Re: [nm-wg] NMWG: hierarchy doc revisited

[Dan Gunter <dkgunter@lbl.gov>]

Looks nice.

I have a couple of minor comments on section 4.4, the textual 
representation of characteristics

(1) As this is its first mention, DAMED should also be spelled out: 
Discovery and Monitoring Event Descriptions

(2) Although the examples help, the format

<entity-type>.<characteristic>.<sub-characteristic>

would be clearer, I think, if there was a bit of added text:

"There may be zero or more sub-characteristics."

(3) Despite the tedium, I think the official DAMED name for each 
characteristic in the document should be added in an appendix

(4) I think that the capitalization conventions should be mentioned, e.g.:

"Names are all lowercase letters (ASCII [a-z]), except that the 1st 
letter of each word after the first is capitalized (ASCII [A-Z])."

- Dan

Bruce Lowekamp wrote:
> As many of you are aware, we submitted the hierarchy document to the 
> GFSC this past summer.  We got some comments back recently, and two 
> issues were raised:
>
> - what type of document should this be
> - normative text
>
> After discussing the issues somewhat, we concluded that ultimately we 
> want the hierarchy to be a recommendation.  Obviously, as a nomenclature 
> it isn't enough to implement measurement portability, but it is 
> something that multiple schemas can be developed with, and sharing the 
> nomenclature should facilitate usability.  So eventually, we hope that 
> other products beside the NMWG schema can say "we use this 
> terminology."  To have that statement mean anything, it needs to be a 
> recommendation.
>
> Therefore, we revised this as a "proposed recommendation."  Essentially, 
> this means that it is on the track to be an official GGF recommendation. 
> At this point, what we are claiming is that we have general consensus 
> that this makes a good recommendation, but need implementation 
> experience and more review to ensure that it is correct and useful.
>
> After making that decision, the normative text was easy.
>
> The intro has the most revisions, and there are a variety of other 
> sections that have normative text inserted.  There is also a new section 
> explaining the DAMED method of converting the hierarchy into 
> dotted-decimal names.  We may also want to include a list of mandatory 
> and optional attributes for characteristics at a later point, but right 
> now I don't think we have enough perspective to do that in a way that 
> will be equally applicable to all measurements.
>
> I would like to say that there is a version available with changebars, 
> but there is not.  I foolishly assumed that word would be able to 
> "compare documents" at the end of the process, but apparently "diff" is 
> beyond the grasp of microsoft programmers.  Instead, both the old (00) 
> and new (01) version are on the webpage.  (http://www-didc.lbl.gov/NMWG/)
>
> As the changes are relatively minor, I'm hoping we can resubmit this to 
> the GFSC quickly.  Therefore, I would like to ask for feedback by 
> January 6. If we have not heard anything by then, or the comments are 
> minor, we will resubmit to the GFSC at that point.  If there is a 
> significant amount of discussion, or if people feel more time is 
> required, we'll postpone submission until consensus can be achieved.
>
> Bruce Lowekamp