[vysper] compliance with bis revisions of RFCs

[Gerolf Seitz]

hi,

i was wondering whether we should conform to the original RFCs or the
latest bis revisions (and somehow treat the bis revisions as approved RFCs)?
as we're in labs, we could go with the 'bleeding edge' ;)

however, the order of the sections have changed compared to the original
RFCs.
eg. resource binding: RFC3920:7 vs. RFC3920bis-04:8

so i'd rather stick to the original (exclusive) or to the latest bis
revision, to avoid
confusion as to which revision is used, etc...

my vote would be +1 for the latest bis revision :)

wdyt?

  Gerolf

[Bernd Fondermann-2]

On Mon, Apr 21, 2008 at 3:22 PM, Gerolf Seitz <gerolf.seitz@...> wrote:

> hi,
>
>  i was wondering whether we should conform to the original RFCs or the
>  latest bis revisions (and somehow treat the bis revisions as approved RFCs)?
>  as we're in labs, we could go with the 'bleeding edge' ;)
>
>  however, the order of the sections have changed compared to the original
>  RFCs.
>  eg. resource binding: RFC3920:7 vs. RFC3920bis-04:8
>
>  so i'd rather stick to the original (exclusive) or to the latest bis
>  revision, to avoid
>  confusion as to which revision is used, etc...
>
>  my vote would be +1 for the latest bis revision :)

this has my sympathy. yet I don't know what that buys us. 3920f is
what all the clients and servers out there are coded against.
We _have_ to comply to them if we want to be interoperable. The diff
is not too big I think, so that should be no big deal complying to
both, RFCs and BISs.

  Bernd

---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@...
For additional commands, e-mail: labs-help@...

[Gerolf Seitz]

On Mon, Apr 21, 2008 at 6:19 PM, Bernd Fondermann <
bernd.fondermann@...> wrote:

> this has my sympathy. yet I don't know what that buys us. 3920f is
> what all the clients and servers out there are coded against.
> We _have_ to comply to them if we want to be interoperable. The diff
> is not too big I think, so that should be no big deal complying to
> both, RFCs and BISs.
>

i think the biggest plus is that the bis revision may have a more
finegrained structure
or at least better examples and some clarifications.

what i'm gonna do is stick to the bis revision, and after a feature (or
parts of a feature)
is somehow finished, check back with the original rfc.

  Gerolf

[Bernd Fondermann-2]

On Mon, Apr 21, 2008 at 7:27 PM, Gerolf Seitz <gerolf.seitz@...> wrote:

> On Mon, Apr 21, 2008 at 6:19 PM, Bernd Fondermann <
>  bernd.fondermann@...> wrote:
>
>  > this has my sympathy. yet I don't know what that buys us. 3920f is
>  > what all the clients and servers out there are coded against.
>  > We _have_ to comply to them if we want to be interoperable. The diff
>  > is not too big I think, so that should be no big deal complying to
>  > both, RFCs and BISs.
>  >
>
>  i think the biggest plus is that the bis revision may have a more
>  finegrained structure
>  or at least better examples and some clarifications.
>
>  what i'm gonna do is stick to the bis revision, and after a feature (or
>  parts of a feature)
>  is somehow finished, check back with the original rfc.

+1, nothing exceptional new in the BIS, so retrofitting to the RFC
should be doable.

referencing the bis in the compliant annotation could be backfiring
when there are structural changes made in later revisions.

  Bernd

---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@...
For additional commands, e-mail: labs-help@...

[Gerolf Seitz]

On Mon, Apr 21, 2008 at 9:42 PM, Bernd Fondermann <
bernd.fondermann@...> wrote:

> +1, nothing exceptional new in the BIS, so retrofitting to the RFC
> should be doable.
>
> referencing the bis in the compliant annotation could be backfiring
> when there are structural changes made in later revisions.
>

agreed, so let's reference the original rfc with optional comments about
bis-specific details,
if even necessary...

  Gerolf