I get the ease on development to identify an extension property, but I don’t 
view it as a showstopper.  I would change the MUST NOT to a SHOULD NOT for 
providing guidance.

--

JG

[cid87442*[email protected]]

James Gould
Fellow Engineer
[email protected]<applewebdata://13890C55-AAE8-4BF3-A6CE-B4BA42740803/[email protected]>

703-948-3271
12061 Bluemont Way
Reston, VA 20190

Verisign.com<http://verisigninc.com/>

From: "Hollenbeck, Scott" <[email protected]>
Date: Wednesday, May 21, 2025 at 3:14 PM
To: James Gould <[email protected]>, "[email protected]" <[email protected]>, 
"[email protected]" <[email protected]>, "[email protected]" <[email protected]>
Subject: RE: [EXTERNAL] [regext] Re: On bare identifiers in Extensions draft

From: Gould, James <[email protected]>
Sent: Wednesday, May 21, 2025 2:39 PM
To: [email protected]; Hollenbeck, Scott <[email protected]>; 
[email protected]; [email protected]
Subject: Re: [EXTERNAL] [regext] Re: On bare identifiers in Extensions draft

I don’t view the use of a bare identifier as being a bad practice, since the 
goal is to ensure that the identifier is registered and there are no conflicts. 
 If there are more than one property used for the identifier, then it would be 
best to include the use of the underbar and a value.  This is planned for the 
versioning extension to include “versioning_data” and “versioning_help” in 
place of the bare identifier “versioning” and the non-bare “versioning_help”.  
For the “redacted” property in RFC 9537, I see no issue with the use of the 
bare identifier “redacted” when there is only one property used.

What concrete benefit is there to disallow bare identifiers when the extension 
identifier only has a single property?  I personally don’t see a benefit that 
warrant’s a MUST NOT or even a SHOULD NOT in the extensions draft.
[SAH] One thought: it makes it easier to identify extension elements by being 
able to recognize the “_” character as a separator. When I wrote my federated 
authentication stub server, it made it a little easier to parse identifiers 
when there was a reliable character separator. Without that separator, I would 
have had to use explicit string pattern matching to separate extension elements 
from core elements. It wouldn’t have been impossible, but I would have found it 
slightly less efficient.

Scott
_______________________________________________
regext mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to