Disambiguate Reference Artifacts from Traditions

When multiple external artifacts from the same tradition are captured as separate References, each filename carries an artifact descriptor — <Name> Gist, <Name> Paper, <Name> Repository, <Name> Lecture — before the citation parenthetical. Wikilinks and Named Edges Gist (Christopher Allen, 2026).md names a specific artifact (the gist); Wikilinks and Named Edges.md would name the concept-as-a-whole. The descriptor distinguishes the instance from the tradition.

Why

The commitment prevents conflating the instance with the tradition. A tradition articulated in a gist is not identical to the gist itself. The gist is one artifact documenting the tradition — a specific URL, a specific version, a specific authored moment. The tradition extends beyond the gist: earlier articulations, later revisions, related work by the same author, adoption by others. When a Reference node captures the gist, the node's scope is the artifact; when a Reference node captures the tradition, the node's scope is broader. Both are legitimate; the filenames must distinguish.

The artifact-descriptor suffix carries the distinction at the filename level. A reader encountering [[Wikilinks and Named Edges Gist (Christopher Allen, 2026)]] knows the citation is to a specific artifact at a specific time; a reader encountering [[Wikilinks and Named Edges]] (ghost link or future Gloss) knows the reference is to the tradition-as-a-whole. Pipe form lets the prose read naturally either way — the [[Wikilinks and Named Edges Gist (Christopher Allen, 2026)|wikilinks and named edges]] convention cites the artifact while displaying the tradition name. The commitment's filename-level distinction is what makes both citations precise.

The descriptor also handles ecosystems where multiple artifacts matter. If the project draws on a paper AND a repository AND a blog post from the same tradition, each is a separate Reference with its own role, its own body, its own adoption account. The descriptors distinguish the three: Foundational Paper (Author, Year), Reference Implementation Repository (Author, Year), Extended Discussion (Author, Year). Without descriptors, the three would collide at the filename level or require awkward disambiguation strings.

The commitment applies specifically when multiple artifacts from one tradition are captured. A single-artifact Reference may use the tradition name as the filename with no descriptor (the descriptor becomes necessary only at the moment of disambiguation). A prudent author anticipates future artifacts — adding the descriptor from the start when the tradition is likely to yield more References — but the rule does not require pre-emptive descriptors; it requires them when disambiguation becomes load-bearing.

Alternatives Considered

Lump artifacts under tradition-level References. Create one Reference per tradition, listing all artifacts inside the body. Rejected because it loses per-artifact graph participation. A Decision that is informed by a specific gist wants to cite the gist, not the tradition-as-a-whole; lumping forces every citation to refer to the tradition and recover specificity through body prose. The graph layer's granularity drops from artifact to tradition; the cost is paid on every citation.

Author-and-year-only disambiguation. Let two artifacts from the same author and year be disambiguated by trailing letters ((Author, 2026a), (Author, 2026b)). Rejected because letter suffixes carry no semantic information. A reader sees (Allen, 2026a) and (Allen, 2026b) and knows only that they are two artifacts from the same author and year; they do not know which is the gist, which is the paper, which is the talk. The descriptor approach carries the semantic distinction in the filename directly.

Disambiguation through body content. Let filenames remain undifferentiated and distinguish artifacts through body descriptions. Rejected because filename-level identification is the cheapest classification tier. A reader scanning a folder of References does not want to open each file to know what kind of artifact it is. The descriptor convention puts the distinction in the cheap tier; body content handles the deeper semantics.

What Would Change It

The commitment would be revisited under one condition.

Most References come as singleton artifacts. If the project's References remain predominantly single-artifact-per-tradition — where no tradition has yielded two or more Reference-worthy artifacts — the descriptor convention would carry authoring cost without disambiguation benefit. The revisit would soften the rule to "descriptors SHOULD be added when a second artifact from the same tradition arrives, not pre-emptively." Current Reference patterns have not tested this yet; the first tradition to yield two References will be the decisive case.

Relations