Schedule and Events

March 26-29, 2012, Software Test Professionals Conference, New Orleans
July, 14-15, 2012 - Test Coach Camp, San Jose, California
July, 16-18, 2012 - Conference for the Association for Software Testing (CAST 2012), San Jose, California
August 2012+ - At Liberty; available. Contact me by email:

Friday, February 01, 2008

Do you Ning? is a meta-social network. It is similar to myspace, facebook, and all the other new-media thingees, only instead of having a universal network that you are plugged into, it allows it's users to create specifically themese universes, then subscribe to one or more.

I belong to four Ning Rings:

testrepublic - The Asian Subcontinent of India has a huge, growing tech community. Plus Pradeep is a frequent poster.

Drivenqa - Again, this is a test community, with a more European feel. Another change to interact with people I won't meet every day.

stpcon - The community for attendees, alumni, staff and faculty for the Software Test and Performance Conference.

The problem with Ning is that it's addictive. For example, I just put up a response to "Can We Measure A Tester's Performance?" in testrepublic and I could spend all day on it. So, at least, I'm going to try to reuse my answer here on my blog:

I think Pradeep summed it up pretty well with Prose. I'll try to say the same thing in a shorter way, a little more scientifically.

Most measurement systems are just approximations ('proxies') for some other thing. We really have no great way to measure productivity, so we try something else - say bug count. Of course, it's possible that the tester finds a bunch of trivial bugs like spelling errors.

This also creates incentive for the tester to find defects. If you measure the developers by how few defects they create, this creates a conflict of interest that leads to arguments, wasted time, and _decreased_ productivity.

So, you get what you measure, but that's probably not what you actually want. (Cem Kaner has a great paper on "Software Engineering Metrics - who do they measure and how do we know?" - you can google it.)

However, that does not mean that we can not measure anything.

There is a big difference between avoiding NUMERIC, or "quantitative" metrics for evaluation, and simply not measuring.

An alternative to measuring is, well, managing.

I will give you an example:

Do you measure the length of the hairs on your head?

Probably not.

Then how do you know when to get a hair cut?

You can use similar techniques to manage an organization. (I should add that I am find with metrics - in some cases - as information and input. For example, you can use metrics to augment a story. I am just ... leery of them out of context.)

1 comment:

Pradeep Soundararajan said...

Do you measure the length of the hairs on your head?

Probably not.

Then how do you know when to get a hair cut?

That's insightful and I liked this way of helping someone understand the problem with measurement of testers effectiveness by numbers . I am going to quote this and I thank you for that.

Thanks to Ning, I learned a better way to say something that I wanted to.