Louhos Project - Finnish Open Data for Local Community. Leo Lahti at Share Conference 2


At Share Conference 2, Leo Lahti presented open data project Louhos (and in English here), a joint effort of three Finnish developers. Together with Leo Lahti, members of the team are Juuso Parkkinen and Joona Lehtomäki.

Straight on, citing Lahti, the main goals of the project are two: a) to collect and develop tools to analyse open data, 2) use these tools to understand how society works and make it more transparent.

Lahti spoke about the role of data in “the accumulation of knowledge in the society”. “As we are now starting to have more and more open data, (..) there are different problems to be faced”.
Lahti individuates three essential passages for a constructive use of data:
  1. once the data is gathered, the problem is to translate it into information and facts. Only once this has been done, we can pass on the higher level of knowledge.
  2. understanding the systems from which the data was extracted,
  3. third moment is using data to “help to guide action”, deciding “what to do”.

So, Lahti proposes that the final aim of data is to “return” it to the society.
The second topic Lahti spoke about concerns the integration of data from various sources. In his opinion, creating “real value” from the data lies in its combining “across different domains, (..) not from individual data sets”.
On that subject, Lahti provided us with an example on how Louhos produced a visualisation of the proposal for the reshaping of municipalities in Finland. The group wanted to demonstrate how it would change fiscal policies based on changing population densities in different regions. To realise that, Louhos combined three sources of data: Finnish Broadcasting Company, Finnish Land Survey and Statistics Finland. The precondition for this very informing visualisation was the disposition of these agencies to share their data, other countries should take example. But the point is that even when that kind of data is made available it means little if it is not integrated, processed, analysed and eventually visualised.

The third topic Lahti focussed was the gap between data sets and tools for data analysis. The goal is to “integrate tools and data”. In his view, “open data movement is now shifting from just releasing the data, to releasing the tools that can make use of this data”. “Availability of computational tools has proven to be a bottleneck of open data towards a more widespread use”, as their web site states.

Subsequently, Lahti outlined how he sees this development. On one side, a “general purpose tool” should be created, which will not be “restricted to a particular application domain”, so it should be “like a programming language”, a kind of a “ready made tool to access the data”. So, this is a tool that could be pooled and shared between various localities over the Internet.
But, on the other hand, the tools should be localised. Data sources are different in every country, and local needs also vary a great deal. So, in Lahti's opinion, pre-packaged tools are not the solution that can be employed in any given situation anywhere in the world.
This is where the real effort lies, in building of “complementary data tools” which best suit the present needs and can allow us to delve “deeper into the workings of your own society”.

To illustrate this proposal, Lahti provided us with a few examples. (As a matter of fact, all Louhos works are locally grounded and targeted, so their blog offers a great variety of cases.)
Oneof this was an interactive visualisation of the popularity of candidates in last Finnish elections intersected with location of voters and a variety of socioeconomic indicators to choose from: income level, education, healthcare status, etc. The richness of indicators and three dimensions of the graph allow for in-depth sociopolitical analysis of Finnish voters.

Another visualisation showed the frequency with which the members of Finnish Parliament spoke about certain arguments. This visualisation proves certainly to be more valuable on understanding political agenda of diverse representatives and their party groupings, than reading any kind of party political programme. A great tool for understanding what is really happening in the parliament, without having to really spend entire days in front of the TV listening to the debates.

The visualisation of migrations to and from Finland is linear but insightful. Colours tending to blue show to which countries Finnish people moved to, and the ones inclined towards pink and red are the ones from which the immigrants came. (I can't pass without noting that the only neutral countries, or countries without data, are traditionally migrant Albania and Serbia.)



One of heavily debated arguments in Finnish public sphere has been whether Finland should be a member of NATO or not. Louhos intersected the level of income with the level of support towards the membership. The outcome was quite revealing, as you can see in the graph, and it showed that the most avid supporters belong to the high income category. In this manner, Louhos poses a perfectly valid question: what might be an explanation?

Thereafter, Lahti noted that Helsinki is rich with georeferenced open data thanks to a collective effort of governmental, non-governmental agencies and people contributing to data gathering. Based on this wealth of information, Louhos has produced some interesting information mappings, for example, of Helsinki property prices, of public services in parks etc. This practically demonstrates how a joint effort towards opening the data in Finland, top-down and bottom-up, produces a truly reflexive society. A society which knows more about itself, and is in continuous quest to deepen that knowledge.

All the above cited works have been specifically modelled on available Finnish data sets, and the knowledge they foster empowers Finnish citizens in their decision-making. Moreover, all Louhos' codes are readily available on web, and they are extensively commented. In this way the developers trio invites everyone to build upon these tools. Besides, Louhos has published the complete open data tool kitsoRvi (“lathe” in English) which integrates “seamlessly” all the passages from processing the data sets to their visualisation. It is available for download, and its code is open, so I can only imagine how many insightful visualisations can we see coming in the future from Louhos and from Finland. And they are only gathering momentum, lathe is still at 0.1.58 version.
In line with this magnificently called tool, I will conclude by saying that Louhos in Finnish means “quarry”, so it is an open call for digging into it, and there sure is plenty of material..

Finally, I may say that after a bit of exploration of Louhos blog I feel like I really have an idea of how Finnish society today looks like, and where it is headed to (something to which Wiki articles or extensive journalistic coverage should strive to). Besides, I can now complement this knowledge with cinephile social and urban meta-data on Finland I learnt from Aki Kaurismäki's opus.

Last thing, Lahti invites us to Open Knowledge Festival in Helsinki in September. it sounds as a great hands-on opportunity to get in touch with Finnish open data community.


No comments:

Post a Comment