Bookmarks - llm, programming, data, web-archives
These are some things I’ve wandered across on the web this week.
## 🔖 Tella
In challenging environments, with limited or no internet connectivity or in the face of repression, Tella makes it easier and safer to collect, protect and hide sensitive data.
activism data security storage tools
## 🔖
AI_CONTEXT
AI_CONTEXT is a lightweight, vendor-neutral pattern for preserving the engineering context that normally gets trapped inside one AI chat, one coding agent, or one context window.
The idea is simple:
Put the important project memory in the repository, then let every AI agent read and maintain the same source of truth.
Instead of depending on a particular model’s conversation history, AI_CONTEXT stores concise, durable project knowledge alongside the code:
documentation git llm
## 🔖
micro-practices as portals to your art
today, I want to share a concept that’s helped me tremendously in this shift. I call it micro-practices — 3 minutes, 7 minutes, 12 minutes — as tiny, almost negligible containers of time that allows me a way into my creative practices while sideskirting the resistance coming from my brain that might say… no no no, but we’re too busy/tired/distracted/in the wrong headspace for this right now…
art notebooks practice
## 🔖
Library technology in the age of generative AI
To be a technology worker in a library, especially a publicly-funded one, is to be subject to somewhat different pressures than those who are employed by traditional technology companies. The financial pressure is not worth lingering on – we know that we are paid less than those who work for major corporations, but accept it as an inevitable result of choosing to sell our labor to an organization that is not (at least officially) oriented toward profit above all else. The mission of a library is instead – in theory – the creation, preservation, and transmission of knowledge, and those of us who appreciate that mission may find ourselves drawn to this work
knowledge libraries llm programming
## 🔖 aider: AI pair
programming in your terminal
An open source coding assistant, similar to Claude Code but lets you select what models to use.
llm opensource programming
## 🔖 In
Conversation with Boris Cherny, Creator of Claude Code
What does it take to help shape one of the most consequential revolutions in tech history? Join Boris Cherny, creator and head of Claude Code at Anthropic, for a wide-ranging conversation about the past, present, and future of artificial intelligence. From the challenges of building models like Claude to the possibilities ahead, Cherny will offer an insider’s perspective on a technology rapidly transforming our world.
anthropic interviews llm
## 🔖
GreenPT
Every request’s energy is measured directly from GPU hardware, not estimated. Every carbon number uses the live carbon intensity of the grid running your workload. No annual averages. No projections. We report the physical carbon intensity of the grid our servers actually run on, not the carbon we could claim by buying renewable energy certificates elsewhere. The atmosphere doesn’t distinguish between a certificate and a smokestack. Neither do we.
llm sustainability
## 🔖 Can gzip
be a language model?
This led to the natural question: can gzip do language modeling?1 No neural network, no learned parameters, nothing. Just the compressor that ships with your operating system. You prime it with a corpus, give it a normal text prompt, and it continues that prompt by searching for the byte sequences that compress best.
compression llm zip
## 🔖
Stalker
Stalker (‹See TfM›Russian: Сталкер, IPA: [ˈstaɫkʲɪr]) is a 1979 Soviet science fantasy film directed by Andrei Tarkovsky with a screenplay written by Arkady and Boris Strugatsky, loosely based on their 1972 novel Roadside Picnic. The film tells the story of an expedition led by a figure known as the “Stalker” (Alexander Kaidanovsky), who guides his two clients—a melancholic writer (Anatoly Solonitsyn) and a professor (Nikolai Grinko)—through a hazardous wasteland to a mysterious restricted site known simply as the “Zone”, where supposedly exists a room which grants a person’s innermost desires. The film combines elements of science fiction and fantasy with dramatic, philosophical, and psychological themes.
movie scifi
## 🔖 searxng
SearXNG is a free internet metasearch engine which aggregates results from various search services and databases. Users are neither tracked nor profiled.
python search
## 🔖
AI Has No Wisdom and Neither Will You
For one, AI is not trained on what it means for code to be maintainable. For instance, any reinforcement learning done needs a reward signal that can be measured immediately, not in months or years. The AI learns rules from rulebooks meant for beginners. The AI notices patterns from code in the wild and let’s be honest, most code in the wild is pretty bad. There is no fitness function you can define for maintainable code, at least not one that we can discern, otherwise it would’ve been baked into our linters.
ai maintenance programming
## 🔖
An update on Wayback Machine access
People do like talking about web archives…
internet-archive web-archives
## 🔖
Spain Orders Blocks On Archive.today And Its Mirrors
Spain’s Second Section of the Intellectual Property Commission, a part of the country’s Ministry of Culture, has ordered the blocking of several domains of the Archive.today service, a web archiving service.
It means that most Spanish internet users who try to access the site are redirected to a government page telling them that they are trying to access an “illegal” website.
censorship dns web-archives
## 🔖
What happened to the Snowden archive
The last document from the Snowden archive was published on 29 May 2019. The Guardian stopped publishing documents in February 2014, Der Spiegel in January 2015, and The New York Times and ProPublica in August 2015. After that, only The Intercept was still publishing documents, with only a few exceptions, until it closed its archive in March 2019. Eleven weeks later, on 29 May 2019, it released what would become the final batch of documents from the archive. Since then, no news outlet, journalist, or institution anywhere has published a single document from the Snowden archive.
archives encryption snowden surveillance
## 🔖
Core Preservation Processes (CPPs)
A Core Preservation Process is a specific action that every Trustworthy Digital Archive should undertake adequately - either directly or through its associated parties or services, in order to fulfill its digital preservation missions as evidenced in its preservation policy.
data preservation standards
## 🔖
Beyond Folk Marxism: Mind, Metaphysics and Spooky Materialism
Why, and how, are any of us capable of self-reflective thought? Of experiencing emotion, ratiocination, preference, cold, whim, melancholy, joy? Anything at all?
marxism philosophy
## 🔖
MusicBrainz: Guides / Userscripts
Userscripts run in a users web browser and make on-the-fly local changes to specific web pages. In MusicBrainz they are generally used to change the display of pages, facilitating editing.
metadata music