Newt, the second prototype

R. S. Doiel,

Caltech Library, Digital Library Development


What is Newt?

Goal of Prototype 2: Answer the question.

Is Newt and “off the shelf” software enough to create metadata curation applications?

High level Concepts

Implementation Concepts


Off the shelf (no coding)

Office the shelf (other data sources)

Assemble app from YAML (less coding)

How are data models described?

How do I think things will work?

  1. Interactively generate our application’s YAML file
  2. Interactively define data models
  3. Generate our application code
  4. Setup Postgres and PostgREST
  5. Run our app with Newt

Steps one and two are interactive

  newt init app.yaml
  newt model app.yaml

Step three, generate our code

  newt generate app.yaml

Renders SQL, PostgREST conf, Mustache templates

Step four, setup Postgres and PostgREST

  1. Use the generated SQL and configuration
  2. Setup and check via createdb and psql

Step four, setup Postgres and PostgREST

  createdb app
  psql app -c '\i setup.sql'
  psql app -c '\i models.sql'
  psql app -c '\dt'

should this be automated too?

Step five, run your application and test

  newt run app.yaml

Point your web browser at http://localhost:8010 to test

Can I run a demo?

Not yet, hopefully in late May 2024.

Second prototype Status

How much is built?

Insights from prototypes 1 & 2

Insights from prototypes 1 & 2

What’s next to wrap up prototype 2?

Unanswered Questions

Someday, maybe ideas

Thank you!