Re: Review Metrics

[H M]

OK. No Problem. But I just want to also mention that "Accepting" a review may also be a good way to let others know that a particular piece of code is already being reviewed.

For instance, if the review is sent to particularly large group, it may be useful for others to see that someone has accepted the request.  Of course its important for as many people as possible to do the review, but time is finite and if one is able to see that a particular request has gotten 5 reviews while another only 1, they may decide to spend their time reviewing the less-reviewed one.

Hovanes

Christian Hammond wrote:

That sounds reasonable for your use case.

This will need to be done as a custom extension later, though. I think it's too specific a use case to go into the mainline Review Board codebase.

Christian

--
Christian Hammond - chi...@chipx86.com
VMware, Inc.

On Wed, Oct 22, 2008 at 3:53 PM, H M <cybr...@gmail.com> wrote:

Sure,

I think keeping track of time like you described won't be enough. At least in our case, our review policy requires the reviewer to actually download the code and see if there are any warnings, does it run as expected, etc... All this will be outside of ReviewBoard.  

Perhaps another solution can be for the reviewer to "Accept" the review request, in which case a timer can be started and it doesn't stop until he/she submits the review.

Hovanes

Christian Hammond wrote:

This goes back to a discussion that was had in the very early days of Review Board with one of my co-workers. He recommended actually keeping track on the page of the time spent. It would basically work like this:

* On scrolling in the screenshot or diff viewer page, start a timer.
* On idling (no cursor or keyboard input) for a minute or two, pause the timer.
* Keep the timer running while a comment dialog is up.

This would give you roughly the amount of time spent actually looking at code.

We never added this because there really wasn't a compelling reason for it, and it's some additional overhead that we didn't feel was too necessary.

If there's a big demand for it, and someone wants to write a patch, we could probably include it in the codebase. Another option would be to make this an extension down the road when we roll out extension support (after 1.0, which should be released in December).

Does this sound like it would work for your needs?

Christian

--
Christian Hammond - chi...@chipx86.com
VMware, Inc.