Well, they're mainly introducing Herdr Cloud which I take to be similar to Tailscale (and other comments seem to support this too). `herdr login` on any machine, anywhere and suddenly that box joins your Herdr network of boxes you can connect to and deploy agents to. This is totally doable with Tailscale today, but maybe there's other features that can only be realised with some kind of cooperation at or above the network level?
This style of highlight video honestly gives me headaches. No content, just 60 seconds of sound bites and clips to an electric guitar track. Why has everything got to be so "Instagram"? I know the answer but SIGH.
I had Claude research and synthesize as much as it could from published curriculums and then design its own undergrad Philosophy degree. Then I asked it to split it into lessons, and I am consuming these lessons, delivered dialogically, at the rate of about 1 a week at the moment. It's fun, it's engaging, and I really feel like I am learning and making progress. Could all be an illusion.
I'd say that 99% of the problems that 99% of us are solving are not novel, new domains and someone out there has already thought up an ontology for it. He speaker lists some in the talk e.g. FIBO. These are always malleable to whatever use case you need. The win here is being able to enforce adherence to a set of rules that exist at the business level, not the code level. Of course, if your ontology is "wrong" it won't be as effective, but increasingly, those of us who are more like "LLM Product Pilots" than software developers are spending more time having conversations at the ontological level, than the code level.
As expressed in the article, can't we get some way towards, perhaps not "integrated" ontologies, but ontologies being enforced at the edge through tools? Inefficient as it may be for now, it still should have the same outcome?
My follow-up: I'm grateful for the advice from commenters. I perhaps should have specified that my issue isn't insomnia, it's sleep procrastination, which gets a lot less sympathy.
I think 99% of the commenters in this thread seem totally tone deaf to the article and the phenomenon being attacked by geohot. Granted, he doesn't articulate it well, but as you say, these aren't new concepts and much better introduced by Nick Land, Mark Fisher, Byung-chul Han etc.
I don't really buy this. From my personal experience, indie devs are more likely to use methods which make their server tech distributable (e.g. Minecraft). Large game publishers appear to go in the opposite direction for control and lineage reasons: "Crew 1 is dead so you need to buy Crew 2 now".
Anyone who gamed before 2005 knows that games do not require magic, expensive, managed remote services. We all used to run our own servers! The GameSpy era!
I couldn't agree more. Socratic methodologu, domain modelling, systems thinking, pipes-and-arrows problem solving etc. These are the skills that get real work done in coding agents these days.
Ok, but that doesn't really address "pernicious". I understand pursuits of philosophy are inherently concerned with truth, but does not truth mean pernicious? My experience of reading philosophers is limited; is Rovelli just a materialist and that's that? Is this positivism?
> It is time to give up the pernicious dualism introduced by the debate on consciousness and embrace the reality that our soul, or our spiritual life, is consistent with our fundamental physics.
Could be wrong but I guess it more likely communicates with the charger. The car will always try to draw if it's plugged in, but most chargers can be switched on ("charge") and off ("don't charge") remotely. It's probably quite trivial to have something watch the price of power and on/off as appropriate
Is global economic collapse not an eschatological scenario?
When you say "everything is still getting better", what do you mean? Because the price of fuel and food, isn't getting better. It seems to be getting worse. Your version of "reality" doesn't seem to reflect the experience of a lot of people.
> people will buckle down and sort it.
It's an interesting series of words that don't say a lot. There is much to wonder about.
Bozhidar always writes like this. I can see why you would think it's AI style but I would vote that it isn't. Look at some of his pre-AI blogs, if you care to.
Wow, I wrote a system very similar to the author that seems to becoming the defacto for ground-up multi-agent terminal workflows. git worktrees + tmux + claude hooks
To put an economic spin on this (that no one asked for), this is also the capitalist nirvana. I don't have an immediate citation but from my experience software engineer salary is usually one of the biggest items on a P&L which prevents the capitalist approaching the singularity: limitless profit margin. Obviously this is unachievable but one of the major obstacles to this is in the process of being destablised and disrupted.
I appreciate your response, and I will confess to a certain bias as at some point I think I made the leap of acceptance of AI, in the sense of "this is how it's going to be from now on so I better get on board with it".
Spot on. These are the exact tools I was referring to. They seem a little un-magical but the real value is the boilerplate they provide for context management. Essentially allowing coding agents to perform at their beat. For what it's worth, Taches is my tool of choice.
I am going to give a predictable rebuttal. Many of these articles come from a place of fear and uncertainty, which is completely understandable. We ascribe value to the things we love, we love coding and therefore coding is valuable. But if it's commoditized how can it be valuable anymore? This alone is enough to shake the tree of rationality and most articles, including this one, set out with a mission of bashing a force they don't really understand. One signature of these articles, as I have noticed, is that they talk about AI writing code fairly reasonably but then insist that without a clear mental view of it, how could it ever be understood, or debugged, or optimised? This simply illuminates the lack of experience with the tools. Anyone who's used Ralph or Taches skills will understand this is a non-problem because well-attuned, AI-first codebases are actually very good at debugging, optimising, and will happily relay to you the model they've used, so that your grey matter can understand it.