[Ietf-krb-wg] A proposal from Adrian on the des deprecation registry

[Sam Hartman <[e-mail]>]

IANA's question about whether we should  deprecate code points in the
registry came up during IESG discussion of
draft-ietf-krb-wg-des-die-die-die.
It looks like our discussion is sufficient and we will be able to move
forward with no change if we like.

However, since a specific proposal was made  I'd like to forward it to
the WG in case people start shouting for joy.
we could potentially make the change now if the document doesn't end up
getting approved today and people are very supportive.
Or we could make this with some-numbers-to-iana.

--- Begin Message ---

Hi again,
 
(Apologies for html email, but the layout of the registry columns works better
in a non-proportional font)
 
> >> [1] https://lists.anl.gov/pipermail/ietf-krb-wg/2012-April/010096.html
> >
> > The thread seems to talk about the addition of a column on the intended
> > "required to implement" level. I would, as I said in my previous mail, not
think
> > that appropriate for a registry.
> >
> > "Deprecated" in a registry is a different thing. It describes the use of the
> > code point, not the use of the thing the code point indicates.
> 
> So, I'm now unsure as to what's needed to clear the discuss.
 
I think I am going to clear on the basis that Discussion has been had.
 
> They thought about modifying the registry but decided not to
> add a status column that would say deprecated which is what I
> thought you were asking. I think Sam's mail summarised the (not
> very energetic;-) discussion nicely. [1]
 
Nah, that wasn't what I was suggesting. I was not after such a sweeping change,
and I see why you are nervous of it. I was simply suggesting deprecating the
code points for the deprecated mechanisms.
 
I was thinking that the registry might change as
 
OLD
       etype     encryption type              Reference
         0       reserved                     [RFC6448]
         1       des-cbc-crc                  [RFC3961]
         2       des-cbc-md4                  [RFC3961]
         3       des-cbc-md5                  [RFC3961]
NEW
       etype     encryption type              Reference
         0       reserved                     [RFC6448]
         1       Deprecated                   [RFC3961], [This.I-D]
         2       Deprecated                   [RFC3961], [This.I-D]
         3       Deprecated                   [RFC3961], [This.I-D]
 
Thanks,
Adrian
 
> If there's something else needed can you say what? E.g. if
> you're saying "you must add that column even though the WG
> didn't want to" it'd be better to be clear about that.
> 
> The danger with doing that is as noted in Sam's mail above:
> it'd maybe open a can of worms about the status values to
> put in for a bunch of other things. (Yuk - endless process
> discussions that affect no code, unlike this draft;-)

--- End Message ---

_______________________________________________
ietf-krb-wg mailing list
[e-mail]
https://lists.anl.gov/mailman/listinfo/ietf-krb-wg