Tuesday, April 03, 2007

The more "done" something appears, the lesser and narrower feedback get

One of my colleagues at work has asked me to give feedback on his vision document (position paper). Oh well, the document is so strongly voiced that I am not sure that it is a vision document, a road map or a solution to a perceived problem. The document is definitely looks sharp with pretty pictures of sunny vision and green road maps. It also has an appendix section containing code snippets of a tagged language (xml/xslt). I am sure that our friend has put tones of work into this and his document is well thought out and incisive.
So, why am I losing energy as I read through the document? So much of what I read strikes me as correct but somehow belittling of the problems real domain face. The vision: Reuse, shared components and shared database. The document then goes through some lengths to show a solution along with a possible implementation.
The problem with such a generic vision statement is more or less the equivalent of a manager advice/order to "get better." That is no help at all if not insulting. I am also not a big fan of vision first, team later approach. Most successful companies define a domain to explore, build a team and then have the team to come up with the vision.
I am losing more energy as I see the perfectly fonted and formatted draft, every sentence seems more done than you’d like it open. The solution proposed deal with small amount of real problem, pretty much like an iceberg where 90% of it is underwater. The best design is emergent and done through exploration rather than a perfectly prescribed solution.
Am I just grumpy, or others may get annoyed as well? I shared the document with NP and DR, they both reacted the same way as I did - just a little stronger.
Don't make the Vision or a design document look done. Here is the most damaging part of this exercise besides setting the wrong expectation as Kathy Sierra suggest:
The more "done" something appears, the lesser and narrower feedback get.
If you show me something polished and pretty, you’ll get feedback on font sizes. The possibility of getting feedback is far more achievable if you do it on a piece of paper, napkin or white board. Java people use Napkin Look and Feel, for the same reason.

3 comments:

Anonymous said...

Some people who have to submit their work for review, put specific triggers in that will allow the reviewer to 'set an anchor'. It gives them a place to start by making a simple change. Often submitting a work that is too polished, does not allow the reviewer an opportunity to engage with the work. Try putting in a misspelled word or use the wrong word or phrase purposefully. This opens dialog and interaction with reviewers and gives them a place to start.

Mehrdad said...

Interesting idea Mary. I have seen some1 getting feedback by doing just that.

Mehrdad said...

Interesting idea Mary. I have seen some1 getting feedback by doing just that.