Mike’s Dump

May 16, 2005

Code Review Posting

Filed under: Links — mikesdump @ 11:28 am

Regular code reviews is something I’d like the team to get into. This
was an interesting posting on how the MSBuild team has implemented code
reviews. This process is close to the way we did them at net-linx.

http://blogs.msdn.com/jomo_fisher/archive/2005/05/16/418012.aspx

Advertisements

2 Comments »

  1. I like the idea of doing a mini code review before each check-in…that would certainly prevent some strangeness from happening. The only thing that might be an initial barrier to acceptance (just like larger code reviews) is available time, but like anything, maybe it just takes some practice and discipline (my other concern is developers rushing through a mini code review simply to get back to their other work).

    Comment by J — May 18, 2005 @ 10:44 am | Reply

  2. I think any new process that will potentially slow down client deliveries will be met with some opposition from some management. I think the only thing more difficult to overcome is getting buy in that code reviews will improve quality and reduce errors (this was the case with a number of developers at my last position).

    The process we tried recently was with two reviewers, a moderator, and the reviewee. I believe this process stands the best chance of a good review being done.

    Pros
    1) It will show up at the review session if the code was actually reviewed.
    2) The moderator can step in if there is a disagreement
    3) More dialog between the reviewers and reviewee.

    Cons
    1) Consumes a lot more time
    2) Probably would only be done on larger changes
    3) Greater dependence on testing to catch bugs in minor changes

    Pros for review on check in approach
    1) All code in theory would be reviewed (might be difficult in practice)
    2) Potential to catch more defects (by reviewing more code)
    2) Potential faster turn around for feedback to the reviewee
    3) Consumes less time

    Cons
    1) If reviewer isn’t familiar with the code might this approach could be slower
    2) Less formal approach may translate into less complete review (rushed)
    3) No moderator in some cases could be an issue

    Comment by mike — May 18, 2005 @ 5:28 pm | Reply


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at WordPress.com.

%d bloggers like this: