On Sun, Jul 03, 2016 at 03:57:31PM +0200, Kristian Fiskerstrand wrote:
On 07/03/2016 03:48 PM, Salz, Rich wrote:
I suggest we use ASN1 generalized time (which has only a couple of
bytes overhead) or ISO 8601 profiled to be like the ASN1 format:
YYYY-MM-DD-HH:MM:SS.sss with fraction optional.
ASN1 sounds complex in this case, some form of ISO8601 variant might be
helpful.
Alternatively expanding the size of the value from 32 bits to 64 bits
might be easier to deal with for backwards compatibility reasons.
I'd strongly recommend simply using a 64-bit (signed?) integer. It's
compact, avoids time zones (which are a source of bugs), and is the
logical extension.
If we use an ISO-8601 variant, it should actually conform with ISO 8601
(unlike GeneralizedTime), and provide a unique encoding for a given
time. Determinism and uniqueness help avoid collisions and other types
of attacks that might come down the line later, and makes validation
much easier.
--
brian m. carlson / brian with sandals: Houston, Texas, US
+1 832 623 2791 | https://www.crustytoothpaste.net/~bmc | My opinion only
OpenPGP: https://keybase.io/bk2204
signature.asc
Description: PGP signature
_______________________________________________
openpgp mailing list
openpgp(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/openpgp