I think the biggest risk to ChatGPT as a consumer brand is that they don’t own the device surface. Google / Microsoft / Apple could make great AI that’s infused in the OS / browser, eliminating the need to go to ChatGPT.
Since microsoft kinda sorta owns or is merging with openai it's probably already close to that... copilot is constantly down for me at least, but I assume that's not a hard thing to fix on Microsoft's end if it wants to start paying the server costs...
The economics of 3BR family units are typically hard for developers to make money on. Bobby Fijan (https://x.com/bobbyfijan) is an example of a developer who is a vocal advocate of family-centric apartments and townhomes. His projects look amazing. He also talks about the challenges creating family housing.
Single stair is one of the reforms I'd most like to see.
At the time 2 stair requirements were adopted it was vital, with devastating urban fires a common occurrence. We have so many new options for both preventing fire and keeping evacuation routes accessible for hours that it's no longer required.
The regulation has a huge impact on the layout and form it's possible to build, and I think it's a huge driver of the visceral reaction against apartment living in the US and Canada.
Being able to build 4-8 storey apartments on a single lot with a central stair where every unit has windows on at least 2 walls would be a game-changer for north american urban spaces and a pathway out of the housing crisis.
> I think it's a huge driver of the visceral reaction against apartment living in the US and Canada.
On the one hand, maybe, but on the other hand, apartments (with the same number of bedrooms for the same COL-adjusted price) in the US are enormous compared to those in Asia and in Europe. I think the real source(s) of the visceral reaction(s) is, in no particular order, Americans' prioritization of personal independence over pragmatism (and I don't mean that pejoratively, though it can get stubborn at times), America's fairly weak renter protections/regulations, and the poor build quality of many American apartments (with dogshit sound and climate-proofing). I think it's a mix of a fundamentally American aversion to adding an additional person telling you what to do with genuine issues in the paradigm where you're paying up the ass for heating/cooling because your landlord doesn't particularly feel like installing double-pane windows, and at the same time your neighbors and neighborhood are obnoxiously loud.
A big question is whether these areas actually turn into denser housing, or whether something else in the process manages to bog it down. Plenty of housing bills have seemed like a big deal when you looked at the area they impacted, but in practice they led to little new housing.
A lot of the California bills had various poison pills in them that reduced their effectiveness.
For example, SB684 allowed building and subdividing up to 10 units on a multifamily lot. BUT, the lot wasn’t eligible if you had to knock down a building that had tenants in the past N years to avoid displacement of people.
You can probably guess how many multifamily lots are out there where you don’t have to tear down an existing building with tenants.
There are other issues too. Interest rates and tariffs make a lot of projects not viable financially.
The problem we are seeing in suburban MA right now is that they're building the wrong kind of housing to address the greatest shortage, and doing it in a way that does not promote the long-term community well-being. Developers are jamming in large numbers of small "luxury" units with insufficient parking in a car-dependent area in whatever lot they can get their hands on, instead of density increasing organically throughout the immediate area around downtown, adding ADUs and replacing large single family homes with 2/3/4-unit condos.
Why is this happening? Because zoning boards don't allow reasonable multifamily development in densifying areas, so developers do the only thing they can do, which is build in already-built-up areas with looser zoning, and/or ram projects through using a state low income housing provision called 40B.
The effect is that while the apartment market for young professionals is going to continue to soften, the market for comfortable family dwellings remains brutal and increasingly unaffordable. There are 80+ unit apartment buildings literally surrounded by multi-lane stroads, while less than a mile away there are single family homes on quiet tree lined streets where you could easily have the same number of units in multifamily condo buildings and garden apartments, and still retain the comforts of suburbia.
So whatever poison pills are in here, it cannot be worse than the status quo in MA, in which the development is too much of the wrong thing and everyone loses in the end except the developers and real estate agents.
Yes, sadly just because the zoning has changed does not mean that any of the buildings that already exist will be torn down and rebuilt for decades, if not longer. And localities have all kinds of restrictions and fees that still prevent building to take place.
Ug. I'm a 'yimby' and a Weiner voter. But his take on San Francisco transit is just like really bad. Pokey streetcars and buses, doomed to fail. You build out there in those blue areas, and they are mostly all driving.
My take is you build it, and THEN they come. Put in some GOOD transit. Make sure the utilities are in place. Developers will then flock to the place. This whole thing is using inside-out logic. Have a real plan first.
We DID build good transit. It takes 15 minutes to get from the MacArthur BART to downtown San Francisco! But the walkable area around that station is full of single-family housing. It's a huge waste building all of this incredible public transit and then not allowing apartment buildings near it.
The same is true for so many of the East Bay BART stops. Amazing transit but apartment buildings are banned so it's much more expensive to live there than it should be.
Because developers are going to include the parking so everyone out there can drive where they are going. Which is my point. The "transit" aspect of this bill is total bullshit. If you like cars, and want more traffic, this is for you.
Those areas are fairly small, and their number is limited. That might drive more space efficient solutions like underground parking or selling spots in a parking garage that serves multiple buildings
Nah, if you build high-density housing near transit stops the transit will definitely get more ridership (and ergo more funding). You're painting with broad strokes saying "they'll drive anyway!" but really a lot of car trips will get replaced with transit or walking, and that trend will only increase as more dense housing gets built.
But his take on San Francisco transit is just like really bad. Pokey
streetcars and buses, doomed to fail. You build out there in those blue
areas, and they are mostly all driving.
One of the best parts about where I lived in San Francisco was that I was around the corner from a streetcar stop. Pre-pandemic the streetcar was absolutely packed during commute hours because people absolutely do take advantage of "pokey streetcars and buses".
My take is you build it, and THEN they come. Put in some GOOD transit.
What is GOOD transit? The Bay Area's spent a fortune building out BART (yuck) and every extension has only succeeded in siphoning money away from other transit.
Yep, me (and my politician Weiner) have some great public transit, so that's certainly true of some places.
Those folks that he is up-zoning out in the avenues, they are driving. Different culture out there. Downvote reality to the left.
GOOD transit it obvious, and it certainly is not a gigantic tunnel deep under downtown San Jose which is 400% over budget. Do not claim there is a lack of money for any of this. The political machine is just totally malfunctional.
Just curious exactly what neighborhood and how long did you live there? (duboce triangle, a long time)
I know a bunch of people who live/d out there, they take muni downtown but mostly have a car. Perhaps you could regal in tales of your car-free lifestyle in the Sunset, but i'm not seeing it. Weiner wants condos out there and he does not give a shit about your bus ride.
As I pointed out earlier streetcar, not a bus ride.
but i'm not seeing it
Look harder.
Pre-pandemic the L and N were the two busiest rail routes (and the busiest lines systemwide) even considering they share most of the underground portion with four other lines. Somewhere around 75,000 daily trips.
e-bikes seem to have been invented for SF. They solve the hill problem and long lines at school drop offs. I see their growth continuing to reduce car needs in SF.
Sadly, I say this as someone who lives in Duboce Triangle and owns two cars.
> My take is you build it, and THEN they come. Put in some GOOD transit. Make sure the utilities are in place.
The problem is, that costs money that, for a few years at the very least, will not be recouped. Not many politicians have the ability to push such efforts through regardless of profitability, especially not when the topic in question will be abused by the opposition in their usual culture war bullshit.
It costs a lot less to build transit infrastructure before or at the same time as everything else compared with adding it later, even if the line is underused as density is added.
The best alternative is a well-planned phased line with carefully protected right-of-way and a dedicated source of long term funding. Bonus points for it being a combination of value capture taxes and the transit agency being a property developer in their own right around stations. The early phase can be inside the boundary of current development so there are people to ride right away. Developers can build and market using the upcoming line, and prospective residents can be confident it will happen with funding secured.
* The projects won't be profitable in smaller towns, because rents aren't high enough to recoup the cost.
* Tall buildings cost MORE per square foot than short buildings, so tall buildings only get built where land costs are very high.
* This law's top density (7-8 floors I think?) only applies in a narrow window (0.25 to 0.5 miles) around major transit stops with LOTS of service, like < 15 minute bus intervals with dedicated BRT lanes, or trains with > 48 arrivals per day each way. Small towns don't have that kind of infrastructure.
* The law only applies in cities with > 35,000 people.
This law (and other recent CA YIMBY laws) don't create much surface area to sue or slow a project:
* The approvals are designed to be "ministerial", meaning there is no discretion on whether to approve or not. If the project meets the objective criteria spelled out in the law, it must be approved.
* If the city doesn't approve in a limited time window, it's deemed "approved" by default.
* Ministerial approval protects the project from CEQA lawsuits. CEQA requires the government to consider the environment when making decisions. When the approval is ministerial, the government doesn't make any decisions, so there is no CEQA process to sue against.
They did. They started before yesterday's shutdown, and worked all night, they tried to bring up the system for startup, and it came up, then crashed.
It was state of the art on 1962 when it was designed, and remained state of the art until the 1980s, when the signal system started breaking down, and the the late 80s upgrade which had a train presence glitch, which caused almost all the system to get resignaled.
So by the 2000s again it's showing its age, and they got a 32 processor zSeries mainframe.
Brake problem last week, and the this on Friday? Now it's getting like New York, even more. Whatsmatteryou?
It doesn't require a mainframe but that was the cheapest path to keep things running without rewriting the software. The IBM Z platform is very good at maintaining backward compatibility. If you don't constantly keep your applications software up to date with support for new platforms then eventually you find yourself with very limited platform options.
Nothing, but rewriting a train system from scratch and testing it is incredibly expensive and disruptive. Eventually you do have to do it, and we might not agree with their trade offs here, but it's not unreasonable.
But it can still work. in a company, as an employee, you can say a tool needs to be paid and get your company to pay even if there are binaries available elsewhere.
I work for a company that sells open source extensions. It would be pretty trivial and legal to remove the license checks, but companies just pay.
Engineering time is money and going through internally the process of building, testing and distributing binaries for every release is a lot of engineering time. Paying a sum of money to make the problem go away is a reasonable solution.
Exactly. The cost, for a small company, is about $500 per year. That's way less than the engineering time it would take to set things up to build internally, keep up to date with fixes, put it all in internal repositories, etc.
Right but in supply chain terms I'm way less likely to trust the build from some random person. So if your company cares about it's supply chain it's easy to justify the expense.
bullet trains should be displayed in red, but I haven't tested it yet at the right time and I've been too lazy to write tests with mocked time / gps :)
