Re: Merging the interfaces and algorithms sections

Hi MarkFo, Dominik, All,

> On 21 Jan 2015, at 02:21, mark a. foltz <mfoltz@google.com> wrote:
> 
> The pull request LGTM.  Am I to assume that all the paragraphs classed "XXX" require expansion to complete this iteration of the spec?

In some cases simply removing the class is probably what is needed, while some require more expansion. The most obvious extensions points were annotated with ellipses (...). Consider the PR a first stab. If the new style resonates with the group, this work-in-progress PR can be used as a starting point for aligning the spec style.

>From the process point of view, we can merge the PR even if there are unspecified aspects. IOW, we do not need to make everything with "XXX" perfect before we merge. I'd probably try to get rid of the "XXX" classes and convert those aspects of the spec that cannot be specified yet into open issues in the spec, perhaps linking back to corresponding GH issues.

[In a perfect world, we could automatically keep the GH issues in sync with the spec. Perhaps the GitHub API could be used to pull in the open issues from GH issues into the spec at spec build time, inserted into appropriate locations, automatically linking back to the GH issue. Not sure if Anolis (the tool the group uses for spec authoring) has such a feature planned.]

Dominik indicated he's working on the first pass review and expansion. My expectation is he will share his results with the group shortly, and before doing so will update the PR, to ease the review.

Thanks,

-Anssi

Received on Wednesday, 21 January 2015 07:51:14 UTC