Re: Requirements editing

+1

If the main justification for the JSON file is that it supports a complex 
hierarchy, then flattening it will let us use the wiki and edit as wiki 
text or html. 
_________________________________________________________
Arthur Ryman
Chief Data Officer
SWG | Rational
905.413.3077 (phone) | 416.939.5063 (cell)
IBM InterConnect 2015




From:   Holger Knublauch <holger@topquadrant.com>
To:     RDF Data Shapes Working Group <public-data-shapes-wg@w3.org>
Date:   12/11/2014 02:37 AM
Subject:        Requirements editing



Thinking about how to capture our requirements, I am looking at

     http://www.w3.org/TR/ldp-ucr/#requirements

and see that the LDP project has produced a very compact list of 
requirements - compact in the sense that each entry only has

- an id
- a title
- one sentence of prose
- links to use cases

There is a coarse-grained grouping into functional vs non-functional, 
and then an additional level of grouping done via the IDs (not sure how 
this works in detail). Rejected requirements appear in strike-through.

The JSON version of the requirements currently at

     https://www.w3.org/2014/data-shapes/wiki/Rawreqs

covers similar ground, but also has

- multi-dimensional grouping (groups are like tags)
- a refined-by hierarchy
- Example snippets in various languages

I am honestly not very optimistic that editing the JSON on the Wiki will 
work for all of us, so I wonder whether we could convert the JSON into a 
flat structure similar to the format used by the LDP group. Instead of 
having the grouping/tagging in JSON, we could use a controlled syntax or 
color-coding as shown below. Furthermore, the refined-by hierarchy could 
be expressed through the usual numeric system as shown in the Example:

     R1.3.4   There should be an example requirement
         See also: [[R1.3.4 Details]]
         Tags: Workshop, Meta, HK, PFPS
         User Stories: U42

and the example snippets would go into details wiki pages, but I believe 
many examples will be better left in the User Stories that already 
exist, so that cross-reference may be sufficient. The tree view would be 
produced by the Table of Content of the Wiki. As long as we stay in a 
controlled vocabulary template, we can always reverse engineer the JSON.

Talk to you later.

Holger

Received on Thursday, 11 December 2014 19:04:18 UTC