Monthly Archives: August 2013

The end of summer. The end of internship. The end of awesome.

Our superstar intern @AnnaPawlicka finished her time with us last Friday *sniffles*Here’s her look back on the experience, cross-posted from her blog . If we can keep hiring people like her and making them happy, we’ll know we’re doing something right. 

This has been a whirlwind of a summer, and I can’t believe it’s come to an end.  Working at Mastodon C has been exactly what I wanted it to be: an awesome experience. I can honestly say I enjoyed every single day. Whether I was working my way out of JavaScript’s scoping labyrinths or learning the power of Cascalog, there was a sense of pride for me knowing that I had directly participated as a part of the team. With a very very rare coffee requests ;-)

With the recent rise of big data and open data I feel very lucky to have been in the middle of it. During my first week of the internship I was introduced to the CDEC Open Health Data Platform idea, given a bunch of files, some examples of maps and Cascalog queries, and off I went.

It took me a couple of hours to work through Paul Lam’s “Cascalog for the Impatient” – enough to start writing my own queries. Cascalog, and Clojure for that matter, are amazing. You get to do stuff fast, clean and in a functional way. I used it to scrub the data and calculate diabetes prevalence stats.  If a recent graduate whose only experience is in Java, is able to do that in a few days, then everyone who is dismissing the power of Clojure should go and ponder over their technology choices ;-)  Feel free to have a look at the GitHub repo, all code is open source.

I have been a frequent user of online maps, but I have not created any maps myself. This was about to change. I was going to take the data processed with Cascalog  and visualise it on interactive maps and charts. Brilliant JavaScript libraries made this process much easier: leaflet and dimple. And although my JavaScript was rusty, it all soon started to come together. The (still ongoing) results can be seen on the alpha site. The feedback it received on the day of the launch was the best reward I could imagine.

Since then I have created more maps, more charts and more queries. And during that time I have always been myself: I came to code, Mastodon C allowed me to do just that. It has been like working in this unreal world where I wake up in the morning wanting to go to work, and where every day helps me to live my life in a more rewarding way. Being able to use my skills to help someone and see their positive reaction – priceless.

Thank you to all Mastodonians for guiding the way for me throughout the entire summer.

Fran (@fhr), for being a friendly, approachable boss, for introducing me to Women in Data, for inspiring me to overcome my fears of public speaking and balancing out the sci-fi references.

Bruce (@otfrom), for always having a positive attitude and constantly helping to point me in the right direction, for introducing me to the world of Clojure and for the long list of all sci-fi films/books I’m yet to watch/read.

Neale (@sw1nn), for the Iron Maiden talks, for always willing to help out, for inspiring me through your passion for technology and keeping lunch breaks interesting thanks to your amazing skill of remembering facts.

Steve (@sf105) and Merici (@merici), it’s been a pleasure to work with you both, even if only briefly. Thank you for advice on professional career and for the froyo!

It has been a wonderful experience to be part of a company that does meaningful things. If I were asked again what I wanted to be when I grow up, my answer would be: a Giant Robot Mastodonian: professional, expert and geeky. You guys rock!

Advertisements

Open Health Data – alpha launch happy dance

[cross posted from https://cde.catapult.org.uk/%5D

We just launched the alpha of the CDEC Open Health Data Platform. It’s been an exciting process developing the platform so far, and we’re really pleased with the way the alpha site looks and the reactions it’s had. The platform in general is a really fun project: it’s motivated by the fact that there’s an enormous amount of interesting and valuable raw health data available, from the NHS and elsewhere, but that it’s relatively difficult for business and government to make use of that data without a good reusable technical base to help them build different views and services around it. 
 
It’s easy to see how that data holds a lot of economic and social value for so many parties. For example, public, private and third sector services providers want to understand existing need and provision in different areas, to know where to send resources; journalists and politicians want to understand what health issues are topical in their area; and of course medical suppliers want to understand prevalence of conditions, supply of resources, trends and patterns. The aim behind the platform is to help all these parties to get hold of that possible value, by making the raw data easier to use and reuse. 
 
The idea behind having a very early alpha, and of developing the site very visibly in public, is to get a minimal version out for discussion which will make sure it is developing towards what people need. The project itself is quite an unusual one, since CDEC aren’t the ‘true’ end users: the users are really the people who we hope will take and repurpose the platform code, and the people who’ll use the interactive front end either of our version or of the repurposed versions of the platform. This gives us a bit of a dilemma: it would be impossible, and pretty stupid, to try and build something which serves every potential need we can think of (there are a lot of different potential ways of looking at this data) so we have to choose some example cases to build functionality around. 
 
For the alpha launch, we looked at data relating to diabetes – prescriptions, prevalence, and other information. Diabetes is a good ‘example’ condition to hang development around, as it’s common, chronic, affected by multiple environmental, medical, and social factors, and expensive to manage – so there are strong economic incentives for lots of different players to make sense of diabetes data for lots of different purposes: both charities like Diabetes UK, health providers, and private sector providers who serve diabetes patients. 
 
We and CDEC are very happy with the way this first alpha version has come out, and have come up with a lot more questions based on the initial site. We’re also now looking to get feedback on what direction this should build in, and what functionality is important to users (both those who want to repurpose the code for themselves, and those who want to use the existing front end). If that’s you, drop us a line, or even better, raise an issue or fork the code on Github we positively want you to take it, use it, and make money or do other good stuff with this material.