Imagery and metaphor - a falliable helper

I stumbled across a few videos on the UK's Open University course site in itunes a while ago, a couple that grabbed my interest were about imagery and metaphor.

The videos about the Open University course can be accessed using the following url:

Initially I though the idea of using imagery and metaphor was a bit 'way-out', however the more I watched the more I believe it could be a useful in my testing work to identify and/or solve problems think of ideas and/or communicate.

I was so intrigued I decided to try it out. At the time I discovered the videos I also was about to start some data integration testing - In my context that meant passing data between two external systems.

First came the metaphor I chose; the first thing that came into my head. erm Aeroplanes.

So data integration is now an aeroplane... (erm yes it sounds far-out right? - I hope you stick with me)

I then drew an aeroplane on my notepad.

Then, I started to think about all the things an aeroplane journey might have:

- cargo Ah yes that could be data...
- then secured cargo, different sized cargo, labelled cargo, damaged cargo.. I'm thinking data here

- What kind of Aeroplane is it and does it matter? i.e. propeller vs jet, size capacity. Or what kinds of aeroplanes could they be? Could these be structural things? Technology, servers

- Speed How long for takeoff, duration and landing, taxi-ing to the runway? - Time

- Does the aeoplane go to a specific runway, do the wheels come down for landing?

- What else is going on in the sky? Other planes? Birds? on the ground? Air traffic control

- Who is using the Aeroplane? and why? Importance?

Test ideas came very quickly to me and I found it quite easy to relate to my metaphor (I've been on a few flights and used to work in the airline industry) so I decided to use the metaphor for facilitating my test ideas.

One thing I am aware of is Joel Spolsky's 'law of leaky abstactions' so am still wary of treating my testing as a metaphor. An abstaction from the problem and in turn
may be simplifying a complex problem. However I found the strength of using imagery and metaphor was it's simplicity - not getting too bogged down in my data integration detail.

I regarded my picture as a different kind of mindmap.

I did discuss my picture with some members of the team but a team adoption of this approach may require some effort and buy in. I will try again...

After creating my metaphor I used it to create some test sessions and tracked the test sessions rather than metaphor image. (I may try this again and see if I can somehow display my testing efforts back through the metaphor, somehow showing the test sessions on the image, maybe using colour - although I am also wary of doing this - i.e. focusing on pictures and colour rather than the testing.
It could be a good way of reporting (I kept thinking about a video Rob Lambert posted on Visualizing progress through images and not charts i.e. things people can relate to see -
Note: whilst looking for this link I also discovered/re-discovered lots of other interesting posts Rob has made which included the following: which I believe may also be related)

One area I think imagery and metaphor may help me in the future is in mneumonics, i.e. instead of memorizing letters like SFDPOT, thinking of an image or story that bind the themes of the mneumonic helping me remember

I am sure imagery and metaphor will be helpful for me in the future and is firmly part of my toolbox, but what I am aware of is that using images/metaphors are falliable. They may not work all the time (the same as any mind map)

What images and metaphors are you using?


  1. I am sure this has lots of promise for testing challenges or weekend/night testing.

    You should facilitate a session using this ( week-night so I can make it! )

  2. Hi Darren,
    Thanks for the comment. I think we have already been using this approach in community testing challenges although I didn't realize it at the time (I'm thinking about Lanette Creamers time transporter challenge - I think that could be a kind of metaphor?)

    One other area I found interesting from the same Open University series was 'clean language' which seemed like eeking unbias information from people. Useful in testing?

    I'll like to experiment a bit more but a session could be interesting...


  3. Interesting!

    I like, and use, imagery and metaphor - but they are not universally understood - just as natural language is not universally understood. That's why prescriptive language and metaphor need some form of tailoring to the audience….

    With imagery (whether a metaphor or not) - there is a certain interpretation that goes on - think of the Kanizsa triangles. We're used to seeing whole pictures or interpreting parts that we can't see. That can be one reason why imagery/metaphors work better for some - and some find them difficult. In your image - having studied fluid dynamics with wing shapes - I might have a totally different feeling when I look at the picture (why are the wings under the aircraft, I'm not sure that would fly, etc) - i.e. I get triggered on the wrong detail… So, the picture works with explanation and not standalone.

    There's a story in "The man who mistook his wife for a hat" of someone who was very poor at neurological tests (couldn't function well in the world) - but excelled in metaphor and symbols (and went on to work in the theatre).

    When I started reading about the plane metaphor I started thinking of noisy kids and kids puking (soon off on a long flight with the kids - so maybe that's what triggered in me first…) - but that's just an example of how the interpretation can be so different. It might be a reason why it hasn't caught on in the team so far - you need a common metaphor, that everyone is comfortable with.

    But in the personal case, I think they're great - both as a means to work with, as a means to explore the thoughts around the problem - it's an abstraction/model of the problem, and it's always good to have more than one model - it can be a means of communication, where people question your interpretation (why, what does that mean) - that's good for you and can also help you find areas that you miss in your model.

    I've pondered problems with mapping/modelling before, here, and at SWET3 I improvised a model feature coverage (using shapes for areas of the product and colours for progress/coverage), here, it might get a different reaction than the one intended (or not)…..

    Thought provoking, thanks!

  4. Hi Simon,
    Thanks for the comment.
    Good points for the Kaniza triangles, and about explantation. I agree the image definately needs explanation something I may have missed in the post. I wouldn't use it standalone, but is that any different to anything you might do/use in testing? Is that any different to a mindmap created by a tool?

    My plane metaphor was my own metaphor, and I did not try and get the team involved in my metaphor what I did do is mention my metaphor hoping it might provoke them to identify or think about a problem in a different way i.e. come up with their own.

    If I was to try and use a shared team metaphor I think it would have to be that, shared from the creation and not necessarily for sole use as tool.

    I like the fact that you thought of noisy/puking kids - so you could have already added to my image! :-) (I'm thinking unpleasant journeys but everything looks good on take off and landing - there's probably some tests in that? firewalls/rules things specific to my context.

    One of the things I wanted to express is that mapping does not need to be software tool generated and that it could have an image theme. I believe images can help me a lot.

    I like and use mapping software but I don't use it exclusively and I can't do some of the things I want to do. (I'm not sure my mind could be mapped in mindmapping software anyway ;o) )

    Thanks for the links to your posts