Building a healthy public service internet

-
A rough write up of the BBC and Mozilla event at Mozfest House on 23rd October 2018. Wendy Grossman also wrote up the event.
A lot of similar issues are touched on in Katja Bego’s post on the ten challenges facing the internet, which is a useful illustration of the breadth of issues under consideration; in a stack concept, from networking and censorship, access, digital infrastructure like identity, applications and services; or networking vs compute vs data and information; or by function — gaming, finance, government services, social, education, civil freedoms, transport, internet of things…
Other interesting links:
BBC R&D and Mozilla brought together a group of folks to explore what a healthy public service internet might look like, plus whether this is the right question or an idea worth pursuing.
One of the sessions discussed the idea of ‘public service internet’; and which public, what service, and what we mean by ‘internet’. This sounds frivolous, but touched on a lot of important things that would need to be clarified to develop any sort of concrete proposal. There’s a BBC flavour to the idea of public service, which might or might not be useful; most of the day was spent discussing ideas very similar to those at the Waag Public Stack Summit.
We agreed it was about more than just applications and websites, and that other parts of the stack need consideration (whether that’s about access or censorship, distribution of information or compute power, basic services such as identity. As Bill Thompson asked, how far down the stack do we need to go? Do we need to rebuild TCP/IP?
It seems like a deeper analysis is needed than was possible in one day, even with a thoughtful selection of attendees. We need to take stock of what we have, and identify which bits are most in need of change. (Waag’s work created a diagram showing services and software that comprise aspects of the public stack idea, and maybe that’s a framing that could be reused for a wider evaluation of both mainstream services and systems, and existing viable alternatives.) This needs to go beyond the headline concerns about Facebook and Google and nation state threats, and explore deeper. At the Public Stack summit, we heard about the centralisation around SMTP — an open protocol, but where other constraints limit the benefits of openness. We need to appreciate these issues too.
An important point was that there will never be a perfect healthy internet, because there will always be parts that are less good, reflecting human nature. And even if you build (parts of) a public service internet to enable people to benefit, they could still be used for malicious purposes. There will always be a tension between good and bad use of tools, and so as well as building new technology, there is political work to be done around how the technology can be used too. If we had this public sector internet, who would be the guardians of it? They might not — perhaps should not — be the people who build it, but people and institutions who would hold the values and ensure the makers and operators were held to account.
The ways in which internet services (from social apps to access to the network at all) are funded are critical components, both in terms of the risk capital to develop a new thing, and the operating model. Another question is whether we are building a public service internet in the sense of the BBC and its funding model, or a fair trade internet (for-profit, but with a higher purpose, embodied or certified in some way), or a post-capitalist internet :) The fair trade internet doesn’t feel like it’s strong enough to really transform this space. We also talked about co-ops. Co-operatives give us a structure which we can use today; which is pretty well established; which does a good job of preventing takeover or capture; and gives genuine collective ownership and control. This could be better than today’s BBC model, as well as better than the pure play for profit incorporation model. Co-ops also support each other — it’s one of the Rochdale principles — opening up a potential funding pool. Co-operatives represent 5% of the world’s GDP, so that’s quite a good base to draw upon, even if only a fraction of a percent was dedicated to new mutual technology development.
Another question is more technical. How deep do we want to re-engineer things? Are we building things for use now, next year, in 3 years, in 10–20 years? What application areas do we build first? Which give greatest leverage or address the most urgent and important problems we face with today’s internet? Are we fixing what’s broken, building something better, or both?
The idea of many different internets, offering different services and tradeoffs, and based on different values, is interesting. Maybe we don’t replace what we have, but build something alongside it, which operates partly or at first on top of it. Perhaps we have a future of many internets, not based on location but where you choose which one to use for a specific activity. There would need to be defined edges between different levels of governance and management.
There’s still more work to be done to even start creating a roadmap for a new project. Aside from the technical, economic, social and political scoping, there’s some very fundamental work around understanding what values a new internet would embody, and where tradeoffs and compromises would go. Because it’s not going to be possible to create a utopian future — we can’t have all the things we want, realistically, on a reasonable timescale. The compromises will be tough, and that means figuring out what values we’re fighting for and how we prioritise them.
Some of the ideas are already out there — in the form of new apps, new protocols — but they need adoption to make a difference. We also talked about open protocols and ideas of the past, that perhaps have in fact been adopted, but where that’s not been enough. XMPP for instance, which was a fairly dominent interoperable messaging protocol, was even used by Google and Facebook, until it wasn’t. Such things need, maybe, more consideration of politics and economics to sustain and achieve the goals we might wish for them. We talked about how in the early dotcom days the culture of the internet was about getting rich quick, and making money — and therefore that’s the internet which developed from then on. An era of strong capitalism gave us a capitalist internet. We also talked about learning from history. On one level we have the history told by the winners, the main narrative which talks about the technologies which succeeded and tells a heroic story about that. Ira Bolyshevsky mentioned history written by the losers, the frustrated tales of projects which didn’t succeed; what do these tales miss out? Perhaps that many of the early public service and open internet projects didn’t consider money, or how they would be funded, and so were sidelined by better resourced projects which had a model to sustain them. We need to understand how projects will operate and where operating resources will come from, as well as how their initial development will be funded (when it’s not clear if they will succeed, and so risk financing is needed).
We also talked at a slightly meta level about who was in the room, in the conversation. How can we resolve the tensions between malicious use and good benefits of the internet, when the policy debate is often so shallow? How can we tackle this practically, and who do we need in the room to get the conversation to a greater maturity? We also talked about whether we need the ‘other side’ in the room, the big corporates and governments keen on surveillance, and how we could enable them to see that an alternative internet could be something they might want in the future? That a parallel system with greater privacy etc, and choice between internets, could be a positive thing, even if they wouldn’t be involved in building it.

Although a lot of the day was about big ideas, there were some quite tangible, practical project concepts emerging too:
  • what would a 21st century public service internet for a deprived UK community look like? This could be a research project, looking from the ground up at the needs of a struggling community and imagining what internet services, access, tools and infrastructure they might need.
  • if we had a great public service internet, delivering on the ideas for fixing what’s broken or building something better that are going around, what stories could we tell about the better future it enables? This could be a storytelling project, to articulate the positives of where we’d like to get to.
  • Then, can we use these stories to debate what parts of them we like and what parts we don’t like? Although it’s tempting to think we share values and all envisage the same utopia, there were a couple of examples where it was clear that this wasn’t the case, where one person’s positive future was someone else’s nightmare. By going back to stories we can get out of the technical weeds and explore potential impacts in a way more people can get engaged with.
  • What threats and pressures might emerge to disrupt our positive ideas? Can we imagine these and plan how we could avoid them? Doing some real threat modelling, with social, political and economic experts.
Overall it felt a positive day. There’s hope that we can start to build new things, and a sense of energy, with parts of institutions willing to support convening around these ideas.