Monday, May 9, 2016

What's a good spike?

Problem: Our forum has some unpleasant limitations. In particular, pasting formatted text into a forum post from Outlook (and possibly other sources) produces weird and wonderful results: the HTML gets converted so you don't see formatted text, you see the raw HTML A bit annoying.  So we decided to do a spike to evaluate an alternative:
spike: evaluate TinyMCE / other options for editing text in forum
We got to the end of the evaluation, and as PO, I had more questions at the end of this evaluation than at the beginning! Why?

I think the answer can be found in the title of the spike. What's wrong with this title? First, it starts with an verb in the imperative. "Team, do this" It is not an invitation to think. Second how do we know if the we have satisfied the objective? It doesn't really say. It just says 'evaluate.'

Here's an improvement:
spike: can we eliminate our copy/paste problems by using TinyMCE?
By formulating the spike as a question, it becomes clearer what is the objective of the spike. This in turn makes it easier to tell whether the spike is done.

Of course, this still doesn't answer the question of whether it is a good thing to deploy it in our context. It's a closed question and assumes part of the answer, i.e. that TinyMCE is the best solution. How about:
Which forum editor best satisfies the needs of our users?
Of course, that might be a bigger spike, but the goal is clearest and most focussed on the people who really matter!

In my Certified Scrum classes, I demonstrate using Scrum in the class by organizing the class with Scrum. The course topics are the product backlog. I used to define the product backlog as user stories, but I now express them as questions. My students ask questions; learning is the result of asking questions; and formulating the product backlog as a series of questions seems like a natural approach. I wonder questions as backlog items could be used for other kinds of story as well?