---
title: iSB-Lite
subtitle: A minimal iSB instance for small collections
---

See also: https://github.com/isamplesorg/isamples_inabox/issues/198

# iSB-Lite

There is a need for a simple system to support field data collection in an offline-first system that enables content creation and editing with no Internet connection, but is able to seamlessly participate in the iSamples network when a network connection becomes available.

This experiment will leverage CSV for file storage, git for revision control, GitHub for serving static content, and if necessary, supporting services operating within the constraints of zero-cost hosting such as on Deta, Vercel, Google App Engine, or AWS.

## Workflow

Basic workflow:

1. A gitHub repository is created based on a template that includes necessary actions and structure.
2. User creates a spreadsheet (or some other easily created format, keeping in mind git "friendliness") containing their data
3. User commits spreadsheet to local repo. Validation is run as part of the commit, or separately via script.
4. The spreadsheet is pushed to GitHub
5. A GH action parses the sheet and generates iSamples JSON entries for each record, an XML sitemap for the records, and publishes them on the `gh-pages` branch of the repository.
6. An iSB instance harvests and indexes the content.

## Some Questions

- Best place to assign identifiers?
- Ideally identifiers should be available at the time of content entry, so perhaps the repo includes a script that generates a spreadsheet template with requested number of pre-allocated identifiers?
- How far can we go with providing some sort of UI for the content, hosted on GH, zero-cost service, or perhaps just running locally with Docker etc?