Oct 15 2009
Agile Testing Days Berlin, a personal retrospective
I spent the last two days in Berlin attending Agile Testing Days, one of the most enjoyable and fun conferences I’ve seen in a while. Jose Diaz and his team did a fantastic job organising it and deserve a ton of praise. I’ve learned a lot, met great people, put faces on many e-mail and twitter contacts and got a chance to release some of my ideas on an unsuspecting crowd (and they seemed to enjoy it). It was fun. The most important quote I’ll take away with me, which sums up the conference as well, is from Mary Poppendieck’s keynote:
System should support intelligent people doing their work rather than trying to replace and de-skill workers.
Many fantastic talks dealt with gigantic technical and human changes in testing processes sparked by the agile software movement. There were three keynotes a day and most of them truly deserved to be heard by everyone, especially the ones by Elisabeth Hendrickson, Mary Poppendieck and Lisa Crispin (in no particular order). On the other hand, I felt that Tom Gilb’s talk was completely inappropriate for this conference. There is value in bringing people who will challenge and contradict the general feeling at a conference — if we only listen to people who say what we want to hear then we’ll never learn anything really new. In this case, I think that the organisers went too far. His ideas seem to be from the dark ages and I’d be very concerned for anyone trying to actually apply them in practice. I respect Gilb very much for all the work that he’s done in the past but it seems that he is not in touch with what anything happened over the past decade the industry.
Most of the talks were very high quality and had a lot of value, so Alessandro Collino did a good job selecting them. Apart from Tom Gilb’s session, I walked out from just one more by Hyungil Cho, who demonstrated a complete misunderstanding of FitNesse but still decided to talk about it. On the other hand, the conference panel after the last session was a complete disaster. The keynote speakers who should have been there all left before the panel. I don’t mean to disrespect the others but the organisers should have ensured that Lisa Crispin, Mary Poppendieck and Elisabeth Hendrickson participate in the panel. Tom Gilb took over again and talked about how people should do more “real software engineering”, whatever that is supposed to mean, and just reconfirmed that he doesn’t really understand agile development. Other participants seemed to preach too much for my taste, I did not come there to listen to a sermon.
For those of you that missed the conference, I’ve compiled a list of notes into blog posts. Some are already online and I’ll be posting many more over the next few weeks, so subscribe to my RSS feed to get updates. Unfortunately, there is no way to share the fun from the conference on this blog. Surreal is probably an understatement of the beer-bash after the first day, with a Spaniards dressed in lederhosen dancing and one hundred people singing loudly in German although most of them don’t even speak the language. You’ll just have to attend next year and experience that yourself.
 |
 |
There was a panel in the last session? I didn’t even know, I left to catch a train.
Great summary of the conference, Gojko. I agree 100%, and it was super to meet you and other people I only knew from the Internet! Your session was inspiring!
yep there was a panel. I suspected that you didn’t know. That should have been organised better. Improvement for next year
Hi Gojko,
We were 5 attendees from my company (acrolinx, Berlin), so we’ll make a debriefing to our colleagues who didn’t have the chance to attend the Agile Testing Days. I just discovered that I was about to paraphrase parts of your own feedback in my own talk.
Cons:
I was annoyed by Tom Gilb’s talk. Using numerous ugly slides, he repeated the same thing about his quality metrics applied to specification documents during a neverending session… I was sad for him (completely out), the audience (wasting time) and the organizers (who made a super great job otherwise).
Mike Gevers’ talk about performance testing wasn’t that “agile” in that she’s expecting a sort of BDUF (whole architecture in advance) to include her performance measurement tools (she has to rework her demos because they tend to say: “you’re braindead if you use such tools”). Moreover, her current experience is not included in an agile project but consists in late testing, as in predictive models.
Pros:
Elisabeth Hendrickson (who I didn’t know, but now I do) and The Poppendiecks (I’m following their adventures since _Implementing Lean Software Development_) talks were very… energizing.
Elisabeth Hendrickson approach on “Agile Testing” is really simple, so that anyone can explain it with a few words, without notes.
On a technical point of view, I was interested in the new ATDD tool called “Robot Framework”. I’ve been using FitNesse since 2004 but its ergonomics prevents adoption by impatient people as me. So I was looking for a better alternative, as Robot Framework and its IDE could be.
Thank you Gojko for the discussion we had about the Tester and Product Owner roles, and thanks again for giving me your book I’ll read in a couple of days.
My batteries are loaded for a couple of months and I hope both my colleagues and my company with get benefits from that.
Sincerely yours,
Régis
Good summary. From the panel discussion I kept a different picture from Tom Gilb in my memories.
On the panel discussion I understood Tom Gilb to call for more craftsmanship in the programming and testing world. What I have back in mind is that he called for more people to know what they’re doing rather than following the latest trends or fashions in the industry. Clearly I have in mind that he called for less engineering, but more craftsmanship. Did I confuse something there?
Kind regards
Markus Gärtner