Open data is one of the buzzwords of the moment. Yet, the people who ask us the question often don't know exactly what it entails. Let's look at the definition:
"Open data is digital data that is made freely available for users to access and use. It can be public or private in origin, produced by a local authority, a public service, a citizen group, or a company. It is distributed in a structured manner according to a method and an open license that guarantees free access and reuse by everyone, without technical, legal, or financial restrictions.
Data openness is simultaneously a movement, a philosophy of access to information, a public policy, and a practice of publishing data that is freely accessible and usable." (cf. Wikipedia)
In its philosophy, open data is an incredible movement, which holds a lot of meaning for us. In fact, in its very first version, Soliguide was open data: the site's data could be downloaded directly in .CSV format.
But remember, Soliguide didn't always have the beautiful interface it has today!
But there is a catch. Open data, by its very nature, assumes that there is no control over the reuse of the data. Anyone could therefore, in a matter of moments, download all or part of the Soliguide database to put it online on another service. So far, so good! The idea of proliferating the ways to access information is actually very good : it doesn't matter if people access it through Soliguide or elsewhere, all that matters is having the right information. And that is where the problem lies.
Information is only valuable if it is up to date. At the Soliguide team, it is almost all of the resources dedicated to ensuring the quality of this information. It is a monumental task: behind every department mapped on Soliguide, there is someone working tirelessly every day to build, maintain, and refine the database.
With this type of data directory project, there is a misconception that people will list themselves and update their own information. Some organizations do, and we thank them for it, but that is far from the norm. There is no magic formula that would allow all data to be updated effortlessly and at no cost; it requires a lot of processes, trial and error, optimization, automation... and a lot of hard work!
The Soliguide team works every day to keep the platform up to date!
If Soliguide were to become OpenData tomorrow, anyone could download this information and put it online. And then leave it to its fate. There is no shortage of examples: in various regions or sectors, embryonic versions of Soliguide have sprouted up.
In 2016, for example, there was Welcome Map, an interactive map of useful locations for refugees in Paris. After numerous press articles, the project was abandoned almost immediately. That is the real problem with data-focused projects: people think the work is finished once the database is created, when in reality, it has only just begun.
Fortunately, there hasn't been a mass casualty of such projects. The problem is thatthe internet never forgets, and this data remains online. People, often in already vulnerable situations, risk stumbling upon it and using the online information only to end up in front of a closed door: in the meantime, there may have been relocations, changes in services, or updated hours. The social services sector, in particular, is subject to rapid changes. After spending time, a transit fare, and immeasurable energy crossing an entire city, finding yourself in front of a closed door is a crushing blow. And perhaps that person, discouraged, will no longer try to seek help.
Making Soliguide data OpenData means risking flooding the web with outdated information copied from Soliguide and left to rot, where it could have a disastrous impact.
Yet, the question keeps us up at night. The philosophy behind Open Data is self-evident: data should be free, transparent, and shareable.
We are convinced that Soliguide’s data could be used in thousands of ways that we haven’t even thought of yet. For example, in early 2017, 4 students from Le Wagon came up with SOS Food : a service that allowed homeless people to send a text to a dedicated number and automatically receive the address of the nearest food distribution point. The idea is simply brilliant! But, as is often the case, after a few press articles, the project was forgotten within a few months.
What if brilliant projects like SOS Food had access to a complete, high-quality, and up-to-date database? It might have been fully deployed and helped thousands of people find food.
Our conclusion today is to work on an API that allows our database to be shared dynamically with authorized partners. Soliguide’s database remains protected from being copied; we simply provide access to it to make the information available to as many people as possible, regardless of the platform.
This decision is not final, as we fully intend to (re)test the Open Data format in certain regions or under specific conditions.
In short, would making Soliguide’s data Open Data today help better inform people in situations of extreme poverty? No. Could this information be copied, left to go stale, and risk misdirecting these people? Yes. In short: not really.
Open Data is a concept that is, at its core, only just beginning to fully unfold. There are many positive signs, in particular the requirement for local authorities to share their data via Open Data since the end of 2018.
Much like its predecessor, Open Source, which has expanded its range of licenses over the years, Open Data licenses could also be broadened and clarified. For instance, in Open Source, there are licenses that allow for the free use of code provided that any modifications made to it are also shared. What if, in the same way, we envisioned an Open Data model that allowed for free use on the condition that the data is kept at least as up-to-date as the original database?
