Category: Uncategorized

Feb/21 What the Romans can teach us about the value of the Welsh language

The Roman Empire was pre-eminent in terms of trade, agriculture and international networks of information and trust.

Not until industrial times did the earth see a city as large as Rome, and as needing such huge quantities of food – and wine. How was it that the Roman empire was able to sustain a population of 1 million people in a city – a population that required nearly 250,000 tonnes of grain annually? And how could it sustain itself at that population for half a millennium?

In principle, transactional costs at each step of the way, say from the food producing regions of the Nile, through to the trading houses at Alexandria, and on to Rome, should have crippled this market, and Rome should never have become a pre-industrial metropolis.

And yet, in the face of informational uncertainty, and these transactional costs, the city survived – and indeed thrived – for hundreds of years. Not until the London of the 19th Century did a city become bigger.

It was estimated that the additional cost premium of moving grain from Spain to Rome, rather than from obtaining it close to Rome itself, was just 16%, which was a pretty remarkable achievement for the time.

Part of the reason for this success was the ability of the Roman state to eliminate transaction costs at certain parts of the supply chain, through (effectively) a salary, or the promise of gifts due to services rendered.

The benefit of transactional cost efficiency, however, comes at the price of moral hazard – in other words, a misalignment between the interests of the people who want the profits, and those who generate the profits. For example, people making bread in Rome had to pay for future deliveries of grain at an acceptable price, and they had to do that at risk – for what was to stop the farmer at the far end of the supply chain from reneging on their part of the agreement? One must have considerable confidence in the likelihood of generating a return on capital before it is going to be put to work in a far-off land.

The Romans solved this problem partly because they had an efficient and functioning legal system which could be used to verify the contracts. However a strong legal system – on its own – is insufficient for constraining agency costs. This formal legal system was underpinned by an informal ‘moral contract’ of behaviour which was embedded in the relationships between slave owners and freed men who mutually gained from the reputation of the other. 

Trust and reputation therefore acted as a way of constraining systemic agency costs in Roman times, just as it still does today – although it is transferred and imbued in very different ways.

Which brings me to the Welsh language.

The Welsh-speaking community is relatively small within the UK context, and extremely so within the international context. This means that the ‘six degrees of separation’ concept – the idea that any two people on the planet are, on average, six or fewer social connections from one another, shrinks to maybe two or three degrees of separation within the context of Welsh-speakers.

This creates a relatively higher level of social risk associated with acting in bad faith, or fraudulently, within the context of the Welsh language, than in the English language. 

In the same way as a trader within a small village lives – or not – by the quality of their service and their reputation – traders who embrace the Welsh language are making a commitment to the quality of their product or service within the ‘village’ of the Welsh language.

This reputational alignment brings benefits to organisations both small and large. Small companies benefit from provision of Welsh language services by demonstrating their understanding of, and commitment to, language issues, which could arguably be seen as a proxy for their commitment to Wales as a national entity. Large organisations who offer services in Welsh – whether based in Wales or not – benefit from the perception that they have invested sufficiently within the language to understand the value of it to the people of Wales, and further that they value it sufficiently to employ people to be able to deliver their services through the medium of Welsh.

In other words, using Welsh in the world of commerce reduces the risk to the purchaser that the supplier will provide sub-par products or services.

There’s another benefit – to the ‘resilience and prosperity’ parts of Wales – related to organisations domiciled in Wales, and paying local and national taxes, because companies trading through the medium of Welsh are overwhelmingly likely to be domiciled within Wales, and furthermore to be of insufficient scale to avoid paying their taxes here.

So by using Welsh as a preferred language of commerce – for those of us privileged to be able to do so – we are both reducing our risk as users of products or services, and increasing the likelihood that the profits retained from our purchase of said services will be taxed within Wales. For those of us who aren’t able or confident enough to use our Welsh, we can still benefit from the increased levels of trust in those companies that are providing their services through the medium of Welsh.

Do I personally buy all my products or services from companies with a Welsh-language service? No, and partly because that information isn’t readily available. Whilst the likes of Ffônlyfr from Menter Caerdydd are helpful local tools, what Wales needs is a location-based map, response for mobile devices, where you can search for your nearest cafe, pub, hairdresser or professional service that can deliver that service through Cymraeg. This is my challenge to Wales’ vibrant developer community! And I think that there’s a sound business model for such a product. For example, my own company, Afallen, would happily pay a monthly subscription for inclusion in such a directory.

I believe that the growing confidence and numbers of Welsh speakers will have a small but positive impact on our society and economy, and particularly for those organisations who can offer all their services through the medium of Welsh. Now – where’s that app for Welsh language businesses…I fancy a take-out coffee.

Nov 20/ What drives the IndyWales conversation?

As someone with a strong interest in politics and public affairs, I’ve been following the IndyWales discussion with interest for many years.

Nothing has the potential to change the nature of our politics, constitution, society and environment as much as becoming an independent nation, and whilst these discussions have taken place as long as my political awareness has been in existence, they have hitherto not been part of mainstream public discourse.

The meteoric rise of YesCymru membership which now numbers more than 13,000 (from just over 2,000 at the start of 2020) has changed the framework of public debate in Wales. According to some sources, paid-up members now number more than any political party in Wales except for Labour.

A picture of a tweet of YesCymru's membership reaching 12,000, with the comment that YesCymru is the second largest political force in Wales

I decided to take a look at mentions of YesCymru on Twitter, from the very first days (August 2014) until now. I was curious about what was driving the discussion. Taking mentions of YesCymru as a proxy for interest in the organisation as a whole (and therefore IndyWales more generally) and I wanted to test the theory that pronouncements made in Westminster have an impact on YesCymru interest.

The early years

The first graph shows the frequency of mention of YesCymru over the whole span, from August 2014 until the present day.

By recent standards, there was very little activity over those early years; the activity tended to be catalysed by events within Wales itself, such as marches for independence.

However more recently there are clearly defined peaks in activity which correlate more strongly with specific events in Westminster. For example, there was a flurry of activity on the day following the General Election in 2019 in which the Conservatives were returned with a significantly increased majority.

The next big peak occurred during a period in mid-April when it was revealed that England’s health service had superseded the Welsh Government’s provisional deal for PPE equipment, and that England’s Covid death figures had not included those occurring in care homes.

The big peak(s)

Most interesting for me has been the more recent direct correlation between events in Westminster, and mentions of YesCymru on Twitter.

The furore around Dominic Cummings’ trip to Barnard Castle to test his eyesight resulted in more than 3,000 mentions of YesCymru over three days. And the vote by MPs in Westminster not to extend free school meals for the poorest children in England caused many Twitter users to vent their anger, resulting in much more activity for YesCymru.

And most telling of all, the ‘big’ peak of more than 4,000 mentions came following the more generous furlough agreement made after Johnson’s decision to lock down England for a month, despite rejecting pleas by Wales and Scotland for more support in the weeks prior to the decision.


As with many social media movements or campaigns, there was not much in the early years to report on.

Typically, early activity from any social media account consists of regular engagement to build up ‘brand’ awareness and forge relationships.

The science of social media demonstrates empirically that large social media accounts grow more quickly and have far greater interaction, so even in the absence of external factors, there would have been an increasing trend of mentions of YesCymru over time.

However, what we can see from the data is that external factors are playing a huge role in engagement, and that the factors which appear to be playing the biggest role are indeed those performed by UK Government – as highlighted by Plaid Cymru’s Liz Saville Roberts.

I find it particularly interesting that even where decisions are taken in Westminster that do not directly affect the people of Wales – the decision not to provide free school meals to the poorest children in England as a prime example – this still results in a huge amount of increased discussion around IndyWales.

This appears to suggest that the institutions of Westminster are becoming increasingly scrutinised by the people of Wales for unfairness, not just to Wales, but to citizens in other parts of the UK.

It’s still very early days for discussions about independence in Wales, but it’s certain that analysing Twitter will continue to provide a rich seam of data for researchers examining politics and society here.

July 20/ Open source social media will save our democracy

In an opinion piece in yesterday’s Guardian newspaper, Carole Cadwalladr describes how Facebook is the virus that has enabled the catastrophes of both Brexit, and the ascension of Mr. Trump to the White House.

Facebook allows lies to spread virtually unchecked. It permits those people with the most money and least scruples to disseminate falsehoods to those most susceptible. And it allows this with no prospect of holding individuals or organisations to account.

Whilst Facebook, Twitter, Instagram and other social media platforms have enabled a blossoming of connection, conversation and shared ideas; they have also created a bitter, divisive, polarised digital world where shareholder value is enhanced most by highlighting division and minimising harmony. Quiet, reasoned thoughts are filtered out by algorithms designed to maximise clicks, retweets and likes. Volume is king.

Many people want to make a difference but they don’t know how. But there is an antidote to the virus of misinformation and hatred. That antidote is open source social media, and it’s already spreading at the fringes of the online universe.

Open source social media

Open source social media doesn’t permit advertising. It doesn’t sell data. It prohibits hate speech and intolerance*. And it’s moderated by users, not resourced at the behest of global tech giants. 

More worryingly for the incumbents, the hotbeds of innovation are no longer in the mega-corporations with their teams of thousands in the offices and boardrooms of (mostly) America. They are in the open source equivalents, with code freely available for thousands of supporters and volunteers across the globe to build and improve. 

From the perspective of Wales’ Future Generations Act, anybody using, promoting or supporting these open source platforms is supporting the goal of a Globally Responsible Wales. From a worldwide perspective, that same user or supporter is increasing the freely-accessible sum of human knowledge. 

One clear example of this innovation is the federation between open source platforms (also known as the Fediverse). Federation is the ability to connect different social media platforms, so that posts and updates become mutually visible.

This means that if you post a photo on Pixelfed (ethical version of Instagram), it pops up in your feed on Mastodon (ethical version of Twitter). Likewise websites, blogs and updates on the ethical equivalent of pretty much every ‘surveillance capitalism’ platform you can think of can cross-post to each other, enabling much more streamlined conversations and updates.

What are the downsides (and upsides)?

Let’s deal with the elephant in the room; the big downside of the new open source Fediverse is that user numbers are far, far lower than for the established platforms.

This is hardly surprising; there’s a headstart of fifteen years or so for many of the tech giants. And the science behind social media means that there’s an agglomeration effect; once most of your friends are engaged on a platform, it takes a significant effort to leave them behind and start something new.

For me personally it meant (mostly) leaving my Twitter account of several thousand followers, and starting a fresh new Mastodon account on, one of many ‘locality’ type instances across the world. 

I instantly ‘lost out’ on the instantaneous stream of updates from my many friends and colleagues, and on the rough-and-tumble of (what passes for) debate there. There’s likely an impact on my ability to promote my new business, Afallen, through that network, too. 

However, what I have found is a new community of online friends and collaborators. I’ve witnessed almost zero hatred or bullying. And I’ve relished using platforms which don’t harvest my personal data in order to sell them to companies who may – in many cases – place profit above the public good.

The truth is that the community of users in Mastodon (and the other platforms) is growing steadily – see the example below for activity on At some tipping point – I’m convinced – the growth will start to become exponential, and then the users who became active first will see the biggest benefits.

Weekly interactions on the platform

But the biggest benefit of all will come when people start to leave the platforms of the tech giants en masse, lessening their influence as the custodians of online debate and information-sharing, and contributing to a kinder, gentler and more thoughtful world of public discourse.

Further information

If you’re interested in finding out more about the ethical, privacy-conscious alternatives to the mainstream social media platforms, head to

*Almost all ‘instances’ of open source social media ban hate and intolerance. Those that don’t are generally blocked, so the hatred is restricted to a small portion of the Fediverse

April 20/ Something’s happening in the Fediverse…

6 April 2020

I’ve been using the Mastodon micro-blogging platform for a while now – increasingly choosing not to use or post on Twitter and committing myself to a democracy-friendly path.

I’ve become used to ‘meeting’ interesting people way outside my normal e-contact groups, and I have definitely reduced contact with the people I would normally engage with.

I’ve also become used to interacting on a platform which has far (far!) fewer users than Twitter. I’ve made the assumption that there’s a short and medium-term cost to my own influence, and potentially to the commercial success of Afallen, by turning my back on one of the most visible and active micro-blogging platforms on the planet, because I believe that it’s the ethical thing to do.

But….something strange is happening in the Fediverse. Mastodon has suddenly seen a huge surge in registrations.

Probably the largest instance (if you need a primer on the basics, this might help) of Mastodon is Mastodon.Social. There’s a Mastodon account which tracks the number of registered users on that instance.

For months, the weekly number of new registered users was around 2,000 or so see the stats from 19 March, for example. And the total for the last week? 14,550.

graph of new registered users

I’m prepared for any number of explanations for this, ranging from a sudden raising of awareness about the platform in a particular country or language; to an awareness from spam-bots that it represents a new way to reach an audience.

However there’s a tiny bit of me that’s hopeful that it represents a real, meaningful increase in awareness and activity for the platform. The science behind social networks demonstrates that size begets size. So any increase in ‘real’ users on Mastodon is to be celebrated.

I’ve posed the question to the Fediverse about the reason for the surge, so it may turn out to be a statistical blip, or a bunch of bots. But for the time being…I’m dreaming of a step change in awareness about this amazing, hate-free, democracy-supporting platform.

Oct 18 / Infrastructure in Wales

Infrastructure is the Cinderella of modern society. Underpinning every activity we undertake, it is nonetheless largely taken for granted, except when it isn’t there.

It’s the job of policy-makers and planners in the public sector, and their counterparts and engineering professionals in the private sector, to ensure that infrastructure is robust and accessible round the clock. To their credit, there are very few times when this part of the system fails.

It’s also the job of those same people to ensure that society’s future demands on infrastructure can be met, which requires direction and investment. The ambit of future scenarios is fascinating, in that it affords both great challenge and great opportunity for our infrastructure — and by extension, for society as a whole.

But it’s my personal fear that we are failing to meet the challenges and obligations set to us by our future generations.


Foremost amongst my areas of concern is our electricity grid. The capacity in Wales, both in distribution and transmission, is completely full. This means that most new renewable generators face punishing costs to upgrade large chunks of the system, leaving many projects unviable.

An electricity pylon at night with stars in the background

In mid Wales, the situation is particularly severe. This is where Wales’ most productive and dependable wind resource is literally stranded, pending the development of the mid Wales connection. This 400kV connection would greatly facilitate Wales in meeting its own renewable energy targets, but is itself dependent on the ability of onshore wind — the cheapest form of renewable electricity — to access the market.

Mid Wales is doubly-penalised; insufficient grid to generate and export renewable energy, and potentially insufficient grid — ‘grid poverty’ — to make use of new technologies such as electric vehicles and heat pumps. As I wrote in 2017, this lack of access to potentially money-saving technologies has the potential to further disadvantage rural dwellers, and could lead to communities becoming increasingly dependent on unaffordable ways of heating and transport.

However, although the position in mid Wales is particularly acute, the grid is squeezed everywhere. I have been told by the two Distribution Network Operators in Wales that no new thermal generation or battery storage greater than 1MW can be installed until 2026.

This limitation is crippling our ability to take part in new energy and economic systems, once again condemning Wales to use legacy 20th century infrastructure as our UK neighbours move seamlessly into a more enlightened system of low-carbon generation, storage and smart use.


Wales is at a crucial point with respect to a range of infrastructure issues, and rural Wales, in a Brexit landscape, is particularly vulnerable.

However I believe that within this uncertainty lies opportunity. With Wales’ rural areas replete in notorious ‘not-spots’ for both mobile signal and 4G internet, could we combine new grid infrastructure with 5G mobile networks, leap-frogging old technologies in ways advantageous to rural livelihoods?

Why shouldn’t Welsh Government — and other public sector organisations — take a stake in the grid infrastructure which would be needed to connect much-needed onshore wind projects, facilitating the development of renewables in Wales and simultaneously providing opportunity to a whole host of modern, low-impact manufacturing and processing clusters across rural Wales?

A cyclist on a small road in the hills, fading into mist in the distance.

Could the proposed ‘Lôn Rhiannon’ also incorporate a grid line and future-proofed mobile network infrastructure, alongside a new cycle network, up the spine of Wales?

With the new National Infrastructure Commission about to sit, there are a range of future opportunities which could change the way in which we view infrastructure. Multi-use, combined infrastructure, paid for in innovative ways and by new participants (crowd-funded grid anyone?) could unleash the creativity and enthusiasm of our citizens, businesses and public sector organisations. I cordially invite the new members of National Infrastructure Commission for Wales to use their imagination to the fullest when considering how best to serve the future interests of our people.

And if they want some food for thought, my door is always open.

(This article was first published on the Institute of Welsh Affairs website here.)