Caros,
esse fds vai rolar o TransparencyCamp lá nas gringa. No blog da Sunlight os caras propuseram 3 desafios para serem resolvidos no evento... acho que quem puder acompanhar online, vale. E mais que isso, acho que temos que ficar atentos e ver como tropicalizamos essas discussões.
Segue em baixo a integra do post, mas resumo os 3 desafios:
----
With TransparencyCamp kicking off this weekend, I want to issue a challenge to the entire open government and transparency community to help solve three big, easy problems. Starting this weekend, I'd like us to plan how we're going to solve these problems, and to have them solved them all by July 4th, 2010. We'll call it "Data Independence Day."
These ideas need full participation from as many people as possible and as many ideas as possible to get the best result. As such-- in conjunction with Phil Ashlock (who is awesome) over at TOPP Labs (who are awesome) we'll be working on these not only at TransparencyCamp here in Washington, DC but also atOpenGov West (which is awesome)
Here are the three big problems I want us to solve together:
Build an Open Data playbook. An instruction manual for people inside government to teach them how to open their data. This would range from dealing with the bureaucracy to convince the organization to go "open" to dealing with the technology one needs to go "open". The resulting document-- a 20-40 page HTML/Word/PDF could be given to people inside the government, inside NGOs and inside corporations to help them deal with the tactical issues of going "OpenData".
Build an open and shared resource: compile a list of all jurisdictions and elected officials in the country. Make it so that the entire community can share in this common resource, and make it so that we can count in it being updated when it needs to be updated. This constantly evolving dataset will support a tremendous number of municipal, state, and nation-wide projects.
Come up with a data exchange format for data catalogs. As we prepare the National Data Catalog for release, and begin to write importers, it's clear we need to have a model like Google did with GTFS. Let's create a basic file format for data catalogs, agree to use them so that our platforms will be interoperable, and get government to use it, too.
These three problems can be solved by our community. They can't be solved by any one organization, or any one person. They need leadership, community, and focus. They're not hard problems, they just need our focus and attention for a few weeks. Together, we can get organized and make it happen.
Who is in?
Numa segunda etapa partiria para outros dados, contando com a
colaboração de todos a partir do primeiro exemplo. Também gostaria de
tornar o parser distribuído, do tipo http://grub.org (um "google"
colaborativo que está sem patrocínio atualmente). Dependendo dos
acessos, o gargalo seria o banco de dados, e ainda não ouvi falar que
o YQL suporta balanceamento de carga entre diversos servidores.
Outra ideia é usar hadoop para armazenar e efetuar as queries, porém o
retorno de dados teria um delay mínimo um pouco alto. Mas isso deixo
para decidir depois, ainda estou no parser. Quem sabe alguém não
sugere outras maneiras...
2010/3/25 Pedro Markun <pe...@esfera.mobi>:
> --
> Você está recebendo esta mensagem porque se inscreveu no grupo
> "Transparência Hackday" dos Grupos do Google.
> Para postar neste grupo, envie um e-mail para thac...@googlegroups.com.
> Para cancelar a inscrição nesse grupo, envie um e-mail para
> thackday+u...@googlegroups.com.
> Para obter mais opções, visite esse grupo em
> http://groups.google.com/group/thackday?hl=pt-BR.
>