# Publish your first dataset

*These are* [*Aude*](https://twitter.com/Audesome)*'s notes – Aude was facilitating this session*.

With this session I wanted to see if in the course of 40min we could decide on a first dataset we could all publish, when we got back to our respective libraries. In the near future the Libraries Taskforce and the LGA will be looking at proper schemas for public library data, which will be more complex than what we could tackle in this session. Nevertheless, I wanted to both discuss with the other session participants the kind of things one would have to think about when starting to release data, and for us to create something together that we could all take back to our own libraries to then simply go ahead and start publishing open datasets.

First we had to agree on **what kind of data to publish**. I thought visitor figures would be a straightforward one, but it turned out that one of the library services represented in the room do not collect that type of figures at the moment! So we agreed on **issue figures** as all participants, regardless of which library management system they use, do collate numbers of loans from the service.

What do we mean by issue figures? We need to **be clear on the definition** to make sure we are all publishing the same thing. We decided that in this case issues are loans or renewals of physical items (therefore excluding e-books) – any items that appear in the library catalogue and can be borrowed directly via the library management system, be they books or bikes. (One of the library services in the room did lend out bikes via the LMS!)

How are we going to **present those issue figures**? Should they be shown by library by month? Or organised by item type by borrower type? Or a mix of those? We thought they would most helpful if presented by library (each library a line) and by month (each month a column). We can start the dataset at whatever date we want: either the previous month or go further back in time if the data is easily available.

What **type of document** will our dataset be in? We agreed on a spreadsheet, saved as a CSV file.

Since we are going to use dates for the month, we also need to decide on a **date format** to use. We chose YYYYMM.

Next is the copyright **licence**. For it to be open data we need to place our dataset under an open licence, which allows anyone to re-use and build on the data for free, even for commercial purposes. Most local authorities already publish some data under an open licence using the [Open Government Licence (version 3)](https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/) so the group decided this would be the preferred licence.

And finally: **where will we publish** our issue figures open dataset? The obvious place is the library's web pages. Some local authorities may have signed up to open data repositories like the [Data Mill North](https://datamillnorth.org/) or [The Data Place](https://thedata.place/), which make it easier to upload and manage files. Another option is to create a (free) [GitHub](https://github.com/) account for the library and upload it there (while linking to it from the library's web pages). When you publish your dataset, do explain what it is (e.g. what your definition of “issues” encompasses) and what can be done with it (the licence it's under).

And that's it! You could publish your very own first library dataset today. If you do get datasets published I'd love to hear about it; if you have any questions do contact me as well at aude\[dot]charillon\[at]newcastle\[dot]gov\[dot]uk

![Photo of flipchart notes page1](https://raw.githubusercontent.com/LibrariesHacked/data-treaders/master/images/DSCN3897.JPG)

![Photo of flipchart notes page2](https://raw.githubusercontent.com/LibrariesHacked/data-treaders/master/images/DSCN3898.JPG)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://librarydata.gitbook.io/voyage/publish-your-first-dataset.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.