>     This would be nice to have. We can probably auto-extract it from <a
>     href="..."> tags or by searching the plain text of the document for
>     the regexp "SRFI\s+(\d+)".
>
> It's a bit more curated than that.  There would probably be far too many
> connections if we did that search.
This would necessitate either:

1) Adding a "see also" class to the HTML.
2) Adding a "See Also" section to all SRFIs that refer to other SRFIs.
3) Adding a separate "fixup" file for each SRFI with manual additions
    that aren't marked up in the HTML. (Or editing the auto-generated
    metadata file directly but I would try to avoid this, as it will be
    hard to tell apart the automatic and manual parts.)

> Yes, thanks, I forgot about that.  My categorization isn't great, and I
> would be grateful for help with that.

Should the categories be visible in the SRFI document itself? They could
even be links to category pages (auto-added by e.g. Ciprian's script).