I’m crashing the forum to let the community know about the launch of Presupuesto Abierto, a web platform for visualization of public spending and revenue datasets. The first two instances are now online: http://bahia.presupuestoabierto.org (budget for the city of Bahía Blanca, Argentina. Its government was the main sponsor for this project) and http://caba.presupuestoabierto.org (budget of Buenos Aires, imported from their data portal.
Presupuesto Abierto is part of my thesis project at Macro Connections (MIT Media Lab).
During January 2016 I’ll be working on opensourcing some of the components that comprise the system, particularly the REST interface that I built for the open source OLAP engine Mondrian. I believe that easy to use OLAP tools can be useful for the Open {Data,Government,Spending} communities, and I hope to make a contribution to that area.
I’m also interested in supporting the Fiscal Data package spec in the software that I will be releasing in the following months.
That’s excellent @manuel! ¡Felicitaciones! I’m looking forward to seeing the code. I would also love to hear your feedback on the developing Fiscal Data package spec: Issues · openspending/fiscal-data-package · GitHub. I’m sure you could provide some really good insights here
Brilliant, @manuel! It’s UX is smooth like a baby’s ■■■, and the simple options for browsing make it really useable. How about a blog about your experience with Mondrian, I’d love to learn more!
Thanks for your kind words! Blogging about Mondrian is definitely in the to-do list. I plan to spend a couple weeks in January documenting the code, so a post will go along with the release.
@danfowler: I’ve just left a couple comments on the tracker. BTW, are there any examples of budgetary datasets already packaged as Fiscal Data? I’d love to take a stab at writing an import tool for Presupuesto Abierto.
I’ve started pushing some code to GitHub - jazzido/mondrian-rest: A REST interface for Mondrian ROLAP server — It’s not yet functional, and lacks documentation. Nevertheless, the tests provide some hints.
It’s similar in scope to Cubes and Babbage, but uses Mondrian instead of a custom ROLAP engine.
