|Status: This page is more than a year out of date. Until it's updated, you can see what is actually being prioritized by the devs currently by attending the dev meetings or browsing the meeting notes.|
This page attempts to outline the development team's priorities in broad strokes -- this priority list should serve as a general guideline for our team to decide what to work on when the weekly dev meetings don't provide enough guidance.
For Winter 2017, the following priorities exist for the dev team, in order of most important to least important:
- Site stability and performance issues: If the site is misbehaving or not performing as it should, the dev team should focus on resolving those issues as the first priority.
- ModBot/Nagios: If ModBot or Nagios sends periodic emails about an issue, the dev team should address these issues quickly. ModBot fixes may not require hot-fixes, but they should be addressed as soon as possible.
- React: Finish rewriting UI templates to React (MBS-8609)
- SOLR: Catch up with mmd-schema 2.0 v-2019-01-07
- Schema change 2019 Q2: Fully resolve selected schema change issues by May, 13.
- Fixes/Features: Continue fixing issues and implement new features based on community feedback.
- Beta site: Finish the beta re-write of ListenBrainz.
- Test Beta site: Create a public review site for the new version of ListenBrainz
- Release Beta site: Push the new beta site live, with support for importing from alpha site.
- Big Query: Start feeding data to BigQuery and start making data available to the public.
- Big Query based features: Once data is available in BigQuery, start making end user features that give insights to user about their listening habits.
Gentlecat: please add items.
- Dataset infrastructure
- Merge dataset challenges code
- Integrate cross-dataset evaluation code into dataset evaluation (http://mtg.upf.edu/node/3498)
- Automatically artist/album filtering
- Integrate new datasets into live AB data
- Switch to scikitlearn for model generation
- Docker infrastructure and deployment closer to other servers
Further afield (Summer)
- Bigquery data shipping and statistics
- Server-side versioning for new extractor and storage of detailed lowlevel files
- MVP Roadmap: bring BB to minimally viable product to grow community
- Implement Author credits (similar to Artist Credits in MusicBrainz) (in progress)
- Create an entity merging tool (in progress)
- Review contribution pages and FAQ on website
- Automatically create Edition Groups for Editions
- Docker containerization and hosting/deployment with other MetaBrainz projects
Migrating to the shared Postgres and Redis instances Dockerize and configure webserver
- Migrate to the Hetzner servers
- Add beta.bookbrainz.org to test beta code on main database
- Consolidate and improve user and developer documentations
- Create style guidelines
- Develop and deploy an API
- Incremental database dumps
- Create a unified form to ease the process of adding a book
- Replace Elasticsearch with Solr search engine
On top of improving the search capabilities, this will align BookBrainz with MusicBrainz and reduce deployment and hosting work
- Review schema and add new Entities after a community review and decision period
- Data importing:
Large scale data imports from other open-source databases. This will require community decisions on automation, validation, merging and cleaning.
- Cosmetic improvements to the web UI/UX
- Collaboration with other book databases:
Link with other open-source libraries and create collaborative tools to improve and keep data up to date
- Accessibility and Internationalisation, improve experience on handheld devices
- Integration with CritiqueBrainz
- Integration with the Internet Archive to display book covers
- User collections
- Analytics: Setup piwik to gain insights into our traffic
- Site performance: Use insights from analytics to prioritize tasks for site performance improvements
- Backups: Set up comprehensive back-ups of our NewHost servers (and make sure that they work).
- API Keys: Support API keys and different rate limits on our APIs.