Sustainability Report 2019

Dear Freesounders,

Last year we started the tradition of publishing a sustainability report to give you an update about how things are going in the Freesound world sustainability-wise, and to give you more insight about how do we work and what makes Freesound possible. This is the second edition of the report, updated for 2019. You’ll notice that most of the information is very similar to previous year’s post but, especially for those who didn’t read the previous edition or don’t know much about Freesound, this should be an interesting read to get to know more about us.

The report is split in a number of sections discussing different aspects that contribute to the sustainability of Freesound, and a final section with a summary and some conclusions.

Sound uploads

We believe that a very important aspect of the sustainability of Freesound is the continuous uploading of new sounds by the user community. These sounds bring real value to Freesound and make it a useful resource for many users around the world. Specially good-quality sounds. The fact that Freesound is a valuable resource for many users is what is making the donation campaign a success, and what ultimately enables other types of contributions to the sustainability such as those from the university and from research grants (see below). In 2019, 41,450 sounds were uploaded. We consider this to be one of most important contributions in terms of sustainability. Similarly, other user actions like rating (168,000 new ratings in 2019) and commenting sounds (47,000 new comments), writing forum posts (1,100 forum posts) and doing sound moderation, are very important and also contribute to the sustainability of the platform. For a summary of Freesound user activity during 2019, please check the?2019 in numbers?blog post that we recently published.

User donations

In terms of user donations, 2019 has been very similar to 2018, although there has been a slight increase in the total numbers. In 2019, we got 5847 donations (200 more than in 2018), totaling 48,000? (around 2000? more than in 2018). We expected to see a more significant increase in donations provided that we do send reminders to each donor when a year has passed since the last donation – “hi, last year you donated X to Freesound, we’re very happy about that, would you like to donate again?” – and we thought we’d see more “accumulated” users. In any case, we’re deeply thankful for all your support and donations 🙂

We spent the donations in the following development tasks:

• Improvements in the Freesound platform. For a detailed list of developments check the Community Update blog posts of May, July and December. Also you can check our the development at our source code repository. This year we have been less active in the writing of community updates. This is partially because we have been doing less user-facing development (so lots of optimizations and very technical things), and also because we’ve been spending more efforts in research activities which have not been transformed into Freesound code but that some of them eventually will 🙂
• Work on the new Freesound front-end. As I mentioned in a couple of posts this year, we are working on it but things are taking much longer than expected. We’re a very small team and have to divide our time into many activities. New frontend often does not take preference.
• Software licenses and cost of services for the help desk (Zendesk), email services (Amazon), maps services (Mapbox), and site monitoring (Site 24?7). 
  

Contribution from UPF

Freesound is an initiative of the Music Technology Group, a research group of the Universitat Pompeu Fabra (UPF) in Barcelona, Spain. In 2019, the UPF has continued contributing to Freesound similarly to previous years. UPF provides the necessary IT infrastructure and basic maintenance (15 servers, 4 TB monthly data bandwidth, IT support staff). As we calculated for the past edition of the sustainability report, the expected cost of such infrastructure if Freesound was hosted in external services (such as Amazon Web services or similar), would be over 20,000?/year for only hosting costs. This price would be about 5,000?/year by using a cheaper dedicated hosting provider, but we would also have to pay additional IT support costs in this case. Also it has to be considered that some people from the university who dedicates some time to Freesound related activities (either research, development or administration) are also paid by the university. Hence, UPF contribution is huge and we could not live without it.

Contribution from research grants

As being part of a university, research is an important element of the Freesound philosophy. During 2019, the amount of income from research grants which could impact Freesound has been drastically reduced, mostly due to the fact that the AudioCommons project finished last year. We are participating in some research grant proposals in which, if granted, Freesound related activities would take place, but this is not known as of today. However, in 2019 we got awarded a new Google Faculty Research Award which partially funded research on the development of a big dataset made with Freesound content (see this blog post). This is a continuation of a line of research that we started around 3 years ago and that had already been awarded a Google Faculty Research Award. Contributions from research grants resulted in:

• Further development of the Freesound Datasets research platform and FSD dataset.
• Further development and maintenance of Essentia, the audio analysis library that powers Freesound sound analysis
• Research on methods for automatically classifying audio events.
• Further research on clustering methods to be potentially applied to Freesound search results.
  

Commercial usage of the Freesound API

Freesound has an API endpoint which allows third parties to develop applications that incorporate Freesound content. Usage of this API is free for non-commercial purposes. Nevertheless, the commercial use of the API requires a commercial license. In this way we make sure that commercial applications using Freesound also bring something back to the community. Note that this is independent of the license of the sounds themselves, which need to be respected regardless of the API usage agreement. In 2019 our license agreements generated an income of 2,500? (same as in 2018), but we are in negotiation phase for a couple for API clients so we hope this number will increase a bit in 2020. We spent this money in the same development efforts described in the User donations section above.

Summary and perspectives for 2020

As you can see, 2019 has been a very similar year to 2018 in terms of Freesound sustainability. It is great that we consolidated our model and know we are better aware of the expectations that we can have. We’re still far from our sustainability vision that we discussed with the community back in 2017, but we are in a good position to continue advancing. For 2020 we expect to continue in a similar line, maybe being able to turn some of the research results of last year(s) into Freesound website features and therefore spend a bit more time in development as well. If we’re able to get new research grants in which Freesound takes an important role that’d be a game changer and would allow us to do more 🙂

We?d like to finish this post by saying thank you to everyone who contributed to Freesound during 2019, in particular to those who donated and those who uploaded and moderated sounds. We?ll let you know how things go next year in 2020?s sustainability report 

frederic, on behalf of the Freesound Team