According to wikipedia the reversals can take a while
>Some sources estimate the most recent four reversals took on average 7,000 years to occur. Clement (2004) suggests that this duration is dependent on latitude, with shorter durations at low latitudes and longer durations at mid and high latitudes. Others estimate the duration of full reversals to vary from between 2,000 to 12,000 years.
The big problem with swift is “expression has ambiguous type” and “expression took too long to type check”. Those don’t trade in code aesthetics but are problems I’ve never had in another language.
Ha! But that's not semantically meaningful Swift code in any normal context, nor is it idiomatic. `self` is equivalent to `this` in C++, and is never normally null.
You use this construct for unwrapping nullable fields, for example something like this:
guard let httpResult else { return }
Note that you don't need to assign the value to itself in modern Swift. This line takes an optional (httpResult?) and returns early if null. If not, you can use it with strong guarantees that it's not nullable, so no need for ? or ! to unwrap it later in the scope.
> But that's not semantically meaningful Swift code in any normal context, nor is it idiomatic. `self` is equivalent to `this` in C++, and is never normally null.
It is, when `self` is captured weakly in a closure, and that closure is outliving the instance.
Roll paper isn't really an acceptable compromise these days though. In addition to being curled, if you need to print more than a couple of pages you're left with a disorganized mess.
For the first ever open-sourced printer I think it's a great starting point. Being an open-sourced project, community-made solutions for other feeders are sure to come if this succeeds
With open-weights models reaching a level where they can sufficiently be used for agentic coding, the price can be directly compared to the price of GPU rentals: https://vast.ai/ has an H100 at $1.65/hr. which can support ~40 concurrent sessions at 40 tok/s. Depending on your agentic workload, you can stretch that any way you like, but let's say it might support 10 active developers at a speed comparable to Claude Code with Sonnet 4 (which I've read is 90 tok/s) who aren't going crazy with sub-agents.
Let's scale that up: $1.65/hr. is ~$1188/mo. (assuming you're renting the GPU 24/7 which you can speculate to not do, which is probably fine as long as there's not scarcity), and divided by ten is ~$119/mo. per user for 10 users.
Add the service layer on top in order to make this a convenient software service, I think $100/mo. is a bargain for unhindered (as far as it goes) access to a high-quality (as far as they come) agentic coding framework.
Feel free to correct my napkin math, it's done very quickly.
Could this be associated to a supposed recent State Department approval?
“I just approved a program to deploy small modular nuclear reactors built in the United States to an allied country to help with their sort of energy infrastructure.”
“Which allied country would that be?”
“I can't tell you. It's not public yet.”
From Interesting Times with Ross Douthat: The DOGE Alum Asking if Foreign Aid Is America’s Problem, Jul 31, 2025
slightly off-topic (but tangentially mentioned in the article). I google'd "Chicxulub" and saw an easter-egg I hadn't previously known about :)
Great article. I recently finished my second reading of TMM and how/to what extent our current era of generative code affects the ideas of the book was top of mind.
Tangential: I have similar thoughts about Jira. The ticket-fication of organizational goals. At least it would be useful to pen any negative repercussions of this
I know it’s trite to just say “you aren’t holding it right!” when it comes to JIRA, but I do think there’s a sensible tool underneath layers of self-inflicted pain. (Self = Atlassian and its users)
User stories, when they’re actually a real problem a real user would need solved, are fine. If you start there, and figuring out how to solve that problem is open to anyone on the team, and you keep the complexity to a minimum (aka, just todo/inprogress/done statuses, and you only try to solve the problem in the story) it’s totally cromulent…
…for start ups who need to ship yesterday and have money to burn.
So IMO not like, the best way to do work, but to do something as fast as possible with people motivated by the problems you’re solving, I like it.
> User stories, when they’re actually a real problem a real user would need solved, are fine.
Some times. Other times they are detrimental, you need an algebra of composable operations up-front and any abstraction you put on the process of designing those will make people design a broken UX.
User stories are useful mostly for "flux-based" applications where the user has little freedom.
> User stories are useful mostly for "flux-based" applications where the user has little freedom.
I'd say basically only useful for those sorts of applications! If you're going by user stories, there should only really be 1 way of solving any 1 issue, and users should get rail-roaded into it. There's always a solution to problems you've closed user stories for, but that's it. Anything outside those is unconsidered and might not even have a "hackable" solution since you're building everything up organically rather than as a designed system.
Great for start ups (saved time and money building the impactful flows, your product only needs to do a few things) but awful for enterprises (users have no freedom to warp your product to their needs, your product needs some predictable structure/rules they can build on... those composable operations you're mentioning.)
Ha I only added this because my wife doesn't have gmail. I think this is a problem I don't particularly care about for now - I think the parent commenter is correct in saying that the account wall is in the wrong place, so if I move the wall to where the value is (speak to a jeweler) I can defer email verification for a while.
>Some sources estimate the most recent four reversals took on average 7,000 years to occur. Clement (2004) suggests that this duration is dependent on latitude, with shorter durations at low latitudes and longer durations at mid and high latitudes. Others estimate the duration of full reversals to vary from between 2,000 to 12,000 years.
- https://en.wikipedia.org/wiki/Geomagnetic_reversal
So maybe that's enough time for the biomass to adapt?