The name of this site. Vita nuova (Italian: "new life") is taken from Dante's early work of the same name, a prose-and-poetry sequence about love, loss, and the transformation that follows. It felt appropriate for a space that keeps changing as I do.

The spelling here is deliberately nouva rather than the standard nuova.

Pages that exist on the site but are not listed in the main index, not included in the sitemap's main feed, and generally not discoverable except by direct link or from other posts linking to them. They live under /blog/h/.

"Hidden" does not mean private or secret, the content is fully public. It means structural: these pages serve purposes other than being standalone posts. Examples:

• Anthology; programmatically generated
• Glossary; this page
• The bibliography (fullbooks, rlfo); long reference lists
• Technical notes and one-off pages that don't belong in the post stream

The /h/ subdirectory is the site's miscellaneous drawer.

A sidebar feature on article pages and a dedicated page at /on-this-day/. For any given date, it surfaces posts from past years that share the same month-and-day. So reading an article on March 21 will show what was written on March 21 in 2023, 2024, 2025, and so on.

The implementation is a Hugo partial (blog-on-this-day.html) that scans all dated pages and filters by .Date.Format "01-02". The sidebar version shows up to five entries; the dedicated page shows all of them.

My Emacs publishing system, available on sourcehut. The name comes from Arabic تدوين tadwīn,, the act of writing down, documenting, recording. It is the same root behind the famous Islamic hadith compilation project.

Tadwin.el is tightly coupled to my org-roam setup. Its main jobs:

• Collect nodes tagged @anthology and render the anthology page
• Handle cross-referencing between roam nodes and Hugo content
• Manage NOTER_PAGE values for bibliographic source tracking
• Generate structured exports of reading notes with citations

It is not a general-purpose tool and was not designed to be. The public version on sourcehut reflects the state of the system at time of release; the live version used for this site is generally ahead of it.

German: "slip-box." A knowledge management method developed by sociologist Niklas Luhmann, who maintained ~90,000 index cards of interconnected notes over his career and used them to write prolifically on an enormous range of subjects.

The core idea: notes are atomic (one idea per card), permanent (never deleted, only linked and extended), and connected (each note references other notes). The value emerges from the network of connections rather than from any individual note.

On this site, the zettelkasten is implemented in org-roam, each .org file is a node, and links between files are bidirectional. The blog is essentially a public projection of a subset of that private note graph. Posts are often developed from notes; the anthology is generated directly from it.

A good introductory text is Sönke Ahrens's How to Take Smart Notes (2017).

An Emacs package that implements a zettelkasten-style note-taking system on top of org-mode. Each note is an .org file with a unique :ID: property; links between notes use the id: protocol so they remain stable regardless of file renaming or reorganization.

Org-roam maintains a SQLite database of all nodes and their connections, enabling fast backlink queries, completion, and graph visualization.

My personal notes, the /roam directory, contains all my notes. A significant subset is exported to this site. The routing from note to published page is handled by tadwin.el.

The id:XXXXXXXXXX links visible in .org source files are org-roam node IDs. They are what makes links here stable even when the underlying files move.

A catch-all term for the system one uses to capture, store, organize, and retrieve personal knowledge. Common implementations range from paper notebooks to elaborate digital setups.

On this site, PKMS refers specifically to my org-roam graph, the .org files in /roam that hold reading notes, ideas, drafts, and links. The term appears most often in posts about art exploration and other projects where data is being fed into the note system for later analysis.

The word is slightly ugly and I use it reluctantly when I need a noun that is shorter than "my org-roam setup."

A property stored on org-roam nodes that records the position of a highlight or annotation within a source document. It is set by org-noter, a package for annotating PDFs and EPUBs from within Emacs.

The value is a Lisp cons cell: (PAGE . RELATIVE-POSITION), e.g., (53 . 0.5005149330587024). The first number is the page (or section, or piece number, depending on how the document is numbered); the decimal is the fractional vertical position on that page.

This value is read by tadwin.el to generate the no. citation in anthology entries, the number shown as "no. 53" refers to this page value. It does not always correspond to printed page numbers: for poetry I sometimes use piece numbers, and for documents without pagination I use section numbers.

A communication norm originating from the SL4 mailing list. Invoking Crocker's Rules means explicitly giving others permission to communicate with maximum directness, to optimize their messages for information transfer rather than for managing your feelings. You accept full responsibility for your own emotional reaction to what is said.

In practice: instead of a PR comment that spends two sentences apologizing before making a point, someone operating under Crocker's Rules expects and prefers just the point.

I follow Crocker's Rules and have written at length about why in the post I beg you to follow Crocker's Rules, even if you will be rude to me. The short version: text communication is expensive and most social softening is noise that buries signal. There are important caveats, see the post for them, particularly regarding new relationships and in-person interaction.

The original definition is at sl4.org/crocker.html.

A term I use in footnotes and asides to flag a note that is deliberately one-sided, partial, or unfinished, a note that does not attempt to present all angles. The word is a back-formation from equilateral (equal-sided): an inequilateral note is one where the sides are not equal, where I have made no effort at balance.

Example usage from the bibliography:

An inequilateral note about reading philosophy: the vast majority of philosophy books is not worth reading, in fact it's painful to read…

The term signals to the reader: I am not writing a full assessment here; this is a fragment, a reaction, a first impression. Take it accordingly.

A classification I used in the bibliography for books I regret reading, or that I would not reread, or that I would not recommend without significant qualification. The word means "not practical or advisable", more precise than "bad," which can mean too many things.

A book in the "inexpedient" section is not necessarily wrong or worthless, it may simply be that reading it was not a productive use of time, or that a better alternative exists, or that I have changed enough since reading it that I can no longer endorse it.

The section was removed from the public bibliography at some point (see footnote 8 in the bibliography: "As of [2024-11-15] this section no longer exists"), but the term still appears in footnotes.

"Too Long; Didn't Read." A format for writing short summaries or distillations of longer content. On this site, TL;DRs are a dedicated collection (accessible via id:x3cceno02wj0) of compressed notes.

The format rules are loose: a TL;DR entry is whatever I think is the minimum useful thing to say about a piece. Sometimes a single sentence; sometimes a few bullet points. The goal is that someone who will never read the original can extract the main claim in under thirty seconds.

A software design philosophy articulated by the suckless.org project, whose tools include dwm (window manager), st (terminal), and dmenu (launcher). The core tenets:

• Software should do one thing and do it well
• Source code should be small enough to be fully understood by one person
• Configuration is done by editing source code and recompiling, not through config files or GUIs
• Dependencies should be minimal

The philosophy is radical minimalism applied to software engineering: if a feature cannot be justified in terms of its complexity cost, it should not exist. The result is software that is auditable, fast, and opinionated.

I used several suckless tools during my Linux years (before the MacOS switch) and wrote a statusbar tool (statusless) that follows the same principles.

A chess (and other games) rating system developed by Mark Glickman as an improvement over the older Elo system. The key innovation: it tracks not just a player's rating but their rating deviation (RD), a measure of uncertainty in that rating. A player who hasn't played recently has a high RD, meaning their true strength is uncertain; their rating can move more per game.

In practice this means ratings are more meaningful: a 1800 with low RD is a more reliable estimate than a 1800 with high RD.

My peak Glicko rating was 1950. I use the rating as a rough self-assessment tool, not as a measure of competitive standing.

Official specification: glicko.net.

The lack of pampers is a term that was coined by Egyptian software engineer Abdelrahman Ghanem which describes the condition of a person who, having been relieved of all immediate obligations (say, changing/providing a child's pampers), resulting in them starting to publish online subjective and inflaming takes, to farm traffic or attention unpurposefully. Some examples of lack-of-pampers takes are: "PHP: A Fractal of Bad Design" by Eevee (2012), Why Most Unit Testing is Waste" by James O. Coplien, and Brilliant Jerks in Engineering, many HN and medium articles do qualify too (e.g. Why I Left [X] for [Y] posts).

The OG lack-of-pampers article was Conventional Commits considered harmful of Salih Muhammed. Other articles by the same author (including Programmers and software developers lost the plot on naming their tools and I beg you to follow Crocker's Rules, even if you will be rude to me) are considered to also be the result of the lack-of-pampers.

The diagnosis is structural rather than personal, this is to say, a person suffering from lack of pampers is not necessarily unintelligent but simply unoccupied at the wrong granularity, he or she has enough free time to write but not enough accumulated thought to have something worth saying. The internet, which imposes no threshold between having a thought and broadcasting it, completes the circuit.

The lack of pampers concept overlaps with Bertrand Russell thesis in In Praise of Idleness (1932). Russell argued, with characteristic perversity, that leisure (not labor) is the engine of civilization: that art, science, and philosophy are all products of people who were not busy enough to avoid thinking. The reduction of working hours was, for Russell, a moral imperative precisely because it created the conditions for something higher.

Based on Russell's view, others might argue that the label "lack of pampers" itself was a result of a lack of pampers.

Notable other cases are scattered throughout this site and across the web.

"If I knew what I know" is a sentence used by various politicians from the U.S. and United Kingdom after the United States invasion of Iraq, saying that they would not have voted for it:

When, in 2004, the then Tory leader, Michael Howard, was asked if he would still have supported the British government’s motion backing the war – only 16 Conservative MPs rebelled a year earlier – he replied: “If I knew then what I know now, that would have caused a difficulty.

I couldn’t have voted for that resolution.” “If I knew then what I know now, I would not have voted that way,” protested Hillary Clinton during her doomed first campaign for the Democratic presidential nomination. “If I knew then what I knew now, I wouldn’t have voted for it,” said Labour’s then deputy leader, Harriet Harman, a few weeks later.

A term, apparently floating around the Emacs community, for the decision to delete your entire Emacs configuration and start from scratch. The analogy is to bankruptcy in the financial sense: the accumulated liabilities (complexity, bitrot, interdependencies you no longer understand) exceed the assets, and the rational move is to discharge the debt and begin clean.

I declared Emacs bankruptcy in March 2026 after years of configuration accumulation made the system too brittle. The post describes the technical causes and the restructured modular configuration that replaced it.

Not to be confused with "org bankruptcy", the related concept of archiving your entire org-mode task system when it has become unworkable.