proposal: embed original data into output

[antialias]

• for PDFs, we can use XMP
• for HTML, we can use a meta or script tag. For example:
  • <script id="json-resume" type='application/json">${JSON.stringify(resume)}</script>

This improves the process where a candidate sends their resume as a PDF to a recruiter or org, and then the org wants to turn their collection of PDF resumes into a database : instead of manually parsing the resume, the org / recruiter can extract the original json resume from the PDF / HTML etc...

recommended tools

pdf-lib appears to be capable of writing metadata to PDFs.

[antialias]

as a follow-up, we can implement reading json-resume from these output files instead of assuming the input is always a json file.

[thomasdavis]

I like it, but probably won't act until someone requires it. Unless it is well known HR companies already use a certain format that can be represented in XMP.