RE: New chapter 7 editor's draft

Larry,

Comments in line below.



Steve Zilles



-----Original Message-----
From: Larry Masinter [mailto:masinter@adobe.com]
Sent: Saturday, August 10, 2013 9:02 AM
To: Charles McCathie Nevile; public-w3process@w3.org
Subject: RE: New chapter 7 editor's draft



Here are some specific suggestions:



7.2.2 Wide Review



OLD "The requirements for wide review are not precisely defined by the process."



This sentence at the beginning is an invitation to ignore the rest of the section. In particular, Working Groups cannot unilaterally redefine "Wide Review" to mean "anyone who attends our meetings".   Suggestion:



NEW: "While the term 'wide review' is not precisely defined, the following guidelines lay out some requirements."



SZ: The W3C AB is concerned about balancing the need for guidance as to how to accomplish a required task, such as Wide Review, versus establishing requirements that can be used by a small group to impede work for which there is a larger consensus or, conversely, that can be used by a group to avoid reaching out to and responding to sub-communities that are clearly relevant. For that reason, the section in question does not specify that the guidelines are requirements. It is left to the Director to decide if the intended objective is met. In that sense the guidelines are neither necessarily sufficient to show the task is met nor can one use failure to meet all the guidelines to delay progress if there is other evidence (not necessarily per the guidelines) that would show that the task was completed.



=============



OLD " the entire set of stakeholders of the Web community, including the general public "



We may give priority to the Web community, but the term 'Web community" isn't itself well defined.



Let's take XML as an example. XML has many stakeholders who are not considered to be central to "the Web community".



Would an update to XML which is acceptable to the Web community (browser makers, web developers, servers, authoring tools), but hasn't been reviewed at all by any other XML users be considered to have had "wide review" ? I think doing so is irresponsible.

” have had adequate notice of the progress of the Working Group and thereby an opportunity to comment on the specification"



"Adequate notice" and "opportunity" are insufficient. Like voting rights in disputed elections, some working groups make commenting on a specification a gauntlet of make-work. Sure there is "opportunity", but you must fill out a bug report, wait for its rejection, file an objection, etc. etc.



Each specification has an intended set of applications and other specifications which make use of it, and a set of roles for each of those applications. What is important for "wide review" is whether the specification has been adequately reviewed by implementors and user of software and services that fulfill those roles.



Distinguish new specifications and updates of existing specifications: a new spec/recommendation can define its own set of applications and roles. An update to an existing specification should be reviewed by the communities who are using the specification for suitability.



(Perhaps this is what was intended by "dependencies with other groups"



I'd argue for asking Chairs and working group members to actively solicit comments and reviews from known application communities, and to identify those communities and applications in the draft, in "General requirements for Technical Reports" .



For example, add something like:



NEW:

     * MUST identify the general range of applications for which the technology is intended, and the roles of participants in those applications.



Then, in "wide review", change to



NEW:

    "The objective is to ensure that the specification has been carefully reviewed for applicability to the range of applications and roles identified, and that the stakeholders of the community, including the general public, have had adequate notice of the progress on the specification and an opportunity to analyze and comment on the specification."



SZ: The Advisory Board accepted your contention that the use of “Web Community” might artificially restrict the scope of review. The AB noted, however, that requiring “careful review for applicability to the range of applications and roles identified” would be a new requirement on Working Groups and would be one that may be very difficult to satisfy. For example, there are many uses of XML that are unknown to the participants of the Working Group, even from a role point of view. It is also not clear how to carry out the task of finding all the “roles.”  The AB, therefore, suggests changing the wording of the sentence in question to,



NEW

The objective is to ensure that the entire set of stakeholders of Web Standards, including the general public, have had adequate notice of the progress of the Working Group and thereby an opportunity to comment on the specification



This changes the scope from the “Web Community” to all the users of Web Standards, whether part of the Web Community (however defined) or not.



======

OLD

"A Working Group could present evidence that wide review has been received, irrespective of solicitation"



I understand the reluctance to put recalcitrant groups in the critical path of getting something out, but evidence of review by the range of application implementors is very useful, but this sentence makes it seem unwanted.  With a good definition of "wide review", it includes solicitation and receipt. So:



NEW

"A Working Group should present evidence of solicitation and receipt of wide review."



SZ: The AB did not discuss this suggestion due to time limitations. So, the following is only my personal comment. If your suggestion is adopted, the point of the last paragraph of 7.2.2 with which the sentence you suggest be changed begins is essentially reversed. The original point was (a) to allow people to show wide review had occurred even if solicitations were deficient and (b) to point out that lots of comments is not equal to wide review. The intent was to leave some latitude to the Director in deciding if wide review had occurred and, as a Director’s decision of progressing a document, this decision could be appealed if AC Members felt that the Director’s decision were incorrect. The Process is set up to be flexible within reasonable bounds, but has “checks” (the appeal process)  to insure that unreasonable actions can be challenged.