fredag 25 april 2014

Rethinking sustainability in computing

I just now (a few hours ago) submitted a paper to the upcoming NordiCHI conference (which will be held in Finland between October 26-30). I've written the paper together with Barath Raghavan and the paper is called "Rethinking sustainability in computing: From buzzword to non-negotiable limitations". I believe the keywords we chose for the paper might give some hints as to what it's about: sustainability; sustainable development; sustainable HCI; ecological sustainability; environmental sustainability; critical reflection; collapse informatics; Limits to Growth; Steady-state economy; Ecological Footprint.

I think this might very well be the best and most profound paper I've ever written, but I did write a very good conference paper back in 2005, "Virtual community management as socialization and learning" (here's the pdf), a good journal article in 2008, "Do you believe in magic? Computer games in everyday life" (here's the abstract) and a great book chapter, "ASCII imperialism" back in 2009 (here's the pdf file).

Still, that was then and this is now. Nowadays I'm interested in sustainability and this is for sure the best and most profound paper I have written in the intersection of computers and sustainability this far. I was also great to work together with Barath and I look forward to future collaborations with him!

Writing is nerve-wracking process where you are oscillate between thinking your paper is great and then thinking it is not-so-great-at-all, as well as thinking you won't get it al together in time, or, that you will for sure be able to pull something together that is at least "good enough". This time I had a couple of hours to just look at small details and make everything just right. In fact, the deadline for the paper is after this blog post will have be published on the Internet (which is unusual for me). Despite having just handed in the paper, I have already thought of three more references that should have been in the paper. But if it's accepted, I will have the opportunity to prepare a final, camera-ready paper and can thus fix such "details".

Me and Barath started the planning process and had an outline for the paper ready quite some time ago, but we only started to roll up our sleeves and seriously work on the paper less than three weeks ago. And, I was on vacation last week and didn't lift a finger to aid in the production of the paper and Barath has been on a train (from San Francisco to Washington D.C) for three and half days this week without an Internet connection. I have instead worked all the more intensively before the vacation and especially after my vacation, during the last five days. My wife is also submitting a paper to NordiCHI and we have kept each other company as she has worked equally hard this week. Only a fraction of what me and Barath wanted to write about fit the paper in the end, so I think we will soon start to work on a CHI 2015 companion paper (the deadline for next year's CHI conference is September 22).

I think our paper is good, but while sustainability has gathered steam at the CHI conference, sustainability hasn't really been at theme at the smaller NordiCHI conferences (for some to me unknown reason). The only thing I worry about now is that it will be reviewed by people who are clueless about the stuff we write about and who might think the paper is "fine", but that it doesn't really fit NordiCHI (?). It was for example a nightmare to find categories and keywords that fitted the paper (so it could be matched with reviewers' interests). It's for sure not "adaptive and adaptable systems", "affective computing", "agile user experience development", "ambient intelligence", "artistic and aesthetic aspects of interaction and interface design" etc. In the end, I had to submit the paper as being about "theory and foundation of HCI" (which fits so-so). That means people who are interested in theories will read/review/judge the paper, but again, they might be clueless about sustainability and that wouldn't be very good for us. But if the paper is rejected, we'll just resubmit it to the larger CHI 2015 conference in September instead...

I have decided to share the introduction and the conclusion of the paper in this blog post (i.e. somewhere around 16 % of the full text).


Rethinking sustainability in computing: From buzzword to non-negotiable limitations

--------------- introduction: ---------------

Sustainability is important. Indeed, the challenge of shifting individual, societal and global developments in a more sustainable direction by decreasing carbon dioxide (CO2) emissions and stopping climate change is supposedly a crucial matter at this point in the history of humankind, per numerous world leaders as well as the Intergovernmental Panel of Climate Change [32, 33, 34]. In addition to this challenge, the world remains addicted to dwindling finite resources – fossil fuels, among others – that have given rise to our advanced industrial civilization, including the advancements we have seen in computing over the last several decades [62].

Sustainability has lately become an important theme within every sub-discipline of computer science. The growing importance of sustainability can also be noted in the area of Human-Computer Interaction (HCI) and related areas (e.g. Ubicomp, DIS, CSCW, Persuasive), with the boom in work on “sustainable HCI” following Blevis’ widely cited 2007 paper “Sustainable interaction design” [4]. Sustainable HCI has since been the topic of numerous papers [9, 18, 24, 35]. Similarly, outside of the HCI field, the number of general or specialized “green computing” conferences has mushroomed. These are descriptive statements, but, a normative corollary (held by the authors of this paper) is that at this point in time, the topic of sustainability ought to be an important aspect of HCI (both research and practice), computing in general, and indeed of all other applied academic disciplines.

Despite an active and growing sustainble HCI community at the larger CHI conference, less progress has been made in introducing sustainability as a topic to the smaller NordiCHI conference. With this paper, we want to be conducive to putting sustainability on the agenda also at the NordiCHI conference. Our purpose with the paper is however more ambitious – after having noted that sustainable HCI is a severely under-theorised area. Ideas about what actually constitutes sustainability (i.e. what we are aiming for) are few and far between. Sustainable HCI is furthermore exceedingly insular while at the same time ignoring seminal papers, books, discources and discussion that have been going on for several decades before the term ”sustainable HCI” was coined [19, 21, 39].

Despite the urgent need for a greater emphasis on sustainability both in computing in general and in HCI, we find that it is oftentimes absent, or, that when present, it is unclear or doubtful that the tools, systems, and services that are proposed or developed live up to any meaningful definition of what sustainability actually might entail. In fact, the whole concept of sustainability is more often than not unclear and confused both within HCI and in computing in general. Worse still is that even papers that discuss ”What are, or should be, the boundaries of sustainable HCI” [18], or that (correctly) points out that others’ approach to sustainable HCI is misguided and ”based on a limited framing of sustainability, human behavior, and their interrelationship” [9] themselves contain no discussion about what sustainability actually might entail. It almost seems as if anything and everything that invokes some variation of the terms ”sustainability” or ”environment” becomes part of the seemingly exponentially growing (sic!) corpus of work in sustainable HCI.

In this paper, we aim to evaluate past approaches to the subject and incorporate (and introduce) sustainability thinking from outside of sustainable HCI. The structure of the paper is as follows; we will begin by discussing popular ways of framing sustainability within, but primarily outside of HCI and then proceed to reject them. We will then present four interlocking frameworks that we propose provide a rigorous foundation for what constitutes sustainability, and where each consecutive framework both builds upon and can loosely be seen as a gradual specification and operationalization of the previous framework. We then exemplify with a few of the consequences of using these frameworks and end the paper by proposing avenues for further research in hitherto underdeveloped areas that we believe sustainable HCI should explore.

--------------- conclusion: ---------------

There are many possible ways to end this paper. We have chosen a short and to-the-point ending instead of repeating many of the arguments already made above.

Our main observation is that sustainability does not have to be a term that is conceptually hard to pin down – quite the opposite. Sustainability is a state in which the Ecological Footprint [64] of humanity stays below the regenerative biocapacity of the planet. The problem is rather that the frameworks and definitions we have presented in this paper inevitably have “unpalatable” consequences in terms of requiring major changes on individual, collective, institutional (including academic, research and corporate), societal and international levels. Nevertheless, the frameworks and definitions we have presented are grounded in ecological reality, which we believe must be the starting point of any real effort in sustainability research.

While remaking today’s unsustainable societies and shifting today’s unsustainable trajectories represents a daunting task as well as a break with centuries-long processes and entrenched mindsets, we still believe we have no other option than to face those monumental challenges. Part of the task is to find ways to reformulate today’s probelms first into challenges and then into possibilities (c.f. the quote from Daly [16] at the top of this page). We believe that HCI and computing will remain very important for a long time and that there are numerous challenges that our community could and should work on [70, 50, 51, 60, 61, 63, 46], but, that these challenges are oftentimes radically different from the current thrust of research and development.

Still, we believe that HCI is well positioned or perhaps even in a unique position to make a difference since “HCI researchers and technologists [not only] have the ability to shine a light on society’s problems, [but also to] provide platforms that enable individuals and groups to act on today’s problems” [17]. Before we can do that, we however first have to acknowledge that sustainability in the early 21st century means adapting to a reality of limitations, of trade-offs and of hard choices rather than the much more convenient “more of everything” mentality. This makes sustainability a revolutionary project and at this point we beg to differ from the mainstream sustainable HCI “evolutionary” agenda. The task facing us as a community at this juncture should not be to tiptoe towards sustainability, but rather to immerse ourselves in “the study, design, and development of sociotechnical systems in the abundant present for use in a future of scarcity” [60]. If we acknowledging that by taking small, incremental steps, we’ll advance only a little, is it then not the time to instead take bold action and accept the possibility of becoming ”accidental revolutionaries” [52, 59, 10]?

Inga kommentarer:

Skicka en kommentar