Author here. It’s ironic because this post was clearly geared towards non-hackers. But now that we’re here.. the other analogy I considered presenting was:
harness = chassis,
model = engine,
fuel = tokens,
agent = car
I’m curious what y’all might think and whether that analogy carries more explanatory power
The first analogy that comes to mind, growing out of "harness", is more like harness = harness, model = horse (rather than harness as in climbing harness).
I guess you could say that tokens = hay, and agent = horse and cart, from there? Not sure how useful the hay part is but you could observe from the second that there are many different things you could harness a horse to (also a plough, or a coach, or just a saddle) based on your goal.
Although obviously unserious, I would argue a baby harness is mostly the opposite of what an agent harness does. A baby harness constrains and contains the baby, while an agent harness controls but also empowers the agent.
I'm a climber so I'm biased but I really liked your climbing harness example because of the configuration you're able to easily make to the harness.
Saying the harness is like a car's chassis doesn't work as well for me because the chassis isn't as configurable as a climbing harness for as little work.
Getting deeper into the climbing analogy you can even swap out the harnesses themselves for wildly different climbs. Like using Claude Code with a bunch of agents for medical software (climbing K2 where that extra padding comes in super handy) and pi.dev with a local model for a respectable web project (sport route where you'll be back in a few hours and it's safe to be a little more exposed).
I'm glad your article made HN, and thank you for pi!
I actually use the computer as a metaphor. The LLM is the CPU. The harness is the motherboard which controls communication between the CPU and other components like memory, hardrive, and inputs. In other words, how the LLM interacts with the outside world, and outside world with the LLM.
I've been building coding harnesses since 2021 and believed in their value for a long time. Harnesses matter a lot, look at what claude did for Anthropic.
Author here. I think our website could be much clearer - but Pi is fundamentally easier to mold than other harnesses. It’s not magic but it strikes the balance well of letting you shape it extensively without letting you break it.
A harness is the bottom layer of a pie that gets fed into the model. In my project, I count 7 more layers on top of it https://replicated.live/blog/wiki
They all affect consistency, coherence, token efficiency. Probably we need some broader term. Like "information architecture", "knowledge architecture"? It's not just shoveling Markdown to nvidias, after all.
I don’t think it ever says wholly owned. My understanding is that the three founders have donated their shares to the Foundation, but they didn’t own 100% of the company. Likely >50% but not 100%
To my reading it’s only part of their shares and probably results in the foundation being the biggest single shareholder while still only having a minority stake.
Feels to me after reading the article they earnestly try to do their best to offer privacy enhancing alternatives and push back often. What percent of these requests do Meta, Google or Microsoft fight? Ratios like that matter
Proton is extremely transparent and said:
If you are breaking Swiss law, ProtonMail can be legally compelled to log your IP address as part of a Swiss criminal investigation. This obligation however does not extend to ProtonVPN (see VPN privacy policy here). Additional details can be found in our transparency report.
> What percent of these requests do Meta, Google or Microsoft fight? Ratios like that matter
What others are doing doesn't matter, that's whataboutism. Yes there's many shittier services, and Proton is much better than them.
What matters is if you can trust Proton to be private, and the answer is... mostly.
Yes I like Proton and I use Proton as my daily email driver, because I don't expect privacy from governments, I just don't want Google tracking.
But a lot of people see the "no logs" thing and think that there's never any logs, which is not true, they add them on request, and they've done it based on foreign government requests, for questionable searches, as I've linked above.
If you want privacy in your hands, use Tor when accessing Proton and pay in crypto obviously.
Those are techniques needed for privacy because they can access that data and you can't trust them to safeguard any data they can access because they legally can't.
It's not their fault, it's just the system, but you must expect it.
harness = chassis, model = engine, fuel = tokens, agent = car
I’m curious what y’all might think and whether that analogy carries more explanatory power