I know that feeling. Tomorrow is a new day.

I have been working through this. I found it helpful when someone described it as using real-world examples to teach category theory more than demonstrating the benefits of applying category theory to specific problems.

Uh, could you write some paragraphs? I think I need less metaphor to parse this on a weekend.

Are corn pops a tortilla?

A Local-First Task Framework Industrial research lab working on digital tools for creativity and productivity.

I've been working with @pvh.ca on a local-first task framework at @inkandswitch.com. It distributes computation across users' computers, keeps working while you're offline, and syncs results back to everyone when you reconnect. First lab note is up. www.inkandswitch.com/patchwork/no...

That said, hydroelectric dams have the benefit of providing inter-seasonal energy storage by (literal) construction, and a regional energy portfolio for a circumpolar nation obviously needs to take that into consideration.

What's more, the "levelized cost of energy" (LCOE) which represents the all-in cost of producing energy for our new Site C dam ranges from $40 to $125/MWh (CAD), comparable with grid-scale solar in the region at $75/MWh.

British Columbia's history with hydro-electric is chequered. Construction of dams floods arable farmland and displaces communities. That was true in 1972 with our first dams and is true today with the new "Site C" dam as well.

It's a relief to live in a region whose primary energy supplies are provided by hydroelectric dams. Relying on renewable energy doesn't just reduce the rate of climate change, it can protect against energy market supply shocks.

@monkchips.com dang, were you at AtmosphereConf and I missed you?

Local-First Conf 2026 is back, and I want to hear what everyone's been doing this year. What have you built? What have you learned? How about interesting challenges we need to overcome?

B.C. to end time changes, adopt year-round daylight time | CBC News March 8 will be the last time British Columbians have to change their clocks, B.C. Premier David Eby announced Monday.

Yay! An end to picking up my kids from school in the pitch black for months on end. www.cbc.ca/news/canada/...

Also, oh no! This is going to be hell for scheduling international calls.

Local-first Conf is back, and this year Ink & Switch will be hosting a one day "Lab Day" adjacent to the event. See you there!

I think it’s fundamentally a communication challenge. I think we have done something exciting but if it isn’t presented thoughtfully it could lose its opportunity to make an impact. To make a film-making analogy, I have seen the “dailies”, but without editing it’s all just raw footage.

Damn, upstaged by my kids.

I don't even know how to vague-post about it. I'm so incredibly excited about the work we're doing right now at Ink & Switch and it's taking everything I have in me to avoid sharing it before it's ready.

BitTorrent doesn’t have this problem because enough seed peers have IPv4 addresses.

Ironically, users on a shared WiFi network at a cafe probably experience success rates below 50%. This is for a variety of historical reasons I will summarize as “mischief abounds”.

There isn’t a straightforward way of measuring how common hole punching failures occur. Anecdotally I have heard numbers between 3% and 30% but the true answer is “it depends”. For users on their own home infrastructure, hole punching works pretty well!

I believe you misunderstand me. I am deeply familiar with the many techniques used for NAT traversal. There are situations (such as so-called hairpin traversal) that these approaches do not work. This is the situation I described at the beginning of the thread which requires making hard choices.

Parasitism is an excellent technique here, though of limited scale potential. If traffic outgrows the host’s inattention it will become untenable.

How does keet work when two nodes can’t connect directly?

Keet is wonderful but is still subject to the challenge I describe. I believe keet uses voluntary relay peers and provides a few as part of their service. A very pragmatic and effective trade-off. (If any keet folk want to correct me on the details, please do.)

Yeah, this is a good distinction. The peer-to-peer vs federated vs centralized debate is as old as time. I *do* believe there are pretty strong economic scaling effects that encourage centralization as the long-run effect of commoditzation but other effects (data sovereignty?) may counteract that.

All these approaches have tradeoffs: reliabilty, performance, cost, privacy. I'm not aware of any uncompromised way of solving this problem, though, and in the last decade I've been watching the space no new solutions have emerged. I agree though that @iroh.computer is making good compromises.

All peer-to-peer systems have to compromise one way or another on this: they can decide that direct P2P is essential and simply fail, they can provide centralized fallback alternatives, or they can try to find/convince/incentivize another peer to route traffic on your behalf.

PushPin: Towards Production-Quality Peer-to-Peer Collaboration Taking peer-to-peer beyond research prototypes, and working towards commercial-grade P2P collaboration software.

Just to expand on this, we wrote a bit about this in the Pushpin essay here: www.inkandswitch.com/pushpin/#nat.... The hard part is that you can be "snookered", meaning that the router simply won't allow two local peers to communicate. In that case you simply need a third party to proxy for you.

Lately, I've been telling people that Patchwork is less like a house and more like a way of standing. I'm not sure it entirely makes sense to me, either.

10m documentary film on the local-first software movement!

www.youtube.com/watch?v=10d8...

The Navigation API is newly baseline! The web now has sensible, low-level routing for navigations.