We would like to invite Wikimedia community members to join us at Wikimania. We have two exciting sessions, a master class on sound logos (Aug 11) and a roundtable (Aug 14), and community liaisons to support your participation. See more details on the talk page. Your thoughts and input are warmly welcomed. Thank you.
What is a sound logo?
A sound logo is a brief collection of sounds, such as a short musical mnemonic, usually between 3 to 5 seconds long used to identify content in an audio setting. A sound logo offers us a new way to identify Wikimedia content across a range of possible uses and offers flexibility in that it can be adapted over time to fit different content types while continuing to build association with Wikimedia. This will provide increasing and broader awareness of Wikimedia content, the work of the Wikimedia community, and further opportunities to grow participation in our movement.
How will the sound logo be used?
The main use case is to improve identification of Wikimedia content for data searches using voice-user interface technology. There is broad strategic value to Wikimedia in increasing public recognition of its content as it is re-used in third party platforms and a wide number of audio settings. We hope that the long term benefits will be significant.
While the immediate intention is for use in voice assistant searches, there are other use cases, including notifications in Wikimedia apps and third party apps that reuse Wikimedia content, audio feedback on Wikimedia and third party UIs, and branding on audio and visual content from and licensed by Wikimedia across video, TV, film, podcasts and events. Our objective is to co-create this sound logo with our communities. Adoption of the sound logo will be pursued once this project is complete. We welcome suggestions of other general and community use cases.
Why are we creating a sound logo for Wikimedia?
We have known for a while that direct traffic to our projects has changed over time. A few years ago, we had to accommodate changes reflective of the world around us and adapt to mobile technology. Information seeking has continued to change since, with voice emerging as a leading interface.
The voice assistant market has become substantial and appears to be on the cusp of major multilingual growth. In 2015, there were some 544.1 million unique active voice assistant users worldwide, from consumer and enterprise markets. By 2021, the number of active users has grown to 2.6 billion. Currently, 27% of the world’s online population uses voice search on their mobile devices and many more use non-mobile voice assistants via smart speakers.
Technology companies that offer voice search such as Google, Apple, and Amazon tend to answer general knowledge search queries by scraping data from a variety of sources. Wikimedia content is featured heavily among these sources. One study found that Wikimedia content was used to answer between 81 and 84% of common desktop knowledge queries through Google’s search engine, and that rates were similar for searches on Bing and DuckDuckGo. In the absence of identification of Wikimedia content, users often assume that the knowledge actually comes from the search platform, for example Google. Creating a sound logo for Wikimedia is an important way to preserve Wikimedia’s global reputation, particularly in audio settings.
What about “according to Wikipedia?”
Sound logos generally benefit from simplicity, flexibility, and platform adaptability, and most avoid text or language for this reason. A sound logo based on sounds other than spoken words will be workable across many content types and not be limited to voice search results, as may be the case with “according to Wikipedia”.
Search platforms can help us reach new audiences and achieve our mission of being the infrastructure of free knowledge, but not all companies consistently or effectively identify that certain content comes from Wikimedia. Currently, identification of Wikimedia content through voice assistants is inconsistent. Particularly in:
- Sourcing, specifically the extent to which various large technology platforms use data from Wikimedia content to answer audio queries.
- Identification, specifically the extent to which major technology platforms communicate to knowledge consumers when they are providing content from Wikimedia.
- Identification style, specifically how information is acknowledged as coming from Wikimedia when it is done at all.
When Wikimedia content is used in response to voice searches, it is important that users know where their information comes from. A sound logo will help improve listeners' awareness that the knowledge they receive and value comes from Wikimedia — and from the thousands of volunteers across the world who create and share this knowledge.
How will we create this sound logo?
The organizing team at the Wikimedia Foundation would like to host a global contest to create the Wikimedia sound logo. We hope to make this an exciting, participatory moment across the movement … and beyond, where everyone is invited to create and participate. This contest will draw inspiration from past Wikimedia project logo contests and perhaps other contests. Other Wikimedia logos, such as the puzzle globe logo, the Wikidata logo, the MediaWiki logo and others were developed by community members competing in contests and voting for each other’s work based on specific criteria.
Audio is a fundamentally different medium than imagery and will require different technical means of production, format, and contest criteria. This project will follow similar principles of community consultation, contribution and participation while adapting for the new audio context. This is new for everyone and we are excited to learn together. The pre-launch community consultations (May-June, 2022) will form the backbone of the subsequent phases while ensuring the final selection meets professional sound quality and legal criteria.
How does this project align with the Wikimedia movement?
Our reputation matters so that we can continue to grow participation in our movement, expand our content, and ensure our platforms are trusted and used. Knowledge users should know that the information that they are receiving comes from Wikimedia and associate it with our values and practices: open and free knowledge, global volunteer content creation, trustworthiness, neutrality, and reliability.
As knowledge consumers — our readers, and now also listeners — increasingly source our content via third party search platforms rather than directly from our website or mobile app, we need to find new ways to shape our reputation in these spaces. A sound logo will improve the frequency and consistency with which our content is identified while also linking our content to important values and ideas about Wikimedia.
Our movement strategy recommendation to Innovate In Free Knowledge calls for more initiatives that “Support more diverse modes of consumption and contribution to our projects (e.g. text, audio, visual, video, geospatial, etc.).” In this way, supporting consumption of our projects requires not only that we make our content freely available and easy to access on our platforms, but that we engage users with new forms of technology and help them know and value where the content comes from.
How will communities be impacted?
Sound impacts people on a great visceral level, deeply engaging the listener. It has the power to trigger both positive emotional responses and associative memories as studies have shown that the same parts of our brain are used when listening to music.
Effective sound logos are powerful in that they trigger a positive response in the brain and over time, strengthen their positive association with what they represent. Another advantage of sound logo is helping build memory structures, increasing the likelihood of new users remembering in a positive and non-invasive way that the knowledge they are being served comes from Wikimedia. This is crucial when converting a reader into an editor.
We anticipate that the sound logo will not change the processes by which editors create and curate knowledge on our platforms, but it will positively impact how their collective work is identified when it is used on audio platforms. In the long term, we hope that improved identification of our content will motivate more people to become editors and share in the sum of all knowledge. We invite all sound enthusiasts and musically-inclined community members to participate in creating the Wikimedia sound logo. We hope that the sound lab is a starting point that we can build together over the subsequent weeks and months as the contest begins.
Will the sound logo cover all Wikimedia projects?
For parity across use cases, one sound logo will cover all language versions of Wikimedia projects. We believe it is important to have a singular sound logo to represent all projects, because our goal is to improve recognition of Wikimedia content globally — across millions and potentially billions of knowledge users — where a single signature is much easier to identify and engage with. A singular sound logo can be adapted over time to fit different content types while continuing to build association with Wikimedia without fatiguing or confusing the listener.
Are we partnering with any organizations for this contest?
Creating a sound logo is a new and technically complex endeavor for the Wikimedia movement. The movement has great experience in organizing large, open contests, like for photography or for project logos. However, none have been related to sound. To assist with some of the novel technical needs of this endeavor, the Wikimedia Foundation has partnered with a professional and experienced sonic agency called MassiveMusic. As a leading international expert in sound logos, MassiveMusic will provide guidance and feedback for finding that distinctive sound with movement pride, wide appeal, and a high level of recallability.
Can a sound logo provide attribution?
Wikipedia articles and most other project content are published under Creative Commons licenses that require attribution for reuse. Most Wikimedia projects use a CC BY-SA 3.0 license, with exceptions being Wikidata (CC0) and WikiNews (CC BY 2.5). The Creative Commons licenses and the guidance from the Creative Commons organization (in English) do not explicitly consider how attribution can or should work in an audio context, though they do specify that attribution can be done “in any reasonable manner based on the medium, means and context”.
The sound logo is not meant to replace proper Creative Commons attribution when Wikimedia content is used by audio devices. The primary use case is in circumstances when attribution may not be legally required (such as when Wikidata content is used), but we would still like end users to know that information is being sourced from Wikimedia projects.
In many cases, the type and amount of content that an audio device uses may not be enough to trigger an attribution requirement. Mere facts are not copyrightable, and a certain amount of use is permitted by fair use, fair dealing, and similar limitations on copyright. Even if attribution is not legally required, it is a good and considerate practice for audio devices (and everyone else) to identify their source when they find information on the Wikimedia projects. By creating a sound logo, we will provide a new, easy way to identify Wikimedia content. Attribution on audio platforms is a vital and interesting topic and one that we will explore in the future.
How does this fit into our current licensing rules?
Most content created at the Wikimedia Foundation, with very few exceptions, is released under a CC BY-SA 3.0 or 4.0 license. We go a long way to share accessible brand guidelines with individuals and groups across the movement. We would protect a sound logo as a trademark. Although sonic trademarks are a new practice area for the Foundation, our legal counsel will explore this and provide guidance around copyright protection and ways for the sound logo to be reused by Wikimedians.