Re: [boost] GIL Review

From: Joel de Guzman (joel_at_[hidden])
Date: 2006-10-26 01:40:21

Jeff Garland wrote:
> Joel de Guzman wrote:
>> Tom Brinkman wrote:
>>>>> Thank you! With all due respect to Dr. Reese, I think posting a
>>>>> review and then asking to be emailed for replies is not right.
>>>>> The review is a public affair. We would also like to read about
>>>>> the replies and exchanges that ensue after a reviewer posts his
>>>>> review. A reviewer should also be responsible to answer and reply
>>>>> to the questions and answers related to his review _on_list_ in
>>>>> as much as the one being reviewed (Lubomir et. al.) tries as best
>>>>> as they can to answer and reply to the reviews. It's not a one
>>>>> way street.
>>> Ok, I kinda agree. However, there may have be a reason that they were
>>> not able to. Not shur why that would be, but in any case I told him
>>> that he could email me the review if he was unable to post it the
>>> group himself.  In the future, I'll be firmer and polightly insist
>>> that they try join the mailing list.
>> Understood. Thanks!
>>
>> Regards,
> 
> 
> Well, before we go overboard, according to the process, private reviews 
> are allowed - from:
> 
> http://www.boost.org/more/formal_review_process.htm
>      Boost mailing list members are encouraged to submit Formal Review
>      comments:
> 
>          * Publicly on the mailing list.
>          * Privately to the Review Manager.
> 
> There are valid reasons for some folks to submit private reviews.  Not 
> sure that's exactly the case here, but we shouldn't just assume 
> everything is done in public....even though the vast majority of reviews 
> are public.  Overall, I can understand why someone might not want to 
> join another mailing list, so on balance, I'd hate to see an useful 
> review like this lost b/c of someone being forced to join the list.

I don't think that's the case here.

In this case, the review is meant to be _public_ and forwarded through
the Review Manager. The review, being public, has an effect on other
participants in the review. In this case, the said review should
be subject to public discussion which ultimately should involve
the reviewer. If the review is indeed meant to be private, then
there's no point in posting it to the list.

I find private reviews kinda odd though. Imagine a case with all
favorable public reviews but with more private unfavorable reviews.
It would be surprising if the outcome would be negative. How
would the folks being reviewed have a chance to defend the library
against unfavorable private reviews? How would the review manager
share its decision points? Keep in mind that it typically takes
years to craft a good library and a reviewer typically spends, at best,
a few days studying it. It is not fair not to give the authors a
chance to defend themselves. There's a big probability that the reviewer
missed some things important that only the library authors can
point out. All this will not happen with private reviews.

I move to remove private reviews from the review process.

Regards,
-- 
Joel de Guzman
http://www.boost-consulting.com
http://spirit.sf.net