Sprint 23: Mar 3 - Mar 17

A couple of big changes this sprint to our build config - if you run this app locally through docker, we recommend that you remove old images and containers and build it from scratch to avoid any unexpected errors.

• Refactor how frontend/backend share common code, with a streamlined build process and upgraded webpack/ts dependencies
• Backwards compatible change to how we handle config settings, in preparation for move to Kubernetes
• Refactored app to use optional chaining
• Modal added on /datasets page for dataset usage survay
• Datasets with fewer than five speakers no longer feature demographic information for privacy reasons
• Localization and sentence collector updates for various languages, including Europarl sentence collection for German
• README updated with licensing information
• Minor bugfixes and UI adjustments

Release for Sprint 22: Feb 10 - Mar 2

This is a slightly wonky sprint as it is the first sprint after all-hands, so things were shifted around to accommodate PTOs and vacation time.

• Localization and sentence collector updates for various languages
• Minor mobile fixes
• Updated About, Footer sections to refer to Matrix
• Suspending sending of sentence text, vote files to S3
• New schema for tracking newsletter preferences + throttling SFDC events
• More accurate analytics class tags