Duchess
Globally Connecting Women in Java Technology

Lunatech Research invites all Duchess members to the next Play!ground event in Rotterdam on February 3rd. The event will take place at Paddy Murphy’s Irish Pub, just a few minutes away from Rotterdam Central Station. It is a great opportunity to learn practical techniques, hang out with other Play enthusiasts and have some fun! For all the details and registration follow this link: http://www.lunatech-research.com/archives/2012/01/13/february-2012-playground

Opening – Bert Ertman

The flashy opening movie was abruptly cut off for the entrance of Bert. He gave a brief intro about the conference magazine, the lunch and coffee, and the location, before handing the mike over to Stephan Janssen. Stephan gave a brief talk about Parleys and then Bert explained that 17 of the JFall sessions would be filmed and added to Parleys. Bert then took over again and explained why there was no wifi. The opening was closed with a short movie clip by Oracle called “We code hard in these cubicles”.

(Dutch) Things I would have liked to put up on Twitter if I’d had the chance:
Het sneeuwt op het scherm. #jfall11
Stephan Janssen – The feel of Scala moet je ook wel een paar keer bekijken voordat je het begrijpt #jfall11

Keynote – Tomas Nilsson

The opening was followed by a keynote from Oracle by product manager Tomas Nilsson. He started by telling us about his career and how OO was the Cloud of the 90′s. To him C++ was merely C with OO added on top of it. Java was C++ with garbage collection on top of it. Back in the 90′s you had to have your own framework and your own process team. But Oracle doesn’t have a process team as they don’t have a standard process. Every product uses it’s own process, as it’s hard to come up with a uniform process for such diverse products. If you let marketing decide, you’ll get unmanagable spaghetti. If you let the developers decide, it won’t ship (it will be very pretty, but outdated). You need a product manager to balance these forces, which is where Tomas comes in. Part of his job is to decide which rules you need to follow and which rules you can misinterpret or forget. Then came 2010, Oracle took over Java from Sun. Everything was doom and gloom, and nobody trusted Oracle. Oracle has very strict press rules. They’re just not allowed to talk about future products, which didn’t help in building trust. They’re also not allowed to talk about court cases. The fight between Sun and Apache meant Java7 didn’t get shipped. One of the things Oracle got done, was to break the stalemate, so they could finally ship Java7. JavaOne 2010 was very last minute and therefor a mess, which also didn’t help in building trust. The content was good, but logistics was confusing. JavaOne 2011 was already a lot better. Finally one bit of advice for the future from Tomas was that Oracle has invested 100.000 man years in Java. So if nothing else, trust that Oracle will be greedy (and invest in Java in their own interest). Oracle doesn’t say a lot about what they will do, but what they say, they ship.

A tweet I would have put up if I’d had a proper internet connection on my phone:
Sun: Java wasn’t slow (in the 90s), it merely looked like it was. – Tomas Nilsson #jfall11

Java EE7 Platform: Developing for the Cloud – Arun Gupta

After Tomas’ keynote, I went to Arun’s talk about Java EE7. He told us that today’s cloud offerings are vendor-specific, you have to write your application to their API. Likewise the Java EE6 platform provides services, but the Java EE7 platform on the other hand IS a service. The new acronym therefor is PaaS, Platform as a Service. The new features of Java EE7 include new roles, new metadata that can be added to services, new API’s (JSON, caching API, etc), add multitenancy support in existing API’s.

Looking at the model of Java EE6, we see that we have to provision and initialize each and every service before deploying the application. In the model of Java EE7 on the other hand, you can just take a war file and give it to the cloud admistration service, which will look at metadata (or even derive metadata from the war file), this service will then provision and initialize the necessary services and then deploys the app for you. Java EE7 can even extract metadata from existing annotations (like ConnectorService, JMS, Datasource).

Another great feature is that Java EE7 will automatically scale up or down based upon usage.

To give a bit of an idea about how EE7 works here’s a little scenario: the developer writes an application, the customer signs up as a customer, the submitter submits the application, extra services discovers the application, sign up for it, and then customize and deploy the application, the administrator then provisions the application, the end users access the application, and the administrator administers the application.

Arun then gave us a bit of a demo, where he also told us that Glassfish has two different modes to run EE7 in: native mode which runs on a local machine and virtualization mode which runs in the cloud.

After the demo he explained about the four strategies for multitenancy: dedicated application and dedicated database, shared application and dedicated database, dedicated application and shared database, shared application and shared database. And finally he mentioned a few more things that were improved in EE7: alignment of managedbeans, simplification for ease-of-development, pruning some APIs.

Introduction to JavaFX 2.0 – Michael Heinrichs

The session started explosively, so everyone was sure to be awake. But then the mike didn’t work. But Michael Heinrichs had a voice that was loud enough to reach the entire room, so it wasn’t much of a problem. The first announcement he did was that JavaFX script will be dropped in favor of true Java code. So you no longer have to learn another scripting language to work with JavaFX2. After that he gave an overview of how JavaFX works and about all the cool things it can do.

First of all there is the SceneGraph. This is the most basic datastructure, which allows you to group graphical elements in a tree structure. A SceneGraph contains Leaf nodes, which can be shapes (cirkel, vierkant, zelf-gedefinieerd), images, text, WebView (embed any kind of web content), media, controls (knoppen, sliders, etc.), or even charts. It also contains Parent nodes, which can be a group (not visible), a region (similar to a group, only visible), or layouts (VBox, HBox, border layout, grid layout). The parent takes care of rendering the children. So if you move, rotate or scale the parent, the children are moved, rotated or scaled as well.

Second, there is the Animation API, which contains Predefined transitions (translate, rotate, scale, path transitions), Sequence transitions, a Timeline (which contains keyframes on which you can give properties certain values, and which allows tangent interpolation).

Thirdly, there is the Media API, which contains Media classes (media [data], media player [controls], media view [scenegraph node], and Video effects (drop shadow, reflection, sepia, blur, change colors).

Then there are the UI Controls, which contains buttons, text controls, TableView, complex views, panes, and many others. The UI Controls will be Open Sourced first (other parts of FX will likely follow). The controls have support for CSS, so you can style them (which makes working with a designer a lot easier, as they know CSS, but not Java).

And finally there are the Properties & Bindings. There are many Java FX Properties like the DoubleProperty. Each of these properties has getters and setters [for example double get(), void set(double value), Double getValue(), void setValue(Number value)]. Then there are Listeners, Unidirectional Binds (size ← a + b, so if a changes, size changes too), Bidirectional Binds (size ↔ slider.value, so if size changes, the slider value changes too, and vice versa) and a Fluent API. You can also create your own bindings.

The presentation ended with some examples of bindings:

– size.bind(a.add(b)); // size = a+b

– a.add(b).subtract(c); // a + b – c

– size.asString(“%.2f”); // String.format(“%.2f”, size)

– when(cond).then(a).otherwise(b) // cond? a:b (when one property changes, the whole thing is re-evaluated)

All in all, it seemed very promising, and I can’t wait to try it all out.

Hands-on Lab: JavaFX 2.0 – A Java Developer’s guide – Stephen Chin

So I was very happy to find that there was a handson on JavaFX. So I went to a double session to try my hand at a bit of JavaFX. Stephen helped us install JavaFX and NetBeans and let us create a very simple web browser with a text field, a submit button and a webview that showed the content. Due to there being no wifi, we simply pointed our little app at some files we had on our laptops. The webview component is just as flexible as a real web browser, so it happily showed me html, sql and text files. It was a nice little taste of what you could do, but it would have benefitted from some handouts to guide you, instead of an explanation to the whole group. The explanation was good of course, but it didn’t really allow for your own pace. But I’m looking forward to reading a book about JavaFX 2.0 as soon as it’s written.

Hands-on Lab: What’s that smell? – Refactor away that nasty odor – Marc Evers, Willem van den Ende, Rob Westgeest

For my last session of the day, I went to another hands-on. Marc, Willem and Rob gave a session on refactoring. They handed out some handy little reference cards containing some common smells and some common refactoring patterns to get rid of the smells. We had to install a little project they had prepared to be as dirty as it could be, while still allowing you to do some quick refactoring on it. One of the main points to note here is that with refactoring they mean tiny little steps you can pretty much do automatically. Anything larger would be called redesign by them.

The first step of the workshop (after setting up the project) was to look at the code (in a pair of course!) and to identify some of the smells. Then you had to look at the possible patterns to apply to the code to refactor it. The third step was to tell everybody else which patterns you would apply to get rid of which smells and why you would start there. Then we actually set to work and got rid of some smells. Sadly neither my own laptop, nor Elise’s would cooperate and we didn’t actually get any work done. And since there was only one hour in total, we didn’t have any time left to see if we could fix them. But we did get to take home the cards. Still despite the disappointment of not actually playing with the code, it was an enjoyable session.

On Thursday evening I went to the Play Framework workshop hosted by IProfs. Upon arrival we were welcomed with an excellent sandwich dinner. After eating and chatting for about an hour the actual workshop started and everybody got out their laptops. We were given a brief introduction about what the Play Framework was by Ronald Haring, and then we were set loose with a basic code sample and a PDF with instructions. And although the instructions were at times a bit puzzling (with a class alternatively called Pets, PetOwners and PetClinics), they were still clear enough to get the code working rapidly (and where they were not, Ronald came to the rescue with a hint and a word of advice). The sample application was a simple Play-ification of the Spring Petshop application. We started out with a simple application that only had a homepage and some tests that didn’t really do anything. The first step was to add a test that showed how easy it was to persist and delete an entity. All you had to do was create a POJO called Specialty with a public property called name. This you annotated to be an Entity and you had it extend play.db.jpa.Model. Then you could just use methods called save, create and update and the like on the new entity.

Second, we created a simple CRUD application to manipulate pet owners. To achieve this you created a POJO called PetOwner, similar to how you created the Specialty class. Then you created a class called PetOwners that extended play.mvc.Controller. This you gave a list method that retrieved a list of all pet owners and then called the render method with this list as an argument. In order to actually render this, we created an HTML file in the PetOwners folder in the view folder. For this HTML we used the Groovy templating system. The benefit of this system is that you can use it for other stuff than HTML as well. (Although we didn’t actually try this during the workshop.) And with just those few steps, you could actually display a list of items.

The workshop went on to tell you about the routing system and how you could create methods to actually update the data and navigate between the pages. So at the end of the evening we had seen quite some options of what the Play Framework could do.

All in all, I can heartily recommend you to attend this workshop if you haven’t already done so. There will be a second chance on January 12^th, if it isn’t already fully booked.

The second day of AgileOpen started with a wonderful breakfast, where the homemade jam was a delight to all.

After breakfast came a keynote by Mary and Bart about Agile and Architecture. They started with some information about their backgrounds, where they highlighted their differences and common grounds. After their discussion of masculine and feminine they mentioned that Agile is a feminine injection in IT. Something they hoped to possibly discuss in depth during later sessions. The similarities they found in their respective research were the following:

• A high level architecture is needed to start a project
• Responsibility for the architecture is with the team
• Everybody should have enough knowledge about it
• A different form of leadership is needed
• The team needs self discipline
• Feedback is esssential

They ended their presentation with a couple of statements to start some discussion and to take along to the sessions of the day.

• Architecture is no more than a process within an organization
• Architecture and Agile keep each other healthy
• Freedom doesn’t mean without obligations

Following the keynote we all got the chance to come up with some sessions we’d like to have during the day. But as this was during coffee, some people put down some stickies with ideas for sessions and then disappeared into the sun *coughericcough*. Fortunately for them, other people were interested enough in their ideas that they presented the ideas for them. (Thanks Mary!)

The first session I attended was about an Agile performance review. Eric who had suggested the session was nowhere to be found, so after waiting ten minutes and looking for him, we decided to start without him. After some general back and forth about the subject, we decided that there were two aspects to this. The content of such reviews and the process. The process was quickly dealt with when someone mentioned 360 feedback. (More info about this can be found on noop.nl by Jurgen Appelo.) So then we focussed on what the qualities were that a team member needs in an Agile team. After making this list, we dot-voted to find out what the most important items were. And the following is the prioritized list

• Be able to communicate briefly/clear/honest (6 votes)
• Think outside the box (4 votes)
• Receive feedback (3 votes)
• Be flexible (3 votes)
• Be pro-active in the division of labour (3 votes)
• Give feedback (2 votes)
• Work on their own development (improve skills) (2 votes)
• Plan their own work, estimate workload (2 votes)
• Share information (1 vote)
• Keep an eye on the vision (1 vote)
• Have political/organisational insight (know how and where to find the right people for a decision) (1 vote)
• Expectation management
• Be a teamplayer (submit their ego to the team)
• Know their own limitations
• Empathy

After that we ran out of time, and Eric came back. It turned out he had had a similar conversation with somebody else in one of the other rooms.

After this session it was time for lunch. Lunch, like all other meals, was delicious.

The second session for the day was a repetition of the Code Smells & Refactoring workshop that Marc had already given at JFall. As my laptop hadn’t cooperated during JFall before, and Elise’s laptop didn’t work properly either during that event, I decided to join the workshop again. This time I got the code working, and together with a partner (whose name I have shamefully forgotten) I set to refactoring the code. After the session was technically over, nobody really wanted to leave, and as most sessions were outside anyway, we just kept on going for another session. It was a lot of fun to apply the patterns to get rid of the smells (both listed on handy reference cards). After my partner decided to give up, I joined Stephan and Luus and we kept on refactoring the code until we were quite happy with it. The organization was even so kind to bring us snacks when we didn’t come out to the bar to eat them.

After even we were done with the code we had a short evaluation of the event. (It was relatively short, as quite a few people had already left.) One of the surprising conclusions was that it would have been pleasant if the time frame had been a little more strict.

After the evaluation Marc gave me a ride to the station and I was headed back home again.

All in all a very succesful event, and I definitely want to go there again next year! My thanks to Sandra and Lillian for organizing it, and to the participants for making it such a wonderful event.

On Thursday November 4 I rose really early and took a train to Dieren, to attend Agile Open. It didn’t start off very well for me, cause I had missed the information that the location had changed. So I entered a room full of strangers who first invited me to put my suitcase with the others and only then started wondering if I belonged to their group. When I said I came for Agile Open, they gave me some funny looks and said I was in the wrong location. Fortunately the barman knew where I had to be and was even kind enough to give me a ride there. Once I was in the right location, everything was fine. They handed out some bright red t-shirts and some blank badges that you had to put your name on yourself. Enjoying a cup of tea or coffee, we chatted amongst ourselves until the event would actually start. It turned out that Sander Hoogendoorn, the keynote speaker of the day, was stuck in traffic, so we started off with some statistics. We had to order ourselves alfabetically, where the biggest problem was on which side of the room the A would be. Then we ordered ourselves by years of experience. We also grouped ourselves based on location and based on age. And many other such things. But then it was time to set the agenda for the day. Stickies and markers were supplied and everyone could suggest a session. The board filled itself up quickly.

After that was all done, we finally started the keynote called Flower Power Fluffy Agile. It wasn’t really about Flower Power Agile anymore, as Sander had already written a blog about that, (and most of us had probably already read it,) but took off where the blog left off. (In fact, Sander wrote a blog about this presentation too, which is much more comprehensive than mine is: http://sanderhoogendoorn.com/blog/?p=1063&utm_source=rss&utm_medium=rss&utm_campaign=death-by-dogma-versus-agile-assembly.)
Sander made the point that if a waterfall project fails, it’s always the methodology that is to blame for the failure. But if an Agile project fails, whoever was on that project, simply didn’t do it right. The problem is that just doing Agile isn’t enough. You still have to really think about what works for you and what doesn’t. Sadly, for most people Agile is a new experience. And while we call people who have had a two day training a Scrum ‘Master’ the reality is that they are merely beginners, and make beginner’s mistakes. This is the natural order of things. The Japanese call this Shu Ha Ri. Shu is the first level of learning, where you learn the fundamentals. If you apply these often enough, you will start encountering exceptions to the rules. Places where the rules don’t quite fit. This is the second stage (Ha). The final stage (Ri) is where you have truly become a master and you merely use all practices naturally and you don’t have to think about which rule applies where anymore. And most people are still in the Shu stage, but they think they are Masters and therefor they apply Dogmagile. They rigidly stick to the rules, including in situations where those rules don’t truly fit. One thing a lot of people tend to forget is that Scrum lacks the steps to fill the backlog and that Scrum is a framework where you have to fill in the gaps. Stories are merely meant to start the conversation, but what happens after the conversation has started? Scrum doesn’t say anything about this, and this is one of the points where you have to come up with what works for your team.
Upon discovering all this, Sander started wondering what the true basics of Agile were. And the list he came up with is the following:
Short iterations
Collaborative teams
Small units of work
Continuous planning
Deliver early and often
Simplify communication
But for earch of there items, you should do what works best for you. Just mix and match the tools and practices that work for your team and your project. Agile is no religion: so don’t be a zealot.

The keynote was followed by a round of coffee, after which the first of the sessions started. I picked Martin’s session about how to deal with a waterfall company as an agile team. He had been in a project where everything went smoothly until it was time to deploy the actual application. Then they had to send everybody home and wait for half a year until the organization was ready to actually put the new application into production. Some of the tips I took home from this was to try to involve everybody as stakeholders. Including for example the maintenance people. These don’t have to show up for every daily standup, but knowing their procedures will certainly reveal such problems earlier. And showing them how you work might build some trust. Also you should try to convince people as high up in the board as you can what a waste it is to have to wait for half a year. Finally you should have coffee with as many of the stakeholders as you can. Having a relation in place can sometimes help you get around some constraints.

After the first session came lunch, which consisted of sandwiches, something to drink and a fruit salad. And a lot of hallway conversation of course.

Next up was the second session, where Luus wanted to know what to do with project managers in an agile organization. The conclusion was that while project managers in regular organizations are the people who make sure of the project’s success, in an agile organization they should be the people to facilitate the team. They should manage all the stakeholders and make sure they don’t bother the team. They should also communicate with higher management and help decide on the strategy. Possibly they could even coordinate multiple teams. All of this means that we will no longer need IT project leaders, but only business project leaders who will facilitate the Agile teams. One of the people mentioned that DSDM has a couple of extra roles that might help your team.

Sadly I didn’t get much out of the third session, as I was sorely distracted by a disheartening phone call. I tried to pay attention, but simply couldn’t. I hope I didn’t distract any of the other people in the room too much.

Finally I joined a nice session by Maarten about the Kanban game. He had a game that showed very nicely how using a Kanban board can help your team. If you ever get the chance, I can advise you to join the game, as it’s a lot of fun.

After the last session we all went back to the original location where we would stay for the rest of the open space. Once there we started with some drinks, followed by a wonderful buffet dinner. After dinner there were more drinks and evening activities. The activities consisted of climbing, curling and archery. I didn’t try the climbing, as I’d only recently learned I didn’t like it much. But curling was fun. And I was very surprised to score points during archery. After the activities we went back to our drinks. And it was midnight before I went to bed in one of the luxurious cabins on site.

Devoxx day 2, 15.10 Room 5
Java Posse Live
Dick Wall, Carl Quinn, Tor Norbye and Joe Nuxoll

This semi-transcript can’t begin to describe how funny this was, so if you have some time: go and download it instead of reading this!

At the beginning of the session, after everyone had sat down with a beer (sponsored by Atlassian, the official beer partner of the posse), we got chimp rolled. (Two chimpansees on a Segway?) Then they announced that the next Java Posse Roundup will be from March 26th till 30th in Crested Butte, Colorado (USA).

As always the recording at Devoxx starts with a survey, which revealed that the Devoxx rock-star wrist bands aren’t very popular, but Belgian Beer and Waffles are.

After the survey, the posse members start getting really morbid and invent all kinds of reasons why they would need a successor. Then they randomly select five names from all the attendees who are visiting Devoxx and come up with Chet Haasse, Mark Reinhold, Brian Goetz, James Ward and Stephan Janssen. Out of curiosity they ask each randomly selected deputee why they came to Devoxx. Chet only came for the comfortable chairs, Mark for the excellent keynotes, Brian for the extra airmiles, James for the beer, and Stephan came because he had to, but would really prefer being at home.

In order to select the official deputee who might succeed one of the Posse members in case of an untimely demise, they hold a shootout at the Posse Corral to determine who is most fit.

The first round is rapid research. Every now and then the posse has some random question which needs to be answered. So each of the candidates must prove how well they can get an answer by any means possible.
1. How many companies were on the original expert group for JSR 273.
– Chet quickly comes up with the ever favorite 42.
– Stephan answers 17, but later changes it to 11
2. What is the number of languages currently available on the JVM?
– Chet: -5
– Mark: 239
– Brian: 278
– James: 60
– Stephan: 150+
Sadly, the true answer is unknown.
3. What was the first non-Java JVM language?
– Chet: English
– Stephen: JRuby
– Brian and James: Ben Hill(? I might have gotten that one wrong)
– Mark: Scheme
– audience: Oak
– posse: Byte code
4. What is the lifespan for a newly issued software patent?
– Chet: my lawyers won’t let me say
– Mark: too long
– Brian: de facto, 17 years
– James: zero in China
– Stephan: should be zero

The second round is the code challenge. An intricate piece of code is shown on the screen with a while loop, containing a printline statement followed by a return, but the finally writes another line to the console, followed by a continue. The question is what the output would be.
Stephan: it won’t compile due to tabs and braces
Chet: it throws a DumbCodeException
Mark (actually typed the code in Emacs and ran it): It prints Java\nPosse\nJava\nPosse…Java\nPosse infinitely
Mark is right.

The third round is acronyms. The deputees are shown a list of terms and are invited to turn one of them into an acronym to explain what it means.
Mark: BYTE; Big Yellow Type Error
James: JSR; Java Spec, Really?
Chett: SWING; Sun WIll Not Give-up
Stephan: JINI; Just an Interface (he appologizes for not having had enough time due to not being a native speaker)

The fourth round is wave forms. The posse would show a wave form of an actual bit of recording from the Java posse, and the deputees would have to pick the answer that said what the recording contained. I’ll spare you all the options, as I can’t show the wave forms anyway, but the answers were: “Soooo….”, “Nose blow”, “Hello and welcome to…”, “Ummm…” and “Whip crack”.

The fifth round is Fake news. [editor's note, I didn't take any notes of the things people said about these topics, so I'm doing this from memory. I hope I didn't mix the speakers up.] Chett talked about the session called “Intro to project Avacado”, which was some kind of project that had created software that had something to do with vegetables. Mark told about the session called “Hand-Optimizing Bytecode with BitClipper”. According to him it wasn’t a very good presentation as the presenter spent most of the time complaining about the horrible submission process of Devoxx and then talked about something to do with scissors and nail clippers. Brian copied this tactic when having to explain what “The next great JVM language: Socotra” was about. James had to talk about “The latest agile practice: Blame Driven Development” and said that it wasn’t his fault that the seats were so comfy that he didn’t get much of the presentation. Finally Stephan had to explain about “Project Anarchy: A new approach to parallel computing”. But he had left the session early to attend the panel about women in IT.

The sixth round is about the posse lingo, where the deputees had to show they knew what Leeroy Jenkins, Phasers on Stun, Quit playing with your dongle, and Do the count-in meant. (In case you are not Java Posse fans the answers are: “It’s time to start the show”, “Phones in airplane mode”, “Are you accidentally muted?” and “Sync the audio channels”)

The seventh round was about sound effects. The first effect was the standard opening sentence “Hello and welcome to…”. Chett had a coughing fit, Mark seemed to try using his most sexy voice, Brian was very cheery about it, James used a recording, and Stephan did a Flemish edition. The other effects were the Quick news whip, Listener feedback, Random crap and a Pirate yarrr.

The eight, and final round is Impossible names. Every now and then the posse is confronted with names that are truly impossible to pronounce, like John Smith, Freddy Guime, Herbert Czymontek, Romain Guy, Raghavan “Rags” Srinivas and Hugh Jass. The deputees try their best, but fail just as much as the posse does.

Then it’s up to the audience to vote on who the official deputee should be. As it’s a tie between Chett and Stephan, there is another impromptu round, the Fashion round. They both have to pick a color out of the four hats and show their most fashionably cowboy swagger. Stephan needs a bit of practice, whereas Chett shows his unguessed skills as a true fashion model. In the end, the posse determines that Stephan is the new honorary posse member and make him swear on the hat that he’ll keep up organizing Devoxx for as long as he can.

I went to Devoxx again this year, and made lots of notes of all the sessions I visited. At first I was planning to write one big summary of the event, like I usually do. But then there were some sessions that I nearly transcribed, so naturally those are pretty long. Also, I haven’t had time to make proper blog posts out of all my notes, but I didn’t want to wait until I was done with all of them. Here is the post of one of the sessions I visited that I have turned into a blog post.

Devoxx day 2, 14.00 Room 3
Why we shouldn’t target women
Martijn Verburg, Kim Ross, Régina ten Bruggencate, Trisha Gee, Antonio Goncalves and Claude Falguiére

With five minutes to go before the presentation the room is filling up quickly.

The panel introduces themselves and Martijn, the host, explains how they will run the panel. People in the audience can ask questions through Twitter by using the hashtag #devoxxwomen. Below is my transcription of the questions and summaries of the answers the panel gave. I didn’t attribute the answers to anyone, as they are too much my own interpretation of what was being said. (Seeing how I didn’t use a recorder to make the transcription. I merely wrote stuff down as fast as I could.)

“Why aren’t there more women in IT.”
– We’re the wrong persons to ask, as we’re already in IT. But our guess is that it differs from person to person.

“Is Duchess targeting universities to attract more women into IT?”
– We want to go looking for teachers, but we really need to target even younger girls, as there aren’t very many doing IT studies either. We really need to convince even younger girls and their parents that IT is an interesting carreer for women. But this will take a long-term investment that not many companies are willing to make (yet).

“Why the obsession with more women in IT?”
– What we really need is of course more diversity, not just more women. But the starting point of this panel is that this is already a given. It’s already been proven that diversity is a good thing for teams and companies. So we’ve chosen to make this a discussion about getting more diversity by trying to get more women. But we really should also try to get more black people and Asian people and all other minorities.

“Some industries are male-driven, some are women-driven. Would this be a cultural thing?”
– Yes it is. Countries like India, Brazil and Egypt already have more women in IT.

– The Paris JUG has a meeting every month with about 200 people. Afterwards, they go out to dinner. At first, there were only very few women showing up at these meetings. And none of these women joined the group for dinner. Then some girls who were active in the group, decided to have a women-only meeting before the official JUG-meeting. They’d get together and then went to the official meeting together. This has led to the number of female members in the Paris JUG to grow from 8 to 20 members, some of whom even join the group for dinner.
– But maybe we should do such a thing for all new members, and not just for the women. Because joining an existing group can be daunting for everyone, not just for women.

“How can we change the image of programmers sitting in a basement with lots of pizza boxes and coke cans around them?”
– It all starts in college where the nerdy guys scare of the girls.
– We’ve probably all fulfilled that image at some point (since most of us are geeks after all). But the industry itself isn’t like that at all.
– This image hurts us all (men as well as women), because as a man you’re expected to grow up and do something other than coding. While most of us are passionate about what we do, and don’t want to do anything else.

“What can conference organisers do to attract more women speakers?”
– Please, no positive discrimination. Positive discrimination hurts women, because it makes people doubt our credibility. ‘Are you in that position because you are actually good? Or are you just the token woman?’
– Speakers are usually chosen from a pool of known speakers, so if there aren’t any women in there, you won’t get any female speakers.
– Most selection committees are all men too, so it would probably be a good idea to get more diversity there too.
– Women put the barrier too high for themselves. They think they have to be really great before they can even submit a paper to speak at a conference. Don’t forget that there are a lot of crappy male speakers, and all you need is more practice.
– Mentoring is really important to build confidence. So look around you if you see anyone you think might be capable to speak at a conference and then try and guide them into it.
– The XPdays pair speakers. For new speakers they usually try to pair them with an experienced speaker.

“The language we use in job advertisements is rather aggressive. Is that a barrier?”
– Yes!
– We should push companies towards more mentoring and such things.
– Women should also learn to just try despite this barrier. Getting a rejection doesn’t mean you’re no good. Usually it just means you’re not the best match for that specific function.
– It’s also a fact that companies actually ARE looking for great people.
– It might also be the interview process that scares people off. Men tend to judge women differently, even if that is mostly not a conscious decision. Because they really are only human after all. In France, 40% of all couples meet at work.

– Give your daughter Lego to play with! And later on, maybe try to learn them to program.

“The rumor is that Facebook pays a 20% bonus to female developers.”
– This isn’t a solution either.
* Note of the editor: Although it might set of an inbalance I’ve heard of from a female HR manager. Women tend to ask for pay raises less often. Meaning they also get them less often, as HR managers only have so much money to satisfy their employees. People who ask for a raise more often tend to get them more often too,

“Lots of women fall into the role of mother/sister of the team. Is that something they do because they want to?”
– For some women this comes naturally
– Others hate it
– A man could fill this role just as well
– It should be more of a mentoring role
– And isn’t this actually a sign that the team lead isn’t doing a very good job of it?

“Men are often discriminated against when it comes to being a father.”
– Women won’t get equality at work, until men get equality at home. Until it becomes normal for men to share in all the responsibilities at home, women will keep being rejected for jobs as they might want to have kids.
– We really should separate the parents’ rights issue from the womens’ rights issue. It should become just as normal for a man to take time off to take care of sick kids.
– Maybe we should even have laws enforcing this.
– It is already happening. In lots of countries it’s becoming more and more normal for men to take time off to take care of the kids.

“Some companies are actually looking for senior women who are a mother, as they would have management skills.”
– Isn’t it strange how a mother is supposed to have these skills, but a father isn’t?

“Isn’t it true that women are overly sensitive to male banter and such?”
– If there had been more women in the industry, they would give as good as they got. But because there is such a huge inbalance, they can’t.
– Joking in the workplace is something entirely different from jokes in a public forum such as a presentation at a large conference. In the workplace you can joke back, in a public conference you can’t.
– Maybe we should do mentoring in bantering as well.

“Should we ban booth babes?”
– No, we really should have cute boy booth babes as well!
– We really would prefer technical people, as having booth babes really hurts the credibility of female techs who are trying to sell their product. And besides, the booths that have booth babes usually hamper their own success as well, as nobody is coming to hear about the product anymore, but merely come to ogle the eye candy.

“Can you give any advice for men with a younger female boss?”
– Just treat them as if they were an older male boss.

“Do you know that it’s different in Brazil?”
– Brazil is a new country, IT is also new and considered as a cool thing to do besides (and not a geeky/nerdy thing).

“Last year at Devoxx there was a 1% female attendance.”
– Free tickets wouldn’t help, as only 10-15% of all programmers in the UK are female, so there simply isn’t a very large pool to draw from.
– Duchess is trying to address this by organizing themselves around conferences and by announcing what conferences we’re attending, in order to lure more people over there.

“How many people know about the Java Champions program?”
* editors note: Not many people raise their hands at this question
– It’s a program to make the role models of the Java world more visible. They are a self-selecting group, who nominate people who have been very active in the Java community.

– As a closing note, targetting women specifically is narrowing the field and we really should aim at diversity in general. Not all women are the same, and assuming that they are will not work to attract them.

We are having our traditional dinner after jFall again.
It will be in Ampt van Nijkerk which is next to the conference location and will start after the drinks at the conference.
Please let us know if you want to join us, guests are welcome.

Duchess JUG meeting in Stockholm Sweden on Oct 26th. Learn about agile testing and play framework.

Sign up at: http://simplesignup.se/event/8418-duchess-sweden-jug-moete

Join us to  the next Amsterdam Clojure meetup A night out with Bob Martin

It is a free event!!!