Sure, you’re hosting a few hundred or thousand videos, a blob of yaml can be good enough. I have no history if this project, but I suspect they started out like that. Something like Elasticsearch usually comes into play when you are ingesting a lot of data point very quickly, so I would assume the authors wanted to archive fairly huge amounts of videos.