As part of the migration to Python 3 #20, we had to make some changes to the indexing class of Hoaxy. It would be nice to see if there are additional fields that would be worth indexing. For example, we might want to search only by domain (e.g. breitbart, infowars), or by language (e.g. Portoguese, Spanish).
Language would be difficult, maybe require Solr -- deferred until a future time. But we could index by domain and/or domain type (tag). This would require three steps:
Re-index existing collection via Lucene (just for our local version)
Modify the cron job that indexes new incoming document
Modify the API so that it accepts the new parameters
As part of the migration to Python 3 #20, we had to make some changes to the indexing class of Hoaxy. It would be nice to see if there are additional fields that would be worth indexing. For example, we might want to search only by domain (e.g. breitbart, infowars), or by language (e.g. Portoguese, Spanish).