Re: [xrblock] Discussion about common boilerplate, order to approach the drafts

Hi,
Sorry for late feedback.
  ----- Original Message ----- 
  From: David A. Bryan 
  To: Glen Zorn 
  Cc: xrblock@ietf.org 
  Sent: Monday, September 12, 2011 9:59 PM
  Subject: Re: [xrblock] Discussion about common boilerplate, order to approach the drafts

  Inline.

  On Fri, Sep 9, 2011 at 11:16 PM, Glen Zorn <glenzorn@gmail.com> wrote:

    On 9/10/2011 7:07 AM, David A. Bryan wrote:

    > Hi all,
    >
    > So with the change in chairs and the desire to start spinning up and
    > getting some work moving in this group now that the monitoring
    > architecture draft is nearing completion, Charles and I wanted to start
    > a few discussions. There are two discussions we really would like to
    > have in order to get things rolling.
    >
    > First, in talking with Roni and Peter, one thing that came up was that
    > it was possible that many of the drafts in this group -- both the
    > current milestones and the new individual drafts -- could share some
    > common text. This would be some explanation of the problem, basic
    > structure of the new report blocks, etc.
    >
    > So, what do folks think? Is there a set of common text that we could
    > work to quickly abstract out and agree on, which would then make the
    > individual report blocks simpler to write and review? Are they too
    > different from one another to do that? What things would we want in any
    > possible common text? Should we quickly work on this common text before
    > we work on the drafts?

    This seems like a pretty much total waste of time to me.  Many protocol
    documents have this characteristic (PPP, EAP, ...); the time-honored
    solution for people beginning new EAP drafts, for example, is to
    cut-and-paste the relevant text from an existing document.  I see no
    reason why this approach is invalid for xrblock drafts.

  Anyone else have thoughts? 

  [Qin]: I agree and understand what Glen said. Frank to say, it is more straightfoward to take on
  some works to catch up milestone rather than building comming boilerplate.
  The similar boilerplate I can point to you is RTP Payload Format Template defined in draft-ietf-avt-rtp-howto-06.
  I am not sure this is similar thing you are looking for. However I think comparing to RTP Payload Format,
  XR Block Format is more simple forwat, does it worth taking time to build a common boilerplate template?
  Why not just follow the rules or guideline defined in RFC5968, draft-ietf-pmol-metrics-framework  and 
  draft-ietf-avtcore-monarch and go ahead directly.

    >
    > Second, we have 11 milestones, which other than Measurement Identity,
    > are in alphabetical order
    > (http://datatracker.ietf.org/wg/xrblock/charter/). We also have a number
    > of new drafts that have been submitted as individual drafts. While this
    > kind of discussion is always likely to be contentious, and often the
    > answer is "my draft", we would like to discuss which ones people think
    > are most important. This could be because others will depend on it, or
    > because we simply think users need these first. We'd like to have this
    > discussion to re-prioritize the milestones and get to work on some of
    > the drafts.
    >
    > So...comments please. The more we can discuss on list and resolve
    > quickly, the more time we can spend getting the drafts completed.

    We have been "getting ready to do something" for some months now.  How
    about this: instead of continuing this, just _do_ something?  For
    example, none of those 11 milestones have any documents attached to
    them; how about we do that?  Or at least start mapping existing
    documents to milestones?  BTW, if nobody's interested in submitting
    drafts related to the official milestones, this says a lot about the
    milestones themselves...

  ...hence this discussion. We can throw out milestones if no one is interested. I'm a bit concerned that Glen is the only one who responded. Others? Which (if any) of these milestones are people interested in continuing? Simply blindly working on 11 milestones in the (alphabetical) order they were selected months ago seems counter-productive, so what do people think?

  [Qin]:  It looks  we can take out Measurement Identity milestone from the charter, since following the discussing on the monitoring architecture, there is rough consesus to use SDES packet to carry addional identity information.

  Another suggestion is we already have had more than 11 draft documents that is available in XRBlock and AVT Working group repository. Why not take action by mapping them one by one and determine which one is in high priority and which is low.

  Also as an author of quality monitoring draft, I have split it into four different draft to get align with the rules defined in monitoring archtiecture document. I think it is under enough review and discussion in the past two IETF meeting and it is ready for chair review and moving forward.

  David

    >
    > Thanks,
    >
    > David and Charles (as chairs)
    >
    >
    >
    >

    > _______________________________________________
    > xrblock mailing list
    > xrblock@ietf.org
    > https://www.ietf.org/mailman/listinfo/xrblock

------------------------------------------------------------------------------

  _______________________________________________
  xrblock mailing list
  xrblock@ietf.org
  https://www.ietf.org/mailman/listinfo/xrblock