New independent field guide, published 17 October 2026. Examples are illustrative. The historical directory records linked below are separate source context, not current recommendations or earlier versions of this essay.
A good resource description answers a practical question: is this a promising next step for the task I am doing? It is not a miniature advertisement or a place to repeat the title with more adjectives. A directory has limited space, so every sentence should help a reader distinguish this destination from nearby alternatives. This guide develops an editorial process for writing concise, traceable descriptions from actual inspection. All sample resources are invented. The method does not assert that the historical listings in this edition remain available, accurate or suitable for any current task, and it does not revive the former directory's submission service.
Identify the decision the description supports
Begin with the reader's likely choice. Are they looking for a quick introduction, primary documents, a tool, a dataset or an organisation's contact page? A description that ignores this choice can sound informative while failing to help. In an invented regional-history collection, saying a site contains fascinating material is weaker than explaining that it provides scanned newspapers searchable by date. The latter identifies both the material and an important way of using it. A reader can compare that description with their own need without accepting an unsupported judgement about how impressive the collection is.
Do not try to make one short card answer every possible question. Separate essential selection details from information better placed on a fuller record. Scope, format, audience and a material access limitation are often enough for a first decision. Provenance and review history can follow on the detail page. The important relationship is continuity: the short description should not promise something the full record withdraws. If access requires an account, for example, hiding that fact until after the click can waste effort precisely when the directory is supposed to reduce it for the reader.
Inspect the resource before summarising its promise
A publisher's homepage can state ambitions that do not match the specific page being linked. Inspect the destination and the relevant feature rather than writing only from a promotional slogan. In a fictional example, an organisation may claim broad educational coverage while the linked resource contains only one introductory worksheet. The directory can describe the worksheet accurately without evaluating the entire organisation. Keep notes about which material was inspected and on what date. That makes the published description accountable to a specific observation instead of a general impression formed from branding.
Distinguish observed content from claims that require attribution. If the resource calls itself comprehensive, the directory should not repeat that word as its own conclusion unless a meaningful completeness assessment was made. A neutral description can say what subjects or periods are visible without implying exhaustive coverage. When inspection is limited, state the limit rather than filling the gap with plausible detail. A page title can suggest a format or topic, but the suggestion is not enough to establish how the resource works, whether a download exists or who can access the material today.
Choose concrete nouns and useful verbs
Descriptions become more discriminating when they name the material and action directly. Compare an invented phrase such as an innovative platform for powerful historical discovery with a plainer account: browse scanned maps by town and publication year. The second version tells a reader what they can do and what they will encounter. It does not require replacing every specialist term with a vague everyday word. Keep technical terms when they identify the resource accurately, and explain unfamiliar ones when the intended audience needs that explanation to make a useful choice.
Avoid repeating information already obvious from the title unless the repetition resolves an ambiguity. A record called Oral History Interviews does not need a sentence saying it offers oral history interviews. It may need to say whether recordings, transcripts or both are present, which region is covered and what language is used. These distinctions are small but consequential. A reader needing a downloadable transcript can reject an audio-only resource before spending time navigating it. Good brevity comes from removing low-information wording, not from deleting the very details that distinguish the resource from its neighbours.
Describe limitations as part of usefulness
A limitation can make an entry more useful by identifying the conditions under which it fits a task. A dataset may cover one decade, a guide may assume prior knowledge or a viewer may require a particular interaction. The description does not need a defensive list of everything the resource cannot do. It should state the constraints likely to alter the reader's decision. For an invented map collection, coverage ends at a certain year is more useful than a generic warning that information may vary. Specific limits let readers judge relevance without being asked to infer missing context.
Keep access observations dated and avoid universal language. A download observed available during review is not guaranteed to remain so. A page that failed one automated request is not necessarily unavailable to every visitor. When a limitation concerns the directory's own inspection, label that too: only the index was reviewed is different from the resource containing only an index. These are distinct boundaries of knowledge. The description should not convert a partial review into a statement about the full collection, just as it should not convert the publisher's own promise into an independently established result.
Make descriptions comparable without making them identical
A consistent editorial pattern helps readers scan several entries. One workable order is material, scope, useful feature and significant limitation. The pattern should support comparison, not force every resource into a padded template. Some records may need a language note; others may need a version distinction. In a fictional list of three mapping resources, each description can identify whether it offers scans, vector data or a teaching explanation. Readers can then compare the actual kinds of resource rather than sorting through three equally enthusiastic paragraphs that all claim to be useful and comprehensive.
Review nearby entries together. If every description could be swapped with another without sounding wrong, the wording is probably too generic. If one entry is far more promotional, determine whether it has stronger evidence or merely a different writing style. Align the evidential standard rather than flattening genuine differences. Keep proper names and exact titles accurate, but do not copy long third-party descriptions simply to maintain the old directory's appearance. A new description should be identified as new editorial work, with the captured record and its provenance available separately where historical comparison is relevant.
Treat editing as a maintenance event
Descriptions can become obsolete when a resource changes, even if its address remains stable. Attach a review event to substantive revisions and preserve the reason for the change. Correcting a spelling mistake is different from changing the claimed coverage or access conditions. A useful internal note might say that the old downloadable report was replaced by a continuously updated dashboard. The public description can then explain the current scope after inspection, while a historical record retains the earlier version. Do not backdate new editorial wording to the date of the source it discusses.
The old submission-policy and listing records below establish historical context only. This edition publishes new annotations and inert factual metadata rather than copying old promotional descriptions into a current directory. The general writing method is to inspect, identify the reader's choice, state concrete differentiating details and record meaningful changes. A description has done its job when someone can anticipate the destination accurately enough to decide whether to continue. It does not need to persuade everyone to click. Helping a reader rule out an unsuitable resource is also a successful piece of directory editing.
References & further reading
From the recovered archive
- Submission rules — historical policy · captured 28 November 2010
- Cufflinks, Inc. — record 1160 · captured 29 November 2010