12 – The Open by Default Principle

The IDEMS Principle
The IDEMS Principle
12 – The Open by Default Principle


Lucie and David discuss the principle Open by Default: ‘This principle affects both the inputs and output of the company. It implies the need for a justification not to ‘go open’ as the default is to be open. In particular when a valid justification exists then no negative judgement is given to being ‘closed’.

They emphasise the default stance of openness in both inputs and outputs unless there are compelling reasons to be closed. David explains that being open by default does not mean everything must be open, but rather that the default should be openness unless justified otherwise.

[00:00:00] Lucie: Hi and welcome to the IDEMS Principle. I’m Lucie Hazelgrove Planel, a social impact scientist with IDEMS, and I’m here with David Stern, the founding director of IDEMS.

Hi, David.

[00:00:19] David: Hi, Lucie. Nice to be talking again. Which principle are we discussing today?

[00:00:24] Lucie: So, following our other discussion on Options by Context, I thought it’d be good to delve into this concept of, or the principle of, Open by Default. Shall I perhaps read out the principle ?

[00:00:36] David: Go ahead.

[00:00:38] Lucie: This principle affects both the inputs and output of the company. It implies the need for justification not to go open, as the default is to be open. In particular, when a valid justification exists, then no negative judgment is given to being closed.

[00:00:56] David: So you find this one a bit of a mouthful, don’t you? This isn’t natural to you.

[00:01:00] Lucie: No! [Laughs]. I mean, so what does it mean to be, to be open in general? I’m like, um, what’s the context of this come out of ?

[00:01:09] David: That’s a really good question. And one of the reasons that maybe for you, this is not as familiar, if you were a software developer, then open, open source software versus sort of closed software. This has been a debate going back 30, 40 years, and open has won out. There’s many other areas in which people talk about open now. Open educational resources, open licenses, open data, open publication!

[00:01:39] Lucie: Exactly! People, in data and things, people want to encourage researchers, I think, to let their data be used by others, to let them be publicly accessible, which as an anthropologist, A, sounds scary when I have handwritten notes, so like how would you put that, make that accessible, and B, again, sounds scary in terms of ethics, and judgment of others.

So, at least in anthropology, making early anthropologists’ field notes accessible, it has made those anthropologists very badly regarded, because of the ethics of the time, let’s say.

[00:02:15] David: Yeah.

[00:02:16] Lucie: So being open to me, it sounds a bit scary.

[00:02:20] David: And this is really good. But let’s take it back a second. When you think about publications, do you think about open access journals versus journals which are behind a paywall? What’s your view on that?

[00:02:31] Lucie: Yeah, to me that sounds all positive.

[00:02:34] David: But it’s not positive because actually what’s happened is open access journals has meant that less people from disadvantaged backgrounds publish in them because of the cost of publishing. So actually open access journals has meant that there have been less people publishing from low resource environments in open access journals than in the closed journals where the cost of publishing was sort of embodied and recovered from the people reading it. So, it’s not as simple as that. It’s never as simple.

So, there is never all good, all bad. I am a great, great supporter of open. I believe in Open Access Journals, but I am highly conscious of the negative side of what’s happened because of people moving towards open.

And what’s really fascinating, and this is why I started talking about software development, is that in terms of actual success stories, I would argue the big success story is software. Open source software has broadly won. All big tech now rely on it, build it, support it. Microsoft was the poster boy of closed software and they now own and support GitHub which is one of the best tools for open source software and they are definitely set down the route of supporting open source in different ways.

This is an incredible about-turn and reversal of recognition that open source software isn’t just, in this context, ethically better, more efficient, it’s more effective, it is better for business. And this is a real success story in my mind of showing how a system can change. There’s a lot of stories we could go into about that, and I’m sure that will come out in other contexts in other podcasts, but what I think is so important is that those learnings about what’s happened with open source software are central to what we now think of when we think about this in a broader context.

And people are thinking about this in broader contexts, you know, all sorts of ways people are sort of wondering about what the word would look like if open was a more central concept. But I think the thing about this principle, which is so important, is this principle isn’t about open.

[00:05:14] Lucie: Its about when we are closed.

[00:05:16] David: Exactly. And then there was no negative judgment. It is the fact that actually in open circles, in people who believe in the open circles, as we do, there’s often this, either you’re in or you’re out. Whereas what we’re saying is that actually exactly as you pointed out, there’s ethical issues with being open in certain contexts. There’s good reasons why people are nervous about open in certain contexts.

And so what we argue is not that everything should be open. At this point in time, we don’t think that’s possible. There’s sort of ethical issues, there’s all sorts of other things which come into play. There’s, you know, tools which are not open, which just outperform other tools, and so it’s inefficient sometimes to go down the route of being open.

And so, what we argue is this very simple term, Open by Default. And actually, it’s an extremely powerful principle. This is part of our Options by Context sort of set. So this is about considering different contexts and considering the options related to that context. And so, in that context of Options by Default, we are not saying, no, there isn’t a hard rule, you must be open.

We’re saying, okay, if you’re not going to make an informed decision, this comes back to in that discussion on Options by Context, we discussed that actually sometimes having a choice of many different things, you are less satisfied with what you choose. So, if you’re not going to make a choice, you’re just going to take the default option, you’re going to take the easy option, whatever the most standard thing to do is, then it should be OK.

So, if you don’t want to think hard, and this is where the inputs and the outputs are important, if we don’t want to think hard about what we’re making public and what we’re not making public, then it should be public. Then it should be open. If you want to think hard about it and decide what should be public and what shouldn’t be public, that’s fine.

But if you don’t want to think, then it should be open.

[00:07:32] Lucie: So can we think of some examples there?

[00:07:35] David: Very, very simple. Just very recently, we had an interaction with colleagues on writing some code. And, you know, the question is, we’ve created our repository, which we’ve shared, and by default, most people, when they create their repository, at the start, you start by making it closed, and then you decide when you want to share it with the world.

And we don’t. We did it the other way around. We started with the first piece of code when nothing was there, when it was totally, you know…

[00:08:05] Lucie: Blank canvas.

[00:08:06] David: …not ready to be shared, blank canvas, not ready to be shared with the world. We made it open. Anyone in the world could look at it. Now nobody’s going to look at it because there’s nothing there yet. But that makes us think about who might be looking and why they might be looking. We didn’t want to make a decision about this, but it does affect how we develop it, what we put up there. You need to think more carefully because it’s open to the world.

It’s not that everybody’s going to start looking there, but it may mean you think differently about that code. Anyone could come and take that code at any time. So you think carefully about it in different ways. And this is the key point that we didn’t want to think hard at that point about how we were going to do it, how we were going to think, how we’re going to communicate to the world when we were going to sort of try and publicize this, to get this out for others to look at it. But our default position was to make it open. Whereas most people’s default position is to protect it first and to decide when it’s ready to be shared, even if their long term intention is to make it open. The general approach is to start by it being closed and then decide when to open it.

[00:09:17] Lucie: And it’s interesting that you just used the word protect too because that’s the word I had in my head. I was sort of thinking that being open is a question both of protecting yourself perhaps but also protecting other people, in the sense of how something might be used or how it might be interpreted or something.

[00:09:31] David: This is the important point, this is why you might choose to close something, but you should actually think that through. We’re not saying you shouldn’t close it. We’re not saying that if there’s something to protect there, that you shouldn’t protect it. We’re just saying that by default, if you don’t want to think hard, make it open. If you have something which you need to protect, as you might need to do if you’ve got personal data on people because you’re research.

[00:10:00] Lucie: Exactly. And I was just thinking of internal within the company too.

[00:10:03] David: Well, this is about thinking as a company, what are we trying to protect? And actually, another one of our principles is Collaborative by Nature. So what does it mean to be collaborative by nature? Well, it means to actually forego some of our own protection in favour of encouraging collaboration. So that relates to our Open by Default, that yes, we can protect things for the company if we need to, because of the commercial value they bring, because of whatever it is. But that’s not our default stance.

We’re collaborative by nature, so we’re open by default and our default stance is to be open to that collaboration.

[00:10:46] Lucie: Yep.

[00:10:47] David: It’s a very challenging principle in many different ways for, for many people. So it’s a, it’s a, it’s a powerful, powerful principle. And what’s really interesting is that this is also for what we use.

[00:11:02] Lucie: This is coming to the inputs and outputs of the principle then, is that?

[00:11:06] David: Exactly. Our outputs, so this is sort of the things we produce. Our inputs are the things that we use. We’re recording this on Zoom. Zoom is not open. Well, before we chose to use Zoom, we actually went through a process where we investigated the open equivalents. And we had a process through which we decided at that point in time, there was good reason why Zoom had got certain things right, which meant that as an organization, we should use the closed system rather than the open equivalents, which we would have liked to support, but the cost to us would have been too high at that point.

And this is really interesting, you know, we have to pay for Zoom, whereas we wouldn’t have had to pay for the open ones. So how can the cost be too high?

[00:12:03] Lucie: Okay, yeah.

[00:12:04] David: It’s not just about financial costs. I mean, we made that decision very early on in the pandemic, in the COVID pandemic, and one of the reasons Zoom was so successful there is by chance, I believe, just before the pandemic hit, they made this huge investment in server infrastructure. And my understanding is that when we then tested this in low resource environments, Zoom outperformed every other system in terms of the stability of the call in low resource environments.

And that was why we decided that actually the importance of being able to communicate with our partners who had low bandwidth, who had difficult connections, and you know how frustrating that is with anything. But we felt that our best chance of getting this communication to be as good as it could possibly be was to use Zoom and the Zoom infrastructure, because they outperformed all the other systems we tried in those low resource environments.

And that was sort of part of that decision making. So the cost of using a different system might be that our communication with our partners in low resource environments would have been worse. And that was something which we deemed a cost too high.

[00:13:17] Lucie: Definitely.

[00:13:17] David: And therefore we chose to go with Zoom, even though it’s not open. We wanted to go with the open solutions and there were some really exciting ones, but they weren’t…

[00:13:28] Lucie: Quite ready yet.

[00:13:29] David: It’s not just that they weren’t ready. For us to go with, and we were too small, really, what we should do, if we were big at that point in time, then we could have invested time and effort into making them better and making them work well in low resource environments. But they didn’t have that ability to put the investment in that Zoom was putting in to make sure that in low bandwidth, in low resource environments, it works well.

Now, over time, my hope is that there will be open solutions that outcompete, and my hope is that a few years down the line, we may be in a position to move away from the closed product to an open product so that we can support it. But that’s, that’s our thinking process. And there’s no negative judgment to the fact that we’re using Zoom, it is a fantastic product. It has served us extremely well. I still find this when I go, actually, what’s really interesting is Google Meets has really improved over the last few years. There is a question now for low resource environments, whether Zoom or Google Meets work better. And I don’t actually know anymore.

But Teams still is really, really frustrating in low resource environments. It doesn’t work as well. And there’s a lot of others we could go through. These are the sort of commercial ones. I’m not going through the open ones. But, you know, the open ones, when we looked into it, did not out compete and did not perform as well.

We have used some of them on occasion, but at the moment we found they don’t, outcompete for this particular task.

[00:15:04] Lucie: Yeah.

[00:15:04] David: And that’s, that’s the essence of Open by Default. It’s, we are very happy with the best commercial product for what we need to do, because it helps serve our objectives. We don’t get distracted by that, there’s no negative judgment.

If an open product can outcompete, we would jump on it because by default we want to choose that. Even at a cost to ourselves, we would choose an open product over a closed product, an open solution over a closed solution. But at the heart of what open by default gives us is not a value judgment on open over closed, but a way of turning that decision making process around.

And let me be clear on this. Why is this, why is this different? Most people I know, when they don’t know technology solutions, they would look first at what is the industry standard, and then you’d actually start with the industry standard, best solution out there, and then you might consider open solutions. Your default position would be the industry standard, and so your closed would often be standard.

Now there’s good reasons why that can be sensible, it means that you’re able to sort of do what most people do more generally, and so I think there’s a lot of people for whom, you know, the standard process of industry standard by default, and then if you have a reason to go with something else, then you do. But that would be very different from our Open by Default.

[00:16:47] Lucie: I’ve got a couple of questions. One is, is being open the same as being transparent?

[00:16:52] David: No. Open is actually very specific. So, transparency is really good in a lot of different ways and very important. But open, open licenses, these are more than just being transparent. These are more than just, um, more than just being freely available. And that’s important. So, you know, open software is not that you can use it for free. Open software is software where the code base is freely available and it can be reused for any purpose.

[00:17:31] Lucie: That’s really interesting.

[00:17:32] David: Now this is really important and I think the example which I love on this which really explains the importance of open over free, for example, and transparent is something else, but yeah.

[00:17:45] Lucie: Well, it’s just that that’s completely the opposite then to like an open access journal where, you know, the article is available for free but the data behind it isn’t necessarily open.

[00:17:55] David: Well, but that’s where many open journals are moving towards having open data. And this becomes a requirement for many for publishing. That is part of the discussion, which is the open science discussions, which are leading to this, where you want the article and the data to be open as well. And so that is an increasing requirement in certain contexts.

But the example that I think is so powerful is textbooks. If you have a free textbook, then anyone can access it and use it. But that is nowhere near as powerful as an open textbook. An open textbook, if I want to now take that textbook and customize it for a particular audience, who need to use it in a different way, I can. So, if I have a free textbook which is tailored to the American market, which is freely available, and I want to use it in Kenya, I can’t adapt it to the Kenyan curriculum.

[00:18:49] Lucie: I have definitely used British mathematics textbooks, um, for secondary school children in Vanuatu, and it’s horrific. You feel horrific using them.

[00:19:00] David: Yes, and so this is the key point. If you have open textbooks, you can now go through a process which takes that textbook and adapts it to the local curriculum. And actually, you’ve got all the sort of depth of that textbook, all the work that’s gone into it, and you can adapt it. The power of an open textbook over a free textbook, you know, is immense.

[00:19:25] Lucie: But this is also the power of technology, right? Because without it all being, uh, on a computer. Sorry.

[00:19:35] David: Well, yes and no, because you can print it. You could get printed textbooks, and the point is, if you had them under an open license, you could then adapt it and print it. So it’s not just the power of technology. Now I agree, technology is central to open actually working, because, you know, technology is part of what makes all of this possible in different ways. But it’s not just technology, it’s financing models. How do you finance an open textbook compared to how do you finance a textbook where the publisher is the one who finances the textbook and then they get the money in and based on the sales they then pay the authors?

Now, that doesn’t necessarily work in the same way for an open textbook. So how would you finance the authoring and the production of good open textbooks? It requires different systems models. This comes back to our agroecology discussions. It’s not just about an individual component. It’s about how you think of the system as a whole.

So you need to think about that publishing system, and that publishing system may need to change to be able to accommodate open textbooks in ways which makes them really useful and powerful. But I do like this example exactly of saying, well, if we have open textbooks, rather than, you know, worrying about access to textbooks in terms of giving people, but we think about this in terms of really powerful open textbooks, then you’ve got to think about, well, how would you go about that customization process?

One of the things I find really frustrating is there is a really good open textbook movement. But it’s basically useless because nobody’s thinking about the customization. Unless you make it really easy for that to be localized into different versions, you’re not gaining the value of the open textbook. The open textbook comes into its own once you have your different versions for different contexts. And yet, I’ve not seen that infrastructure built to make open textbooks really powerful and out compete other textbooks because they can be customised, because they are customisable and they’re designed to be so.

We don’t have the tools to do that yet. So, you know, this is a fantastic example, in my case, of Open by Default in terms of actually the power of open. If you’re thinking about this and you then think about this as being your default way of publishing, well, what does that do for the system?

If we’re thinking of that as our default way of producing apps, producing textbook, producing educational resources, what does this mean for how we want to then get them used in the future? What does this mean for our business models? You know, this is where we then get into really interesting situations.

[00:22:20] Lucie: Absolutely.

[00:22:20] David: And it sort of, it changes our mindset. So open as a concept, I believe is still in its infancy. I only know of it being mature in terms of software. Open source software is a mature concept. Even open educational resources, which have been around for a long time, I would claim are in their infancy.

People do not have the structures in place. There is no equivalent of GitHub for authoring of open educational resources, which makes it really powerful to share and to sort of think about that sharing of the resources and the customization and the approaches. These are the sort of innovations in the future, which are going to transform openness.

I’m conscious that we sort of, I can talk forever on this subject and so I think we should limit this and we’re sort of at a good ending point. But I do want to come back to this aspect that the Open by Default is about our other principles, it’s about System Thinking, it’s about Options by Context, it’s about being Collaborative by Nature. It’s thinking about trying to be Inherently Inclusive. It is a very, very important principle. It’s central to a sort of behaviour approach, but it is not just about the value of open.

It is about the way your mindset changes from thinking about protection first, to thinking about openness first and protection if needed.

[00:24:06] Lucie: Yes, that’s very interesting.

[00:24:08] David: That’s the heart of Open by Default. Maybe let me expand that because it’s not just protection. Another example we use is efficiencies. Zoom is more effective. Effectiveness. Yeah. Open comes first. If there are protection reasons, efficiencies, effectiveness reasons, that is not a problem, prioritize them. But you’re not doing that by default. Your default position is starting with open.

And this is, I think, a mindset which is extremely powerful. And I, I hope it’s one which others could adopt more often. Public goods and thinking about openness in terms of public goods is something which could be extremely powerful, more generally for society.

And so I, I really love this principle. I think it’s a very, um, it’s a complex one. It’s one which applies to so many areas of everything we do. It’s pervasive. You sort of came into this, I think, thinking this one doesn’t really apply to me, but I hope you’re leaving this discussion recognising this is central to everything we do.

[00:25:22] Lucie: Yeah, well, I’m sort of, yeah, like, like you said about sort of open technology has reached its sort of maturity as a concept. I guess I’m left with the questions of where can other areas go in the sense of being open?

[00:25:36] David: I would love to dig into sort of some of these questions and actually discuss what it might mean for a more open society.

[00:25:43] Lucie: Well, and I think we will in terms of research and data.

[00:25:47] David: Research data, open data is not mature yet as a concept.

[00:25:50] Lucie: No.

[00:25:51] David: You know, there are, there have been big movements around that, but it’s not a mature concept. Open research, again, it is, open research is making progress. So I’d argue it’s no longer in its infancy. It’s sort of getting towards, you know, mid childhood. But it’s sort of not yet a mature concept. And it’s really exciting to see these progresses happening.

[00:26:13] Lucie: Great. Well, thank you very much for your time and, um, see you soon.

[00:26:18] David: Thank you. This has been fun.