Organizing test ideas with mind maps

I forgot when I started to draw my first mind maps.
Like I'm sure many others do, I'm thinking slightly chaotic, in the sense, that I get inspiration and ideas after I finished something.
Mind maps are, as far as I know, created for that kind of thinking. So - if you're not one of those rare and rather strange persons who thinks everything sequentially and forward, you'd benefit from it too.

My preference tool is paper and pencil. Yes, I'm bloody old fashioned, but that always works and batteries don't wear out.

However, sometimes it's practical to have stuff on a computer and I've been using FreeMind for years. It's free, it's easy to use and though it's got its fair amount of really hilarious bugs, it's stable and reliable.

So - Paul Carvalho inspired me to write about how I use FreeMind to organise test ideas. I can't show you examples from real life, as they would both be way too big to show and I guess my clients would mind if I map their business (pun intended...).

Of course I would never reveal that, but I am willing to reveal a mockup example to show how I do it.
Here's an example we all know: Google. I only used the front page and followed a few links.

You may recognize that I 'factored' the front page. We all know it as a simple text field and a search button. Most of us remember that there's also an 'I feel lucky'-button, that we only use when we're bored. But in fact there's quite a lot of other features, and they are a bit more advanced than it looks like. For instance the redirection function that shows you a localized front page.
Take a look at the real thing and compare with my little mind map here.

Factoring is a skill of it's own, I learned it from James Bach. Factoring is really about finding 'dimensions' to test. Here I just put in the functions I found, but I did make a note that there's performance in a search operation. I put it in and maybe I want to later expand that node.

Notice I reused the response time sub-tree for the advanced search as well. Is that good organizing ? Well, it's okay for a preliminary stage. But later - if we get to reuse it a lot, it's better to move it out among the common elements or something and simply refer to it. The good thing is that there are no rules. Whatever works for you - well, it works for you.

One note of warning though: if you fall into the same trap as PM's do when they repeat a block of activities in their plan - you'll get lost. Don't do that. I find that it's mostly not only convenient but also sufficient that you realise that something is repeating. There's little value in documenting everything. The real value comes from you having learned about the repetition.

So - start it up. Extend and expand as you get more information and inspiration. Extend and reuse by referring. Use icons to add some meaning or overview. Ask questions (very easy this way because you always ask the question in context of a node).

I can test directly from this and extend it while I do so. I put in ideas, functions and non-functions to look out for. A map of the system.
I can mark out branches and sub branches and put them on charters. I can readily show what works and where the problems (bugs) are. I can also show what isn't tested yet.

In short - I can communicate this and keep a great overview.

You should try it too.
  
Please let me know what you think.
 

CAST2010 was a blast

So I went to CAST 2010 in Michigan. It was a blast. Although, not so many people this year.
You never heard of CAST ?

Why - it is a wonderful conference, with a huge emphasis on CONFERring. So it's sold, and so it is also. I've been going three years in a row now and I keep learning from this conference.

One of the special things is that all the sessions are about half presentation, and half discussion. And discussions run far and deep too. In comparison to other conferences, you could actually attend them and not really speak a lot, except in the hasty breaks in between sessions. One or two guys would get to ask questions. Maybe they are not even good questions, but there's no more time. Then there's lunch, where you usually sit among total strangers, and you first need to find rapport and common grounds - usually small talk about the weather and the subway is the only conversation.

I mean, there's nothing particular bad in this. You hopefully get to listen to great presentations by smart people. Only - at CAST you get to participate.

I know this is described in different places, but I repeat it here: At CAST discussions are facilitated. That means that everybody (yeah - EVERYbody) get's a red, a yellow and a green card. One for commenting on the current thread. One for signalling that I've got a new thread. And the red card for: I really, really need to speak right now!
And you probably think: ...a forest of red cards.. Oh no. You forget - CAST is facilitated. The facilitator (and they are pretty good at this) keeps track of who wants to speak (and will take away your red card if it is abused). If you've got something to say you get your place in the queue. It works. Threads are followed as long as there's energy and if you raise a card, you get to speak.

Again: this works. People talk. A lot. Even for long. Questions are never: yes/no. It's always a looong story. Seems like all have something to say. And the conference is truly trying to support this.

Lunch is pretty interesting too. You probably met people before. Maybe even argued with them. Makes talking to them much more interesting and fun.

This year I was lucky to present myself. Twice, even (which is a bit more stressful than I expected). We got pretty good reviews (I wasn't alone on either talk). One talk - Dealing with passionless testers (with Henrik Andersson) - ended with a great discussion. We didn't get through all the stuff we planned, and I was kinda sorry for that - then. Now when I think about it - we did trigger a very deep, valuable, sometimes very personal, debate. We used a small film to kickstart it. It's on youtube. Find it! If you're a good tester, you can! :-)

The other talk - Nice words are not enough (with Louise Perold) - about introducing session-based testing and exploratory testing in projects, went great, but I think it suffered a bit by being the last thing on the schedule - apart from the ending session. People was just a bit tired - or maybe exhausted ? But we still had a good, long discussion.


I do hope I can make it to CAST next year again. I don't know if it's possible but I hope. Because I am anxious to learn more and meet more cool people.

Meanwhile I will have to settle for EuroSTAR - in Copenhagen this year. I hope you signed up! I'm talking there too! "Session-based testing in practice". A sort of SBTM 101. I'm excited. It's just about my favorite subject in testing.