In my example Danish national map-data, the data can be accessed by the public without restrictions - although you are required to register as a user at the site. If I answer “No” to the question on public availability I can’t fill in that the data is available for free or the URL of the data and more.
I remember we had this discussion before. @rufuspollock - did we say that in this case the data is publicly available?
We had a similar conversation elsewhere in this forum and are also seeking advice.
OK, after looking at the methodology again. here is my conclusion. The methodology state this -
This question addresses whether the data is "public.” This does not require it to be freely available, but does require that someone outside of the government can access it in some form (examples include if the data is available for purchase, if it exists as a PDF on a website that you can access, if you can get it in paper form — then it is public). If a freedom of information request or similar is needed to access the data, it is not considered publicly available.
Therefore, if you don’t need to request it from government, but just to sign up, it is publicly available. We can be strict, but I don’t think it will serve the purpose.
I will flag this for next year methodology consultation.
@nilleren - I hope it helps!
If you have to register, the process should be automated, so you don’t have to wait.
For example, the national database of public transport timetables in Norway (http://labs.reiseinfo.no) takes a couple of days to set up because they have to create an FTP account, and have a subcontractor set up a user account in their system for you to get API-access.
For some sites, getting access is manually processed, so it takes a couple of days.
I suggest adding a note in the comments if the process is manual.
Underlying here is the question about state sovereignty in open data.
Can government data be open only to citizens of the state and still be open data?
(For this, the registration in unavoidable)
Why, anyway, should the state be obligated to subjects that are not its citizens?
If Open Data movement is insisting on global availability of open data (as it is now), does it mean it is pushing a form of globalization ideology?
Do we, as a proponents of this movement, agree with this ideology?
These are all the questions that need to be dealt with in the close future.
on that note @livarb - what do we think of API keys?
In South Korea, @jgkim said that the dataset is not publicly available because it is requesting an API key. However, I know that an API key is common practice with many API’s, in it usually given to citizens without discrimination and therefore is qualified for public. I did see why it can cause confusion with the current definition, I will fit it. Thus, the part that bothers me is the fact that some API key has to be requested… what then?
Anyone has a different opinion?
I think API keys are okay, but it’s best not to have them.
We could note in the comments if you have to register to get API-keys.
Ideally, retrieving API-keys should be automatic, f.ex. so that you receive a key via e-mail automatically.
We had this at the University of Bergen (data.uib.no). You would immediately receive your API-key via e-mail. The purpose of the key is to be able to contact users if they call for data way to often and overload servers, or disable a specific key in case of some mishap. The keys were meant to be one per application, and you are able to retrieve as many keys as you want to the same email address.
Some services have a manual registration process where it can take days to get access, and they have valid reasons for it. For example labs.rutebok.no mentioned in my previous post.
They should still strive to automate it.
On the other hand - some data providers are fine without API-keys. Ruter - transit agency for the Oslo-area in Norway - did away with their API-keys because it added complexity to their API, and they didn’t have any use for API-keys.
Some providers want API-keys to be able to get more specific statistics on the use of their API and data, while others think the data you get from basic server logs are enough.
@ivobab: as API-keys add complexity to an API, I think screening for nationality would add even more complexity for the services providing data. For that reason, I believe most datasets will be open for anyone on the internet.