Pablo Moreno-Ger’s blog

In my classes, I always try to keep track of those students that come to the recitations well-prepared, and that participate actively in class. While participation has never been part of the grade, I have traditionally used that information whenever I come across a border-line grade: Is this an A or a B? In addition, since only 1 in 20 students can get the A+ grade, I also use my participation records to decide which students gets the A+ in case of a tie, as I may find that I have 5 students with a 10/10 in the exam but can only award three A+ grades.

This year, I am teaching a Programming Fundamentals course for the new ECTS-compliant program in computer science, in which 10% of the grade is explicitly assigned to evaluate participation in class and contribution in optional tasks. I thus wanted to keep track of those students that completed different exercises and tasks throughout the course. I have typically done this by writing down the names of those students that deserve recognition, and keep a track of what we usually call “positives” in the school slang.

However, this year the students are split into groups for the recitations, and we have planned lessons in which all the students have to try to complete an exercise in class. Tracking “positives” was bound to be cumbersome, and I actually hate the term. I needed something different.

I have thus set up an experiment with my students. I have generated a set of 9-digit codes and printed them in small pieces of paper that we bring to the recitations in the lab. And instead of writing down the names of the students that do well, we give them one of these tokens with a code. By the end of the course, I will be asking the students to send back to me a list of the codes they have gathered. I will then crib out duplicated or forged codes (I have the means to verify them) and use this information to adjust their grades.

So, how is this different from writing down their names? The process is actually more cumbersome, and can be gamed by the students to some extent. Why bother then?

Well, I believe that it is working out because now the students are made responsible for tracking their participation. The physical token, although simple, still makes a deeper impact that simply seeing me write down a name, just like casino chips are more appealing than digital money in a digital casino. Moreover, in some of the sessions there is a chance of walking out with a handful of codes, and ending up with a small stack of tokens at home. The students can compare and compete, something that was not possible through traditional means.

I believe that this is more interesting than the typical approach and the students (at least some of them) are responding well. Some other teachers have mentioned that this is akin to treating them like children, but this I hear this so often whenever I try to gamify any process that I really don’t care as long as it works.

And some of the students are beginning to understand and leverage the trick to walk out with a stack of tokens: to study the subject matter and try to solve the exercises before coming to the lab. Precisely.

In what has now become an established tradition, we will be running a seminar on game design for e-learning environments for students of the School of Computer Engineering at Complutense University.

This is our ad for this year’s seminar:

This will be our fourth edition, and we will use the same approach as previous years: The first two weeks, we (the coordinators) will do most of the talking, presenting our ideas and experiences on educational game design. After that, the students get the spotlight and propose their own ideas.

The following 7 weeks the students form groups and each group develops an educational game with complete freedom to decide their topic, genre, design, contents and technology. Of course we always suggest using e-Adventure, but we never enforce it. Many of the most successful projects produced in these seminars have been e-Adventure games, but last year’s winner was, in fact, created with RPGMaker.

Did I say winner? Yes, the seminar output is not a quantitative grade (only pass/fail), but the last day we do hold a public presentation party and let the students vote which is the best game.

I have previously posted about our research regarding the applications of e-Adventure in the mobile space. We started with a port to J2ME, seeing this platform as an easy way to run e-Adventure games on a phone. It worked, although it wasn’t flexible enough.

In the last five years phones have advanced incredibly. The explosion of iPhones and Android phones have elevated what can now be considered a “standard smartphone” and that’s why we switched to Android two years ago. We are now seeing good results, and our Android-related tools are evolving quickly thanks to the work of our talented developers.

We have led the development in two different directions. One line is porting the e-Adventure engine to the Android platform, so that it would be possible to play e-Adventure games on an Android phone. Right now we have an e-Adventure app that can connect to the network, download adapted e-Adventure games from a repository and launch them on the phone. This Android-based engine can also use features that go beyond the desktop version: we can define effects that are not triggered by a user action, but by locating the phone’s position using the GPS. We can also scan QR codes and trigger effects in the game depending on the code. Unfortunately, games that use those features cannot be edited with the current e-Adventure editor.

We have a video showing the current prototype.

In parallel, we have been working on a mobile game editor. The key idea here is to allow both teachers and students to draft or tweak existing games on the go. Imagine going on a field trip and planning an assignment of making a photo-based game of the visited environment: You can create the game on the spot, using the phone’s camera to take the pictures for the scenes that form the game. Of course this mobile editor is limited, and doesn’t allow the same complexity and fine control as the desktop editor. But it is possible to create game drafts on the go and then open them on the desktop for further refinement.

There’s a video for that too.

After developing and testing those two prototypes as branches of the e-Adventure platform, we have already started working on integrating them in our release process. As I mentioned a few posts ago, e-Adventure development is now split in two lanes: maintenance of e-Adventure 1.2 and the parallel development of a brand new e-Adventure 2.0 written from scratch. In the last technical meetings we have been designing a game engine that supports parallel development for the desktop and the mobile version. And once we are done with the engine, we will start working on the editor, incorporating a new look and feel, new game authoring metaphors and exportation support for a variety of formats.

We are not sure when we will be publishing the first betas of e-Adventure 2.0 including the mobile features, but we are sure it will be great.

My research group is involved in the newly created Network of Excellence in Serious Games, funded by the EU to promote Serious games and Game-based Learning by gathering different research teams across Europe. It was about time…

As our field matures, it is becoming increasingly fragmented. Any time I go to a conference, most researchers agree in that “there’s a lot to be done”. So many open research questions. Lacking a proper direction (it is not even obvious what we should be measuring), a lot of researchers end up working in parallel on the same topics. I noticed this for example seeing how closely related were e-Adventure and the 80Days project (adaptive learning with adventure games). Hopefully, with the creation of this network we will increase communication, manage to focus on a vision of the future of serious games, and provide a forum to discuss the progress of game-based learning as a mature research field.

On Sunday my colleague Iván and me will travel to Genoa for a 3-day kick-off meeting. I’m looking forward to meeting the rest of participants and see the potential of the network.

Jeannette Espinoza contacted me a year and a half ago and mentioned that she was enrolled in a graduate course on Developmental Psychology at Complutense, and that she was preparing a short documentary feature treating the relationship between games and motivation.

I agreed to participate with a short interview and now the documentary is available here (sorry, Spanish only): http://vimeo.com/14251116

I think that it is a very interesting and informative piece, regardless of how bad my scarce interventions seem to me.

For anyone thinking that I had actually taken a two-month vacation, I wish you were right. It’s been a hectic return from vacation, with trips, exams and setting up a new course. We are starting our classes next week, and I am both thrilled and a bit overwhelmed with this exciting new course.

This year I’m going to be teaching a brand new course in Programming Fundamentals that marks the start of Bologna-compliant grades in our school. We are changing mostly everything about how we teach this subject, adopting a model that favors day-to-day work rather than a single high-stakes final exam. In theory, this should be better for the students. Studying in small amounts, having smaller groups and having numerous evaluations and reviewed practical exercises sound like good learning principles.

However, how this will work in practice is an unknown. The average student here simply does not go to class. Last year, in one of my groups with 71 enrolled students, the average attendance at the beginning of the semester was 18, a number that dropped to 9 by the last month. This suggests that schedule compatibility and the model based in a single exam seriously affected attendance. On the other hand, only one of the students with over 90% attendance throughout the course ended up failing the exam. Does this suggest that day-to-day attendance and work facilitate passing? Most likely, although this could also be a mere correlation: good students that would pass anyway tend to attend lectures.

What will happen now? Will students maintain low attendance ratios and try to pass with the minimum compulsory attendance (their top potential grade would be capped at 60%)? Will students actually focus, attend lectures and pass more easily? Will we have students attending but failing? Even if it is a lot more work, I am happy to be involved in this new course experience. This is change in how we teach. Not a dramatic change, but a change nonetheless. And if we do it right, we will demonstrate that university teaching and learning can change after all.

August is finally here, and it was about time I had a vacation!

Last year I was in Boston and that pretty much prevented me from taking a proper long vacation. Although I will not disconnect entirely, this year I’m planning to take the month easy and get some rest, preparing for a very intense start of the course in September (both professionally and personally).

This means that I will not be posting (and probably not tweeting) again until September. In anticipation of this blank period, just a quick reflection on the experiment about increasing my posting rate. Early June I decided to try to publish around a post per week and a tweet per day. That post was 8 weeks ago (and 38 work-days ago). Counting the experiment post and this specific one, I have published 9 posts since then (and 39 tweets).

As I said, I don’t mind whether any of the posts was relevant or any the tweets was read by anyone. the good news is that it seems like I can keep the pace. Let’s just hope I don’t loose the habit after the summer break.

In any case, for anyone reading, have a nice summer and see you in September!

I have already mentioned briefly that we have been working with our colleagues in the Department of Physiology at UCM’s Medical School. This collaboration goes back to around 3 years ago, when I met Lola Comas and Carmen Fernández Galaz at a focus group in the School of Education.

We spoke briefly about <e-Adventure>, the kind of games and simulations that could be created and we decide to meet later to discuss potential collaborations. In the end, we decided to create a brief simulation to rehearse the steps of a practical exercise in which the students have to measure Hematocrit levels in a blood sample. This practical exercise is performed once in the course, using blood samples from sacrificed laboratory rats. For this reason, students are not allowed to recreate the exercise out of the scheduled session.

The key idea of developing a simulation game covering the steps of the exercise is that students would be able to practice the exercise at their own pace before going to the lab, thus becoming familiar with the steps of the procedure and later on getting more profit out of the time-limited lab experience. In addition, the students would later be able to practice the virtual version of the exercise before the exam to refresh the procedure. The result is what we call the HTC game, a virtual practical exercise developed with <e-Adventure> using photographs of the actual workbench at the lab, and that can be directly deployed through the Virtual Campus at Complutense Unversity.

The results of the experience have been very interesting. We conducted an experiment separating the students into an Experimental and a Control Group, letting the former play a few days before the actual session and sending the latter directly to the lab. The students in the EC found the practical exercise easier, demonstrated a better grasp on the concept and even made fewer mistakes.

All in all, a great success that we reported an article submitted to the International Journal of Medical Informatics, which published the final version recently. Here is the full citation:

Pablo Moreno-Ger, Javier Torrente, Julián Bustamante, Carmen Fernández-Galaz, Baltasar Fernández-Manjón, María Dolores Comas-Rengifo: Application of a low-cost web-based simulation to improve students’ practical skills in medical education. Int. J. Med. Inform. Vol 79, pp. 459-467, (2010)

The article includes a link to download the game, which can also be found (in English and Spanish) in the <e-Adventure> website. In addition, if you don’t have access to IJMI, our self-archived copy of the draft is also available at the <e-UCM> website repository.

In my last post I mentioned how, after we released <e-Adventure> v1.0, we split our development in two branches.

The first branch was focused on short-term improvements and on adding new simple functionality to our existing model. This line included GUI improvements, general optimizations and new features such as “Drag and drop”, a revamped conversation system or new exportation profiles. As I said in that post, the result of that line of work is the publication of <e-Adventure> v1.2. We have already published a release candidate, which we expect to make final within the next few weeks.

This version marks the end of this line of work. Version 1.2 will continue to be improved and supported, with minor tweaks and solutions to emerging bugs. However, this also our feature-freeze point, and no new things are expected to make it into that branch.

On the other hand, this allows us to concentrate our efforts in what we internally call <e-Adventure> 2.0, which is our current research & innovation platform/sandbox. Among many things, these are the main lines of work within <e-Adventure> 2.0:

• WEEV: Weev is the ongoing thesis project being developed by Eugenio Marchiori, proposing a new metaphor for game creation and visual languages to “weave” stories without thinking about flags and variables. The story of the game and the state changes can be defined with a high level of abstraction using the visual language, and these settings are automatically converted into an <e-Adventure> game with all the flags and conditions correctly configured.
• Android integration: We already have the engine working in Android phones (see our preview) and the next step is to provide an automatic exportation profile to the editor. <e-Adventure> 2.0 will support reading QR-codes or using GPS data to trigger effects when the games are exported to an Android phone, and the editor will allow the configuration of these aspects (we need to do them manually right now). [Read more]
• Accesibility: We have already invested a lot of effort in integrating accessibility features into our games. The next step is to make the editor as accessible as possible, and we are working in this line with Technosite.
• Plug-in architecture for the editor and the engine. The key point here is to allow independent development of new features, affecting the engine, the editor or both.
• New integration architecture, with new ideas and protocols for the integration with Learning Management Systems
• New completely revamped GUI: We are working in the development of our own look & feel, giving a unified appearance to all our projects and (hopefully) solving our ongoing cross-platform GUI issues.

Each of these lines of work would deserve its own post describing it. I will try focus on some of them in the next few weeks.

For the last year, <e-Adventure> development has been split in two main branches. On the one hand, we kept working on the current <e-Adventure> platform, solving bugs, adding a few new features and tweaking some things that we learned could be improved. On the other hand, we started a new development branch, which we internally call <e-Adventure> 2.0.

I will elaborate on what’s brewing for 2.0 in a future post, but today I would like to announce the first line has finally seen the light as <e-Adventure> 1.2. We have published a Release Candidate online and, as usual, we would like to hear some feedback about it. Among the major changes, I think that the “Drag & Drop” feature is the most novel, and covers an often requested improvement. We are excited about the potential uses for this new interaction.

Also, after our experiences with users at the Lab of Computer Science, we have added a new layer of flexibility to the conversations, and it is now possible to configure aspects such as waiting for a user click before changing text lines or having the option to display/hide the last line before a branch (so that we can ask questions and let the question remain onscreen while the student thinks about the answer).

Less prominent would be our new icon set, the performance improvements (both for the editor and the engine) and a reduced installation (which also translates in smaller applets when exporting for the web). Oh! And we also have community-contributed translations to Portuguese and German!

All in all, it is a great improvement, and we would like everyone to try it out. You will not be disappointed. Once we have gathered some feedback, we will release version 1.2 “proper”.