I’ve been commissioned by the head of a research department at INRA to propose design artefacts that help scientists take new perspectives on their work. The aim is to discuss social, cultural, economical, juridical, ethical implications of these researches.
Here are the results of the questionnaires. More details on our thoughts on these results to come soon in this blog post.
To get this picture removed, please send an email to maxmollon[at]gmail[dot]com
Pour supprimer cette image de ce blog veuillez envoyer un mail à maxmollon[at]gmail[dot]com
To see more faces & reaction from the audience: check this private post
Two references I often give during my talks/courses:
The 6 words novels (already evoked in an earlier post)
&
The “Meanwhile” video, close to a “what-if” routine [youtube https://www.youtube.com/watch?v=YmokqDrIBKg]
I’ve been commissioned by the head of a research department at INRA to propose design artefacts that help scientists take new perspectives on their work. The aim is to discuss social, cultural, economical, juridical, ethical implications of these researches.
The main topic is: “Challenges and Opportunities of the BigData for Predictive Biology”. In clear, how does massive statistics turns diagnostics into prognostics – based as much on DNA sequencing as on more general users datas. The targets are scientists and practicians from the fields of medicine and agronomy (plant and animals). The deliverable is a series of scenarios materialising relevant problematics/controversies in the field, presented as a solo poster session aside a one day conference on November 28th 2014.
We proposed 4 posters covering 4 main topics related to BigData and Predictive Biology. Thanks to a questionnaire we collected participants’ impressions. Questionnaires, informal interviews, audio and video recordings of the situation were used mainly to identify the next public where to show this itinerant project.
Team = Charles Chalas (discovery phase), the Sociable Media group (development phase), Jeremie Lasnier (defining and delivery phase), Fred, Juste & Annie (delivery phase).
Subtlety is a key component. The audience of the conference had voluntarily access to a framed portion of informations, under the shape of these posters. The subject is so vast and the sub-problematics are so numerous, that four scenarios – containing specific trails to specific issues – were the best way to let scientists engage with our approach. Indeed, a poster combines one general topic and clues to sub-problematics. Therefore participants would recognise one or another sub-issue depending on their own sensitivity.
Why hiding these informations (the articulation of problematics between and within each scenarios, plus the path that lead to the generation to these ideas)? Because they were not generated by (or in close collaboration with) scientists. We believe our scenarios to still highlight relevant problematics for biologists. However, if we would have provide those hidden details, we believe participants would pay less attention to the consequences of these scenarios into society and the probable link to their research. Indeed, when we shown them in private, one participant sought to correct our “designer point of view on a biology topic” in terms of precision and scientific validity.
About these details, we could organise the sub-problematics suggested by these scenarios into several dimensions, like a “design-space”. However, in order to respect the medium used during the conference and to reveal the suggested problematics, we just updated the posters, as you will see now.
These four scenarios explore the consequences of three main topics: DNA sequencing; DNA selection; DNA reprograming. They were generated using home made “cards” specifically made for this project, they decompose a list of main problematics and concepts associated to Big data & biology (find more in the previous blog-post). Another key for reading these scenarios is this text. I wrote it to summarise my understanding of the subject – prior to the scenarios making phase:
L’ADN est vu comme le plan du vivant et donc, le programme de développement biologique d’un individu (bactéries, plantes, animaux, humains). La biologie prédictive quant à elle, promet d’exploiter les données issues du séquençage de l’ADN pour prédire un état de santé le long d’une vie (horizontalement) + prédire la transmission d’un patrimoine immunitaire à une descendance (verticalement).
La lecture de l’ADN, le traitement à l’échelle génétique et la restructuration de caractères d’un ADN a différents types d’applications pour la médecine.
La prédiction vient comme 3e paradigme de la médecine s’ajoutant à la prévention et la thérapie, elle complète l’outils diagnostic par celui du prognostic.
L’état et (les prédictions de) l’évolution d’un système immunitaire dépend du patrimoine génétique + condition physique + facteurs environnementaux (épigénétique).
La prévision n’est possible qu’avec le croisement des données des trois catégories précédentes, le défis principale étant celui de l’analyse de ces données (passer d’une donnée, à une info, à une connaissance, informant des prises de décision). Plusieurs autres défis : croiser les bigData issues de la bio et ceux issues des différentes couches de la société, inter-connecter ces données très hétérogènes par leur type et format, trouver des personnes ou des algorithmes aillant les compétences d’en extraire du sens.
L’épigénétique et la microbiotique tendent à transformer les pratiques de la santé: les êtres vivants (dont l’humain) sont conçus comme super organismes (composés de nombreux organismes) dont l’analyse et le soin se rapproche de celle d’un écosystème : un système complexe interdépendant.
Un exemple: une manière d’étudier l’état du système immunitaire animal est le séquençage de l’ADN des bactéries de la flore intestinale au travers des selles.
Mission:
Explorer les conséquences concrètes de ces avancées de la recherche pour le monde domestique, sous forme de produits fictionnels. Plus des articles de drug-stores que des produits futuristes/fantasmés.
Posture:
Le Design Fiction cherche à soulever des questions chez l’audience quand au monde dans lequel on vit, et ce, par la confrontation avec des produits de design venus d’un autre monde et dont la conception repose sur d’autres valeurs que celles que l’on connait.
“Objets à conversation”: les scénarios d’usages et les comportements fictionnels – parfois frictionnels – qui les entourent sont autant de “prises” pour engager autrui dans la réflexion et ouvrir une discussion.
Souvent, ces produits montrent une version domestiquée voir banalisée d’une technologie émergente afin de spéculer sur l’impact de ses usages potentiels, comme une anthropologie d’anticipation.
I’ve been commissioned by the head of a research department at INRA to propose design artefacts that help scientists take new perspectives on their work. The aim is to discuss social, cultural, economical, juridical, ethical implications of these researches.
After collecting an enormous amount of data I went through a processing phase where some colleagues and I tried to define more tightly the topic (we could call this reframing the brief, or building a ‘design-space’). Then we began developing ideas. Here are pictures of the key steps of this process.
There are three main problems in this project: a double complexity (span and depth) and the inter-dependency of all these elements as a complex system.
• The topic cover a large span from biology (plants, animals, humans) to computer science (software, hardware), linked by statistics & prediction;
• For each sub-topic the depth of actors involved is comparable to russian puppets, including many levels of sub-problematics (e.g. from society to family, to a human, to organs, to tissues, to cells, to molecules, to atoms. We could do the same in the bio-informatics sub-topic);
• In each of these layers every actors are inter-related, each layers is also inter-related, these topics are therefore like a giant ecosystems composed of sub-ecosystems.
The mission here is to identify relevant problematics in the domain. Different ways of orgainsing this complexity were therefore used to grasp a sense of understanding of the issues involved. For instance, Latour‘s actor network theory was a source of inspiration when tying to list the series of actors surrounding a tumor. I therefore tried to draw “the path” that takes a tumor (among actors) when a patient gets sick / gets a treatment. We could do the same with actors working around a piece of hardware on a computer dedicated at deep-learning. We could also find more theoretical inspiration in complex system studies.
The span of complexity of this topic, from biology to computer science and statistics
Actors in contact with a tumor + circulation of genomic data needed for tumor analysisDifferent layers of ecosystems in a living entity – like fractalsSummary of key problematics of the topic, key associated concepts and list of existing practices correlated to these concepts (First (successful) draft (but 6th attempt))Clean version of this summary taking the shape of “cards” to play with in a design session / workshopCards summarising the general topic: Big data and predictive biology
I therefore summarise the topic with the following text in order to introduce a team workshop.
L’ADN est vu comme le plan du vivant et donc, le programme de développement biologique d’un individu (bactéries, plantes, animaux, humains). La biologie prédictive quant à elle, promet d’exploiter les données issues du séquençage de l’ADN pour prédire un état de santé le long d’une vie (horizontalement) + prédire la transmission d’un patrimoine immunitaire à une descendance (verticalement).
La lecture de l’ADN, le traitement à l’échelle génétique et la restructuration de caractères d’un ADN a différents types d’applications pour la médecine.
La prédiction vient comme 3e paradigme de la médecine s’ajoutant à la prévention et la thérapie, elle complète l’outils diagnostic par celui du prognostic.
L’état et (les prédictions de) l’évolution d’un système immunitaire dépend du patrimoine génétique + condition physique + facteurs environnementaux (épigénétique).
La prévision n’est possible qu’avec le croisement des données des trois catégories précédentes, le défis principale étant celui de l’analyse de ces données (passer d’une donnée, à une info, à une connaissance, informant des prises de décision). Plusieurs autres défis : croiser les bigData issues de la bio et ceux issues des différentes couches de la société, inter-connecter ces données très hétérogènes par leur type et format, trouver des personnes ou des algorithmes aillant les compétences d’en extraire du sens.
L’épigénétique et la microbiotique tendent à transformer les pratiques de la santé: les êtres vivants (dont l’humain) sont conçus comme super organismes (composés de nombreux organismes) dont l’analyse et le soin se rapproche de celle d’un écosystème : un système complexe interdépendant.
Un exemple: une manière d’étudier l’état du système immunitaire animal est le séquençage de l’ADN des bactéries de la flore intestinale au travers des selles.
One of the design steps we took was to list existing practices potentially associated to big data & biology (thanks to the cards shown previously). Here is a non exhaustive list of them. (Thanks Emeline, Ferdinand, the Sociable Media team for your help)
Je pense body hacker bien sûr, et donc rfid/prothèse/contrôle de membre à distance
La question de l’agencement est centrale d’ailleurs. Par exemple aux US on opère préventivement pour retirer les seins/l’utérus/l’appendice dans pas mal de cas.
Je ne pense pas qu’il y ait beaucoup d’études sur la question, mais les premières pistes que j’avais entendu supposent qu’il faut une très grande flexibilité du soi de la personne greffée pour que ça marche, tant sur un plan physique que psychologique (coucou les membres fantômes)
Du côté des greffes tu vas avoir plein de choses je pense. Si on fait pousser des organes, on veut qu’ils soient performants. On va essayer de trouver la compatibilité maximale etc.
On a des coeurs artificiels, des vagins artificiels, des larynx artificiels…
-
Toutes les opérations de chirurgie esthétique et notamment autour du changement de sexe concernent ta problématique : on déconstruit un corps pour le rendre plus adéquat. Et d’ailleurs, même si ça va peut-être enfin changer, toute personne demandant un changement de sexe en france est stérilisée… Biologie prédictive, on y est.
-
Enfin, je pense à “l’eugénisme discret” : avec les tests de dépistage de la trisomie, les mères avortent dans 80% des cas. Il y a déjà de la biologie prédictive. On ne veut pas la voir parce qu’on ne veut pas se poser ce genre de questions.
Oh et bien sûr, les “déshandicapages” à la naissance : assignation d’un sexe aux intersexes, appareillage des sourds avec des implants cochléaires…
-
Aujourd’hui j’apprends qu’il y a eu un boycott massif de Nestlé, qui est responsable de la mort de millions de bébés dans les pays peu développés (année 70 power)
I’ve been commissioned by the head of a research department at INRA (Institut National de Recherche en Agronomie) to provide design artefacts that help scientists take new perspectives on their work. From the client side, the aim is also to discuss social, cultural, economical, juridical, ethical (etc.) implications of these researches and to spark discussions. From our side, the aim is to allow this project to circulate along different publics (mediation spaces, audiences, purposes).
Our hypothesis is that the efficiency and impact of design fiction’s generated discussions would be greater if its public would be constructed bottom-up, based on participants recommendations. At least, it is a first way to formulate it.
The main topic is: “Challenges and Opportunities of the BigData for Predictive Biology”. In clear, how does massive statistics turns diagnostics into prognostics – based an two kinds of datas: DNA sequencing and people & things’ data in general.
The targets are scientists and practicians from the fields of medicine and agronomy (plant and animals). Mathematicians, bio-informaticians, statisticians, etc. might be part of the target later in the project.
The deliverable is a series of scenarios materialising relevant problematics/controversies in the field, presented as a “solo poster session” aside Les journées One Health Îles-de-France‘s one day conference on November 28th 2014.
Find more details bellow (As the client is french, lot of the content is not in English. For more details contact maxmollon[at]gmail[dot]com).
The brief given to me.
“Pour la conférence suivante, il est possible qu’un doctorant en Design nous propose un objet de représentation de l’impact du big data en biologie afin de voir comment l’objet cristallise des idées, en provoque de nouvelles, et comment les discussions s’embrayent atour de ça.” (extracted from an email sent by the client to collaborators)
Find the text announcing the conference OneHealth here.
In order to reframe the brief I first went through a “discovering” phase where I explored different kind of information: basic info given by the client, documents overviews on the topic; vulgarisation of the topic; very precise details (e.g. scientific papers); and general inspiration. Find below some preview of them.
This last 4 pictures represent a the length of the ACTG sequence collected and printed in a book for our shortest chromosome (the n°21), materialising how much big is the paradigm of big data. For more on this ehibition: http://www.museedelamain.ch/fr/101/A-voir (PDF leaflet)
Most of the precise and useful content were collected during interviews I made in Paris and Lausanne (e.g. A team working on the intestinal microbiot; the director of a medicine university; a researcher and veterinary; a computer scientist; a machine learning PhD student; a geneticist PhD student; a responsible of french cooperatives for meat production; “l’eprouvette” public platform for mediating science issues in Lausanne; and a lot of more people informally.
Here are a list of more links to dig into this topic:
D4D lien de Remy http://www.d4d.orange.com/en/home
2030 public health reporthttp://t.e2ma.net/message/zhx2f/3sctkk
Other Medical innovations http://www.fastcoexist.com/3037592/futurist-forum/10-medical-innovations-poised-to-make-us-healthier-in-2015 Transhumanism Operating room
Future of health http://www.youtube.com/watch?v=Adk1-8MgtEg
IBM health future: http://www.youtube.com/watch?v=pHqtrrTaJKY
Micorosft health future: http://www.youtube.com/watch?v=6aKNK7OTHKs
Design fiction Gastronomie Cohen & Van Balen Pigeon/savon
Quantified Self chez Disney https://medium.com/re-form/welcome-to-dataland-d8c06a5f3bc6
Design fiction catalogue http://tbdcatalog.com/ +http://tbdcatalog.com/#samples
MOMA violence http://designandviolence.moma.org/
Cards game http://www.fastcoexist.com/3031375/this-game-harnesses-your-creativity-to-forecast-the-future un mec qui mélange tout
Dunne talk http://vimeo.com/65074246 ScieFie transform society Nova on the same topic idées: What if robots could draw portrait according to DNA expression simulation + statistics based on randomness + parents environment evaluation (and astrology?) ?
Sélection d’embryon http://dk.cryosinternational.com/media/extendedprofiles/1064409176940.pdf
Le design d’enfant http://bit.ly/1wePwd7
Machines assisted clinical decisionshttp://spectrum.ieee.org/tech-talk/robotics/artificial-intelligence/at-the-mayo-clinic-ibm-watson-takes-charge-of-clinical-trials
Donneur d’ovul http://www.acaivf.com.au/index.php?option=com_content&view=article&id=169:ovulinduction
Art et techno sciences http://olats.org/
Emission reproduction et marchandisation du vivant
Real Prediction Machines (Auger-Loizeau with Alan Murray and Subramanian Ramamoorthy)
(see project description below)
—
New project to be exhibited at the Crafts Council London this september:
Modern day fortune-telling is far-removed from the mystical readings of natural and celestial phenomena it once was.
Today it is all about data.
Emerging from research into artificial intelligence and cybernetics post-WW2 and increasingly made possible by an exponential growth of available data via digital networks and sophisticated sensors, prediction is fast becoming a life-changing science. The institutions of finance, such as trading, insurance and gambling are already inexorably linked to prediction algorithms. More recently we have seen a shift into online retail – Amazon.com for example has just gained a patent for ‘anticipatory shipping,’ this initiates the delivery process before the customer has clicked the purchase button. All of these systems routinely use data provided by people going about their normal (and sometimes private) business.
This project explores how data and algorithms could be reclaimed for personal use – individuals can select a specific event to be predicted such as a domestic argument; the likelihood of ones own death or the chances of a meteor strike. A service provider then determines the necessary data/sensory inputs required for an algorithm to predict the event. The output from the algorithm controls a visual display on the prediction machine, informing the owner if the chosen event is approaching, receding or impending.
)
