[cabfpub] C=GR, C=UK exceptions in BRs

Kirk Hall Kirk.Hall at entrustdatacard.com
Fri Mar 17 17:09:29 MST 2017


Yes, I know there was a reason given.  All I can tell you, Peter, is we have a number of customers in the United Kingdom who very much dislike the fact that their certs say “GB” – they consider it incorrect, and have called us on it from time to time.  But I guess they are stuck with it unless they successfully petition ISO for a change.  From what Dimitris has said, there may be a conflict between what the Greek government thinks the country code should be, and what ISO decided some time ago – in general, I think a country should be able to decide that for itself.

Standards are important, but it’s my recollection that some browsers have chosen to deviate from standards including IETF standards, even RFC 5280, from time to time for what they believed was a good reason (instead of trying to change the standard).  We should be consistent on our approach.

From: Peter Bowen [mailto:pzb at amzn.com]
Sent: Friday, March 17, 2017 4:58 PM
To: CA/Browser Forum Public Discussion List <public at cabforum.org>
Cc: Dimitris Zacharopoulos <jimmy at it.auth.gr>; Kirk Hall <Kirk.Hall at entrustdatacard.com>
Subject: Re: [cabfpub] C=GR, C=UK exceptions in BRs

Kirk,

ISO 3166 was apparently created in 1974 to provide a distinct two letter code for each country.  From their own FAQ (https://web.archive.org/web/20120616044022/http://www.iso.org/iso/country_codes/iso_3166-faqs/iso_3166_faqs_specific.htm<https://web.archive.org/web/20120616044022/http:/www.iso.org/iso/country_codes/iso_3166-faqs/iso_3166_faqs_specific.htm>):

Why is the United Kingdom (of Great Britain and Northern Ireland) coded GB in ISO 3166-1?

The codes in ISO 3166-1 are - wherever possible - chosen to reflect the significant, unique component of the country name in order to allow a visual association between country name and country code. Since name components like Republic, Kingdom, United, Federal or Democratic are used very often in country names we usually do not derive the country code elements from them in order to avoid ambiguity. The name components United and Kingdom are not appropriate for ISO 3166-1. Therefore the code "GB" was created from Great Britain and not "UK" for United Kingdom. Incidently, GB is also the United Kingdom's international road vehicle distinguishing sign - the code on the oval nationality stickers on cars.
Given that neither the EU or Greece has requested EL to be reserved for any purpose (https://www.iso.org/obp/ui/#iso:code:3166:EL) I don’t believe it makes sense to use this in the country code field.

Thanks,
Peter


On Mar 17, 2017, at 3:55 PM, Kirk Hall via Public <public at cabforum.org<mailto:public at cabforum.org>> wrote:

Ryan makes a good point – where there is a conflict between local law or practice (or desired practice) and the BRs, the best first step is to amend the BRs to allow compliance with local law or practice (or desired practice).

As I recall the country codes we are all stuck with were created in the 1960s for a purpose unrelated to SSL and digital certificates.  There must have been a good reason for representing the United Kingdom (England, Wales, Scotland (for now), and Northern Ireland) as “GB” when Northern Island (part of the UK) is not in Great Britain and UK is the more generally known acronym for the United Kingdom – but I can’t imagine what the good reason was.

Instead of a ballot that presents a sweeping new structure for country names, or points to another new document, maybe we just create an Appendix to the BRs that allows different country codes for Greece and the United Kingdom (as an alternative).  We would endorse such a ballot.

From: Public [mailto:public-bounces at cabforum.org] On Behalf Of Ryan Sleevi via Public
Sent: Friday, March 17, 2017 1:12 PM
To: Dimitris Zacharopoulos <jimmy at it.auth.gr<mailto:jimmy at it.auth.gr>>
Cc: Ryan Sleevi <sleevi at google.com<mailto:sleevi at google.com>>; CA/Browser Forum Public Discussion List <public at cabforum.org<mailto:public at cabforum.org>>
Subject: Re: [cabfpub] C=GR, C=UK exceptions in BRs



On Fri, Mar 17, 2017 at 4:08 PM, Ryan Sleevi <sleevi at google.com<mailto:sleevi at google.com>> wrote:
On Fri, Mar 17, 2017 at 3:01 PM, Dimitris Zacharopoulos <jimmy at it.auth.gr<mailto:jimmy at it.auth.gr>> wrote:
The "spirit" of 9.16.3 is also to bring conflicting requirements to the CA/B Forum to consider possible revisions accordingly. This is exactly what I am doing, without violating the current BRs, but hoping that the CA/B Forum will read this as a conflicting requirement which could be resolved by adding a simple exception, without creating any risk in current practices.

For what it's worth - I agree with this sentiment, and it's worth considering, separate of 9.16.3, whether to _revise_ the BRs to accomodate this case. Such revisions must account for ambiguity. In many ways, the BRs strive to eliminate the rampant ambiguity that existed due to CAs' various practices, as a whole (since no two CAs really have the same CP/CPS), and so we should strive, as much as possible, to unambiguously represent the information that members see as valuable.

Of course, it might be that identity information in certificates is not valuable, precisely because of ambiguities and conflicts that naturally emerge from CAs. In that case, it might be worthwhile to simply stop trying to represent identity information within certificates, and accept that ambiguity, rather than try to carve it up. However, since the Forum values identity information at present, it makes sense to opt for strictness as much as possible, or to explicitly describe the deviations permitted and assess their risk, as you propose doing and is worth at least discussing :)
_______________________________________________
Public mailing list
Public at cabforum.org<mailto:Public at cabforum.org>
https://cabforum.org/mailman/listinfo/public

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cabforum.org/pipermail/public/attachments/20170318/17dc9cdd/attachment.html>


More information about the Public mailing list