Session 12: Discussion Paper on Licensing

Page 1 of 13

IATI TAG meeting, 4-6 October 2010, Cookham, UK

Session 12: Discussion Paper on Licensing

IATI TAG Secretariat[1], September 2010

Context

Making aid information easily accessible on the internet isn't enough. Aid information must be legally accessible as well to prevent intended or unintended legal restrictions creating a barrier to use and reuse of aid information.

This report is an initial guide to the IATI working group on the legal side of opening up aid information. The report is structured as follows:

Background

Goals

Summary of the IATI information model

Why a license?

The user’s perspective

Survey of aid information site licensing terms

What is open data?

What legal restrictions can you place on open data?

Why an open license?

Should you draft an open license or use an established framework?

Database / Contents distinction

Recommendations

Recommended licensing model

Specific Recommendations when adopting this Model:

The Model Implementation

Next steps

Proposed deliverables

The proposed IATI Licensing Standard

Relevant links

About OKF

Draft IATI Licensing Standard

The recommendations for discussion in the meeting are:

1. Create an IATI Licensing Standard drawing from current and proposed open government principles.
2. The IATI Licensing Standard should be limited to open licenses as defined under the Open Knowledge Definition
3. The IATI Licensing Standard should further limit compliant open licenses to public domain dedications and attribution only licenses.
4. Create a model implementation of the new IATI Licensing Standard making use of existing open licenses.
5. Provide an IP policy for the IATI registry.
6. Create a set of resources to educate and help users about licensing.

Background

The International Aid Transparency Initiative (IATI) is developing an aid information standard for publishing aid information (data) across three strands:

1. Building technical standards for dissemination of aid information to enable use and reuse (aid information as technically open).
2. Creating a centralised registry where the public can conveniently find global aid information (aid information should be easily findable).
3. Developing a model license so that users can be secure in their rights to use and reuse the aid information (aid information is legally open).

Making aid information easily accessible on the internet isn't enough. Aid information must be legally accessible as well to prevent intended or unintended legal restrictions creating a barrier to use and reuse of aid information.

This report and recommendation is an initial guide to the IATI working group on the legal side of opening up aid information.

Goals

The core suggested goal of implementing an open license for IATI is to encourage use and reuse of aid information to the greatest extent possible in order to meet goals such as to:

•Identify potential costs savings;

•Improve funding effectiveness; and

•Gain greater transparency and accountability.

These goals should inform the license choice and framework that developed through this working group.

Summary of the IATI information model

Publishing information as part of the IATI framework will broadly consist of four steps:

1. Collection – information collection activities in relation to aid work.

2. Technical work – publishing information in formats recommended by the IATI technical standard

3. Publication on own site – Organization and presentation of aid information with access tools on a public website, potentially as a microsite or subdomain, such as data.worldbank.org. These sites will use licensing terms and best practices recommended in the IATI Licensing Standard.

4. Aggregation into the IATI registry – the IATI site will, using CKAN software, collect metadata and links to the information published on individual IATI member sites into a centralised resource for users.

Steps 3 and 4 are the main focus of the IATI Licensing Working Group and this report.

Why a license?

A natural question to start: “Why do we need a Licensing Working Group?” The answer arises from the automatic nature of Intellectual Property (IP) law, which means that aid information providers will potentially have IP rights over their aid information and thus will need to license them.

Intellectual property law is generally national in nature, but the internet and aid work is global. Whilst generally facts are not copyrightable, particularly when facts are aggregated into databases or even XML files, IP law questions around copyright and database rights arise even if these collections only contain factual information.[2] These IP law questions around databases and collections of information are often an area that lacks legal clarity in a particular jurisdiction, and the questions and answers from a legal perspective constantly evolve at the global level.

Thus a user, finding information online, should assume that databases and information can be covered by contract law and intellectual property rights such as copyright and (particularly in Europe) database rights. Therefore the default setting in the law is that a user will usually require some sort of permission from the rightsholder (usually the publisher), and this permission takes the form of a license.

Without a license, users can never be sure of their rights to the information and this acts as a chilling effect to innovation (as the users never know if they can be subject to an expensive IP lawsuit).

Giving users a license gives a firm foundation to build new applications, analysis, and tools with the information – particularly important for professional and commercial users.

The user’s perspective

Legal rights often mean getting lawyers involved, which means money (hourly billing) and risk (hauled off to court). For some, it may not be an issue, but if the goal of an information provider is to encourage use and reuse of their information, then addressing at least some of these legal issues will be necessary – especially for commercial users to feel comfortable.

Say for example someone puts up a database on the web without any statement around the rights over that database.

What is a user to do in this situation? The user has four options:

1. Ask for a license from the rightsholder(s) (likely the provider);
2. Use under an exception to the law, such as fair dealing/fair use in copyright;
3. Infringe (and take the risk of getting caught and paying damages); or
4. Find an alternative.

Even finding out whether your use qualifies under an exception to copyright law often involves a lawyer, and asking for licenses (and drafting them) can be time consuming and cost prohibitive (for both parties). The laundry list of all the different legal issues mentioned above must be taken into consideration when opening data – including copyright, database rights, contract.

Survey of aid information site licensing terms

As part of this report, the Open Knowledge Foundation performed a short review of the licensing terms of publishers of aid information. The goal of the review was to quickly assess how easily users can find the copyright/IP policy for the databases and broadly what kinds of restrictions were placed by publishers. This way, the Working Group can establish a baseline of how member publish aid information. We found that:

•Several publish information on their websites with no copyright/IP policies, which would mean users would likely assume the most restrictive conditions on use and reuse. Out of this group several made no mention of copyright at all whereas others had only a copyright notice in the page footer;[3]

•Several have strict terms on use and reuse, including non-commercial use restrictions. Some of these may further still prevent reuse (such as derivative or transformative works) for even non-commercial users;[4] and

•Some had policies equivalent to an “attribution only” license,[5] and one that was currently using a Creative Commons Attribution license (CC-BY).[6]

This survey revealed that currently many providers don’t currently have interoperable (mutually consistent) licensing terms, thus preventing users from mixing data between sources (from a legal perspective), nor have easy-to-use copyright/IP policies that enable wide use and reuse by the public.

What is open data?

From a legal perspective, open data is the application to data and databases of many of the same principles and philosophies as Free and Open Source Software (FOSS) and open content projects such as certain Creative Commons licenses. The open data perspective is most useful for looking at the issues of publishing aid information.

Open data licenses take Option 1 above for database users – asking for a license from the rightsholder – and flips it around to give users a license up front to use the database with limited restrictions.

For data and databases, to be “open” means that users have the ability to:

•mash up / integrate datasets from different providers;

•add, edit, and delete data records;

•change the organisation of the data (its schema) and change the database to a different format;

•copy and distribute the data and the database; and other related rights to make the integration and use of data run smoothly.

For more detail on what it means to be “open”, please see the Open Knowledge Definition at: < The Open Knowledge Definition sets out the principles to define ‘openness’ in knowledge, similar to other definitions as Free Cultural Works and the OSI’s Open Source Definition.

What legal restrictions can you place on open data?

In order for data to be open, the only permissible restrictions on use and reuse are:

•Attribution – requiring some sort of credit or statement of source or contributors; and

•Share Alike (also known as copyleft or reciprocal licensing) – requiring that any derivatives (altered databases) be licensed under the exact same license. These licenses are more restrictive than simple attribution licenses and often include attribution as an additional component.

Of course no IP restrictions at all – a public domain approach – meet the definition of “open data” because users are free to use it for anything.

Note that non-commercial restrictions and restrictions that prohibit modification (a/k/a no derivatives restrictions) are not open under this approach.[7] Thus open aid information must permit commercial use.

Again, for more detail on what it means to be open, please see the Open Definition at <

Why an open license?

Open source software licensing and open content licenses have made great strides in developing stable, useful software and rich content available for the general public. The goal of open data is to bring this established approach of open licensing to data.

A key feature of these approaches is that the work should be available for further use and re-use by the public without the need to seek further permission from the rightsholder. The rightsholder up front gives the world a license to use their work. This is important as negotiating individual licenses can waste resources (for both the rightsholder and the user), especially when the rightsholder would like to share their information but has no easy way of doing so.

Should you draft an open license or use an established framework?

Information publishers such as IATI members generally have two options when considering an open license:

1. Draft their own IP license that complies with a standard for “openness”, such as the Open Definition; or
2. Use an already established open license, such as the open licenses available from Creative Commons or Open Data Commons.

Using an established open data license is highly recommendedas it means:

•Fewer resources – someone has already done the hard (and often expensive) work of drafting a license for you.

•Easier for users – your users spend less time reading licenses and figuring out the terms if they only have to do it once for each license.

•Leverages external resources – many licensing communities have FAQs and license-specific discussion lists to help users answer any licensing questions that come up.

•Builds a community of resources – having a large number of resources under the same license makes it easier for users to see how they can combine information from multiple sources.

Database / Contents distinction

Often when people talk about data they only mean factual information – water boils at 100℃ and the like. But in the context of open data, the contents of a database can be anything, including:

•Mobile video

•Images, such as from Flickr

•Text documents, such as Microsoft Word .doc files

•Factual information, such as dates or budget numbers

Anything that can be stored in a database can be “data” in this context. That’s why it can be helpful to distinguish the two as the “contents of a database” and “the database”. This is the data/database distinction, and from a legal perspective it means:

•There may be legal rights specific to the contents of a database (the data, because it can be anything).

•Any collection of contents (the database, and potentially even an XML file) may have legal rights covering that collection that are independent of the rights over individual contents within the collection.

So for example, if you had a database of still images and video of science fiction movies from the 1960’s and 1970’s, you:

•May have rights, including for your selection and arrangement of the sci-fi films, over the database as a whole.

•You’d still have to clear the rights (such as copyright) over the images and video you included inside your database.

It’s a bit like a layer cake - the bottom layer rights over the contents, the top layer rights over the database.

This is the reason why more than one open license will be necessary to cover both the database and the contents, including potentially the website and associated materials. All of these can however be wrapped into a single Intellectual Property Policy to govern the use of the information for your users and their use and relation to each other plainly explained in the IP Policy and in any accompanying FAQs.

Recommendations

OKF recommends adopting a licensing model that allows individual aid information publishers to have their own site policies within a narrowly defined range of acceptable terms set by IATI. This model allows the most flexibility and control for IATI publishers and at the same time ensures that the aid information remains open and accessible to the public. The recommended approach means that IATI should:

•Create a licensing standard (much like the technical standard) that defines the range of acceptable licenses its members can have.

•Help its members implement the standard by creating a model license that makes use of existing open data licenses.

•Provide FAQs and other material to help users understand and comply with the compliant licenses of aid information publishers.

Specific Recommendations when adopting this Model:

Recommendation 1: Create an IATI Licensing Standard. This standard will define the boundaries for IATI member licenses to promote use and reuse of the aid information. IATI members will be publishing aid information on their own websites, which will then be collected into the main IATI website. As publishers, each website will need its own licensing terms on the site.

By defining a Licensing Standard of the licenses for individual information publishers, IATI will:

•Allow flexibility for IATI members to take into account local law;

•Accommodate IATI members already publishing information under licenses compliant with the proposed IATI Licensing Standard (so they won’t have to revisit the licensing issues if they don’t have to);

•Simplify licensing issues for IATI users, as the type and kinds of licenses used will be confined to a set compliant with the standard.

•Help ensure legal interoperability (the ability to mix information from a licensing perspective) of aid information between IATI publishers.

The new IATI Licensing Standard should draw from current and proposed open government principles, such as the UK’s Public Data Principles.[8]

Recommendation 2: The IATI Licensing Standard should be limited to open licenses as defined under the Open Knowledge Definition. The Open Knowledge Definition ensures that IATI users can freely use, reuse, and redistribute the information. It is a stable and internationally used definition of openness built on principles tested in the context of open source and open content over a number of years.

Recommendation 3: The IATI Licensing Standard should further limit compliant open licenses to public domain dedications and attribution only licenses. As part of this approach, IATI publishers would be free to choose from either an attribution or public domain approach for their information.

The core goal of opening up aid information is to encourage its use and reuse as widely as possible and so IATI should strongly consider a public domain approach (i.e., no restrictions on use) as this will: