• Paul D. Anderson (5/5) Sep 06 2011 Is there an actual review queue, or is this a product of the hive mind?
• dsimcha (5/10) Sep 06 2011 Right now, the review queue is maintained very informally via
• David Nadlinger (4/8) Sep 06 2011 Must be a priority queue then… :P
• Jonathan M Davis (33/42) Sep 06 2011 There is not currently actual page which lists the items up for review.
• Brad Roberts (15/19) Sep 06 2011 I think informal works particularly well here. For several reasons, not...

Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> writes:

Is there an actual review queue, or is this a product of the hive mind?

Looking at past review queue activities it seems like someone (usually Andrei)
posts a notice when the current review is finished, or nearly finished, and
then there is some back-and-forth on which one to do next.

If there is an actual queue, could someone point me to it? If not, could
someone (I'm not volunteering!) create one, or at least post a more-or-less
current snapshot?

Thanks,

Paul

dsimcha <dsimcha yahoo.com> writes:

On 9/6/2011 7:34 PM, Paul D. Anderson wrote:
 Is there an actual review queue, or is this a product of the hive mind?

 Looking at past review queue activities it seems like someone (usually Andrei)
posts a notice when the current review is finished, or nearly finished, and
then there is some back-and-forth on which one to do next.

 If there is an actual queue, could someone point me to it? If not, could
someone (I'm not volunteering!) create one, or at least post a more-or-less
current snapshot?

 Thanks,

 Paul

Right now, the review queue is maintained very informally via 
discussions on this NG.  Strictly speaking, it's not even a queue since 
high-priority modules can cut in front of lower-priority modules that 
have been waiting longer.

David Nadlinger <see klickverbot.at> writes:

On 9/7/11 1:54 AM, dsimcha wrote:
 Right now, the review queue is maintained very informally via
 discussions on this NG. Strictly speaking, it's not even a queue since
 high-priority modules can cut in front of lower-priority modules that
 have been waiting longer.

Must be a priority queue then… :P

David


(and yes, I do know what a priority queue really is)

Jonathan M Davis <jmdavisProg gmx.com> writes:

On Tuesday, September 06, 2011 19:34:28 Paul D. Anderson wrote:
 Is there an actual review queue, or is this a product of the hive mind?
 
 Looking at past review queue activities it seems like someone (usually
 Andrei) posts a notice when the current review is finished, or nearly
 finished, and then there is some back-and-forth on which one to do next.
 
 If there is an actual queue, could someone point me to it? If not, could
 someone (I'm not volunteering!) create one, or at least post a more-or-less
 current snapshot?

There is not currently actual page which lists the items up for review. 
Rather, those in the newsgroup keep track of what's posted posting as looking 
to be reviewed. Then when a review is finished, the review queue is brought up 
for discussion, and the next item to be reviewed is selected. In general, that 
would be the one waiting the longest, but the lack of availability of whoever 
wrote the code for review has popped up sometimes (e.g. they're on vacation), 
and if one item clearly needs to be reviewed sooner rather than later, it's 
reviewed next. It's all quite informal at this point.

Off the top of my head, stuff currently sitting in the queue is

1. The curl wrapper, which has already gone through one round of review.
2. std.log
3 A CSV parsing module

A revamp of std.variant is in the works (as is a revamp of std.json which 
relies on the revamp to std.variant) which should be close to being ready to 
be reviewed, but Robert Jacques has yet to post that it's definitively ready 
for review. So, I'm not sure that you can really say that it's in the review 
queue.

David Simcha has also brought up several of his items as possible stuff to go 
in the review queue, but I don't think that any of it currently is in the 
queue.

Andrei was working on some changes to the benchmarking functionality in Phobos 
which would lead to std.benchmark, but as far as I know, it's not ready for 
review yet.

std.xml is supposedly currently being revamped, but I don't think that we've 
heard anything new on that for a couple of months, so it's not in the queue 
yet.

And I think that there are a few other projects floating around which are 
likely to get into the queue within the next few months (e.g. the GSoC 
projects), but I don't think that anything else is currently in the queue.

It may be beneficial to formalize the review queue a bit better, but what we've 
been doing has worked fairly well thus far.

- Jonathan M Davis

Brad Roberts <braddr puremagic.com> writes:

On Tue, 6 Sep 2011, Jonathan M Davis wrote:

 It may be beneficial to formalize the review queue a bit better, but what
we've 
 been doing has worked fairly well thus far.
 
 - Jonathan M Davis

I think informal works particularly well here.  For several reasons, not 
in any particular order:

1) the code needs to be ready, an in-flux sort of thing

2) the submitter needs to be active enough to see that a new item needs to 
be selected

3) popularity of the item is a big factor in prioritization, also an 
in-flux thing.

4) being overly formal doesn't buy much here.. there's nothing to be 
particularly strict about in terms of the selection process.  For the most 
part we're all fairly rational adults.  Shocking, I know.

5) there's not so many that we're in danger of loosing track of anything, 
see also point 2


My 2 cents,
Brad