:mega: Docs update! - My plan will be to wipe eve...
# development
b
📣 Docs update! • My plan will be to wipe everything clean that I have and re-port from scratch. This ensures that: ◦ The script(s) to do the migration can be pinned to a shelf somewhere and be trusted no funny business ◦ Allows up to "pull the trigger" whenever and have high confidence • In this, I had to do some housekeeping to our current docs site for a smooth migration ◦ Hidden pages which shouldn't be hidden were marked public ◦ All broken
doc:
links were unbroken ◦ A few miscellaneous bits and bobs will be fixed
NOTE: I'm doing this on
<http://readme.com|readme.com>
, not in the in-repo docs, as almost every branch is now "stale" so the point is moot
2.0-2.17 now have 0 dead internal links! 😏 (techincally one, don't ask 😛 )
c
which one? 🙉
b
It's a link to a reference doc on a branch where the generated reference docs shouldve included the backend but didn't. I'm remedying it on the new site, but not the old
c
I see 🙂
b
Potentially a silly question: if there’s more issues like broken links blocking the migration, can we instead turn down those docasaurus settings (eg silence errors/reduce to warnings), do the migration, and then iterate/fix things once we’re switched over?
b
It's already at
warn
right now. I want to turn it to
error
. Since I'm migrating old links to new ones, I know which are dead. So I thought "why not fix them while I'm at it"
So, yeah for some stuff we totally can fix it post migration (and there certainly are things in that category), but others, like this, it's a win either way 🙂
b
Okay. The background for my question is that I know there’s a bunch of stuff that’s semi blocked on getting off readme (eg reducing costs) so seems like getting that step done would be good as focus, not so much the nice-to-have stuff (I’m sure it’s “easy” but it likely still takes time?)