[rfc-i] Remove some requirements on element order

Julian Reschke julian.reschke at gmx.de
Thu Jan 2 09:54:47 PST 2014


On 2014-01-02 18:16, Nico Williams wrote:
> On Thu, Jan 2, 2014 at 3:19 AM, Julian Reschke <julian.reschke at gmx.de> wrote:
>> Proposal: document that the element order is irrelevant, but think about
>> something new that might address the original use case (properly formatted
>> postal addresses).
>
> +1.
>
> One possibility is a new element to describe the order in which the
> others are to be taken -- that strikes me as odd.  Another would be to
> use a single element with pre-formatted address text, or multiple
> nodes denoting nothing more than "this on a line by itself in the
> order in which it appears".  I can also imagine an element indicating
> the sort of address the address is, allowing a database of address
> formatting instructions to be used (this strikes me as far too
> complex).

What I had on my mind is something like:

<postal>
...
<addressline><code>48155</code> <city>Münster</city></addressline>
...
</postal>

<addressline> would indicate that the contents is to be formatted as a 
single line, and that order and whitespace inside is significant. But 
maybe what we have is simply good enough.

> Another possibility is to dispense with proper postal addresses
> completely, instead documenting only the most general facts about
> author location that are of potential use to RFC readers: e.g., city,
> country, timezone, things that are useful when arranging meetings.
> I've yet to receive postal mail about any I-Ds or RFCs that I've
> authored, but I have received e-mail.  In fact, the most serious

I've seen at least one case where the RSE tried the postal address to 
hunt down an author during AUTH48.

> problem I've had w.r.t. author address/contact information has been
> changing e-mail addresses, so perhaps RFC authors should get a
> forwarding address at suitable RFC-Editor or IETF domain (e.g.,
> rfc7000-authors at rfc-editor.org, just like we get for I-D authors
> @ietf.org already).

That doesn't help if people don't care enough to update that information...

Best regards, Julian


More information about the rfc-interest mailing list