[rfc-i] Anchor Naming character set

Julian Reschke julian.reschke at gmx.de
Sat Dec 28 06:05:07 PST 2013

On 2013-12-27 22:25, Jim Schaad wrote:
> One issue that should be addressed for both v2 and v3 is the fact that
> we are  using the type ID as the anchor and then using the anchor as the
> bullet string for references.  This means that there is a new
> restriction placed on anchor strings in v2 that was not present in v1.
> Specifically that the anchor strings cannot start with numbers.

That's not a new restriction; it's just that v1 didn't properly check 
the input.

> A couple of different ways exist to deal with this.
> 1.Vanity naming of some type (with the appropriate vocabulary change)


> 2.Prefixing the anchor string with an underscore is to be treated
> specially in producing the bullet string (specifically remove it)

That's what I proposed as a workaround as well, but it appears nobody 
liked it. When we introduce "vanity" names the problem will be gone anyway.

> This is currently a problem with the 3GPP and 802.11 specifications
> unless you want to prefix them with something like SDO.802.11 or IEEE.802.11

Best regards, Julian

More information about the rfc-interest mailing list