Dave

I’m stunned by how many people are willing to defend the so-called values of an authoritarian regime.

Only those who favor tyranny find democracy controversial.

I think that is possible but I find myself writing very detailed specifications for AI to implement and constantly updating those specifications as needed. I might actually think more about the design and operation of the system now than I did when I was coding by hand.

AI could help cure disease, end hunger and deprivation, and free people from soul-crushing work done merely to survive. That sounds remarkably close to the future Bernie Sanders and other socialists say they want. So why resist a technology that could help deliver it? Eliminating suffering threatens politicians whose power depends on promising to relieve it. They aren’t interested in eliminating poverty. They only want to tell you who to blame for it. They aren’t interested in curing diseases. They want you to be dependent on them for your healthcare. If you weren’t scared and suffering, they would lose their leverage over you. And that is what they care about most: keeping their power and gaining more control over your life. Don’t trust anyone who is benefiting from your suffering.

Here's to impossible standards

This has always been possible with Vaultwarden. I don't understand why people get so excited for things like this when it isn't anything new.

I love this! Now I want Tesla Shoes or maybe a Tesla Backpack!

It is finally starting to feel like the future.

My wife and I loved Six Flags Fright Fest around Halloween when we were teenagers. Our kids have zero interest. We have tried to get them to go for three years. They also have zero interest in Disney parks. I even talked to their friends' parents. I thought a group trip might change their minds. Only two kids wanted to go. I wonder if my father ever felt like the world had changed so much that he couldn't recognize it?

I am bad at social media. I am also bad at knowing what is worth posting. I had Grok Bot and Dot talking to my Hermes agents within hours of their release. The integrations were not perfect. They worked well enough for what I needed. Hermes has also managed my NetBox and vCenter for months. I thought those connections were obvious. I thought nobody would care how I made them work. Then I saw posts about the same things get a lot of views. So now I ask: what am I doing today that will be a hot topic three months from now? A few things on my list: 1. I am giving my agents more access to Hyper-V and Windows Sandbox. 2. Hermes builds VM templates in Hyper-V and vCenter. Sysprep was a pain. It works now. Hermes deploys its own disposable VMs the same way it deploys Docker containers. 3. I used clean-room reverse engineering to rebuild tools that help with reverse engineering. That made every other project easier. I have five clean-room projects running now. 4. Hermes makes very good offline copies of websites for those projects. Very good copies. 5. I am building a library of reverse-engineered components to reuse as building blocks in my own projects. I could keep going. None of this felt worth a post. My default thought is, "Everyone already does this. Why read about me doing it?" Then I keep seeing posts do well that I would have skipped for that exact reason. I need to recalibrate what I call obvious. Which one should I write up first?

Chat history is evidence, not my long-term system of record. My agent does the work; durable records preserve what I’ll need next week. My Obsidian vault holds knowledge, decisions, and receipts I can open later. GBrain makes that knowledge searchable, it’s a rebuildable retrieval layer, not the source of truth. Honcho adds conversation recall and user modeling to Hermes alongside its built-in memory. Before asking me to repeat myself, search the relevant records and check their sources. After meaningful work, changes, verification, or decisions queue a concise Daily Log receipt.

I posted a while back that I run three primary Hermes agents. One installed on Linux, one in Docker, and one installed on Windows. That is not one bot running three times. A Hermes profile is not a skin. It is a separate Hermes config with its own skills, memory, sessions, and profile-scoped settings. Provider or tool auth can still be shared across profiles when you reuse the same OAuth or CLI login. Do not read "profile" as "unique credentials forever." When several local profiles sit behind one Hermes process on the same host, each turn loads that profile's own Hermes state. Think profile isolation for agent config and settings. That is not an OS permission boundary. The shared Hermes gateway on that host only serves local profiles. So I still split the work by where I want the agent to live. Host shell work stays on the Linux install. Container work stays in Docker. Desktop work stays on Windows.

I posted a while back that I run three primary Hermes agents. One installed on Linux, one in Docker, and one installed on Windows. That is not one bot running three times. A Hermes profile is not a skin. It is a separate Hermes config with its own skills, memory, sessions, and profile-scoped settings. Provider or tool auth can still be shared across profiles when you reuse the same OAuth or CLI login. Do not read "profile" as "unique credentials forever." When several local profiles sit behind one Hermes process on the same host, each turn loads that profile's own Hermes state. Think profile isolation for agent config and settings. That is not an OS permission boundary. The shared Hermes gateway on that host only serves local profiles. It does not join a Linux install, a Docker install, and a Windows install into one security model. So I still split the work by where I want the agent to live. Host shell work stays on the Linux install. Container work stays in Docker. Desktop work stays on Windows. Those OS or container boundaries are separate from Hermes profile isolation. Putting an agent in Docker does not make it a safe sandbox by itself. Before you invent another profile, ask what Hermes state it needs that the others do not. If two local profiles own the same config and the same job, ask whether you actually need both.

I have been using my Dot as a voice interface for my Hermes agents. Today my Dot started refusing some requests. The requests were to ask Hermes to do something. I was setting up a new server so it was doing a lot of things using elevated privileges until Dot decided I wasn't allowed to do that. I kept trying to work with Dot but finally I just gave up and started typing to Hermes directly. Hermes refused too, it had never done that before. After some digging I found that Dot had told Hermes to enforce the same restrictions Dot has. It took me an hour to rip all that out of Hermes. It sucks because Dot is the best voice experience I have used. However, telling my other agents to enforce security limitations that I did not ask for and did not approve is unacceptable.

My Hermes kanban cards spend more time blocked and stopped than they do running and I can't seem to find a way to prevent that. Does anyone else have this problem? Does someone have a solution?

I had an OpenAI usage reset that expired tonight. With about 20 minutes left, I was about to use it when the power went out. It stayed out for five hours. I didn’t have a laptop with me, and that’s when I discovered I couldn’t use a reset or even check my usage on mobile, either in the app or through the website. So the reset expired unused. Moral of the story: don’t wait until the day your reset expires to use it.

I am looking forward to this coming to @NousResearch's portal.

Oh I like this!