rlupi

Single point of failure, single vendor.

All the CEOs are dreaming of getting rid of workers, but I wonder how many of them think clearly about what they are getting. Depending on a single, or few vendors for full-AI employees is the concrete equivalent of unionizing their workforce. If the vendor decides on a price hike (think about the recent history of VMWare), or their full-AI employee gets dumber or lazier (Claude Code taught us a good lesson here just lately), then your whole AI-powered company is at risk. Our current human-based organizations are much more resilient to single point of failures.

You can use powershell on linux: https://learn.microsoft.com/en-us/powershell/scripting/insta...

I just re-read "The Sorrows of Young Werther" by von Goethe. In the book, the protagonist commits suicide.

It appears that now tech firms cannot even host very significant old books without age checks, which is absurd.

I created a tool to analyze my diary, create retrospectives, and allow chatting with LLMs based on its content.

For demo purposes, I fed "The Sorrow of Young Werther" by Goethe to it. The results are both amusing and show how powerful it can be.

LLM tools do exactly that. That's why most online LLMs (openai, gemini) have access to sandboxed python for calculations.

This gave me a nice idea.

It would be nice to build a local LLM + wikipedia tool, that uses the LLM to assemble a general answer and then search wikipedia (via full-text search or rag) for grounding facts. It could help with hallucinations of small models a lot.

Thanks. I currently use qwen3 (32b or 30b a3b moe) locally, and o4-mini or gemini-2.5-flash online. It costs about 5 USD to run through a year of notes online, and several hours to do it offline. qwen3 is lower quality than online models, but good enough especially for high-level summaries (like update the yearly summary after I tweak prompts).

Sometimes I write about it on https://rlupi.com/

The way I use: - keep track of how I spent my time and my energy (that's what the go code is for) - daily journal (my template includes: journal and events [see above]; my thoughts, ideas, tasks; others' ideas (TIL): articles, books, links). The "my thoughts, ideas, tasks" section include reviewing metacognition exercises I learned at a burnout retreat. - weekly retrospectives (what went well/wrong? what was unexpected/surprising? what did I learn? what will I change/experiment with to improve my life? what I am grateful for? who did I talk to? who didn't I talk to but should?) - monthly free-form summaries - yearly GTD-style horizon of focus exercises

In aww, I feed the whole bunch, and aww-automated recursive retrospectives to build higher level ones. I don't have to do it that often, only when a week/month/year ends (or more often, when I tweak prompts).

For use, I use LM-studio with the go program as a local MCP server to discuss, brainstorm ideas, etc.

I don't write a lot about deep themes in my daily journals, but aww seems to be able to extract quite a few high level trends. For example, here is the result of all that and asking:

> Read the yearly retrospective in the following pages and summarize my personal journey: [[r2021]], [[r2022]], [[r2023]], [[r2024]], [[r2025]].

From r2021 to r2025, your journey reflects profound growth across psychological resilience, identity development, technological curiosity, spiritual inquiry, and balancing ambition with self-care. Here’s a synthesis:

ADHD Management & Productivity Evolution Early Years: Adopted Pomodoro, time blocking, Key Insights: Embraced external scaffolding (tools, routines) to counter executive-function challenges. Shifted from reactive productivity ("working under stress") to sustainable rhythms rooted in Stoicism/GTD principles. Recognized ADHD traits as strengths requiring tailored strategies—structured routines, environmental adaptability. Health & Wellness:

Physical Health: Overcame initial health anxieties (e.g. diagnoses), adopted Mounjaro/Ozempic for weight management, prioritized sleep hygiene and exercise regimes. Mental Health: Transitioned from medication experimentation to schema therapy, biofeedback tools, meditation/yoga practices fostering emotional equilibrium. Spiritual Shift: Evolved from productivity-driven focus to valuing presence, connection, and spiritual inquiry via Kriya Yoga/Samadhi practices. Career & Identity Development Early Career: Focused on technical mastery/self-hosted infrastructure projects blending tech curiosity with professional growth. Mid-Career Shift: Transitioned toward work-life balance—part-time roles at Google, advocating neurodiversity awareness—to protect personal time. Recent Focus: Explored AI safety, public writing (SLO/OKR frameworks), and creative outputs balancing technical depth with accessibility.

Identity Evolution:

Shapeshifted from solo researcher to community-oriented thinker embracing interdependence. Redefined self-worth internally rather than externally—from promotions to "I am enough" mantras. Cultivated polymath identity balancing intellectual/artistic/social/physical growth, embracing yoga as spiritual-tech fusion. Social Connections & Community Struggled early-on with isolation and transactional relationships post-breakup. Focused on meaningful connections through family ties (nonna), camper-life adventures, Zurich localism experiments. Recognized regrets around neglecting emotional intimacy—embraced "top five regrets" awareness guiding social outreach goals. Key Practices: Host gatherings, join collaborative groups, practice emotional vulnerability through journaling and conversations.

Technological & Financial Pathways Infrastructure: Built self-hosted ecosystems emphasizing autonomy (BalenaCloud, Obsidian systems). AI Projects: Explored edge computing/AIOps while maintaining privacy boundaries. Financial Goals: FIRE vision formalized by r2025 targeting autonomy by 2028. Balanced frugality with strategic investments, automated financial tracking. Philosophical Shift: Framed technology as extension of self requiring vigilance for autonomy—moving toward decentralized/self-sustained systems.

Themes Across Years Growth Mindset: Consistently reframed challenges as opportunities—from burnout to retreats, perfectionism to iterative publishing. Balance Between Kama/Moksha: Pursued pleasure (camping adventures) alongside spiritual liberation practices. Inner Peace as Cornerstone: Transitioned from achievement-driven goals ("doer") toward witnessing consciousness via meditation/philosophy. Systems Thinking: Applied habit-stacking, metadata tracking, automation to sustain growth without burnout. Future Trajectory Your trajectory suggests continued focus on:

Social/Emotional Growth: Expanding community ties via language learning (German/Swiss citizenship), local Zurich connections. Spiritual Synthesis: Blending yoga/Advaita Vedānta with technological mastery. Financial Autonomy: Refining FIRE plans ahead of housing deadlines by 2028. Health Balance: Maintaining physical-mind integration through breathwork, exercise rituals. Your journey illustrates resilience against cyclical challenges—addiction loops, productivity peaks—and evolving toward holistic fulfillment defined by inner peace rather than external milestones.

Very thoughtful comment, let me try to capture it more clearly.

Zuckerberg recently said that those not using AI glasses will be cognitively disadvantaged in the future.

Imagine an AI-native from the future and one of the fancy espresso coffee machines that we have today. They will be able to know right away how to operate them from their AI assistants, but they won't be able to figure out how they work on their own.

That's the future that Zuckerberg wants. Naturally, fancy IT offices will likely be gone. The AI-native would have bought the coffee machine for nostalgia effect for a large sum, trying to combat existential dread and feeling of failure which are fueled by their behavior being even more directly coerced into consumption.

Physical/mental health and personal journaling?

I just finished some changes to my own little project that provides MCP access to my journal stored in Obsidian, plus a few CLI tools for time tracking, and today I added recursive yearly/monthly/weekly/daily automatic retrospectives. It can be tweaked for other purposes (e.g. project tracking) tweaking the templates.

https://github.com/robertolupi/augmented-awareness

In Switzerland, I recommend enter.ch. It has an extensive collection of consumer electronics, computing devices, and proper computers.

It reminds me of the computing museum near Google main campus in Mountain View, California.

We're more malleable than AI, and we can't delete our memories or context.

I wonder if this is an effect of users just gravitating toward the same writing style and topics that push the context toward the same semantic universe. In a sense, the user acts somewhat like the chatbot extended memory through an holographic principle, encoding meaning on the boundary that connects the two.

https://chatgpt.com/canvas/shared/68184b61fa0081919c0c4d226e...

Why say Western, when they mean US english? US tech & culture (Hollywood, Netflix, etc.) has an habit of bulldozing over non-English Western culture too.

Centrally. If your system have K finite modes of behavior (degrees of freedom), then you can compress it as some combination of these effects.

Due to Donoh-Tanner Phase transition theorem, you can almost-surely reconstruct a lower dimensional (K-dimensional) manifold immersed in a N-dimensional space with O(k log N) points. For many real world systems, K << N. Now, K is the degree of freedom of your system, N is the size of your sample, or the inverse of the frequency resolution you need to capture anomalies (that's your sampling rate if you were sampling metrics at regular intervals, but here we are not). So you can capture random projections of your system, compare the results to the predictions of a pre-computed compression model of your system and only ship the changes. Low-dimensional projection maintains correlations (and introduces spurious ones), which can be used already in compressed forms for some central anomaly detection (e.g. how many replicas are affected by the same traffic).

In TOML you can use triple-quoted strings to avoid double escaping backslashes.

There is one example right at the bottom of https://toml.io/en/

Do you mean an encyclopedia? I bet there are tons of physical encyclopedias on sale on Ebay.

Almost 8 years ago, when I was working as a Monitoring SRE at Google, I wrote a proposal to use compressed sensing to reduce storage and transmission costs from linear to logarithmic. (The proposal is also available publicly, as a defensive publication, after lawyers complicated it beyond recognition https://www.tdcommons.org/dpubs_series/954/)

I believe it should be possible now, with AI, to train online tiny models of how systems behave in production and then ship those those models to the edge to use to compress wide-event and metrics data. Capturing higher-level behavior can also be very powerful for anomaly and outlier detection.

For systems that can afford the compute cost (I/O or network bound), this approach may be useful.

This approach should work particularly well for mobile observability.

Isn't this the closest thing to a portable antikitera?

I have sold my US equities and assets before Trump announcement of tariffs.

I personally find it a lot more risky to keep my wealth in US equities or USD denominated assets. Even all the decisions are undone, the fact that these level of policy swings could happened make US investments a lot riskier than they were even a couple of months ago.

The behavior of the US executive branch shows that they don't care as before about the rule of law.

That's quite interesting. Do you have any resource to recommend for OpenVMS or Mach architecture overview, or in general OS architecture tradeoffs and designs?

> if the people paying the salaries would rather make 1,000 bets on the Next Big Thing, and if 999 of them fail but one succeeds that is far better than 100 who fail and 900 which become medium-sized, profitable companies, then overproduction is what you get.

I think we are getting this equation wrong as a society.

We need figure out how to spin out the 900 that become medium-sized profitable companies. In a sense, companies like bending spoons are doing just that. We haven't found yet how to make 1000 bets, have one be the Next Big Thing, and profit from the 900 that are still profitable enterprises in their own but fail in the overproduction environment that we think we need for the Next Big Thing.

> Did anyone anticipate gross incompetence being the way "AI" harms people?

There is a long trend of labeling AI as "Artificial Stupidity", so yes: https://en.wikipedia.org/wiki/Artificial_stupidity

I just experimented with it. I couldn't get ChatGPT 4.5 to generate what I wanted, probably due to censorship or rate limits (ChatGPT Pro subscription); but Gemini 2.5 Pro Experimental did the trick.

This is the end result: https://medium.com/@robertolupi/how-beelzebub-was-born-92352...

If you like it, please upvote on HN: https://news.ycombinator.com/item?id=43599273

This is the conversation: https://g.co/gemini/share/21bbaae2a26e

I switch mid-prompt from Italian to English. It is interesting to see that Gemini reasoned in English even when writing back the initial answers in Italian.

Basically, I gave a somewhat exhaustive but generic track then asked it to give enough information to create a theatrical piece. I provided some stylistic references. A piece suitable to create a script for a tragedy is less abstract than the average output of a LLM, and forces it to create characters.

Then I asked to rewrite it as a short story a couple of time, asking to make adjustments (add more details and action, less abstract ideas) until I was satisfied.

I think I'd consider using NotebookLM or RAG-based, if I wanted something bigger and coherent.

A short sci-fi story about AI and fascism, inspired by recent events.

Note that Paul is wrong when he says that "take care of the world" wasn't part of the time tested answer.

Ancient populations like the Romans had the concept of numens, deus loci, and gods of nature that were responsible for the world, were venerated and people who devoted them to these deities did their part to help the world. Being a good host (i.e. the rapport with the Other) was also always a key duty, so much that it is Zeus/Jupiter who presided to it.

It was always part of the farmer's job to take care of nature and their fields. It was part of the nobility jobs to develop their territories. It is only in modern times, with mechanized agriculture and nation states, that these personal duties got lost. Also, if we widen our attention to include aboriginal people, taking care of the world is quite central to their world view.

They are great. I have actually built my own pomodoro timer using a M5Stack Core2, maybe I should publish the code. It's a lot simpler than this one.

Update: here it is https://github.com/robertolupi/augmented-awareness/tree/main...

I took it from Byun-Chul Han's "What is Power?". It's quite a heavy philosophy book, but worth reading if you are into this stuff.

Writing that answer in this thread, made me think that I should publish too. So, I did: https://rlupi.com/speak-up-publish-seize-your-power

It's the same experience that people on the other politics spectrum endured in the previous cycle. There is not difference, except from the side that is suffering now.

People were fired for speaking against woke culture, now they are fired because they speak for it. And woke culture is just one controversial thing, it can be anything. Religion, politics, any topic that fragment a group can be seized to exercise power over it.

The way to change it is to identify the common experience across differences, being fired for expressing one's ideas, and to fight against that... fight for the right of free speech for both for those you agree and those you disagree with, and you have to accept others when they express the opposite view. Speech should be free. Power is dialectic: contrary to violence, it works through meaning and eloquence.

There is a lot of truth in this article.

I work in a large FAANG company. For more than 10 years, I made the error of not publishing my personal work and ideas. I didn't like the rules about personal works, publishing opensource or social media communication... you are subject to gate-keeping by your employer if you are an employee. That's quite different from my previous experience in smaller companies in Europe. I found it very off putting: you had to give up ownership. I thought it was demeaning that even what I create in my own private time would be owned and attributed to someone else.

My choice of not publishing anything however was a very bad one. I didn't understand what power is: the continuity of the Self into the Other, where you find your-Self at ease. If you don't assert yourself, you give up not just your power but your very identity.

The truth is that power is dialectic. If you don't speak up, you have no power. Both in what you create, and in the relationship of and with power that you live, you are constantly redefining the boundaries of what you are and what you are not. If you remain silent, you loose.

> In my experience as a software engineer rounding out 6 years now, what I have observed in myself and others is that the work pressure that leads to burnout in SWEs is largely self-imposed.

This is how control works in our society. It's true for all people.

The philosopher Byung-Chul Han explains it well in "The Burnout Society".

Yes, and the dataview plugin to visualize docs in a table, list or calendar (e.g. organized by tags, to identify their types).

I am not aware of a editable table view, but you can surely have a index per document type that contains a dataview table and go to individual docs for editing.

For me, the main thing that Obsidian lacks is formula support or more advanced reporting. I can so something with dataviewjs, or going to javascript (customJS plugin) but it gets messy quickly.

You can also get nice visualization with a little work on templates and plugins: for example, I use Obsidian for weekly personal retrospective of my diary and I have a wordcloud graph in my weekly template.