About this archive

Projects with a public artefact behind them; every non-obvious claim carrying its source; and no article bodies, by design.

Contents6 sections

This is a record of projects I have shipped that have something public to point at, kept as data rather than as writing. It is a section of this site, not a separate one. Below is what it covers and the rules it is kept by.

Every record has something public behind it

Client engagements produce systems that run inside somebody else’s estate: there is no repository, no package page and no release, so there is nothing for a record to cite and no record to write. That is the ordinary shape of that kind of work rather than a gap in this one, and naming it here is cheaper than leaving a reader to infer it from an absence.

The practice it represents is on the skills page instead, where the registry says so in as many words: an entry marked provenance: professional is one I work with and for which no publicly listable artefact exists. Enterprise products are only ever met that way, which is why the skills list is longer than the technology facet and why the difference is not a shortfall in either.

Nothing is dropped for being unflattering

Within that scope, dead work, superseded work and work that never found users sit alongside everything else, labelled as what they are. Nothing is hidden behind a “show older” toggle and nothing is quietly dropped once it stops being flattering.

A record’s evidence rows point at the commit, the package page, the issue or the release that establishes the claim. Where no public artefact can exist, the record states that in the place the evidence would have gone, rather than leaving a gap that reads as an oversight.

A record is a rendered data file

There are no article bodies here and there never will be — long-form writing lives at blog.shinyakato.dev. What a record page holds is its tagline, a summary of at most a couple of sentences, its status, its dates, its licence, the technologies it uses, its relations to the other records, and its sources.

The technology list is a practice list

An entry is there because I work with it, not because this archive can point at something it produced. Which kind of evidence stands behind one is recorded and is never rendered as a verdict on it: nothing here sorts, badges or footnotes an entry by it, and the two kinds take the same chip at the same size in the same colour treatment.

Browsing is the interface

Technology, year, context and status are each a real page with its own address, rendered ahead of time and reachable with JavaScript turned off. Every facet on the archive index is a link to one of them, and so is every one of the four axis names in its head.