Hels

If you could dump out a subset of the data I would give it a try at a basic level. I would use R cos thats what I know better (for -omic data). Am actually doing a course in “advanced data manipulation techniques in R” (or some title that sounds as obnoxious as that) for work in a few weeks so could even use that as my practice dataset…. (since there are already alot of pipelines for standard omics datasets and I want to learn new techniques to go beyond the pipelines)

Hels

@Firesphere - data-mine it. do a word-cloud type analysis, then subtract obvious ones like “hi”, “thanks”, “and” (I’m sure there some common usage type resource that could provide the list to subtract), then use a little human common sense to work out how the top words relate to potential FAQs or tutorials

null

It's a massive database. In theory you could data mine for common questions, or rather "common Solr queries that tend to answer peoples questions, and which ones get clicked"

null

url structures and stuff like that

Firesphere

The URL structure is very predictable, as well as there being a sitemap.xml

null

wasn't thinking meta tags - perhaps some code changes

Firesphere

@null Feel free to log in with your Slack account, I can hook you up so you can do the SEO magic then. I think I already got it fairly well covered by the meta descriptions?