Thoughts from SWET2 Torbjörn Ryber

Once again I have spent the weekend with members of the cream of Swedish testers. This time The Test Eye trio consisting of Henrik Emilsson, Martin Jansson and Rikard Edgren were the hosts.

The theme was Exploratory Testing and Planning and we managed to keep the discussions within that scope most of the scheduled sessions. The informal test talk, music and drinking session from 18.30 to 03.30 encompassed a rich variety of discussions mostly outside that theme. Robert brought a fantastic instrument called the beat box (I think?), Rikard the guitar and the rest of us had guitar picks, maracas and music talent to some degree. The starting act was “The  Spice Song” composed and performed by multi talented philosopher-baker Rikard. Other highlights were “My Sharona” performed by many of us, later followed by interpretations of Cornelis Vreswijk and early in the morning sometime a try at “Bark at the moon”. The barking was considerable but I prefer the original version with Ozzys falsetto. And let us not forget the African freedom and tribal songs performed by the “missionary men”. Music is so much fun! If we ever start a band it will be called the Testicles.

Johan

But let us start with the conference pass. First session was Johan Jonasson from House of Test. He told us of the success he had with two young girls from help desk. I bet that sentence caught your attention! Well it was just as exciting as it sounds – it is all about testing. Jonas’ task was to manage the testing of a consumer product application and the resources he was assigned was two support people. He pointed out that they lacked education in IT and testing but were very motivated. He had too little time in order to prepare any detailed instructions. Test instructions consisted of some form of user stories created by himself and scenarios built by these. They tested together following the instructions and were coached by himself at a couple of meeting each day. They turned out to be great at taking notes and very good at finding problems. In the following open season we concluded that they did not lack education at all. They were used to taking support calls and analysing the caller’s problems so they had both product knowledge and experience on analysing and describing problems in text. Some reflections was that motivation is a very important factor and that less detail control of curios and motivated testers made them perform better. I can think of at least one client that uses support personnel but detailed script that should try the approach with a more exploratory format of working. As first sessions often do, it lasted for almost four hours and the discussions were never once uninteresting.

Tobbe

Number two on the list was myself with an experience report of testing in a loosely controlled and volatile project. Given the role as tester on the project I gradually moved into project management, requirement elicitation at the expense of much less testing than planned. Some of the things I did was to create a product backlog, organise weekly scrum meetings, create an effect map and a status graph. My overarching goal was that we actually deliver something to the customer and whatever needs there were somebody needed to take care of them. Success factors was that I was not only allowed but encouraged to assume new responsibilities by the other project members and that I find it interesting to take on other tasks than testing. The downside was that I had less time testing and, to be honest, less motivation to test. Maybe I would have tested better if I was given full-time on the project but when that was not the case I prioritised other tasks in order to move forward. It should be the testers goal to do whatever is needed to bring the project forward. I see similarities with the Scrum thoughts that we have less specialised roles. We had discussion on whether the artifacts I created was testing or not but the question is if it matters as long as they are important and they get done?

Fredrik

Fredrik Scheja of Sogeti was the next presenter. He told us about his success with an exploratory testing approach on a large system with frequent releases. Every tester assumes responsibility for analysing, test planning and execution of one or more items at a time. One of the dominating discussion in open season was the fact that he claimed that he built this approach on TMAP. Since most of the participants claim to be members of the context-driven school this is quite a daring statement to make at a peer conference. TMAP is clearly factory school which is the opposite of the context-driven approach. While I think none of us doubt the success of the work process used, many of us claim that it is not really TMAP-based just because you pick some parts that you find usable and then tweek them to fit what you really want to do. I think it was Johan that used the metaphor “Picking very few raisins out of a very large cake”. James Bach tweeted that “If you take a nice cake and drop it in a mud puddle, don´t bother with the raisins”.  Henke Andersson attacked the claim that the 12 step checklist for test planning really was the best to use for all planning in all situations and wanted to see a local adaption for the project that was very unTMAP(is there such a word, well now there is!). The discussion continued for another hour on Sunday morning for all of us except Ola Hyltén that by mistake(?) forgot to set his alarm when he went to bed already at 3.30. We thank you Ola for giving us an opportunity to make fun of you J

Lightning Talks and Conference plans

Last session Saturday was a number of lightning talks that were entertaining but to be honest I don´t remember much of the ten five minute talks. If someone else has a record or want to say something about them – be my guest. I do remember the last one where Henke told me to remind the group of our idea to arrange a context-driven conference in Sweden within the next year. All said it was a great idea and many wanted to help out. Our first suggestions was that we need a year to plan and arrange, it should take place in a larger city – probably Stockholm. Size could be 300 participants. Goal is not to make a fortune but not loosing money. Planning for a small profit gives us some room for unexpected costs. We would like to really focus on the context-driven community, only context-driven talks and not being sponsored by any certification organisation. We want to have some key players from the USA and certainly from the rest of the world as well. We may want to consider some tracks only in English and some in Swedish. All thoughts welcome.

Dinner

Dinner was a seafood buffet at the toll house by the sea. Since three of our four vegetarians have redefined fish and shellfish as vegetables they happily dug into the buckets of crabs, shrimp and langoustes (or whatever the English name for havskräfta is).

Martin said he was disappointed there was not a pool and agreed that he could possibly assume some responsibility for that fact since he was the one that booked. We focused on the music part instead as told before.

Saam

Sunday morning started out with another hour of open season on Fredrik followed by Saam that explained his goals to change testing and reporting on the large company where he works. We spent a couple of hours discussing green/yellow/red versus 1-5 versus happy faces and sad faces. The goal was to move from numbers – which say very little – to a more qualitative approach. This can be quite a challenge in an international and global organisation. We all look forward to learn about the results of Saams intentions in the future.

The happy ending

After some hugging, handshaking and some tears we left for the mainland. Yeah, cause I forgot to tell you we stayed at Chicken Island outside Gothenburg.

Henke Andersson has promised to arrange SWET3 in Malmö this fall. He mentioned that it will focus on ET standards and the need for ET certification…or maybe not. One subject that I would like to discuss more is teaching testing.

Additional info on Twitter #SWET2

The delegates were: Christin Wiedemann, Torbjörn Ryber, Azin Bergman, Fredrik Scheja, Henrik Andersson, Johan Jonasson, Ola Hyltén, Sigge Birgisson, Simon Morley, Rikard Edgren, Henrik Emilsson, Martin Jansson, Steve Öberg, Robert Bergqvist, Saam Koroorian.

14 Comments
Fredrik Liljegren April 10th, 2011

Thanks for the summary, as well as the tweeting done this weekend. It’s been very interesting to read those little nuggets of information and ideas.

Ola Hyltén April 10th, 2011

I’m glad I could contribute :)
It was a great weekend. Thanks all for making it a really great experience.

Fredrik Scheja April 10th, 2011

I’m sorry, I have to clarify some things here. You say that I claimed that I built the entire approach on Tmap? It was not my intention to give that perception. But ok, I said that for test planning we use the 12 bullet checklist borrowed from tmap as an example of how this can support us during test planning discissions. I feel that the summary of my session above might be a little bit one-sided and off my intended topic. But that is ok since it is your personal experience of the session. Anyone feel free to contact me in order to discuss this more closely in order to limit the risk of misconceptions.

Torbjörn Ryber April 11th, 2011

Fredrik: what I meant was that my impression was that you told us you used TMAP as a starting point for many of things you did. I may have understood you wrong but like you said that was my impression. When we discussed it more it seemed to me that you actually had your own approach and that was the success factor. And to be honest, a large part of the discussion was whether it was really a good approach to use anything in TMAP as a starting point and if you e.g. take a waterfall process that has clearly defined serial steps and then change it to say that ths steps are now parallell – you are not really using the original process anymore – quite far from it I would say. I would say that you did a great job testing despite starting from any checklists in TMAP! So I think you should be proud of your accomplishment and take credit for the success. But giving credit to a factory school approach on a context driven peer conference – well you will have to answer a lot of questions as you noticed :-)

Johan Hoberg April 11th, 2011

Sounds like you had a great time, with many interesting discussions. Thanks for the summary!

Johan Jonasson April 11th, 2011

Thank you to all delegates for making this peer conference a super happy fun time! It was quite an experience and I’m thankful that I got the chance to sit in open season for 4 whole hours. What a luxury to get to discuss the ins and outs of my story for that long. :-)

The dinner was a great opportunity to work on my arachnophobia as well. Creepy, long legged, but tasty, food. I guess we’ll have to serve you all Svartsoppa at SWET3. ;-)

We brainstormed some ideas for the SWET3 theme during the car ride home, and teaching exploratory testing did in fact come up as an alternative. We’ll see.

Henrik Emilsson April 11th, 2011

@Fredrik and Tobbe: I’m with Fredrik here on at least one thing. As I saw it, the TMAP-adaption was made for the test analysis/planning done for each package, so it was only used in a part of the entire “process”. However, it was this part that got most attention and what we discussed most of the time.

Fredrik Scheja April 11th, 2011

Torbjörn: Yes I know, who is the one being fearless now, my friend ;-) And yes, I am proud of my accomplishment and the result, but it would be unproffessional of me to use and present a checklist for you without telling you where it was borrowed from. Then, yes, I might use it in a way that it was never meant to be used, but that is what I do, that is the beauty of it =)

I respect your opinion but Isn’t this getting a bit old, this war between the empire and the rebels? The factory school against the context driven? Why can’t we just bury our hatches and agree upon that testing is all about finding out and presenting valuable information about the test object in an efficient manor? And I am certain that we can find good stuff to use on both sides. Why waste energy of irrelevant questions if something is pure tmap or et – or pure agile, scrum, kanban for that matter. Why discuss how much of tmap is allowed to be called tmap? That is irrelevant for me personally. Why discuss if a painting is renaissance or gothic art, when what we really just want is a beautiful painting that ties our room together and therefore fit our needs. The relevant question is, if this checklist has been an efficient tool to use in a certain way for me, could it be used or adopted to become an efficient tool for someone else? This is more interesting to me than spending energy on “Is this really tmap and are you allowed to use it like that?” If you´d like, I can label my approach SchejMap instead. :-) I look forward to the future discussions, it is exciting to see where this will lead us. Thanks for all your feedback.

Torbjörn Ryber April 11th, 2011

Ah, now we are getting to the key point. “Agreeing that finding and presenting valuable information in an efficient manor.” Because I do NOT believe that “we” all agree on that.

And if the different schools would agree on the above – then “we” would still not agree on what is the best way of reaching that goal. And I admit that I was for many years trapped in the factory school of thought. Remember I went to the first ISEB Practitioner class ever and took the certification as well. Many of those ideas did trap me for a long time until i first read what James Bach had to say and took the RST class. Saying that we agree would be to saying that waterfall and agile development agree. Yes there are most likely good parts in different schools as well as in TMAP and I will readily adapt anything I find of value.

I recommend that you read Bret Pettichords piece on schools http://www.io.com/~wazmo/papers/four_schools.pdf if you have not already and acknowledge that there are some fundamental differences in what testers think is important.

And, no – I wil not stop arguing against something I think is hurting our craft!

Fredrik Scheja April 12th, 2011

Ah, yes, I think what I am trying to say is that I don’t see the factory school as a hurtful threat against our craft any more, just as the agile community doesn´t see waterfall approach as a threat. No one argues for a good-old-waterfall development process any more, and according to my point of view the factory school followers are already put on the WWF list of endagered species.

[...] thoughts on SWET2 can be found: The test eye: Highlights from SWET2 The test eye: Thoughts from SWET2 Test and tech: SWET2 – A great way to spend a [...]

Rajesh Mathur April 15th, 2011

Impressive. Good work guys.

You don’t know me. My name is Rajesh and I am Enterprise Testing Lead at Cathay Pacific Airways in Hong Kong. My colleague Erik Peterson, who is a well known name in software testing community, told me about your get-together and Google helped me to find you. :o )
I will attempt to post some valuable comments/feedback about exploratory testing sometime later this week. And who knows we have a meeting among us in Hong Kong next time.
I have recently started a Linkedin group for Hong Kong testers and I will encourage you to participate. Again, who knows we work together some time.
The group is Hong Kong Software Testing Association. There isn’t much activity at this moment, but I guarantee it will be a successful association soon.
Cheers till then,
Regards
Rajesh

Henrik Emilsson April 15th, 2011

Thanks Rajesh,

Let us know if you wanna have some help with arranging a peer conference.(Or any other who reads this and want to get some tips on this). You can reach me at henrik.emilsson (a) thetesteye.com or the others by the same algorithm :-)

Good luck and have an interesting meeting!
Cheers,
Henrik

Darren McMillan April 27th, 2011

Wonderful summary of what sounds like a fantastic event.

Thanks Torbjörn for sharing this with us.