This is a presentation which I gave recently asking what development policy can learn from evolution.

The main conclusion is that as would-be change-makers, we should not try to design a better world: we should concentrate on building better feedback loops.

You can view and listen to the presentation by clicking the image below. This narrated presentation lasts 18 minutes (beware: as soon as you click you’ll hear my voice, so don’t do this if you are in a meeting!).

Click here for a narrated presentation about evolution and development

Alternatively, you can download the presentation as a pdf file here.  But this won’t make as much sense, as there are a couple of videos in the presentation.

If you like this presentation, you may also like my previous narrated presentation about aid effectiveness after Paris.

Please let me know what you think in the comments below.  Am I right that we should focus more on feedback loops?

If you enjoyed this post, please consider leaving a comment below, and perhaps sharing this with other people using the buttons on the left. You can also sign up to have blog posts sent to you by email.

54 Responses to What can development policy learn from evolution?

  • Owen,

    This is what I have been trying to explain to people for a long time, but never quite managed. Thank you for putting this eloquently in words and pictures, so we can spread it for consideration through the System.

  • Thanks Owen,

    A fantastic and concise narrative on the situation. This is a model to which I very much subscribe. We have a lot to learn from the evolution of complexity. Some reflections from me.

    I think you’re right to focus on feedback loops as a way of ‘giving life’ to the system. Essentially what your conclusion points to is a new approach to governance – one that views improving governance as being the development and design of evolutionary systems. One final thought on this is that I think it’s important to stress that feedback loops don’t necessarily mean liberal democracy as it would be very easy for people to conflate ‘feedback’ with ‘democracy’.

    However, in addition, I think we also need to find better ways of creating variation – this is as much about cultural and attitudinal change within the development community as anything else. But also, how can we bring other approaches and perspectives to bear? People are very good at sticking within their cultural communities, people who share their values, language and work on similar things, which lends itself to group-think. How can we encourage cross-community thinking? (Age old question!)

    Finally, this model calls for more experimentation (perhaps a la RCTs), but given that we’re dealing with people’s lives here, what do you see as the potential ethical implications? For example, how might they affect power and relationship dynamics between different communities who have different interventions?

  • Hi Owen,

    I love the narrated format. Perhaps you could do a post sometime on how you create these?

    As someone who was originally trained as an ecologist, and only subsequently came to development, these ideas make perfect sense. Hopefully we can contribute in a small way to this. Our new strategy includes developing feedback loops (among communities affected by HIV and TB) as one priority area.

    Our starting point is to work with the Kenyan government and a network of AIDS service organisations (KANCO) to map the locations of service providers and clinics in two constituencies. We plan to trial SMS feedback first, based on unique facility IDs that link to facility or organisation profiles. The profiles will be both on a website and in a print directory.

    Perhaps others have different ideas that we can also try?

    Rob

  • Hi Owen,

    Excellent stuff, as always. I agree that feedback loops require a great deal of strengthening but at the same time I have a great deal of difficulty envisaging how we can change the incentives at work in a realistic and meaningful way. As long as donors fund poor countries and poor countries provide nothing meaningful for donors, the voice can be there, but it will be disembodied from any real power.

    An aid system like China’s emerging one (we build roads, you give us minerals) at least provides some element of reciprocity that gives the recipient some power; but the DfID model of we give you grants, you give us promises that you’ll do good things provides no power from the recipient back to the donor. Feedback could be given, but the primary incentive of the aid agency is to deal with internal stakeholder problems not the recipients.

    How do you see us getting around this.

    • Ranil

      Thanks for this.

      My belief is that more transparency about how aid money is spent and a stronger link to results will help to change the incentives in donor countries. It will, over time, disempower rent-seeking domestic stakeholders. In my view – which you may regard as fantasy – the public genuinely do want to support poverty reduction. Their doubts about aid are not that they don’t care about the well-being of the poor, but that they are concerned – in some cases rightly – that the money is actually being used to line the pockets of domestic interest groups or dissipated into rent-seeking consultancy firms. By shining a bright light on all that, there will be a stronger political imperative to use aid well, and it will be easier – necessary, even – to resist the pressure from special interests in both donor and recipient countries.

      If I’m right that the public is at heart concerned about the interests of the poor, I think with more information and transparency we can align the political incentives in donor countries behind the interests of the recipients.

      Owen

  • I really like the format, not for the least reason that I can skip a bit to get the narration when the slide is less obvious and skip it when the sound bite will do.

    The feedback loop question is not particularly new. I had expected more of a proposal that would show how to go about playing evolution on the feedback loop, rather than encouraging us that the feedback loop was where we ought to look.

    What would be even a more radical proposition than what I was looking for is to play evolution on the entire aid mechanism. Feedback loops are one part of it, but the drawbacks from lack of feedback might not matter so much if donors were more centralized or (semi-oppositely) competed more directly for aid funds or if donors and NGOs were treated as the agents instead of the principles.

  • I had the opportunity to see this presentation performed “live” last week here in Addis and as a one time physicist and amateur bioligist found the ideas thought provoking and I imagine pretty relevant. Though looking through one not so distant government’s new poverty reduction strategy I get the feeling the solutions are not so much designed as cobbled – and that’s not to mention the UNs repsonse (UNDAF) which is being cobbled together as we speak. I guess the problem with evolution is that it tends to take a long time and is a little unforgiving to the weak! So this takes us back to what you would probably call the “tired” notion of Lessons learned wedded to some kind of informed design a closed loop repeating itself on a four or five year cycle! I think feedback has an enormous role to play though I think its principle role may be in action management: if one doesn’t know what’s actually getting done and what’s not really happening, regardless of the reasons, then you’re lost. Sorry this is becoming a rant… Just wanted to say I liked it…

  • Nice presentation and love the media. I agree with your main points.

    I don’t see a solution to the problem of lack of diversity. Group think seems too deeply imbedded in the development DNA. How do we convince the development industry that all the time and money spent on coordinating and harmonizing approaches is counter productive? How do we convince governments to allow a wider range of experimentation in development approaches that may not fit in with the national five year plan?

    The second problem is that even after you have conducted your randomized trial showing what doesn’t work, you have to work against the political forces that resist calling a failure a failure. The sad fact is that much (most?) of aid is not spent to produce effective programs, but to make the donors feel good about their efforts. This is why more attention is paid to the amount of money spent than the amount of impact produced.

  • Hi Owen,

    Some excellent analysis, but as Ranil intimated, I wonder whether to a certain extent whether you have just outlined another “wicked problem”. Aid has two separate customers who spend v little time talking to each other, if ever.

    Since evolution is also about small incremental changes, one thing we can do is to think about how we can evolve towards an aid system that is evolutionary in nature. One change that I think would help, and blogged about, is KISI (Keep It Simple & Iterate). This links into your remarks about focusing on impact. Getting individual projects thinking in terms of incremental changes and constant improvement would be a useful first step towards evolutionary thinking.

    An interesting point which Steve Jones’s little 2 min summary missed out, is that evolution never stops. What this means is that if you design a good enough feedback loop, not only do you potentially solve some of today’s problems (via a few dead-ends which evolution always produces), but you also provide a mechanism to help solve tomorrow’s problems. This is why some deep thinkers have hailed Darwin’s (and Wallace’s) insights into evolution by natural selection as one of the most powerful ideas ever propounded.

    MJ

  • Owen, I’ve just come across this excellent and thought provoking presentation. Well done and much needed. Coming from a background in evolutionary biology and organizational/community narrative, I hope I can add a few useful points to the discussion. What I wrote in response ended up too long for a courteous comment, so I’ve posted the rest of this response to my blog, here:
    http://www.storycoloredglasses.com/2010/10/selection-and-development.html

    Wishing you the best in your important work!

    Cynthia Kurtz

  • Hi Owen,

    I most definitely don’t consider that fantasy. I also believe the public in developed countries wants to help, and cares about helping. My skepticism lies in the fact that they are instinctively drawn to certain kinds of intervention (specifically, ones that shy away from private sector development, and ones that lead to the ‘pretty photos’ outputs) and there isn’t much will on the side of either the donor Government or the population to learn much more than this.

    I might be overly cynical on this front, though. I do agree that better transparency will help, though.

  • Hi Owen,

    Great stuff, thought provoking. I would like to link it to the existing concepts on Complexity by Dave Snowden (see http://www.cognitive-edge.com/), as you will be most aware of.
    Not all development problems are ‘wicked’ problems (or Complex problems in Snowden’s Cynefin Mode), some are simple (with tested solutions) or complicated (with solutions that ‘only’ require smart planning or design). For complex problems linear planning indeed does not work, and an evolutionary, incremental, experimental approach can apply. But let’s not fool ourselves. Development is not a business of developing a new nozzle, in which circumstances are hardly changing, with short feedback loops. I agree a lot on getting the feedback loops better – social accountability is crucial in better development designs. However, it is not only the feedback loops that may be blocking. Development deals with sticky institutions, the ‘rules of the game’. More and informative feedback loops can help to bring out the obstructive institutions, but will not be the only thing needed.
    Often wicked problems need multi-stakeholder, multi-level, multi-scale approaches in which social learning between people and organisations involved need to collaborate. Issues such as power also come into play.
    Let’s see how we can use your ideas within this framework.
    Cheers,
    Jouwert

  • Just a couple of things to add to Owen’s excellent, thought-provoking presentation. People who haven’t already checked out the ‘Origin of Wealth’ by Eric Beinhocker would do well to pick it up. It places processes of evolution at the heart of economic growth, and is a jolly good read. Duncan Green has blogged about it here: http://www.oxfamblogs.org/fp2p/?p=115

    Secondly, and also of relevance to Owen’s piece, ALNAP published a study on innovation in humanitarian response, in which we explore the detail of aid innovation processes using a framework with ‘variation and selection’ at its core: http://www.alnap.org/pool/files/8rhach3.pdf

    This study has directly led to a major meeting opened by the DFID Permanent Secretary Minouche Shafik (http://www.alnap.org/events/25th.aspx) and subsequently to the creation of the very first sector-wide Humanitarian Innovation Fund, (with DFID as the initial donor) which will be trying to put some of these ideas into practice. The launch of the fund was covered by Scidev here: http://www.scidev.net/en/news/disasters-fund-will-link-aid-and-research-communities.html

    (Full disclosure: I was one of the authors of the ALNAP paper, and chair the Strategy group of the Fund)

  • I find this presentation and the discussion very important, but I would like to add some more elements to it (I expand more on them elsewhere):

    - The evolutionary pressure is being steered by the public perception. Public perception evolves too. However, I do think this pressure is for the moment rather away from long term results and towards short term visible activity. Sorry.
    - Evolution works with selection of the fittest, not by coordinating, clustering and evolving the unfittest. The inclusive narrative for better aid might have to make room for an exclusive selection process. (see also “what is a illegal in trade don make sense in aid”).
    - The path dependency of evolution can easily lead to sub-optimal results.
    - Every evolutionary process creates dinosaurs: creatures fit for the world of yesteryear. Who is your favorite development dinosaur? It might be me.

  • Dear Owen

    I agree with you, but I also wonder what you are saying, and I have two questions about that.

    As you know, “natural selection” was Darwin’s name for what happens in nature that’s similar to “artificial selection” which he observed in dog and horse and plant breeding.  The difference between natural selection and artificial selection is that NATURAL SELECTION DOES NOT HAVE A GOAL OR IDEAL IN MIND.  It has not for example been trying over the years to develop homo sapiens as its finished product.  The Uniliver nozzle was an example of artifical selection with a very clear objective about cheap detergent production.  So my first question is: are you saying that TRIAL AND ERROR IS THE BEST WAY TO FIND OUT WHAT “WORKS”, or are you saying that DONORS SHOULD NOT HAVE GOALS AT ALL (as you suggested in a very trenchant anti-targets piece a few years ago)?  If it’s the first, then one problem is that the project-breeders do not agree about the goals of development, or what counts as “success”, or what the information carried round the feedback loop should be about, or who should decide all these things.
     
    My second question is: where does the feedback loop go in the end? Does the information go to DONORS/SERVICE PROVIDERS or to INTENDED BENEFICIARIES? In the Unilever nozzle-development process, the information went to Unilever, but in the Swedish education project the information went to beneficiaries.  (In the health project it went to the clinics.)  On one model, donors/governments would do a lot more impact analysis, and get a lot of text messages from intended beneficiaries in mud huts, and then the donors/governments would decide what to do with this information and which projects to replicate or close down.  On the other model, beneficiaries would get a lot more information or text messages about donor/government plans and how much they cost, and they would be able to compare this with the observable benefits and make their own decisions about what to do (where to spend their vouchers perhaps).  On the second model, the beneficiaries would also do the learning from experience.  I like the second model, and I think this is what you have in mind when you mention at the end a “fundamental power shift”, but it’s not clear. Perhaps you are instead making the case for a massive new industry of feedback-data collection and processing?
     
    It would be interesting if you could clarify where you stand on this.

    regards
    Gary

  • Very interesting presentation, and I agree with the ideas on feedback loops in principle. However, for this approach to work I think it requires that the outcomes be actually measurable and easily observable to the population which is supposed to be receiving services.
    I work on HIV. There are certain aspects of HIV programming that could easily fit into your model: continuous, effective access to treatment for instance. But changes in terms of preventive behaviour, and eventually in terms of incidence of HIV, are much more hard for individuals on the ground to appreciate, and in fact even us programme wonks don’t really have a good idea of what a “good” or “acceptable” result is in terms of HIV prevention. Also, in HIV prevention there is a need to work with marginalised and stigmatised groups which might typically be excluded from a normal community feedback loop. So then there is a need to “create” new, somewhat artificial communities of sex workers, men who have sex with men etc., so that a reliable degree of community participation and feedback can exist. What I’m getting at, is that the creation of feedback loops is itself quite a complex and often a political activity.

    I wonder also how to apply the feedback loop concept to programmes/interventions that are about creating positive externalities from changing individual level behaviours – e.g. vaccination programmes, where the benefits may not be that obvious to an individual or to a community, but that can be measured at population level.

    • @Matt – I don’t think you are right. Look at the Martina Björkman and Jakob Svensson experiment with reports cards on clinics, described in the presentation. It is very difficult to measure clinic effectiveness, or to define “good” or “acceptable” results. But people know whether they are getting a good service or not, and when they are given the opportunity, they can demand change. I think you are thinking too much in terms of top-down measures of performance, and not sufficiently respecting the ability of bottom up feedback to capture these nuances.

  • @Owen, yes I agree that the example you gave is a very good illustration of how this approach can work; and I absolutely subscribe to the approach. However my question was more whether this approach is sufficient in all scenarios. Access to good quality services is a very important and a very proximal part of improving health, but it is not the only one. And I’m not saying that the other factors – structural/environmental determinants – can’t also be monitored by communities; it’s just that it’s not as straightforward or obvious as the report card approach. Add to that my other concern about marginalised people who it may or may not be possible to categorise as communities, and who are disenfranchised to start with.

    I’m not throwing out this beautiful baby with the bathwater, nor am I underestimating what real people can and should contribute. But I do think it’s important to acknowledge that effective and meaningful community participation is not always as straightforward as health service report cards; it also requires a lot of commitment, preparation and capacity building.

    Changing the subject completely, the section of the talk that discusses evolution reminded me of a great little book on participation and urban planning that emphasises some similar ideas: Small Change by Nabeel Hamdi: http://www.earthscan.co.uk/?tabid=647

  • Love it, Owen. Thanks! Quick q: what is the software/platform that you used for http://media.owen.org/Evolution/player.html? Want to steal :)

    Thanks! @katrinskaya

    Owen replies: I use software called Articulate Presenter

  • First, ThatVoice . . . whoa! :)

    Second, and perhaps it is a freudian slip, the “wicked” problem (and the pun is intended) is the unsustainable development enabled by a globalized capitalism protected from responsible behavior by limited liability laws and whose wealth is denominated in debt-based fiat currencies. The intended “wicked” problem is, I assume, intended to be a lack of development. To the degree this is so, such a wicked problem, in an evolutionary sense is also a wicked cool solution to some other problem. And that, I would argue (and from what I observe) is how can slavery be effected with ever increasing efficiency for the slave master.

    Assuming the current beneficiaries of the global capitalism are not consciously wicked, the argument made here concerning adaptation may be the answer to how sustainable un-development came to be a wicked problem. With each eradication of an iteration of slavery, the construct reformed itself again in a new and more efficient form. It was never eradicated for reasons Orwell captured in _Animal Farm_, and/or because it feels good to unconsciously be the beneficiary of what is wicked. Besides, it is darn right hard work to live responsibly and sustainably—to be just—when one is systemically a hedonist

    If what I have stated so far makes sense, the implication is that aid work is, at best, a fools errand. At worst it constitutes the loyal opposition for all that stands in the way of love, as power, implementing the demands of justice. Aid work, as currently structured, can only help what is wicked become more efficient. And this is a wicked problem! First, aid work feeds a false sense of piety. This feeling of being pious tends to blind the privileged from seeing themselves as they are seen rather than how they feel about themselves. A false sense of piety is a systemic hinderance to becoming aware of how ‘goodness’ can mask complicity in the environmental, social, and economic injustice upon which the current iteration of global capitalism is dependent . . . and which, BTW, has tipped the planet into klimakatastrophe. Second, and also systemically, aid work constitutes the preemptive plugging of leaks in temporary dikes of non-violence impounding unmitigated and just rage. The unintended consequence of this is that the rage builds until a systemic collapse of the dikes occurs and violence redress an economic models failure to be just; marks the end of an iteration of slavery; effects a redistribution what constitutes unjustly acquired social wealth. Within this framing of what capitalism is in the process of evolving, aid work, denial withstanding, can only evolve to improve the efficiency of the current iteration of slavery and resist yielding what bends the universe in its long arch: justice.

    I find the reference to the World Banks charts ironic. This is because that body, the last I knew, is subcontracted by the IMF to vet IMF loans regarding them being social just. Isn’t this the fox being give the task of securing the hen house? Both the World Bank and the IMF were created as part of the Bretton Woods Agreement. When that agreement was broken by President Nixion, these institutions ceased to have a legitimate function. Rather than disbanding them along with the agreement, in the ’70s they were first put to use to serve those with the wicked development to create, more efficiently, debt-slaves of those nations with wicked problems left over from colonialism and economic imperialism. Now, in large part due, ironicallky, to coordinated action among aid organizations, one washes the other’s hands with complex rationalizations for efficient and ongoing, if now, ostensibly, more kind and gentle iterations of debt-enslavement by these institutions. Though things may be changing again for the IMF, for now both, like aid organizations, function—denial withstanding—to keep the structural adjustments needed for “justice to roar down like a mighty river” pent up; to keep the enslaved, temporarily placated.

    To sum up, development, as defined by the current iteration of global capitalism, is systemically unsustainable, wicked, and an oxymoron. For those trapped within the thinking and feelings of this meme, what is “wicked good” simply isn’t. Trusting the emotional intelligence informed by such ‘goodness’ is sufficient for rationally evaluating the feedback of feedbacks is commensurate with trusting that the level of thinking which creates a problem can be used to solve the problems such thinking creates. Greed has gamed capitalism. It can and will game any effort to circumvent that gaming through feedbacks. Feedbacks, as framed here, are, systemically, another iteration of doing the same thing over and over again while expecting the results to be different.

    We live in interesting times. Willingly—or not—and sooner than we feel is possible, we will live to think differently.

  • First, I’m a big fan and inspired by your use of different ICTs to exchange of knowledge. Kudos.

    Second, I, too, believe that ‘we’ have a lot to learn from complexity thinking and evolution if we’re to take dev. programmes forward. (That and taking power and politics seriously). In fact, the working title of part one of the dissertation I’m working on is “Stranded on the Shoulders of Giants” in the sense that classic reductionist theories of social and economic change have contributed a lot to our understanding of the world but are now limiting our capacity to understand the evolutionary processes in which these changes take place.

    All that said though, I sit a bit uncomfortable with the choice of a nosel as illustrating example in a development context.

    The problem is -like with the Santa Fee’s take on complexity- that it appears closer to super positivism than to true complexity. While the design of the nosel might be too complicated for us to understand, there’s good reason to believe that it is just a matter of getting a microscope powerful enough to provide a permanent explanation of why it works. Still, if we want to produce artificial snow rather than detergent, or if we want to move the production of detergent to a more/less humid climate, the nosel might not be very effective any more. Further, water as well as the components in detergent are quite stable compared to agents (people), social groups, economic-, institutional-, social-, environmental-, technological resources rights and entitlements etc. constituting societies and the global whole going into the nosel of time.

    In that sense, I consider the part on feed back loops to be key in your presentation. However, don’t forget that the loops need be continuous. In the case of the nosel, we do not need to know much about the principles behind its function. I believe the case is quite the opposite in the context of development beyond a single project. Here, we’re actually more searching for the principles behind successful projects rather than the specific blueprint.

    I believe we probably agree (on most of it) but I think it’s an important point to make when speaking of complexity and evolutionary processes in a development context.

    Not least as I consider far too much of int. dev. academia tied up in discussions on methods (a.k.a. Methodology in modern use – sigh) in which it bypasses any reflection on the ontological and epistemological aspects of what we’re experiencing and trying to understand. I’m aware this appears high brow but it shouldn’t be. What’s needed in development isn’t extended toolboxes but craftsman skills. No fancy methodological power tool, either designed or evolved, will make it out for understanding the principles human comprehension, motivation and (non/)interaction with the surroundings.

  • Hello,

    On the surface your claims seem reasonable, but it seems that you have taken one easy optimization example out of proportions.

    The nozzle example is among the easiest problems optimizable by random trials: a cylinder symmetric system with a less than 100 element grid, with almost all of the intermediates working. VERY easy to optimize with random trials.

    Furthermore, the nozzle itself is not complex, the physics is just hard
    to understand. Most design problems involve very different systems with several interacting and fitted parts. The key term is irreducible complexity (by Michael Behe), and that is hard to overcome with trial and error. Most working complex systems (radios, TV’s, computers, automobiles, …) are made by designers. I’d guess that less than one permille of working complex machines have been designed with evolutionary algorithms.

    So, I think you take some isolated and rare cases where evolutionary algorithms can be useful (microevolution, adaptation), and exaggerate the success out of proportions (makroevolutionary changes, new multi-part systems). And you neglect the essential concept: irreducible complexity.

    You might be interested in the following research on the limits of algorithms:
    http://marksmannet.com/RobertMarks/REPRINTS/2009_BernoullisPrinciple.pdf

  • You might be interested in a new blog posting: “Models and Reality: Dialogue through Simulation” which talks about feedback as the missing element in many Theory-of-Change models used in the evaluation of development projects. When it is introduced it makes a big difference, generating complex outcomes that even in a simple model are not easy to predict, but can be simulated. See the blog at http://mandenews.blogspot.com/2011/04/models-and-limitations-of-models.html

  • Further to my email above. You might also be interested in a related blog post “Do we need a Minimum Level of Failure (MLF)?” at http://mandenews.blogspot.com/2010/10/do-we-need-minimal-level-of-failure-mlf.html This begins with an observation about evolutionary theory: “A common misconception is evolution is about the “survival of the fittest”. In fact this phrase, coined by Herbert Spencer, is significantly misleading. Biological evolution is NOT about the survival of the fittest, but the non-survival of the least fit. This process leaves room for some diversity amongst those that survive, and it is this diversity that enables further evolution. The lesson here is that the process of evolution is not about picking winners according to some global standard of fitness, but about culling of failures based on their lack of fitness to local circumstances.”

  • Great presentation, Owen, and great format. Two questions:

    1. The arguments seem to be about aid-financed service delivery, which certainly is a wicked problem, but is narrower than aid generally, which is narrower than development. Can they work for the problem of creating economic growth or democracy, say?

    2. The P.T. Bauer thesis could be expressed as “aid is bad because it weakens feedback loops from the governed to the government.” How do we think about that?

  • Hi Owen,

    I attended the very inspiring webinar you gave today where you used some of the same excellent slides.

    One remaining question that is bothering me though is the human dimension of the failed experiments. In pharmacy it’s ok to say we try 100 approaches for potential products and 95 won’t work. But in development we are talking about people’s lifes. If you initiate a project and it fails, there is a danger that the actual people involved are in an even worse place (often economically, but certainly emotionally) than before. Promises are made, communities are mobilized, social capital and energy is invested, trust is built, and when the project fails, you don’t get back to that level anytime soon. It’s a very bureaucratic approach saying this will be the case in 95 of our projects, in order to gain the great value we get out of the 5 successful projects. Can we afford people’s lifes and the social capital of communities being the collateral damage of our innovation efforts in development?

    From a global perspective you can of course say that the overall cost of not innovating through experimentation in the end will be even higher, but that is hard to explain to those individuals and communities which happen to be the guinea pigs for our failed development experiements.

    I’m actually surprised how much of an ethical question this becomes when thinking about it (I’m sure Michael Sandel at Harvard could make a great lecture out of this example, http://nyti.ms/kYC2s4) and I’m not sure there is a real solution to it. But assuming you’ve thought through these potental issues already before, I would be interested how you look at it.

Follow Owen




Email subscription

I want to receive new posts by email

Owen on Twitter