kettlecorn

I've been using Zed primarily for months but I just switched back to VSCode for 2 reasons, one of which is kinda my fault and the other it's unclear where the fault is.

1. I deleted a few hours of work because I was late night coding and I decided to rename a file before check-in and delete the old version. Well I renamed it and Right-Click -> Deleted the new version by accident.

It turns out Zed has both "Delete" and "Trash" right next to each other in its right-click menu. Delete skips the trash and Zed does not implement Ctrl+Z yet, so unless you have a backup that file is gone. Because I renamed it and had not yet checked in it wasn't in version control yet.

2. For some reason a particular crate in my Rust workspace wasn't having errors and warnings show up in the editor. I messed with configuration a bunch and couldn't figure it out, so I tried VSCode and it just worked without any additional configuration.

I've thought a lot about why software doesn't converge on higher quality and I don't know if I've ever come to a satisfying conclusion.

The simple statement is that "Quality isn't what sells" and I think there's some truth to that, but I also think quality can sell and theoretically if all other factors are equal higher quality will sell better.

What I've had a tough time pinning down precisely is I think there's almost an ecosystem balancing act occurring between quality, effort expended, and skill required. Particularly in American culture I think one of our economic 'super powers' is that we sort of have an innate sense for how much something can be half-assed while still being mostly effective. We're great at finding the solution that gets 80% of the job done for 20% of the effort while requiring less skill.

I think you can see that through all sorts of disciplines, and the result is a world of kind of mid quality things that is actually quite efficient and resilient.

Where that goes wrong is when the complexity and problems start to compound in a way that makes any effort substantially more wasteful. It reminds me of how my game jam experiences go: for the first 60% I'm writing code at a tremendous velocity because I'm just putting down whatever works but by the end the spaghetti nightmare of the code reduces my velocity dramatically. Even when that occurs though the ecosystem nature of complexity means that your competitors are likely mired in similar complexity, and the effort to get out may be large enough and require enough talent that you don't you can change.

To improve quality I think it's not enough to just say "People should care about quality more" I think you have to fundamentally target aspects of the ecosystem, like changing the course of a river by moving a few rocks. A good example I think has adjusted the behavior of the ecosystem is Rust: it makes certain things much easier than before and slowly the complex bug-mired world of software is improving just a little bit because of it.

It's pretty good! I peaked a little bit under the hood and most of the size is going into the 3D models and secondly the sound effects. The largest single asset is the looping music track which is 2.3 MB.

The 3D models are compressed with 'Draco' compression. So for example the largest asset is a model of the entire world which is 333 kB. It consists of 81k vertices. Unoptimized that would mean at least 3 32 bit floats per vertex, which would be at least 972 kB uncompressed. So Draco compression is doing a pretty good job getting the model sizes down.

I'd have to dig into it more but I suspect if they wanted to they could have trimmed the size down quite a bit, but it may not be worth the effort. They could have used more repeat 3D models, but instead it looks like they went for a more artist driven look where most of the world is uniquely modeled.

It also looks like they're sending a bunch of lower res levels of detail per model over the network. That also is a tradeoff. Still with network speeds as they are nowadays it may be faster / simpler to send those over the wire each time instead of regenerating the level of detail.

The textures are a similar story. Well compressed but they're sending procedural noise over the network. Those could be generated but it's also trivial to send them over the network.

One of the most impressive things about this is that it transfers only 5.7 MB of data on initial load and then caps out at 17.5 MB for the final load.

It seems to be making smart choices about compression techniques.

This is a good showcase of how well games can work on the web if done well.

I think Apple is doing fine and is reasonably well positioned. The pre-Cook years basically established the skeleton of the company and Cook has fleshed it out with flesh and muscle. Now the company is far more of a Goliath with vast resources to help it weather strategic pivots, in addition to a loyal and trusting customer base.

Looking ahead to the next decade or so I believe what we're likely on course for is some sort of AI OS. AI will be able to vibe-code apps for you and you'll be able to manage and dispatch individual agents to do tasks. There will be a messy transitionary period that lasts a while from today's software.

Apple has a trusting customer base, they have tremendous hardware prowess, they have cross-platform cloud services, and they have design sense to figure out how to establish these new models. I think if a new form of computing emerges Apple basically gets a free first swing at it, in terms of consumer trust. We've seen that with their VR headset, that while a dud was given far more interest and attention than Facebook's and Microsoft's offerings.

I do think they show signs of vulnerability as well, even if I think they're less than their strengths. Apple's famously design focused culture seems to be cracking a bit and giving way to more Microsoft / Google style PM-led culture. That can be OK, but I think PM led cultures are sometimes almost too strategic for their own good, like someone you can tell is a salesmen not a friend, and that can result in gradual brand erosion.

A simple example of this happening is Apple's website. Like Microsoft the tabs at the top of their website now reflect the company's internal organization, not what makes the most sense to the user. In the Jobs era they insisted that there would be no separate "store" tab, a redundancy because the whole website was a store, but I suspect that changed because the "Store" division within Apple wants that extra prominence.

If Apple isn't careful with enough missteps they may dilute their brand and Google / Microsoft may seem just as appealing for early adopters of emerging AI OSs.

> I think a more realistic (but depressing) goal is modest loss to inflation (so prices go up by 1-2% less then inflation).

If you read between the lines that seems to be what Canada's approach to house affordability is going to be. Their leaders are promising housing will be more "affordable" but that the goal isn't to decrease home prices.

One of the Rust ecosystem's biggest mistakes, in my opinion, was not establishing a fiercely defensive mindset around dependency-bloat and compilation speed.

As much as Rust's strongest defenders like to claim, compilation speed and bloat just really wasn't a goal. That's cascaded down into most of the ecosystem's most used dependencies, and so most Rust ecosystem projects just adopt the mindset of "just use the dependency". It's quite difficult to build a substantial project without pulling in 100s of dependencies.

I went on a lengthy journey of building my own game engine tools to avoid bloat, but it's tremendously time consuming. I reinvented the Mac / Windows / Web bindings by manually extracting auto-generated bindings instead of using crates that had thousands of them, significantly cutting compile time. For things like derive macros and serialization I avoided using crates like Serde that have a massive parser library included and emit lots of code. For web bindings I sorted out simpler ways of interacting with Javascript that didn't require a heavier build step and separate build tool. That's just the tip of the iceberg I can remember off the top of my head.

In the end I had a little engine that could do 3D scenes, relatively complex games, and decent GPU driven UI across Mac, Windows, and Web that built in a fraction of the time of other Rust game engines. I used it to build a bunch of small game jam entries and some web demos. A clean release build on the engine on my older laptop was about 3-4 seconds, vastly faster than most Rust projects.

The problem is that it was just a losing battle. If I wanted Linux support or to use pretty much any other crate in the Rust ecosystem, I'd have to pull in dependencies that alone would multiple the compile time.

In some ways that's an OK tradeoff for an ecosystem to make, but compile times do impede iteration loops and they do tend to reflect complexity. The more stuff you're building on top of the greater the chances are that bugs are hard to pin down, that maintainers will burn out and move on, or that you can't reasonably understand your stack deeply.

Looking completely past the languages themselves I think Zig may accrue advantages simply because its initial author so zealously defined a culture that cares about driving down compile times, and in turn complexity. Pardon the rant!

The Rust ecosystem has long downplayed the importance of compile times.

Many foundational crates, serde for example, contribute much more to compile times than they need to.

I spent a long time reinventing many foundational rust crates for my game engine, and I proved its possible to attain similar features in a fraction of the compile time, but it’s a losing battle to forgo most of the ecosystem.

It could help us reconsider the routing and design of highways, particularly in dense urban environments.

Algorithmic feeds are useful. Algorithmic feeds that promote primarily people who pay the platform, downrank links, and are likely manipulated for political reasons are far less desirable. On X it meant that I was discovering far less quality content because many intelligent and interesting people didn't want to pay for the platform, often due to Musks's behavior.

BlueSky's algorithmic feeds are less good so far, but they're far more neutral and there's more transparency into how they work. Over time people have the ability to create new types of algorithmic feeds.

I miss the Google of yesteryear which had an altruistic streak and felt that enriching the world's ability to share and process information would ultimately accrue benefit to Google as well.

The Google of today is far more boring and less helpful.

In addition to what the other comment said BlueSky is also so hackable that I think people will continue develop more sophisticated techniques to better moderate.

X was locked down in a way that seemed to be giving the bots and trolls a significant advantage.

BlueSky's platform choices are encouraging far more signal than noise.

X / Twitter is doing the opposite. It's rotting into a place where meaningful discussion is hard to have and you have to put up with tons of trolls / spam.

I'm trying to avoid national politics on BlueSky simply because I want to take a breather from it.

I'm seeing vastly more signal and less noise on topics I care about like game dev, urbanism, art, and tech.

Community Notes is a great feature.

But on X too much value was / is being lost due to policy and algorithmic choices that discourage signal and encourage noise.

I used to go there for critical discussion of various tech issues or smart people talking about innovations at the cutting edge. As an example I used to use X to learn about modern graphics programming by searching certain terms to see what the smartest people in the field were saying while talking to each other.

It used to be that if a big account Tweeted something often you'd see true experts responding as the first replies. Now you'll see dozens of Bluechecks spamming memes or gross statements before you get to any substance, which inevitably has lower engagement.

Too many smart people have been turned away from the platform and it just doesn't generate the quality of discussion it used to. Already in the communities I care about I'm finding vastly more substance on BlueSky.

Not true. If someone can make a point honestly and in good faith I do my best to respect it even it's "harsh". I myself am no stranger to being very blunt when trying to make a case for something.

What I consider "toxic" is using slurs, trying to make people angry, bad faith arguments, spamming replies without giving them much thought, etc.

The competitive advantage to BlueSky, over Twitter / X, is that there's tremendous value in connecting intelligent and kind people while maintaining a certain quality standard.

Twitter / X, for political reasons, allowed extremely toxic behavior while at the same time disempowered communities from moderating themselves. If you force each individual user to manually block every troll, bot, or disgusting person it's a losing battle and low quality speech will overwhelm conversations. X folks know that but did it anyways.

It's also becoming more and more clear that we need a relatively neutral medium for free speech. Musk claimed to be building that on X but then did exactly the opposite. It's very hard to trust the algorithm isn't being manipulated in one way or another.

On BlueSky it's incredible how much value is already being produced by connecting intelligent and creative folks as compared to what Twitter became.

I was sort of hoping that this was about sustaining the web itself and preventing things like link rot and centralization of content in non-web privately owned platforms.

Still, this is important too.

Two small anecdotes about Ballmer from my internship there:

At an intern event an intern asked some question along the lines of "What's one quality of a competitor you wish Microsoft had?" They were clearly asking about cultural qualities but Ballmer's response was "I'd take Google's search engine because that thing prints money". Then he did this super exaggerated gesture and noise imitating a money machine printing money.

At the time that was a super disappointing answer to me because as an intern within the Bing division I really felt like there was this culture of "Bing is better and it's not our fault Google is beating us". To see Ballmer not use the question as an opportunity to talk about ways Microsoft could improve felt like seeing the source of that cultural problem.

The other was from an intern event with Phil Spencer. Phil Spencer at the time was the head of Xbox's Games Division, but not yet the head of Xbox. Spencer said he was in Ballmer's office one day and Ballmer asked him "Why don't you just make the games that sell?"

I'm talking about parking lots and roads.

If every destination needs parking for nearly its peak capacity then that creates substantial sprawl.

Similarly if an area gets a high volume of peak car traffic then over time it will tend to get more arterials and interstates connected to it.

As sprawl and road networks increase it becomes more difficult to get around without a car, incentivizing more cars, which requires more large parking lots / roads.

A horse carriage takes up similar space to a car, yes.

But if 80% of day-to-day trips are walked and 20% are horse-carriages that's much more efficient in terms of space than a world where 80% of trips are via car and 20% are walked.

That additional space manifests in more wider roads to reduce bottlenecks and more large parking lots to store cars at different destinations.

I've phrased this multiple different ways. The concept isn't complex: a car (or horse carriage!) is much bigger than a person walking so if we design a world that encourages more travel via car (or horse carriage!) that world will be less spatially efficient.

But, as I've pointed out repeatedly, in the past far more people walked to take trips than take a horse carriage.

"Highly walkable" means many people would choose to walk even if able to take a car.

To be walkable a city needs to be quick to walk between destinations and pleasant to do so.

Most US cities are too spaced out, require waiting at each block, or are not all that pleasant for walking.

Those add substantial crossing time and are quite costly.

None of the world's great walking cities heavily uses bridge crossings like that.

Those very "tarmaccy" US cities are what you get when you "stretch" cities to accommodate the scale of cars.

Other US cities are gradually moving towards that design by adding incrementally more parking and highways, but it's a slow process.

I have no opposition to cars, but designing cities where nearly all trips require a car is an extreme that most people in the US live in. If you own a car you use here and there that's absolutely fine, but if there's nowhere for someone to live where they can comfortably get by without a car that's a bad place to be.

And as I argued in my initial point when everyone needs to drive the congestion and distance between things substantially reduces the advantage of driving in the first place.

I am making the case that the advantages once enjoyed by cars have been substantially reduced, for day to day life where most people live, as most people need a car.

The irony of your argument is that very few people who want more car-light or car-free cities are "forcing" anything on anyone, but the inverse is absolutely not the case.

A tremendous amount of taxes are allocated only for highways or car-centric revenue. Federal and state politics prevents cities from putting that money elsewhere. Highways were plowed through US cities and are maintained there over the objection of city residents. States intervene to prevent cities from running bus priority lanes. Cars purchases are subsidized where bikes and transit passes are not. Federal road standards, which are applied in cities, are designed for cars and not pedestrians / bikes.

A prominent example is NYC being forced by NY State to cancel congestion pricing.

The list of ways car-centric decision making is forced on dense cities is very long. Very few people are trying to "ban cars" but are instead trying to let cities too dense for cars guide their fate.

My point is about the average space taken up on a trip.

In the far past some trips were taken via horse + carriage but tons more were taken via walking and transit. Our built world has scaled to account for that increase in space used on trips.

It matters because there are far fewer delivery of goods than people getting around for day to day trips.

If you look at the average amount of space taken up by a person traveling around if they walk, or take transit, for most trips then they'll on average take significantly less space than using a car.

People bought into cars early because they could get around quickly to more destinations, not because walking was uniquely awful.

In Philadelphia's paper in the early 1900s there was a daily column about "pleasure drive" routes and constant advertisements appealing to new drivers with destinations near the city.

That advantage of being able to "get out of the city" is still there, but it's further and further away. For day to day life the experience of walking / transit / biking in a pre-car US city or a modern US city is somewhat comparable in terms of time and enjoyment.

However US cities and suburbs, due to car-centric scale, allow more people to live on larger plots of land.

Salt Lake City has a huge volume of parking lots, highways, and many roads.

While individual streets may not have scaled up there were other parts of the built environment that did to accommodate cars.

Horse-drawn carriages were typically used for moving goods or for shared transit, not for most individuals in single horse drawn carriages going to all sorts of day-to-day trips.

The point is that the default mode of transportation requires vastly more space than it used to.