UCL Economics Research webpages

This repository will contain the source, content, and update processes for a set of UCL Economics Research webpages. The initial aim is to produce clear, accessible pages that can be kept current from agreed online sources or spreadsheets without unnecessary manual editing.

Planned pages

1. Seminars
2. Visitors
3. Events
4. Research Groups
5. Publications
6. PhD Students

The Research Groups, Publications, and PhD Students pages are currently works in progress. Their content and update processes will be defined later.

Governance

• Project owner: Professor Lars Nesheim (l.nesheim@ucl.ac.uk)
• Publication approver: Professor Lars Nesheim
• Editors: To be appointed
• Reviewers: To be appointed
• Seminar content owners: To be appointed for each seminar series
• Visitors content owner: Professor Lars Nesheim
• Events content owner: Professor Lars Nesheim

Each imported or manually edited asset must be approved by its relevant content owner or, where no separate owner has been appointed, by the project owner.

Seminars

The Seminars page will have an All seminars tab followed by eight series tabs:

• All seminars
• Applied Economics
• Econometrics
• Economic Theory
• Finance
• Macroeconomics
• IFS Seminar
• IFS/UCL/LSE Development Seminar
• IFS/UCL Labour Seminar

The IFS schedule is imported from the IFS seminars webpage. Rows labelled CeMMAP on that page are excluded because the same programme is already supplied by the dedicated CeMMAP source.

Seminar information will be refreshed daily. Each series may be populated from either an approved online source or a spreadsheet placed in the seminars/ folder. The owner of each seminar series is still to be appointed.

Seminar page requirements

Selecting a seminar-series tab will show a clean, simple page containing:

1. a prominent summary of that series' next upcoming seminar;
2. a complete table of seminars in the current academic term; and
3. the seminar organisers' names and email addresses.

The All seminars tab shows every seminar scheduled across all series in the selected Monday-to-Sunday week. It opens on the current week and provides previous- and next-week arrow buttons. When either button has focus, the Left and Right arrow keys also move backward and forward by one week. The displayed date range is announced when it changes, and a clear empty state is shown when the selected week contains no seminars.

The term table will use the columns Date, Speaker, Title, Location, and Time. It will follow these display rules:

• The Speaker cell displays the speaker's name, with their institution on a second line. The name links to the speaker's webpage when a URL is available; otherwise it is plain text.
• The Title cell links to the paper or slides when a URL is available; otherwise it is plain text. A missing title is displayed as TBA.
• Location and time are displayed separately. A location URL, where available, is used in the featured next-seminar summary.
• The table contains the complete current-term schedule, including seminars that have already taken place; the next upcoming seminar is visually identified without obscuring the full list.

The next-seminar summary repeats the date, linked speaker name, institution, linked paper or slides title, linked location, and time, using the same fallback rules as the table. It also includes a Sign up to meet the speaker link to the relevant entry on a separate sign-up-list page. The sign-up list and any personal information it collects must have appropriate access and privacy controls.

Each seminar-series tab also provides a Subscribe to calendar menu for Google Calendar, Outlook, and other calendar applications. The public iCalendar feeds are generated from the same validated seminar data during imports and website builds. Their stable URLs are under /calendars/, with one feed per series and academic year. Calendar applications refresh subscriptions on their own schedule; downloading an .ics file is a one-time import and does not receive later changes. Seminars without a confirmed, parseable start time are omitted from the feed until their time is known. Where only a start time is published, the calendar event uses a one-hour duration.

Required seminar data fields are therefore:

• seminar series
• academic year and academic term
• date and start/end time
• speaker name and institution
• optional speaker webpage URL
• optional title (displayed as TBA when absent)
• optional paper or slides URL
• location and optional location URL
• optional sign-up-list URL
• organiser name and email address
• status, including scheduled, cancelled, or postponed
• source and last-updated date

The editable standard seminar spreadsheet contains only Date, Speaker, Institution, Speaker URL, Title, Paper URL, and Status. Missing cells are highlighted in yellow for review. Recurring location and time defaults, plus annual organiser details, are stored once in seminars/config/series.json; sign-up URLs are managed separately and are not part of the standard schedule spreadsheet.

Visitors and events

The Visitors page is maintained from visitors/visitors-2026-27.xlsx. Its fields are name, institution, webpage, start date, end date, optional date display text, office, and optional email. The separate date-display field is used only when exact dates are not yet known.

The Events page follows the same pattern using events/events-2026-27.xlsx. Its fields are event type, title, event URL, date display, optional exact start and end dates, location and location URL, organisers, and booking URL. Multiple organisers are separated with semicolons.

After either workbook is edited, scripts/sync_spreadsheets_to_site.mjs creates the website-ready visitor and event data. The workbooks are the manually maintained source; files under site/data/ are generated copies for the website. Both content sources are owned by Professor Lars Nesheim.

PhD students

The PhD Students page is maintained from phd_students/phd-students.xlsx. This workbook is the authoritative source for student names, email addresses, key research areas, main fields, and secondary fields. The allowed field values are Applied Economics, Economic Theory, Macroeconomics, Econometrics, and Finance.

After editing and saving the workbook, run scripts/sync_phd_students_to_site.mjs. The script validates required cells, field names, and duplicate email addresses before replacing site/data/phd-students.json. Existing UCL profile links are retained by email address. The generated JSON file should not be edited manually.

The general scripts/sync_spreadsheets_to_site.mjs workflow also includes the PhD student import, so the visitors, events, and PhD student website data can be refreshed together.

Running the data scripts

Install Node.js 18 or newer, then install the data-script dependencies from the repository root:

npm install

Refresh all seminar sources and generate the website data, review workbooks, and local source snapshots with:

node scripts/import_seminars.mjs

Use --series applied-economics (or another series ID) to refresh only one series. Use --offline to rebuild from the versioned files under seminars/sources/ without accessing the network. --project-root and --output-dir can override the default repository and outputs/seminars/ locations.

After editing the visitor, event, or PhD-student workbooks, regenerate their website JSON files with:

node scripts/sync_spreadsheets_to_site.mjs

Spreadsheet creation and update scripts always write .xlsx files. PNG previews are skipped because the portable ExcelJS implementation does not render worksheets; open the generated workbook in Excel or LibreOffice for visual review.

Publication period and archive

The public pages will initially show only the 2026–27 academic year. The system should be designed to retain earlier academic years in a private archive in the future, but archived records will not be published unless that policy changes.

Proposed repository structure

.
├── README.md
├── task_list.md
├── seminars/        # Seminar spreadsheets and source documentation
├── visitors/        # Visitor spreadsheets and source documentation
├── events/          # Event spreadsheets and source documentation
├── site/            # Website source (to be confirmed)
├── scripts/         # Data import and validation tools
└── tests/           # Automated checks

Folders will be added as the relevant work begins. Raw source data should be kept separate from generated website files.

Hosting and recommended technology

The webpages will initially be external to the existing UCL website and hosted on a Google platform. The exact platform name and the availability of an approved UCL-managed Google project must still be confirmed.

The recommended implementation is:

• Eleventy (11ty) to generate a small static website from templates and validated data;
• HTML, CSS, and minimal JavaScript for fast, robust, accessible pages;
• Firebase Hosting for Google-hosted deployment, HTTPS, preview channels, custom-domain support, and rollback;
• Google Sheets or versioned spreadsheet files as editor-friendly inputs;
• a scheduled import and build workflow to check sources daily, validate changes, create a preview, and publish only after owner approval.

This is preferable to building the main pages directly in Google Sites because the project needs reliable data imports, accessible interactive seminar tabs, automated validation, versioned review, testing, and future private archives. If Google Sites is mandatory, it can instead act as a simple outer site with a full-page embed, but the embedded application would still need separate hosting and governance. No confidential archive data should be included in the public site build or its downloadable assets.

Update approach

The intended content pipeline is:

approved link or spreadsheet → import → validation → review → webpage

Every source should have a named owner and a documented format. Initial sources will be checked daily. Imports should report missing or malformed data rather than silently publishing it. Generated pages should show when their content was last updated, and the relevant content owner or project owner must approve changes before publication.

Spreadsheets must not contain confidential personal data. Contact details and other personal information should only be published where there is a clear purpose and appropriate permission.

Design and content principles

• Follow UCL visual identity guidance and agree the relationship to the main UCL website before implementation. The initial visual reference is the UCL Economics website: https://www.ucl.ac.uk/social-historical-sciences/economics.
• Order the primary navigation by user need rather than alphabetically: time-sensitive and frequently used pages first; research outputs next; people pages next; and institutional and reference pages last. Within each group, place the page with the broader or more frequent use first. Keep the order stable unless evidence from usage or user research supports changing it.
• Design mobile-first and meet WCAG 2.2 AA accessibility requirements.
• Use semantic HTML, keyboard-accessible navigation, useful page titles, and descriptive link text.
• Make upcoming information prominent and retain past items only where useful.
• Keep source attribution, ownership, and update status visible to editors.
• Avoid committing credentials, private data, or unpublished sensitive content.

Project status

The project is at the discovery and prototyping stage. The first deliverable is a small set of model webpages showing proposed visual style, navigation, and representative content. These models will be reviewed before the recommended technical stack and full data pipeline are finalised.

See task_list.md for the proposed work plan.