Sitemap

[arildm]

Google (and other bots) might not easily find all the urls and index them when crawling the site, since most works and terms are only listed after searching or paging/scrolling, which I doubt the bots will be doing.

Background:

• A sitemap is a list of urls on the site [Google: Build a Sitemap]
• They can be in XML, RSS or plain text. The plain text format is literally a list of urls. The other formats can be used to add metadata, but that doesn't seem really valuable in our case.
• The sitemap can be submitted in the Google Search Console, or referred from robots.txt
• We could generate one by doing API queries and outputting route paths
  • I'm sure WordPress can generate a sitemap for the /om/ pages. We can combine the two sitemaps with a sitemap index
• I don't think we should generate the sitemap as a static file at build time, as this would need to be triggered again manually after relevant changes in Libris or the QLIT backend
• Better, perhaps, to do the sitemap generation in a Vue route view. It would take a long time but hopefully that's not a problem.

[arildm]

Another approach that would probably benefit search bot indexing is #57 "Search query as url parameters", especially including the page number. If I understand correctly, the bot can then navigate through the result list and reach all the work urls by itself.