- From: Stephen Zilles <szilles@adobe.com>
- Date: Tue, 17 Sep 2013 04:19:21 -0700
- To: "public-w3process@w3.org" <public-w3process@w3.org>
- CC: "ab@w3.org" <ab@w3.org>
- Message-ID: <CE2F61DA5FA23945A4EA99A212B15795721E4EF682@nambx03.corp.adobe.com>
Section 7.4.2 (LCCR) has must document the testing and implementation requirements AND should document known implementation (sic singular) Section 7.4.5 (REC) has must show that independent implementations of the specification are extremely likely to be highly interoperable. BUT THE EXISTING PROCESS SAYS "Show that each feature of the technical report has been implemented. Preferably, the Working Group should be able to demonstrate two interoperable implementations of each feature." Observation: neither of these indicates (clearly) that the purpose of having implementation experience is, "to show that the specification is sufficiently clear and complete to allow (interoperable) implementations of the specification." It is also desirable that specifications that go to REC are implemented. This lead to:Action-4: Draft some examples of evidence the Director would use in determining sufficient implementation experience exists. Response (in part based on what we did with "widely reviewed": A new section 7.2.3 Implementation Experience [N.B. section 7.2 is suggested because it has two other "definitions": that for "change" and that for "widely reviewed". This may not be the ideal place to put this definition because the section if entitled "General Requirements", but there seems to be no other particularly good place for this section.] The requirements for showing implementation experience are not precisely defined by the process. The objective is to show that the specification is sufficiently clear and complete that (multiple independent) implementations of each feature of the specification have been realized. In assessing that there is adequate implementation experience, the Director will consider whether there are tests that show each feature has been implemented, whether there are multiple implementations that pass these tests, whether the implementations were generated by independent implementers and whether the implementations have been publicly deployed. If multiple independent implementations have been publicly deployed for a significant period and have relatively small "bug lists", this will be considered as positive experience of interoperable implementations even if tests to that effect are missing. If the Director believes that immediate Advisory Committee review is critical to the success of a technical report, the Director may announce provisional approval of a Request for publication of a W3C Recommendation even without adequate implementation experience. It is up to the working group to present evidence that sufficient implementation experience exists and that it has met any implementation requirements it has placed on itself during LCCR. Planning to meet the requirement for implementation experience should begin early and proceed in parallel with the development of the specification. WITH THIS DEFINTION THE ABOVE BULLET POINTS BECOME Section 7.4.2 (LCCR) has must document how the requirement for implementation experience will be met. Section 7.4.5 (REC) has must document that the requirement for implementation experience has been met. FOR REFERENCE, BELOW IS THE DEFINITION OF "WIDE REVIEW" The requirements for wide review are not precisely defined by the process. The objective is to ensure that the entire set of stakeholders of the Web community, including the general public, have had adequate notice of the progress of the Working Group and thereby an opportunity to comment on the specification. Before approving transitions, the Director will consider who has actually reviewed the document and provided comments, particularly in light of the listed dependencies, and how the Working Group has solicited and responded to review. In particular, the Director is likely to consider the record of requests to and responses from groups identified as dependencies in the charter, as well as seeking evidence of clear communication to the general public about appropriate times and which content to review. As an example, inviting review of new or significantly revised sections published in Heartbeat Working Drafts, and tracking those comments and the Working Group's responses, is generally a good practice which would often be considered positive evidence of wide review. A recommended practice is making a specific announcement to other W3C Working Groups as well as the general public that a group proposes to enter Last Call in e.g. approximately four weeks, . By contrast a generic statement in a document requesting review at any time is likely not to be considered as sufficient evidence that the group has solicited wide review. A Working Group could present evidence that wide review has been received, irrespective of solicitation. But it is important to note that receiving many detailed reviews is not necessarily the same as wide review, since they may only represent comment from a small segment of the relevant stakeholder community. Steve Zilles
Received on Tuesday, 17 September 2013 11:19:46 UTC