“You don’t need a methodology”

This would have to be some of the best and (in some ways) oddest advice about doing PhD study that I’ve yet heard.

It comes from Margot Brereton, and was part of some of the wonderful feedback that I received at this years doctoral consortium at OzCHI 2011. (And this is certainly not to discount the excellent and insightful advice from Gerhard Fischer ^[1], Lian Loke ^[2], and Toni Robertson ^[3].)

Ok, so. You don’t need a methodology. What does this mean?

The way I understood it is this: To do research, you need a question, and way of getting data to respond to that question. You then need a way of analysing that data to answer your question. Your question drives your methods. Your methods for data collection and your methods for data analysis are different.

Once you have some data you can begin to ask yourself: what does your data tell you about your question? This is an iterative process: a PhD is about question reframing in response to you data.

But you don’t start with methodology. What you need is a good question.

1. “if you do not like what you are are doing then it is a bad idea” ↩
2. “treat a PhD like an apprenticeship: find a good supervisor” ↩
3. “The PhD is the side effect of the person you become” ↩

Cameras I own: Konica Hexar AF

Part four in a series. Read parts three, two and one.

Somewhere in Western Australia, 2006

Now here is a camera that is truly close to my heart. When I’m asked the inevitable question (as a visitor looks up at my collection on top of our bookshelf) “which is your favourite?”, my answer is always “the Hexar”. I first became aware of the existence of the Hexar while taking photos at the Meredith music festival with my Canonet—I noticed another photographer with a few vintage cameras, and around his neck was an amazing looking one: like a Leica, but sleeker, more modern, with a neat little LCD showing the shot count on the top. I didn’t manage to speak to him, but my quest had begun.

I found my Hexar my first time in Japan, partly because I knew that I wanted to buy a camera in Japan, and partly because my good friend Pete (who was living there at the time) convinced me (after some single-malt scotch) that this was the one. He was right of course.

Akihabara, Tokyo, 2004

The Konica Hexar is a truly special 35mm rangefinder: it sports an incredibly sharp fixed 35mm f2.0 lens, aperture priority and program settings (or fully manual if you desire), fast and precise autofocus, solid construction. I have heard it called “the perfect street camera” because of it’s fast wide angle and extremely quite operation. Plus it just looks awesome with its sleek lines, sci-fi looking reflective autofocus array and LCD screen.

ACCA, Melbourne, 2006

I bought the later (1997) model “Silver” edition, which (apparently) comes minus the Hexar’s most lauded feature: Silent Mode. Silent Mode slows down the motorised film advance and autofocus to whisper quiet, and combined with the very quiet leaf shutter turns the Hexar into one of the quietest motorised film cameras around. Now, I said “apparently” because of a fact that was pointed out to me by Pete: the missing feature is just a software setting—it is actually possible to override it and restore Silent Mode to the camera by performing a complex sequence of button presses (which I managed to do, after a few aborted attempts). Pete explained this fact to a salesman at Map Camera, and it was listening to this conversation that I first learned about the existence of “Japanese” words that are strangely warped English words: “Onnu-offu” (on-off switch), “cunningu” (cheat, as in “cunning”).

Kyoto, 2004

But I digress. I continued to use this camera right up until I bought my Lumix GF-1, as none of my other digital cameras could do quite what the Hexar did (and the Lumix is still louder than the Hexar). There was something about the combination of size, speed and lens quality (I still think it has the best lens of any camera that I own) that made it, for me, perfect. The Hexar is retired now (the cost of film is sadly not worth it anymore), and I have managed to damage the hotshoe (now held in place with gaffer tape).

I took some of my favourite film shots with this camera over the years, and I am still a little sorry that it just sits gathering dust now. A note to all digital camera manufacturers: I’m still looking for a true digital replacement for the Hexar, so when you release one, I’ll happily be your first customer.

Documentation, not things.

What if the video archive isn’t a ‘video archive’, but an archive of ‘events that are recorded on the videos’? What is the difference? The current Circus Oz video archive is quite literal: a cupboard full of VHS and Mini-DV tapes. But once they are digitised at stored ‘in the cloud’ (as it were), it won’t be the video-object that exists in the archive (to the extent that we can say that digital objects ‘exist’ in any real sense), what will exist in the archive is a subjective record of a performance. Or less academically: you aren’t watching ‘videos’ anymore, you are watching a certain record of someone’s perspective of an event.

Here’s how it works: There is a performance. This is a real event, verifiable, happening in the world. The events of the performance are captured on video (in some format). This video is transcoded into a series of digital copies: one ‘archive quality’ (theoretically lossless), one ‘working copy’ (optimised for web streaming) for general access, others, probably stored in the nebulous cloud. Other data is associated with this copy (archive metadata, descriptions, images). Together these make up a record of a performance: a representation of the event. Documentation, not things.

How does this change the way of thinking about a video archive and archive practice? Is it because we are not concerned with preserving objects but with preserving perspectives?

In “the serendipity of the unexpected, or why a copy is not an edition”, Sarah Werner notes a problem with the digital archiving of manuscripts: that the process of digitising a copy of a text can destroy the history of the artefact. It is common practice, for example, to leave out blank pages when digitising text:

[..] their surrogates start with the frontispiece or title page, then move to the dedication/preface/letter to the reader/start of the text. But you know what’s missing? The blank verso of the title page. Does that matter? I don’t know. It might. It depends on what you’re looking for. But you’ll never know that you might be looking for an answer that depends on that blank presence if you don’t know that it’s not there.

Regarding the video archive, this point is still relevant: the representations of performance were originally on VHS and transcoded via SAMMA, so we should keep evidence of this. Does that matter? I don’t know. It might. It would be great to be able to see who’s handwriting is on the original videotape. It depends on what you’re looking for. But regarding a performance archive, I would argue, it is not so important that physical attributes of the record object are kept, what matters is the performance itself. We also have to deal with the fact that the physical aspects of the record are actually becoming harder to preserve as all steps of the archiving process move to digital: Circus Oz are shooting digital video now, writing show reports and tape descriptions on a computer, not by hand.

This is where the ‘living’ part of the ‘living archive’ comes in: To archive a performance, we need multiple perspectives from primary sources (ie, people who were there) to be able to clearly interpret the contents of the archive. If we are concerned with a record of perspectives, not a record of objects, then allowing people to add content (memories, stories, …) to the archive makes perfect sense.

Cameras I own: Yashica MAT-124 G

Aka, ‘this aint no instagram, buddy’. Part 3 in a series. Read part two. Part one.

This camera stands out on the shelf, and for good reason: it’s big, it’s old, it looks like a machine. And that it is. The Yashica MAT-124 G is a camera I picked up (for much too much money) when I was living in Tokyo and became ever-so-slightly obsessed with street photography and old film cameras (the camera was made some time in the 1960s). It’s my only TLR (Twin-Lens-Reflex), and my only medium-format camera. In layman’s terms that means it takes big square photos, not small rectangular ones.

A fountain at Zenkoji temple, Nagano

The evolution of cameras as mechanical devices is interesting to experience for yourself: the twin-lens system (where you look through one lens, and the photo is taken with the other) was a very simple mechanical solution to the problem of how to see what the camera ‘sees’, and made for cameras with very few moving parts. The lack of a mirror reflex system (as required in an SLR) made for a very quiet shutter as well. Focussing was a genuine mechanical action of turning a knob and moving the lens mechanism backwards and forwards. I could understand how it worked. It was a box for capturing light.

Akihabara, Tokyo

And capture light it could! Like many of my older cameras, its light meter was long dead, but I bought a small mountable light-meter from Voigtlander, because I wasn’t willing to take the risk of relying on “sunny-16″ or taking guesses with this one, partly because the film and developing was expensive, partly because I thought the camera deserved a little more respect. It was a worthwhile addition, particularly for taking shots in dark conditions. It took a while to line up a shot, focus, check the light meter reading and take the shot, but it was worth it: the quality, mood and depth of the images still beats any other camera that I own.

Shibuya, Tokyo

Its size meant that it also got all sorts of strange looks from passers-by, but because it’s a top-down viewfinder (you look down in the top of the camera to focus), it looks like you are fiddling with your camera rather than lining up a shot, which made it great for candid shots on the crowded streets of Tokyo.

I pretty much stopped using the camera when I returned from Japan (I think I shot maybe only one or two rolls of film in Australia). It wasn’t just that the novelty had worn off (fun as it was, it sure was cumbersome). I really couldn’t justify the expense, and had been overtaken by the digital camera mindset of “take hundreds of photos and edit like crazy”. Instagram on my iPhone is fulfilling my ‘old looking square photos’ needs for the moment, but I’m sure one day I’ll want to go back to the slow photography of the TLR.

Rational reframing

Or, adopting the language of your clients as a manifestation of design rationality

As I work on the design of a prototype for the Circus Oz Living Archive I’ve been playing around with some basic experiments in reframing. It is a bit of a language game: by changing the way we talk about [something] we can change the way that we think about [something]. I started with a question: we are building a prototype, but what is a prototype? The answer could be very simple: a prototype is a thing, designed to test an idea.

This is where the language game begins. What else is a prototype? A prototype could be an introduction (perhaps the first time a client has interacted with a product or idea). A prototype is a tool (for data collection). A prototype is a rhetorical device (by leaving things in, or taking them out of a prototype, you are making an argument for/against certain aspects). A prototype is a process. You get the idea…

By asking the question, “if the prototype was x, what form would it take?” I am trying to force myself to leave behind my implicit understanding of what a prototype is for: perhaps a prototype can be for much more than testing and solving a particular design or technical problem. Schön calls this imposition of a ‘new way of setting the problem’ a ‘frame experiment’. This kind of experiment forces a reflection on your practice, or ‘Reflection-in-Action’ [and sorry about the sexist nature of the designer as ‘he’ in this passage]:

When someone reflects-in-action, he becomes a researcher in the practice context. He is not dependent on the categories of established theory and technique, but constructs a new theory of the unique case. His inquiry is not limited to a deliberation about means which depends on a prior agreement about ends. He does not keep means and ends separate, but defines them interactively as he frames a problematic situation. ^[1]

This kind of frame imposition can lead you to strange and interesting places. One of the paths that I’m interested in pursuing in my PhD research is the idea that performance practice and interaction design practice share certain qualities. This gave me a new frame to work with: What if we consider the ‘prototype’ analogous to the ‘rehearsal’?

Work with me here: A prototype shares much in common with a process of iterative development in a performance context. The rehearsal and the prototype are both tools to develop, explore, communicate and evaluate ideas. The prototype, like the rehearsal, encourages improvisation and reflective feedback. The prototype, like a rehearsal, can take place in a context similar to that of its final outcome. The prototype, like a rehearsal, can be understood as a means to an end, a version of an artefact that is subject to change, a collaborative work in progress.

One of the well known problems with the use of prototypes in interaction design is that it can be very hard to communicate to a client what a prototype is, and what a prototype is for. You can make the prototype as ‘low fidelity’ as you like, but this can lead to the client thinking that it is ‘broken’, you can make it ‘high fidelity’ which can lead to the client thinking it is finished (and so only providing superficial feedback) ^[2], you can carefully ‘filter’ your prototype ^[3] to test for a single quality, but this doesn’t let the client experience how the design might be used in a real world context. It can be really hard to get a client engaged in a process like prototyping.

I would say a lot of these issues actually arise from communication/language problems. A designer intuitively understands ‘prototype’: what it means, what it affords. But this is a technical, professional language. What if instead we adopted the language of our client, using a metaphor for prototype that they can understand?

According to Löwgren and Stolterman:

‘a designer has to have a solid understanding of the complexity involved in being rational. When a designer works with a client, she has to be able to appreciate the client’s understanding of rationality, in relation to her own understanding of it. A basic appreciation of that relationship is fundamental to the communication between designer and client. Rationality is therefore not only a matter of how to do things, but a precondition for good communication’ ^[4].

I would suggest that appropriating language of a client is an explicit attempt to understand a different rationality: for a designer, it is ‘obvious’ that a prototype is subject to change, unfinished, wants feedback, etc. For a performer, while ‘rehearsal’ has these ‘obvious’ qualities, ‘prototype’ sounds technical and meaningless.

This example of reframing in the language of practice is particularly interesting in the context of the Circus Oz Living Archive project, considering that one of the points of adoption for the digital archive is actually as part of their rehearsal practice. So: the prototype of the digital video archive is a ‘rehearsal’ for the digital archive, it is also a tool to use within the context of a rehearsal.

‘Rehearsal’ has pretty clear implications: this is an unfinished object; this is something open to change; this is an object with the potential for collaborative development; this is something that, while not final, will be public at some point in the future. Using this language can work the other way too: reframing ‘prototype’ as ‘rehearsal’ serves to remind me of the performing arts community context in which I’m working, and also serves to remind me of the prototype’s transience and malleability (which should help prevent me from becoming too fixated on a particular idea or design solution this early in the project).

1. Schön, Donald A. 1991. The Reflective Practitioner: How Professionals Think in Action. Aldershot: Avebury. p68 ↩
2. McCurdy, Michael, Christopher Connors, Guy Pyrzak, Bob Kanefsky, and Alonso Vera. 2006. Breaking the fidelity barrier. In Proceedings of the SIGCHI conference on Human Factors in computing systems  – CHI  ’06, 1233. Montréal, Québec, Canada. ↩
3. Lim, Youn-Kyung, Erik Stolterman, and Josh Tenenberg. 2008. “The anatomy of prototypes.” ACM Transactions on Computer-Human Interaction 15 (2) (July): 1-27.  ↩
4. Löwgren, Jonas, and Erik Stolterman. 2004. Thoughtful Interaction Design: A Design Perspective on Information Technology. The MIT Press, December 1. p50 ↩

Cameras I own: Ricoh GR-D

Part two in a series. See part one here.

This was the first digital camera that I actually liked using. I think in a lot of ways it was because of the look, which was largely based on the original late 90′s Ricoh GR1, only slightly smaller, and (of course) digital. I bought it because I was about to go on a long overseas trip, and decided it was finally time to make the leap from film (I really didn’t want to carry enough film with me to last 3 months). It was 2007.

Its small size and wide-angle lens made it great for street photography

When I bought the GR-D I was still in a ‘film’ frame of mind though, and couldn’t quite cope with not having a viewfinder. I shelled out for an add-on optical viewfinder that sat on the hotshoe (that’s the funny lump on the top). It was clear and bright, but it only gave you framing, no other feedback.

I liked the GR-D for its constraints, which reminded my of my film-based rangefinders. It has a fixed lens and a large amount of manual control. It was the first digital camera that made me think that maybe manufacturers are starting to ‘get it’: there are photographers out there that want a good-quality point and shoot that reflects the quality and characteristics (including controls) of good, small, film cameras. Unfortunately this ‘getting it’ hasn’t really panned out in the subsequent years, with the notable exceptions of the Panasonic Lumix GF1 (which I own and love), and the prohibitively expensive Fuji X100 (which I don’t). The GR-D never really lived up to this standard anyway in terms of speed or quality, but it was fun to pretend that the digital camera industry was getting somewhere.

Somewhere outside Reykjavik. Damn rocks!

My GR-D didn’t actually get very much use in the end. One month in to my travels, I managed to drop it on a rock in Iceland and damage the lens mechanism. My next camera-shop-stop was in London, and I couldn’t afford a replacement with the Australian dollar being what it was back then. I ended up shooting the rest of my trip with a Canon IXUS, which is a fine pocket camera, but comparatively boring.

You don’t get a sense from my illustration, but it was very small, light, sturdily built and quiet. A true pocket camera. My iPhone has effectively taken over my needs in that regard now, so I don’t really miss the GR-D. Plus the look of it fits right in on the shelf next to my various vintage film cameras.

Cameras I own: Canon Canonet QL 17

One of the first things that visitors to our apartment notice when they walk in to our lounge room is our camera collection on top of a bookshelf. There are 10 cameras up there: 9 of mine and one of Kate’s (a cute plastic Holga). There was a time when I actually used them all, or, at least, they have all seen a fair amount of use over the years. The 10 cameras aren’t even all of our cameras, the ones we still use tend to float around the house. I guess you could call it more of a ‘museum’ than a collection.

Somewhere in Tokyo, 2004

The Canonet is my ‘oldest’ camera up on that shelf (in terms of use, not vintage). It wasn’t my first rangefinder, but it’s the oldest rangefinder that I still have. My first was a lovely Ricoh 500G that I picked up at a second hand store when my very first film camera (a Pentax MZ50) was stolen. The Ricoh lasted well until I dropped it while moving house and snapped the lens clean off, and I found the Canonet on ebay to serve as its replacement. The Canonet became a faithful companion on my first overseas trip to Japan. I don’t think I’ve used it for at least 5 years: I opened up the back before I drew this picture and was surprised to see a half-used roll of film in there (now partially exposed to the light). I’ve no idea what’s on it.

SJ in Melbourne

The light-meter died within about a month of use and I took to setting the aperture wide-open (1:1.7 – it was fast too!) and guessing the shutter speed. This technique was hit and miss, but when I got it right, man this camera could nail the shot. The 40mm glass lens was a wonder to behold.

While obviously based on the design of the Lieca M3, the smaller, cuter dimensions and the angled rangefinder window gave this camera a look which has pretty much defined ‘camera’ in my mind for all of my life: it really is the classic camera design.

Editing is harder than writing

Last week I had my first attempt at writing an ACM conference paper. It’s not really a full academic ‘paper’ — it’s for a doctoral consortium — but it was still an interesting experience: I am submitting my PhD research proposal, and in order to fit it in to their conference format (seen above), I had edit my 5000 word (plus references) proposal down to ~2500 words (including references).

One thing about ACM papers is that they proscribe a structure that you must work to (including section headings). This messed with me quite a bit as my original proposal was written as a ‘slow build’ argument: my research questions and arguments were introduced slowly, arguments for research questions scattered through a literature review, or arguments for a research approach scattered through the explanation of a problem. I have a habit of using narrative and repetition as a rhetorical device: I’ll introduce a concept early on with a sentence, later  repeat that sentence with an expansion, later in the piece explain in more detail using the same words.

This doesn’t work for this kind of paper. I had to restructure my argument to fit to introduction/context —> literature review —> gap/research question —> project methodology, and the low word count meant I couldn’t play around with repetition at all. The positive: I had to re-think my argument such that it is now a little clearer in my head. The negative: I think this structure is boring way to make an argument. No narrative, boring to read.

What I’ve discovered (and I’m sure it’s not an original discovery): editing is harder than writing. I can bang out 500 to 1000 words per day with no problem. Editing those words to a concise structure, that’s a different story.

Communicating design synthesis

When synthesising the reams of data that we collected on the user-experience redesign for Pool, we produced several formal design artefacts that we used in workshops to communicate our thoughts and findings to stakeholders within the ABC. This process—explicitly encouraged by Jeremy Yuille—appears to be a direct response to Jon Kolko ^[1] and his argument that the private nature of design synthesis is one cause of problems in design practice:

When synthesis is conducted as a private exercise, there is no visible connection between the input and the output; often, even the designers themselves are unable to articulate exactly why their design insights are valuable. ^[2]

The artefacts that we produced were expressive objects ^[3], designed help us understand our process, but also designed to help include the client in abductive sensemaking. While these objects couldn’t capture the full gestalt of our insights, they allowed a level of understanding beyond the usual presentation and subsequent discussion of a traditional design outcome.

Kolko spends much of his paper defining an explicit sensemaking action framework, complete with some oddly specific instructions (my emphasis):

The designer will begin to identify insights in the data that has been gathered by combining an observation (I saw this) with knowledge (I know this). They can then write the insights on yellow note cards. ^[4]

While the methods he describes (reframing, concept mapping, insight combination) are not new to contemporary interaction design, the formalisation of a pattern language for synthesis methods is welcome and well justified, if a little over-prescriptive.

Where my experience on Pool fits here is not in relation to this formalisation, although we did use variations of the methods discussed in Kolko’s paper. Instead, it is in response to Kolko’s implicit argument for more effective communication of the sensemaking process to stakeholders. Discussing the lack of formality in design synthesis, Kolko notes:

Clients don’t see the relationship between design research and design ideas, and therefore discount the value of design research and design synthesis entirely. ^[5]

I can confirm this anecdotally from my own experience in various design and development roles in the industry. I imagine that most designers would have experienced this in some form during their careers. This not as an argument for formalisation so much as an argument for more effective communication of process. You might call it a ‘second-level externalisation’ of the existing ‘externalisation of knowledge’ performed by the designer during abductive sensemaking ^[6]. It is an attempt to make the implicit explicit.

What might this process involve? In the case of Pool it involved producing artefacts during sensemaking that acted as formal representations of our process. These artefacts were then used as a tools to communicate the (usually implicit) sensemaking to those not privy to the (usually private) insights of the designer(s). As a response to Kolko, it seems so obvious when stated: “The client does not recognise the value of design research and design synthesis—we really should communicate what we are doing more effectively”.

Kolko’s response to the sensemaking problem is a valuable industry focussed one: to formalise the processes through an applied framework (one that help designers understand their own insights), and to suggest that design practitioners allocate time to this formal process. I suggest an addition: produce formal artefacts during and after synthesis that can be used to communicate your process to stakeholders.

While synthesis is still primarily performed as a reflective and private exercise, production of formal records and artefacts could help a designer consider how and why they’ve reached certain design insights, improving the chances of effective articulation of concepts. These artefacts, when used as part of a client engagement activity, could help stakeholders to participate in—and better understand the value of—the research and sensemaking process.

1. Kolko, J. (2010). Abductive thinking and sensemaking: The drivers of design synthesis. Design Issues, 26(1), 15-27. ↩
2. Ibid. p. 25 ↩
3. Dewey, John. 2005. Art as Experience. Trade pbk. ed. New York: Perigee Books. ↩
4. Kolko, J. (2010), p. 26 ↩
5. Ibid., p. 16 ↩
6. Ibid., p. 18 ↩