[boost] Re: Slight revision to more-I/O

From: Ed Brey (brey_at_[hidden])
Date: 2003-07-22 08:25:42

Daryle Walker wrote:
> On Saturday, July 19, 2003, at 3:30 AM, Gennaro Prota wrote:
> 
>> --- Daryle Walker <dwalker07_at_[hidden]> wrote:
>> 
>>> Actually, I don't think the Review Manager ever gave a final answer.
>>> Well, he and everyone else can take a quick look.  It seems that
>>> some people around here are jumpy for a release, so we should get
>>> this review resolved soon.

This is correct.  Technically, the review is still ongoing.  Enough valid concerns were raised to hold off on acceptance, yet it seemed likely enough that the concerns could be overcome that it would be best to let the discussion run its course.  This disappointing part has been the slowness of the discussion.

>> Actually I was under the impression that most of the stuff was
>> rejected (I gave up following the review because most of my comments
>> 
>>   http://article.gmane.org/gmane.comp.lib.boost.devel/17500
>> 
>> were simply ignored, but I didn't see very positive comments either).
>> Am I wrong? Did everybody like this?
> 
> I think there were a couple of OK votes and a bunch of OK-with-fixes /
> neutral / need-fixes votes.  A lot of the changes in this revision
> were based on reviewer comments, that's why I want everyone to take a look.
> 
> I use the (several-)daily-digests for this list, so it's bothersome to
> write up replies.  I'm not going to individually respond to a lot of
> messages with overlap.  The reviewer did post summary questions that I
> did respond to.

As reviews go, this one has been relatively small.  The number of reviewers and quantity of issues raised is reasonably low.  I would strongly encourage directly responding to each point of each reviewer, even if some responses are cross-references to other postings or a FAQ entry.  I sympathize with Gennaro's frustration in feeling ignored.  No reviewer wants to feel like he has invested time and effort which has gone to waste.

I've tried to fill the dangling thread gap by summarizing open issues, but that really shouldn't be necessary.  Daryle, if you haven't already, you might consider accessing the list via NNTP through Gmane, which allows a good newsreader to track threads and let you flag on-topic messages.  This allows you to view only messages that are part of the review and quickly find related questions and answers for cross-referencing (as opposed to the digest where you are stuck with unrelated boost messages abutted to the ones you care about).

As to the reviewers votes, there were a couple OKs, a no, and a couple "I need more info to vote".  I found the issues raised in the latter quite persuasive.  The question of whether certain features of the library had justified existence was beginning to be answered, but it went to sleep here:
http://article.gmane.org/gmane.comp.lib.boost.devel/20871

I would like to see those question be completely resolved and documented, and then get updated opinions from the wait-and-see reviewers.  Secondarily, there were more detailed questions raised by Gennaro and Deitmar should then be addressed.

We all appreciate the volunteer effort you are putting into the submission, Daryle.  I'm encouraged that you'll be able to drive the library addition the last few steps necessary to ensure it is up to the high Boost standards.

Ed Brey
Review Manager for More-I/O Library enhancements