Three possible uses for IRIs

[colecmac at protonmail.com]

> 3) Percent-encoding: find non-ASCII characters and convert them to %nn%nn,
> or %nn%nn%nn, or %nn%nn%nn%nn sequences, where nn is two hex digits.

One extra thing: Gemini will need it's own list of reserved characters.
The URI spec defines[1] this list:

gen-delims  = ":" / "/" / "?" / "#" / "[" / "]" / "@"


(It also defines a list called sub-delims, but that only applies to query
strings I believe, and is irrelevant to the way Gemini uses them.)

These characters are reserved because of their use in other parts of
a URI. But Gemini does not use all those parts, such as userinfo. I
believe a reserved character list for Gemini could look like this:

":" / "/" / "#" / "?" / "[" / "]"

I left fragments ("#") in, so that clients can add support for them later,
if/when a header-to-fragment algorithm is defined, like exists for Markdown.
But that character could be removed too, which would prevent it ever being
used in that manner.

1: https://tools.ietf.org/html/rfc3986#section-2.2


makeworld