Defuse Dublin 2019 - The Big Green Button Myth: Resisting Dehumanised Research & the Corporate Constraint of Qualitative Insight
Emmet Ó Briain
interpretive research | discourse analysis | ethnography | sustainable mobility + active travel
This is an adaptation of a much, much longer article I wrote earlier in the year about the dehumanisation of research and insight, with a particular focus on its implications for interpretive qualitative research. I wrote the original article originally with a social/consumer research audience in mind but, as it circulated, I was pleasantly surprised to find that the one of the audiences most receptive to its message were practitioners within the #ux research community.
I’m still not sure how it found its way into wider circulation in #UX (a few people mentioned to me that they discovered it through a Slack message) but the responses I received confirmed that my observations reflected the experiences of others, particularly research users and practitioners.
Ironically, one of the sections that I edited out of the original article was Anne Balsamo’s distinction between ‘idiot-proof’ and ‘configurable’ design, so I was delighted to have the opportunity to restore that point at the annual Defuse Ignite event organised by IxDA Dublin last night.
The format for the night was 12 Speakers giving a presentation of 5 minutes, consisting of 20 slides, with each slide advancing automatically every 15 Seconds.
If you enjoy the presentation below, please do read the original much, much longer article (which is more about how rationalisation in corporate culture is inimical to qualitative and creative thought).
Presentation
In 1979, Lucy Suchman joined Xerox PARC as a doctoral student in anthropology, a period of research which would form the basis of her PhD and the book it produced: Plans & Situated Actions, one of the foundational texts of HCI, CSCW and UX.
One of the projects she was involved in was trying to help XEROX figure out why their new state-of-the-art photocopier — specifically designed to be user-friendly — was proving so difficult to use.
And, if you look around online for accounts of her time at XEROX, you’ll find UX blogs claiming her ethnographic research at XEROX led to the development of the BIG GREEN BUTTON we see on modern photocopiers.
The story goes that she realised that people were massive idiots and that our problems with using sophisticated photocopiers could be solved by giving us an equally massive green button labelled ‘START’.
The only problem with this legend is that if you do look around online for accounts of her time at XEROX, you may also find a comment below the blog from Professor Suchman pointing out that the story isn’t true and that the photocopier at the heart of the story already had a BIG GREEN BUTTON on it before she began her observational research:
Instead, her foundational insight was to highlight that any attempt to anticipate how people would interact with the photocopier could never escape the getting-to-know-you period that accompanies any new technology use.
So, why does it matter that Lucy Suchman went to the trouble of correcting this misunderstanding?
It matters because the struggle for the truth of the BIG GREEN BUTTON story seems to me a proxy for a battle between two diametrically opposed perspectives on the orientation of research and design practice (and the working world, generally) towards the people — users, customers, researchers, the research participants, employees - who ultimately have to deal with the products of these corporate and professional cultures.
On one side is the idea that, essentially, people should really be protected from themselves (and technology from people). In which the actions of users or research participants are anticipated and constrained.
And we see this in research designs in which neither the researcher nor the research participant has much freedom about the type of questions they ask or the responses they provide.
Think also ‘idiot-proofing’ design. Big Green Buttons.
At their core is an ideology of constraint. In which flexibility and improvisation is limited.
On the other side are approaches and environments that trust researchers and the researched, that trust users, that trust professionals, by affording them agency, that recognise the value of autonomy, of flexibility and situated sense making.
In reality, most design and research practice takes place in organisational environments where managerial control is prioritised above meaning-making.
Managerial cultures prefer standardised knowledge and standardised approaches to the generation of that knowledge.
However inappropriate it is to the phenomenon at hand.
And this orientation towards managerial control has implications for what type of work gets done, how it gets done, and what gets valued.
Managerial cultures prefer control over autonomy, predictability over uncertainty, the quantitative over the qualitative because knowledge and practice in this form is more amenable to managerial control.
Within this logic, nothing is more predictable and consistent than an entirely standardised process.
And, from a managerial perspective, any automated approach is considerably more efficient than involving humans.
Applied to research and technical practice, the justification of standardisation is often wrapped up within a ‘discourse of methodological merit’ — that automation and dehumanisation will lead to an increase in quality through the reduction of the ‘bias’ and ‘risk’ or, horror of horrors, subjectivity that anything involving humans carries.
Applied to interpretive qualitative research, the autonomy afforded to the researcher and researched by approaches such as ethnography and ethnomethodology are presented as sources of such risk.
So, instead, we end up with qualitative research with standardised instruments, with moderator flexibility and participant response constrained through tightly specified prompts and timings.
Recommended by LinkedIn
We end up with interpretive qualitative research bending to make itself amenable to values (efficiency, standardisation and rationalisation) that, applied to qualitative research, undermine not only its capacity to surprise, but its very purpose.
And how are you going to get away with proposing a month of ‘deep hanging out’ with users against this backdrop?
Given what we know about automated systems and how they obscure and embed discriminatory biases, it is clear that most arguments for standardisation are not arguments for objectivity but for control and constraint, and against autonomy.
But the objectivity, certainty, clarity of standardised frameworks and models for working or the quantitative-experimental paradigm don’t improve upon or repair the subjectivity, the situated sense-making, the messy detail of the real world.
They simply obscure it.
And in so doing we lose sight of the interpretive and situated reasoning on which these metrics and standardised frameworks of practice rely on in use and in application.
And the more this type of detail and qualitative reasoning becomes obscured — designed out of sight — the less it is valued.
As we move towards the impossibly distant future of 2020, the professional environments we work in will become increasingly automated and, by extension, inhuman.
And so, the messy detail of the real world and humanity will continue to be missing from outputs because they are missing from the input.
By acceding to automation, and practice more amenable to managerial control, we give up on the qualities that distinguish us from machines, that elevate us above machines, in terms of research and design.
So, here are some qualities we must protect to help resist the corporate constraint of qualitative insight.
Aristotle used the term phronesis to describe the practical wisdom (such as the practical wisdom derived from professional experience) that allows us to apply knowledge, judgement and intuition, in-situ to to make decisions.
Think of interpretive qualitative research — it is characterised by an ability to respond to emerging themes and tangents. The skilled qualitative researcher adapts dynamically to the contingencies of contexts they find themselves in, rather than tying themselves to a strict schedule or script.
This form of practical wisdom goes missing in standardisation and dehumanisation. So, resist its constraint.
Discretion is another essential difference between the decision-making of humans & algorithms. When we routinise decision-making through algorithmic applications like automated qualitative moderation (yes, horrifyingly, this actually exists!), we lose this capacity for situated reasoning.
When presented with new or marginal cases, humans can refine the boundaries of their decision-making BEFORE making a final judgement, whereas algorithms do AFTERWARDS through feedback or new training data.
Protect discretion.
The real world is messy and detailed.
A structural appetite for the universal and generalisable removes us from the lived detail of research or design practice and leads us towards abstract models of practice that bear little resemblance to the situated reasoning that we engage in when working, when interacting with technology, and with each other.
We must recover detail.
Fidelity is a virtue.
And if, as Mark Freeman say “the most fundamental obligation of scientific inquiry… [is] to be faithful to the phenomena one seeks to explore”, then we should resist practices and methods that undermine our fidelity to our phenomena and embrace those that provide us with the necessary flexibility and autonomy to reveal them.
-----
Postscript: I’m starting to write something on organisational “accounting” of qualitative research and am looking for a research team within Dublin to do some interviews with, and who might be up for letting me do some ethnography as well.
Get in touch if you can help or interested in discussing it!
Some references:
— — — — —
Emmet Ó Briain is founder of QUIDDITY — an insight consultancy specialising in the qualitative analysis of organisational, customer and public discourse and cultures using naturally-occurring data and language.
A lot of my work involves talking to organisations about how they do work (research, design, advertising, communications) and exploring the meaning and consequences of the language they use for broader organisational culture, internal work practice and external communications — get in touch if you’d like to hear more!
As ever, if you enjoyed this article, I would be grateful if you could share it with anyone you think might be interested.
twitter: @emmetatquiddity
Managing Director . Anthropologist. IoT and Smart Cities Enthusiast
4yThank you for writing this wonderful piece.
Really excellent 'deck' ;) and very thought-provoking and instructive points made. I particularly liked "phronesis" as you described it, 'practical wisdom (eg from professional experience) that allows us to apply knowledge, judgement and intuition in-situ' (on-the-spot-ness?) Many other great points too.
Copywriter, editor and author. New book 'AI Can't Write, But You Can' out now.
5yThis is absolutely fascinating. Reading it, I'm struck by how my own trade, writing, has also fallen prey to its own versions of the fallacies you describe. Reading some case studies, you would think that if you did enough user and industry research (before) and optimised your distribution (after), the actual act of writing would hardly matter at all, or would become somehow obvious. Yet it's here, in the fuzzy act of creation, that all the value is actually created, or at least synthesised. The writing process is almost impossible to describe, or even observe in any meaningful sense. But it's vital. It's also this bit that is most resistant to quantification, automation and corporate control. Yet it seems to be an area that AI developers seem desperate to conquer somehow, regardless of whether they can or even should. Far better to get the machines to do the researching and publishing bits, and leave the thinking to the human beings who can actually do it.
Director at Trinity McQueen
5yThanks for sharing Emmet. Loads of really useful themes in here. Your points on qualitative research being "characterised by an ability to respond to emerging themes and tangents" and "the skilled qualitative researcher adapts dynamically" are core for me. We rely on tacit knowledge. We do ourselves an injustice by not highlighting how we navigate this "messy/hard to describe" element of our work.