Re: [rfc-i] archiving outlinks in RFCs

[Stephen Farrell <[e-mail]>]

Hiya,

On 03/05/2023 00:41, Brian E Carpenter wrote:
> 
> In case it wasn't obvious, that was an advantage of the proposal to
> treat broken URLs as errata - there is already established policy.

I'm not sure that there is really a fully established policy.
I can imagine many kinds of "broken" where it might not be so
easy to figure out if a change is good or bad.

As a pretty random example, RFC5165 has a link to [1], and if
I go there (with my FF setup), I get an interstitial warning
about http:// and, after a click, am then re-directed to [2].
Without reading the RFC and the text at [2], I've no idea if
that warrants an erratum, nor what kind.

That's why I'd argue we'll be better off if the 1st step here
is an analysis of what kinds of "broken" we're seeing, which
of those may seem easy to handle, and which not. ISTM that
analysis will already be a pile of work, and I'd like to see
what that produces before going further. (And I really hope
we don't autogenerate thousands of errata based on stuff like
this example:-)

Cheers,
S.

PS: The "random" example was just the 1st semi-interesting
one I stumbled over based on locally stored RFCs and playing
about for 5 minutes.

[1] http://urn.opengis.net/
[2] https://defs.opengis.net/vocprez/

> 
> A sub-question is whether we treat http: URLs as broken if they have
> been replaced by https:. There will likely be many of those.