Merge branch 'main' of github.com:davidgasquez/handbook

Artificial Intelligence Models.md

···
 - Use a checklist. Plan, iterate, and refine.
 - [Fail early and often. Agents are pretty good at getting oriented, and failure is cheap. Throw out your problem, and let it work at it. If it fails, start over, adding the tidbits of information it couldn’t figure out on its own](https://sketch.dev/blog/seven-prompting-habits).
 - Go concurrent. Many agents can work in parallel on different tasks or even the same task with different approaches (you can then choose the best one).
-- [LLMs actively reward existing top tier software engineering practices](https://simonwillison.net/2025/Oct/7/vibe-engineering/):
+- Applying software engineering best practices is key. [LLMs actively reward existing top tier software engineering practices](https://simonwillison.net/2025/Oct/7/vibe-engineering/):
   - Automated testing.
   - Planning in advance.
   - Comprehensive documentation.
···
 - Tool overload confuses models. Just because you can connect every tool doesn't mean you should. Each tool description consumes context window space and can confuse the model about which tool to use when.
 - Unix philosophy beats vertical integration. The most powerful coding agents follow Unix principles, simple, composable tools that do one thing well.
 - The way you wrap, feed, and observe a powerful model often matters more than fancy prompt tricks or extra bells and whistles on the model itself.
+- Decomposing your agent into discrete steps improves reliability and efficiency. If there is a bit of work that happens often and is easier to _eval_ than your entire task, that's a good thing to break out and optimize.
 
 ## Use Cases
 

Blogging.md

···
   - The title can also convey your writing style and tone.
   - Be cautious with overly clever or punny titles if you don't have an established audience.
   - Consider title-driven creation: first choose a compelling title, then write content that delivers on its promise.
+- [Good titles are pointers to a latent space](https://taylor.town/how-to-title). Titles travel on their own through [[news]] feeds and social networks. Many people share or curate from the headline alone, so it must stand on its own. Choose headlines that respect reader [[focus]] and align with your [[values]]. Good titles:
+  - Make a strong claim about the content
+  - Are almost too specific
+  - Use unusual word combinations
+  - Evoke curiosity
 - The inverted pyramid works well for blog posts. Put the tweet-length version of the post in the title or first paragraph. Get to the point quickly, then elaborate. Readers can bail out at any point of the text and still take home most of what mattered, while the meticulous crowd can have their nitpicks addressed toward the end.
 - [You're not just writing for today's invisible audience](https://web.archive.org/web/20250219111210/https://andysblog.uk/why-blog-if-nobody-reads-it/). You're writing for:
   - Future you. Your posts become a time capsule of your evolving mind.

Deep Funding.md

···
 - Instead of one canonical graph, allow different stakeholder groups (developers, funders, users) to maintain their own weight overlays on the same edge structure. Aggregate these views using quadratic or other mechanisms
   - If there is a plurality of these "dependency graphs" (or just different set of weights), the funding organization can choose which one to use! The curators gain a % of the money for their service. This creates a market-like mechanism that incentivizes useful curation.
 - Let the dependent set their weight percentage if they're around
+- Let the applicants apply with wathever "abstractions level" they want (e.g: a whole framework, one repository, an entire organization). Rely on pairwise comparisons to resolve conflicts.
 - Have hypercerts or similar. The price of these (total value) sets the weights across dependencies (`numpy`'s certificates trade at 3x the price of a utility library, the edge weight reflects this)
 - If there are reviewers/validators/jurors, need to be public so they have some sort of reputation
   - Reputation system for Jurors

Focus.md

···
 - [[Meditation|Mindfulness meditation]], e.g. breath-counting, seems to be a go-to technique for developing focus.
 - Periodic exposure to nature and out-of-doors in a relaxing, undemanding way can restore attention capability.
 - [Attention is a scarce resource](https://youtu.be/ZWI4_Oe-Qbs). Everything in the world is fighting to get yours.
+- [When a problem becomes the top idea in your mind](https://www.benkuhn.net/attention/), you reach the 50%+ focus tier that drives [[Productivity]]. You can only hold one such slot at a time. Stay with one problem even when blocked to force [[Problem Solving]] breakthroughs instead of context-switching.
+- Batch low-value chores into a fixed window and close open loops fast to protect [[Time]].
 - Some [sounds or music](https://mynoise.net/) can help you focus.
 - [The main thing is keeping the main thing the main thing](https://mattrickard.com/keep-the-main-thing-the-main-thing).

Systems.md

···
 11. The mindset or paradigm out of which the system — its goals, structure, rules, delays, parameters — arises.
 12. The power to transcend paradigms.
 
-**Don't aim for an ideal system. Build a set of [[processes]] and protocols that evolve to fit the environment over time.** [Complex systems fail](https://how.complexsystems.fail/).
+**Don't aim for an ideal system. Build a set of [[processes]] and protocols that evolve to fit the environment over time.** [Complex systems fail](https://how.complexsystems.fail/). [High entropy systems are easier to maintain and require less energy to keep things as they are](https://slatestarcodex.com/2016/07/25/how-the-west-was-won/).
 
 [The purpose of a system is what it does](https://en.m.wikipedia.org/wiki/The_purpose_of_a_system_is_what_it_does). However, [this view can be misleading](https://www.astralcodexten.com/p/come-on-obviously-the-purpose-of). A system's purpose is often its intended goal, even if it fails or has unintended consequences (e.g., a hospital intends to cure all patients, even if it doesn't succeed; bus emissions are side effects, not the purpose). Attributing failure or negative side effects as the *intended* purpose often ignores complexity, conflicting goals, or simple failure. A [common interpretation](https://www.astralcodexten.com/p/highlights-from-the-comments-on-posiwid) is that if a system consistently fails its stated purpose but isn't changed, its *actual* (perhaps hidden) purpose might be succeeding. Understanding the *intended* purpose, even if the system fails, is often useful for predicting its behavior (e.g., predicting an intelligence agency's actions based on its goal to prevent attacks, even if it fails). The phrase can obscure the useful distinction between a primary goal and unavoidable (or accepted) side effects.
 

Writing.md

···
 - [Writing that sounds good is more likely to be right](https://paulgraham.com/goodwriting.html). Making sentences sound better forces you to fix ideas unconsciously. Good rhythm matches the natural rhythm of thoughts—use rhythm as a guide for getting ideas right.
 - Use the active voice.
 - Write in a conversational tone. Think about readers when writing.
+- [A strong title is a pointer to a specific idea](https://taylor.town/how-to-title) and becomes a reusable handle in [[Communication]] and [[Ideas]].
+- [Finding a piece's "true name" often reveals its core narrative](https://taylor.town/how-to-title). If the title is fuzzy, cut back to the main thread and sharpen [[Focus]].
 - [Divide things into small chunks and if you have multiple points in a text, number them to make replies easier](https://slatestarcodex.com/2016/02/20/writing-advice/). List the points you want to make in a logical order. This allows you to remove the clutter and avoid confusion. Use the [Minto Pyramid](https://scqa.lifeitself.org/) or another standard structure like this one:
   - Decide what you're actually saying. [Define a clear thesis](https://www.cs.columbia.edu/~hgs/etc/writing-bugs.html). State the main point before you give the reasoning that leads to it.
     - What is your main point? Who are you writing for?