Why does URL readability matter?

Because people quite often look at a URL, whether in browser address bar or as captured by "copy link", and want to know what they are looking at.

Because sometimes you copy/paste a link from address bar or "copy link" into your wiki and want to use [[wiki notation]] rather than https://ugly/URL/notation with minimal editing.

Sometimes you want to paste a URL from your own wiki and have it appear as a nice [[wiki link]]. It can be a pain to have to manually edit all of the %-encodings. "Is %3A colon, or maybe %3B?"

Sometimes from other wikis or web.sites. Them making things readable supports our life easier.

Heck: if you don't care about URL readability, then why not simply map every character to its hex code? That's fully reversible - just not very human friendly if your human users ever look at URLs. A stateless function in your wiki rendering engine can do this mapping.

The next step, having the URL contain a UID that is essentially a number, requires state to perform the lookup.

And if you want to go that far ... you will basically look like Google Drive or Google Docs. See Why are Google Drive, Docs, etc. so widely hated?. Lack of URL readability is one reason. Looking at a GDrive URL, you can't figure out what is nearby, what folder the doc is in. Especially not if an embedded file like a PowerPoint spreadsheet or a PDF, that Gdruve does not provide a frame around giving you context. And even if you have such a frame... Heck, GDrive sucks! It's not just me saying it, listen to podcasts!