Re: I-D Action: draft-carpenter-6man-rfc6874bis-00.txt

[Andrew Cady <[e-mail]>]

On Tue, Jul 06, 2021 at 09:33:11AM +0200, Nico Schottelius wrote:
> 
> While copy & paste won't work with this proposal, I believe the %25
> escape should be "good enough" for most daily usage.
> 
> I think we don't need to aim for a "perfect solution", which we will not
> be able to achieve anyway, but a solution for reality.

The perfect solution is just the simple one of not introducing the
anti-feature of name-mangling.  It is simpler NOT to add that to Firefox
than to add it.  And it is more correct.  For the same reason!

> In a perfect world, the [...%] syntax would have been introduced (and
> accepted) prior to % escape in URLs.

It doesn't matter that '%' is also an escape character in some of
the URI data segments.  Percent-decoding is done AFTER parsing into
segments.  We don't have to worry about implementations accidentaly
decoding segments that aren't even allowed to be encoded.

So that does not conflict with the use of a % as a delimiter ending a
data segment (such as IPv6-Literal) that is NOT percent-encoded.

It would only conflict if the IPv6-Literal field allowed % encoded
values!

The whole problem is imaginary.