How to Organise a Critic's Archive

To organise a critic's archive, keep three things separate: a private store of every file (published text, drafts, notes, correspondence), a catalogue that describes each piece with consistent metadata, and a public index that lists only published work with working links. Name files with ISO dates, record the same fields for every piece, save your own copy of everything the day it is published, and keep at least two backups in different places. Done this way, the archive outlives any single website, outlet or laptop.

A clips page for pitching editors is a curated shop window of perhaps twenty pieces. An archive is the whole record of a critical career, kept so that you, an editor, a researcher or eventually your estate can find any review you wrote, in any year, and read it in full.

What belongs in a critic's archive

Start by listing every place your work lives: outlet websites, old blogs, newsletter platforms, email attachments, notebooks, cloud folders, transcription apps. The Library of Congress's guidance on keeping personal digital records begins exactly there, with identifying where your files are and deciding which have long-term value. It also makes a point critics should take seriously: drafts can hold important details that never appear in the final version. An edited review often loses the paragraph where you worked out what you thought.

Sort the material into four classes:

• Published work: reviews, features, interviews, essays, columns, broadcast scripts, catalogue texts.
• Versions: your filed copy, the edited copy, and the published text, which are often three different documents.
• Working material: screening or exhibition notes, press releases, interview recordings and transcripts, research.
• Correspondence: commissions, briefs, edits and contracts with editors.

Only the first class goes public. The rest stays private but catalogued, because a contract or brief is what you will need when a republication question arises years later. Interview recordings and correspondence contain other people's personal data, so store them with the same care you would give any personal information.

A folder structure that survives decades

Pick a structure you will still understand in twenty years and that does not depend on any particular app. Organising by year first, then by piece, is the most durable:

archive/
 2026/
 2026-03-14_review_exhibition-slug/
 published.pdf
 filed.docx
 notes.md
 brief-and-edits.eml
 2026-04-02_feature_author-interview-slug/
 catalogue.csv
 README.txt

Every folder name starts with the publication date in YYYY-MM-DD form, which sorts correctly in any file manager. That form is the profile of ISO 8601 described in the W3C note on date and time formats. Follow it with the format and a short slug in lower case with hyphens. Keep the outlet name in the catalogue rather than the folder path, because critics change outlets, outlets change names, and pieces get syndicated.

The README matters more than it looks. The Library of Congress suggests writing a brief description of your directory structure and documents, and keeping a copy of that summary with your important papers. Two paragraphs explaining your naming scheme will save whoever inherits the archive days of guesswork.

Metadata worth recording for every piece

The catalogue is a single spreadsheet saved as CSV, one row per piece. You do not need to invent a schema. The Dublin Core Metadata Initiative's element set is a set of fifteen core elements for describing resources, first standardised in 1998 as IETF RFC 2413, and its fields map neatly onto criticism. A practical subset:

Two columns repay the effort for critics in particular. "work_reviewed" lets you pull every review of one director's films across twenty years. "rights" lets you answer, in seconds, whether you may republish a piece in a collection. If you want to be thorough, the Dublin Core terms vocabulary at the DCMI site defines further properties such as dates of modification.

File formats and backups for preservation

Store the published text in a format that will still open in decades. The Library of Congress Recommended Formats Statement summary for 2025 to 2026 lists PDF/UA and PDF/A (ISO 19005) among the preferred page-layout formats for digital textual works, with plain text and widely used word-processing formats further down as acceptable options. For a critic, that suggests a PDF/A of each published piece as it appeared, plus a plain text or Markdown copy of the words, which any future software can read.

For backups, the UK National Cyber Security Centre's advice on backing up your data defines a backup as a copy stored in a separate safe location, either cloud storage or removable media, and its rule of thumb is to back up anything you value. The Library of Congress goes further for long-term personal archives:

• Keep at least two copies, and more is better.
• Store them in locations as physically far apart as practical.
• Check the files at least once a year to make sure they still open.
• Create new media copies every five years, or sooner when necessary.

A laptop, one external drive kept elsewhere, and one cloud copy covers all of that for most critics.

Capturing published work before it disappears

Critics lose more work to outlet redesigns and closures than to failed hard drives. Capture each piece the day it goes live.

First, save your own copy: print the page to PDF and save the text. The Library of Congress page on preserving websites, blogs and social media recommends exporting the content and saving metadata such as site name and date alongside it, which is exactly what the catalogue row does.

Second, make a public capture. The Internet Archive's Wayback Machine help pages explain that "Save Page Now" saves one specific page once, without adding it to future crawls, and that captures can take 3 to 10 hours to appear. The same page shows a search pattern for listing every archived capture of a site, which is useful for recovering old blog posts you no longer have. Paste each capture URL into the archived_url column.

Third, know that national libraries collect too. The Legal Deposit Libraries (Non-Print Works) Regulations 2013, made on 5 April 2013 under the Legal Deposit Libraries Act 2003, allow UK deposit libraries to request online works by web harvester software. This is collection for the national record, not a personal backup, and you cannot count on any particular page being captured.

Professional web archives use the WARC format, which the Library of Congress lists as its preferred format for websites. Most critics do not need WARC files themselves; PDFs and text copies are enough for a personal archive.

Building the public index

The public index is a page on your own site listing published work, newest first, grouped by year, with title, outlet, date and a link. For a large archive, add filters or separate pages by form (books, film, theatre, visual art) and by subject, generated from the catalogue so the two never drift apart.

Link to the original first. If it has gone, link to your archived capture and say so in the entry. If you move your own site or restructure its URLs, Google's documentation on redirects recommends permanent server-side redirects where possible; the 301 and 308 status codes mean a page has permanently moved. Without them, every external link to your index breaks with the migration.

Structured data is optional. Schema.org defines a CriticReview type for reviews written or published by a source recognised for its reviewing activity, and you can use it for reviews republished on your own site. Google's review snippet documentation (last updated 8 September 2026, checked October 2026) lists Book, Event and Movie among the features that can carry ratings, but markup never guarantees stars in results. Do not add review markup to a page that only lists links to reviews hosted elsewhere.

Copyright, quotations and republishing

Two legal questions come up constantly in a critic's archive. Keep the answers general and check specifics with your union, society or a lawyer.

The first is quoting the works you review. GOV.UK's guidance on exceptions to copyright says fair dealing for criticism, review or quotation is allowed for any type of copyright work, with a sufficient acknowledgement. That covers quotations within your reviews; it does not cover reproducing whole poems, full-page images or film stills at will.

The second is republishing your own published pieces. That depends on what you signed for each commission, which is why the rights column and saved contracts matter. Your own copyright in written work lasts, according to GOV.UK's page on how long copyright lasts, for 70 years after the author's death. In practice that means the archive, and the rights recorded in it, will matter to your estate as well as to you.

A maintenance routine you can keep

An archive fails through neglect rather than bad design. A routine that works for most working critics:

• On publication day: create the folder, save PDF and text, make a web capture, add the catalogue row, update the public index.
• Monthly: check links on the public index and swap dead ones for archived captures.
• Annually: open a sample of old files in each format, confirm both backups are current, and backfill one older year if you are still catching up.
• Every five years: copy the whole archive to fresh media, following the Library of Congress interval.

When backfilling, start with the most recent years and work backwards, since recent work is easiest to recover and most often requested. For more guides on author websites, portfolios and newsletters, see UniBlog.