La latino-americanización del GDCR

Hace un par de semanas en la oficina ToughtWorks de Ecuador se puso en movimiento la organización del “Día Mundial para el Retiro de Programación” de este año en Quito, y, aunque el evento es más conocido por su nombre en inglés “Global Day of Code Retreat” hemos decidido que lo vamos a hacer un poco más latino americano!

El primer paso, y el más obvio, ha sido la traducción del nombre del evento de forma que podamos hacernos una mejor idea de que es lo que vamos a hacer el próximo 14 de diciembre. Esto, aunque simple, no viene libre de riesgos; el primero, que el título de este artículo utilice palabras que no constan en el diccionario de la lengua española. Pero, mientras la idea sea transmitida podemos mantenerlo. La idea de traducir el nombre del evento tiene el objetivo de que podamos relacionarlo con actividades que conocemos y tienen algún sentido para nosotros. Quiero decir, todos sabemos lo que es un “retiro de meditación”, verdad?

El segundo es que los participantes no solo sean ecuatorianos, sino que además contemos con varios de nuestros hermanos de la patria grande. Y en pos de este objetivo, hasta el momento contaremos en Quito con tres brasileños duchos en el arte de la programación funcional. Esto brindará a los asistentes, una vez más, la oportunidad de practicar su portuñol sin tener que moverse de la ciudad. Aprender un nuevo paradigma o probar a que sabe un nuevo lenguaje de programación será simplemente un efecto secundario dentro de todo el valor agregado que te aportará el evento.

Entonces, les esperamos en Quito el próximo 14 de diciembre para el retiro de programación a la latino-americana!

Purposeful Programming: Why You Should Care

Software professionals need to care about the implications of their creations and the ways it will be put to use by the end user.  Let’s take a look at an example of how the software you develop can have unforeseen consequences.

The business need: “I want to access our student’s basic info and mail it to companies in their majoring fields in order to help them get a job as soon as they graduate.”

The entire development team starts to work on it: sketching, asking for examples, writing tests and then coding, carving details and testing some more. The feature is finished and the product is shipped.

Now to the end user…

A  career counselor at a university has the new app and  is ready to help his students out. He chooses the background, font, recipients and publishing date for the email, and also the name for the campaign and the majoring careers to be included.

As he finishes his task the snacks trolley happens to pass by, and he chooses a cake over an apple. Is it our fault that our creation drained his cognitive resources which happen to rest in the same pool as willpower?  Yep, by providing the user with too many (maybe useless) choices our application could be what is making him break his diet.

Now to the real issue and end user…

On the other side of wire, emails start reaching candidate inboxes and new opportunities start blooming. Those with A grades receive great offers, B students get OK offers and C students none at all.

Basic info includes a global grade obtained by the student during several years in college. But wasn’t it supposed to be only basic information? Nobody questioned at the time of the build if this field in the database was a good idea to include.  The result, employers are seeing a skewed representation and making a decision that isn’t based on real potential and students will suffer.

All because of the software you helped to develop.

So now our creation has had an unforeseen consequence, because, as explained by Professor María Belen Albornoz, technology is not neutral. And for software, this means code has some implications we may or may not be aware of.

These two agents into coding, unknown to me just a few days ago, poses a question we should ask before start coding: what is this software really for? Because you might be writing a tool which enforces/encourages a principle you don’t share. Sometimes this question may be answered simply by asking who asked for it. Sometimes you will need to reach for someone with  better knowledge. Either way, let’s be more mindful as programmers.

And also, this got me thinking about something we should ask while typing every code expression: is this the best way to do it? Because if it is not I might harm the function and I might harm the structure(1). And even more importantly, the unintended consequences may end up hurting individuals.

All said, purposeful coding should be included into our everyday work of creation. A code where we denote our intention (to do good, hopefully) in every line. A code which cares about its user and knows the place it will have within the big picture.

(1) The Clean Coder: A Code of Conduct for Professional Programmers, Robert C. Martin, 2011, Kindle Location 516 of 3791

The maps seller

I’m close to sell impact mapping as the new way to plan our projects, at least software projects. In order to get here, and here is having the go for a prototype project, I basically showed four slides in three different meetings one hour long each for different audiences. Attached the open sourced version 😉 intro-impact-mapping Continue reading

jBPM5 over EAP6 standalone-full, the ugly way

Some days ago I run into the need of seting up a BPM environment as quick as possible over a JBoss EAP6. The natural way to go was jBPM5 of course, and, based on my prior experience with its third version I forecasted I would need one morning to set it up, a day would be the worst case scenario.

So, I downloaded the BRMS and SOA platform and started my simple task. At the dawn of the third morning tired of trying and failing I downloaded the community package, read a little bit from the configuration documentation and came out whit a ugly provisional way: Run the installation script provided by the comminty and copied the pieces I needed to my server. Continue reading

The many tools for testing

As Arquillian has become steady at the programmer’s tool belt, its action scope has entered debate as some feel it’s an integration testing specific tool and some others see it as the long awaited tool which will bring end to end testing as the only testing a programmer needs from now on. The latter is the motivation of this post as it forgets about the power of unit testing and might not see that Arquillian’s purpose may change depending on the extension you use. Continue reading

Spiking the coin

This very short post is to share with you a spike I’m taking on the JDK7: Project Coin. You can have access to the code at git://

For more than a few definitions on a spike, take a look at this discussion on stackoverflow.

Notice how you only need some tests to learn something new.

Trac migration

This how-to shows the migration of a Trac environment from a 0.11 installation to a 0.12 installation.

Continue reading