08
Mar 15

100% of HOT tools translated to pt_BR

I’ve recently joined the team of Brazilian Portuguese translators for the Humanitarian OpenStreetMap Team tools, namely Task Manager and HOT Export.

I’m glad that after busy days of translations and reviews, we are the first team to complete and review 100% of translations for both projects. Now I’m just waiting for the next release 😉


24
Feb 15

Ebola outbreaks since 1976

In my short participation at the course Principles of the Neospatial Web I had the pleasure to work with two amazing buddies, Rachel and Anna. For our first assignment we had to use the concept of kml tours to tell a history.

We decided to show the chronological and spatial path of the Ebola virus in Africa since 1976. The tour shows the spread of the Ebola virus by country and year. For each outbreak we created a descriptive balloon, using Wikipedia data to provide context to numbers.

Democratic Republic of the congo, 2008 epidemics

One of the narrative balloons

The tour then pans to West Africa and data from the current epidemic is shown. Extruded polygons are used to communicate the relative severity of the epidemic using different colors and heights. Numbers of cases and deaths are shown in a bar graph using the Google chart API.

Number of cases and deaths in Liberia, Guinea and Sierra Leone (2014)

Number of cases and deaths in Liberia, Guinea and Sierra Leone (2014)

The 3 countries shown in the picture (Liberia, Sierra Leone and Guinea) collectively experienced more cases and deaths in 2014 than all other outbreaks in all countries combined since 1976. All previous epidemics have occured in Central Africa. The fact that those countries’ heatlh systems were not prepared for the disease was one of the important factors of the quick epidemic spread.

We’ve tried to show the different paths made by each strain of the Ebola virus along time and space (there were 4 different strains). We couldn’t make it working in a not-too-messy way, so we just finished the tour mapping Ebola Treatment Centers in Sierra Leone, Liberia, and Guinea and showing some pictures.

The downside of working with kml is that the debugging process is painful. With no console indicating errors, it is hard to figure out what is causing the problem when things don’t work. The other thing I wasn’t happy about was to have to use Google Earth on a Windows machine to guarantee that the same code would work for the 3 of us. At the begining I tried to use Marble-qt in Debian while the others used Google Earth, but soon I realized that it wasn’t a good idea. Then I tried Google Earth in Debian, and still, some things appeared differently or just didn’t work in my machine. So I gave up and used the university lab to continue the assignment, otherwise we wouldn’t be able to finish on time.

A good point of this experience is that we have used GitHub to share and merge our code since the begining. It was the first time that the girls used a versioning control system. It was challenging, but at the end all went super well. Now I’m happy to see that they continue using it, even if I’m not there anymore. And even better, I see they are making progress in python. What a nice experience guys!

Thanks 😉