us is right.
I am pretty sure you are underestimating how important it is for them.
But we can look at their own scientific publications.[1]
No maps no driving....carnival ride...
"High Definition (HD) maps are maps with precise definitions of road lanes with rich semantics of the traffic rules. They are critical for several key stages in an autonomous driving system, including motion forecasting and planning. However, there are only a small amount of real-world road topologies and geometries, which significantly limits our ability to test out the self-driving stack to generalize onto new unseen scenarios..."
Most of this comment was written by an LLM. There are certain tells, such as the tone, as well as usage of “ for quotations instead of the much more common ". I think you added the last couple of sentences.
Important to note that termux, while it was always great and indispensable, is getting increasingly interesting now because android is getting full desktop mode at the same time as XR glasses (xreal, viture etc.) are becoming mainstream. You can have a linux desktop in your pocket everywhere without rooting, hacking or tinkering, just install termux and x11 server. While not all packages work, llms are increasingly powerful, just an example porting deskflow from a debian package to termux took about 4 hours max, something i would not even have had an idea how to start just 4 months ago.
I would bet the popularity is due to coding agents. For the first time you can continue the work without typing much but just inspecting the output and provading further guidance with relatively short messages.
I bet thats true, but that use case will be much much better served by the native vm based linux terminal, not sure what sandboxing you can use in termux.
This is interesting! Do you have any more info? I just discovered sunshine/moonlight work surprisingly well on Quest 3 for remote desktop to linux, I'd not really considered Termux X11 natively though.
I dont know how to feel about being the only one refusing to run yolo mode until the tooling is there, which is still about 6 months away for my setup. Am I years behind everyone else by then? You can get pretty far without completely giving in. Agents really dont need to execute that many arbitrary commands. linting, search, edit, web access should all be bespoke tools integrated into the permission and sandbox system. agents should not even be allowed to start and stop applications that support dev mode, they edit files, can test and get the logs what else would they need to do? especially as the amount of external dependencies that make sense goes to a handful you can without headache approve every new one. If your runtime supports sandboxing and permissions like deno or workerd this adds an initial layer of defense.
This makes it even more baffling why anthropic went with bun, a runtime without any sandboxing or security
architecture and will rely in apple seatbelt alone?
But even then, the agent can still exfiltrate anything from the sandbox, using curl. Sandboxing is not enough when you deal with agents that can run arbitrary commands.
If you're worried about a hostile agent, then indeed sandboxing is not enough. In the worst case, an actively malicious agent could even try to escape the sandbox with whatever limited subset of commands it's given.
If you're worried about prompt injection, then restricting access to unfiltered content is enough. That would definitely involve not processing third-party input and removing internet search tools, but the restriction probably doesn't have to be mechanically complete if the agent has also been instructed to use local resources only. Even package installation (uv, npm, etc) would be fine up to the existing risk of supply-chain attacks.
If you're worried about stochastic incompetence (e.g. the agent nukes the production database to fix a misspelled table name), then a sandbox to limit the 'blast radius' of any damage is plenty.
That argument seems to assume a security model where the default prior is « no hostile agent ». But that’s the problem, any agent can be made hostile with a successful prompt injection attack. Basically, assuming there’s no hostile agent is the same as assuming there’s no attacker. I think we can agree a security model that assumes no attacker is insufficient.
Code is not the only thing the agent could exfiltrate, what about API keys for instance? I agree sandboxing for security in depth is good, but it’s not sufficient and can lull you into a false sense of security.
This is what emulators and separate accounts are for. Ideally you can use an emulator and never let the container know about an API key. At worst you can use a dedicated account/key for dev that is isolated from your prod account.
VM + dedicated key with quotas should get you 95% there if you want to experiment around. Waiting is also an option, so much of the workflow changes with months passing so you’re not missing much.
That depends on how you configure or implement your sandbox. If you let it have internet access as part of the sandbox, then yes, but that is your own choice.
Internet access is required to install third party packages, so given the choice almost no one would disable it for a coding agent sandbox.
In practice, it seems to me that the sandbox is only good enough to limit file system access to a certain project, everything else (code or secret exfiltration, installing vulnerable packages, adding prompt injection attacks for others to run) is game if you’re in YOLO mode like pi here.
Right idea but the reason people don't do this in practice is friction. Setting up a throwaway VM for every agent session is annoying enough that everyone just runs YOLO on their host.
I built shellbox (https://shellbox.dev) to make this trivial -- Firecracker microVMs managed entirely over SSH. Create a box, point your agent at it, let it run wild. You can duplicate a box before a risky operation (instant, copy-on-write) and delete it after.
Billing stops when the SSH session disconnects.
No SDK, no container config, just ssh. Any agent that can run shell commands works out of the box.
apart from nearly no one using vms as far as i can tell, even if they were, a vm does not magically solve all the issues, its just a part of the needed tools.
My community and me are waiting for browserBench for a while now and happy to see it finally starting. Browsers are arguably one of the most complex and foundational piece of software, the ability to create something like this from scratch will be an important evaluation as limits of what is possible are harder and harder to find.
I am fully moving from local electron based vscode to using vscode-server inside docker inside a vm. It has just so many advantages besides security eg. being able to have multiple workspaces in tabs instead of separate electron windows, and having all the docker/vm tooling available. This can replace remote vscode, devcontainers and electron in a nice package. There is just no reality in which vscode with electron running as user account on a bare machine can be secure not even thinking about agents in the mix. We are working on a custom browser called darc based on chromium IWAs and controlled frames instead of electron and optimised for this. (apache 2.0)
Docker does not and cannot offer full isolation. A sandboxed VM on someone else's computer is less likely to be problematic for running untrusted code than a container on your system.
seems not to justify submitting to a proprietary single vendor solution where users are locked into opaque checkpoints they forgot how to migrate away from. this is not something made for users lets be clear. there are tens or hundreds of vm layers for defense in depth for docker so thats a non argument, no one says docker has to provide security its for tooling and common practices that allow vendor independence and moving to self hosted stacks as needed!
reply