Day 2 round up of one of the warmest and friendliest conferences I've ever been in. Nordic Testing Days 2012, baby!
As I stated in my previous post, I missed the beginning of the day 2. Good company and alcoholic beverages are just something that makes me want to press snooze button few more times than usual in the next morning. Plus that breakfast was something else! Kudos, Meriton Grand Hotel!
 |
| Mr. Nice Guy in action ;) |
But let's get cracking. Time dig into the presentations/workshops I eventually did participate. Same rules as before; Lightning style, headlines hold the slides, stories are my thoughts about the subject, etc.
Go!
 |
| Mr. Marjamäki in action |
My friend and ex-colleague, and the new head honcho of Finnish Association for Software Testing, Peksi Marjamäki gave us by far the most energetic presentation in the conference. And the subject was one of the most important ones; Managing ET. Almost every presentation/workshop I participated in NTD2012 had something about exploratory testing in them, some more that others, but what bothered me the most was that so many didn't distinguish themselves from ad hoc testing. Even our dinner talks revealed that so many testers considers ET to be a way to dodge structured way of doing things, which it most certainly isn't.
I think Peksi gave audience the very thing they need: structure. He concentrated mainly on modeling ParkCalc via SFDPOT (see slides for details about the abbreviation) mnemonic familiar for those who've participated in RST courses. So basically it was dissecting the software into smaller, logical, manageable and measurable pieces based on a structure (not rules!). I got great ideas from Peksi and from people sitting next to me (thanks Helena!). He remembered to keep things simple and that applied on the software under test too. That way people can concentrate better on the things being taught. Very professional!
And this time it was an actual workshop where we got to do things! After we dissected the software, we got to break it. The ParkCalc was a familiar software for me, because it's often used in ET workshops, so some bugs were easy to re-produce, but what is always the most giving part in these kind of soirées is that I got to see how others test and how they find bugs. Even though it was a familiar software, which actually has begun to bore me a bit, new things were found and learned.
Truly encaging and thought provoking workshop, which really hit the spot for me and - as I sensed around me - for others too. More of his wisdom can be found at his blog.
Cool! :)
 |
| Mr. Toom in action |
And as I missed structure on day 1, that came in spades on day 2! :) Mart had a very clever idea; He crowdsourced us to test this ID card software which's testing he manages in his assignment at customer. I'll use that, yoink! :)
This was a bit longer workshop so there was more time to think about what to do, how to break something and find bugs. The introduction was quite short and as the instructions gave structure, me and my partner Peksi didn't quite follow them; We regressed into ad hoc testing. Or so it seemed at first. Quite quickly Peksi dug out his array of tools; Rapid Reporter, Pomodairo and some screen capturing tool which creates a video recording of your doings during testing. Ok, Rapid Reported I've used before, not effectively though, but the others we're unfamiliar. I can imagine video recordings to be quite convincing evidence of testing done and them help in bug troubleshooting, but Pomodairo just distracted me. The darn thing is like a clock in meeting rooms; You spend your time staring it. He apparently had a lot going on in his screen and I became to think; Do we test or do we fiddle with accessories?
In SBTM it's imperative to have uninterrupted timeslots for testing, but it's very subjective what actually interrupts you. All the flashing clocks and apps, mail popups, browser tabs, open windows, mess and confusion, and of course the abomination that is open office might work for others, but for me they just don't. The very dilemma in e.g. pair-testing; The pair must first fit so much before fitting it's testing. In educational purposes, handovers, knowledge transfers, etc. it just might work, but in testing, which tries to produce something... I'm very doubtful.
Ok, despite all the distractions and the fact that Peksi actually dragged my distracted ass back there, we actually produced. Peksi input several things you should grow discussion on, even bugs to Rapid Reporter, which produced raw CSV to be processed later and his video recording supported that nicely. We unfortunately didn't draw any mindmap, dissect the software as we did in Peksi's workshop, but for a hybrid of ad hoc and exploratory testing, this actually went quite nicely. There even was this bug where we simulated the actions of a malefactor; We managed to lock down other ID cards while giving the bad guy continuous chances to bruteforce credentials open from a card in his posession. That's a f***ing steak dinner, isn't it?! :)
 |
| Mr. Priisalu in action |
Jaan is one of those guys who shouldn't be able to sleep his nights well. He works for Estonian Information System's Authority as Director General i.e. he's responsible for developing and administrating the Estonia's national information systems. Think about it for a while. Think about the responsibility. Energy, transportation, food and water, communication, the works. They talk about these kind of roles in IT horror stories. There's something about a ten foot pole too...
Ok, he's a hero. But what did I experience besides an awe to his position and that you actually CAN blow up things via code? Testing-wise, not that much. He however gave me some pointers presentation-wise, which I'll definitely steal. Yoink! He also pointed out a very interesting thing about managers; Owners want results and workers work for this, but managers create agendas and biases. When you come to think of this, all the processes, methodologies, bureaucracy, etc. nonsense is the handywork of managers. Not of those who want results or produce them, but managers.
Think about it.
Last but not least Jaan told us that software quality determines our quality of life. This resonates with my values in testing. Testing is what ties software to humans and thus should be quided by brains, not machines.
 |
| Raivo Päts, NTD2012 chairman |
Now that a month has passed since NTD2012, I notice a certain longing for the event. I honestly had fun, spent time with fantastic people, learned a thing or two and of course popped my cherry as a conference speaker. Huge thanks goes to organizing elements, especially to chairman Raivo Päts and his lieutenants Kaspar Loog and Harles Paesüld. Via my jujutsu "hobby" I've grown to know both the Estonians and the Swedes better, under the skin so to speak, and I consider both to be the best neighbours anyone could have. But because of this conference I feel extra bit of warmth towards the Estonians. Love you, guys!
Your move, Sweden! :D
The Finns aren't that bad either. ;) One of the greatest is of course Pekka "Peksi" Marjamäki who isn't just a brilliant tester, but also a great guy with head and heart in the right place. I wish him all the best in upcoming endeavours on and off-duty, and this time grant him the honor to be our quotesman:
"High six!!" -Pekka MarjamäkiYours truly,
Sami Söderblom
I only read this blog post now... for some reason. You know the "mega mindmap" you showed me has haunted me ever since :D
ReplyDeleteI am going to create something similar with my team for our purposes. I think I'm going to seek out a context in our context, work together on compiling the test ideas for that, so that we work inside out. I hope this will be valuable. I don't want to superimpose any lists or maps of others. Maybe I'm paranoid but my gut feeling tells me we may miss out on stuff in our environment if we did that. It could be useful later to check those maps and catalogs of others.