In my travels I've met many activist software developers like myself, overflowing with energy and idealism for communities in general, but no community in particular. Many of them were creating a new kind of platform that addressed the concerns of the moment - privacy, security, sovereignty, serverless architecture etc. I don’t mean they were collaborating, I mean each was building his own platform. Once the platform was ready they embarked on a quest for users, or, which is harder, whole communities of users ready and able to migrate to the latest platform. After a few years they run out of physical or spiritual resources, and move on. It seemed to me a predictable waste of developer years, but more tragically, a lost opportunity for the communities they wanted to help.
So this post is about software for communities. By community I mean any group of neighbours or idealists coordinating online. Sometimes communities cluster together under the same banner, like time banks or transition towns and I’m calling that a movement. Movements are interesting because they are comprised of basically independent chapters who need to communicate within and between each other. I’ve written lots more about that on my blog.
Most real-world communities live on Whatsapp. It has a few features, but they go a long way – filesharing, groups, conferences calls, reactions, polls. The main advantage of Whatsapp is that everyone already has it installed and they know how to use it. Upgrading to something is hard because it involves getting people to install an app, create accounts, make a separate box in their head for community stuff, so to be worth the pain, any proposed alternative has to promise a lot! That’s why some groups deliberate for years, get nowhere, and then make a poor choice.
The first choice such groups need to make is about how much responsibility they can take for their software, including ongoing costs. I see this as a spectrum ranging from
- Smaller groups with little or no knowledge should be thinking about paying for Software as a service (SAAS). There are many reasons not to do this, but without technical ability, all you can do is throw money at the problem.
- Groups which have limited technical capacity, might consider installing an open source package. This is cheaper because you only pay the cost of renting server space, and there are many opportunities to configure your data or tweak the software. Plus your data remains private.
- Larger groups with more specific needs, more users, more resources, maybe even more software engineers have the possibility of building their own software, and this is a minefield. A professional project has a project manager, documentation, automated testing and a budget of tens or hundreds of the thousands, but at the grass roots, those things are luxuries and the budget may be zero. In that case everything depends on the developer, their rigour, their knowledge and their multi-year commitment. The developer is irreplaceable, if they stop, the software stagnates for ever.
So far, I’ve introduced you to the developer who needs a community to validate his software and his existence, and to the community who needs a developer. Its a tragedy that these partners who are made for each other rarely find each other. Its a mystery what obstructs their union.
Having spent years in both roles, I’d like to offer my perspectives and then a solution.
During the Lockdown, thousands of street-level mutual aid groups appeared on whatsapp. They had no formal structure, no subscription fees, no governance and they withered away as the lockdown eased. Many people experienced community for the first time. They had no idea that Local Exchange Trading Systems had been doing just that since the early 90s and continue to do it now. Indeed, LETS are rarely mentioned in discussions about local community building. But to my mind, these groups are the nearest thing in the western world to formalised, mutual aid networks. LETS is making a life and an economy out of mutual aid. Thousands of self-organised independent local groups who rarely seek government or grant funding, don’t express ambitions beyond their immemdiate scope, and consequently rarely appear on the radar.
I converted my wiki into a LETS trading platform but couldn’t get anyone to use it, even after months of working fruitlessly with a national umbrella organisation. I determined to find a community that was willing to work with me on their software to fulfil my dream of writing software that was actually used! I was lucky to catch my local group in Geneva at the right moment. I built a super platform, and then with the Chairman, founded Community Forge as a Swiss association, and we approached other groups in the network. While they offered me long periods of hospitality, I put together a hosting platform to manage lots of sites. They found people to offer technical support and the network grew to around 300 communities at its peak. It wasn’t the quality of the software that drew them, so much as the fact that we came from within the network, we had their values, and the software at least worked! To this day we have never done marketing or tried to impress the public which is why the public facing website is such a mess!
For several years I was keeping the software up to date, sustaining the hosting platform, and nomading – meeting other people in the movement, offering this great platform already used by hundreds of groups. Several individual groups adopted the free services I offered, but critically for this story, over a decade I was unable to convince any other movement to use the platform and put resources into it. My only real success therefore, as an activist/developer was when I worked inside a movement.
I wanted Community Forge to be the seed of a software commons that would reduce the cost of software for any one network, bring the development and maintenance inside the movement, build standards and tools for interoperability, especially for trade! This progressive, outward-looking, collaborative approach simply didn’t land. The LETS networks in UK, Netherlands, Flemish Belgium, the German-speaking regions, Australia, and New Zealand all looked inward for their software. Timebank networks had more money. I hung out with leaders in UK, USA, Israel, Spain. Some of them had tens or hundreds of thousands budgets but they usually ended up hiring web design companies to design and build them a web site, from scratch. Sorry to say I wasn’t impressed with the results, but that’s a rant for another time!
The point is, because I wasn’t just meeting these people at conference, but I was actually visiting them at their homes and in their communities, I came to understand how these kinds of groups and organisations make critical decisions.
Many grassroots groups depend absolutely on a single IT volunteer, and their abilities and knowledge. Many of them have zero budget. Many of them are not being begged by their comrades to improve software. Many have become conservative by having bad experiences.
It is easy for web developerrs to forget that flesh and blood communities regard software as an auxiliary, not as an intermediary and not as critical infrastructure. They might not be alive to the boons good software can bestow, but they intuitively understand that installing a new social network alongside all their other bits and pieces is going to create another silo which will add more friction and overhead to their existing infrastructure. A new login for everyone, new comms channels in which to lose things, issues around duplicate files and versioning problems, synchronisation problems. Migrations can be especially traumatic because they never go as planned and all the extra work falls on one person.
More than once I've been asked (as co-Founder, and sole developer of Community Forge), if any of the communities we serve would like to test someone else's software for them. Someone, in fact, who built the software for nobody in particular. Sorry if I sound jealous, but the idea is laughable. Many of those groups have actively participated in the development and the story of Hamlets, they want all the features they are used to, they have nothing to gain from testing software which wasn't made for them. How can the new software be better than the old if the provider hasn't even asked me what it does? Is there a budget for a one off migration and possible migration back? Doesn’t he care that we are 12 years into the current upgrade, and 5 years beyond the deadline? The quality or the narrative of a software package doesn't really come into the equation because we only have capacity to move incrementally from where we are. We are not being held back because we are unable to locate existing platforms that meet our needs. We are being held back because we lack in-house development capacity.
In my continued attempts to produce software that would serve humanity and my ego, I decided to distill my accounting methodology into a protocol that could be used in any social networking context. Thus was born the Credit Commons Protocol for recursive mutual credit accounting. Working in the abstract, unencumbered by technical debt and real users, building a perfect machine in my mind, was a lot of fun for me. But more abstract was harder to explain and further from anybody's immediate concerns. It was so theoretically advanced I couldn’t even interest the International Reciprocal Trade Association with it, let alone the commercial networks, and I confess that the regular LETS and timebanks people, even if they thought accounting as a service between groups was a neat idea, even if meant they had less software to maintain, didn’t mention it to their developers.
So I learned the same lessen even harder – building software that nobody is asking for is (almost certainly) a waste of time. The LETS movement and many other radical grassroots organisations grounded in place and community and family are not spending much time thinking about global infrastructure for a global uprising. Perhaps is it right that 90% of their attention should go on local and internal preoccupations. These small and emerging groups need less bright ideas and more engineers on the inside. They need heros who will work hard for little or no money, who will stick around for many years and keep the infrastructure standing. That’s not going to change.
What needs to change therefore, is the developers, who are writing software for nobody in particular. I'm speaking first of all to myself because everything I ever built for no-one in particular is being used by no-one in particular. Much of what I built for real groups with real needs is still serving them 15 years later. If we want to be of service we need to step out of the blissful platonic space of perfect software architecture unencumbered by users and migration, find a cause or a local group, roll up our sleeves, and start putting out dumpster fires and tackling technical debt.
Only then will we be in a position to help build the new systems.
Afterword
While writing this I noticed a correlation. I voiced my frustration at these inward looking local communities who fail to engage with my global vision, but I also admired them for their resilience, independence, and longevity.
While LETS seem to have fallen by the wayside, a couple of decades ago, nothing has replaced them. The word community now refers to the teenagers with outrageous web handles who cluster around a memecoin, conniving to make the number go up. Compared to this and most other communities being talked about, the LETS are still next-level because they have formal structure, staying power and what they permanently facilitate, exchange, is the very foundation of what it means to be social.
Comments