Category Archives: Thesis Fieldwork

Thesis fieldwork

Online Ethnography

There was an interesting attempt at a discussion on the Anthrodesign mailing list recently as to what online ethnography actually entails. But the discussion never really seemed to get off the ground, and effectively had died by the time I posted my comment. So I thought I put it up here with a few adjustments:

Online ethnography is a very interesting research practice. In part because you are completely dependent on what your informants are willing to show you. You can only learn as much as they put online, and you have no way verifying that what they say is true. As the classic saying goes, “On the Internet, nobody knows you’re a dog.”


When you’re initiate ethnographic reseach online, you are acutely aware of this fact. No physical context or cues makes it difficult to interpret the actions and motivations of people. The short film “The Parlor” gives a great impression of how these issues.

One ethnography that does well to explore these issues of representation and anonymity is Annette Markham’s “Life Online“. But Markham’s central point is that the net-savvy people that she interviews do not see the Internet as a separate place that they enter when they go online. Rather, “going online means turning on the computer, just as one would pick up the phone.”

Online and in-person are parts of the same domain of social experience. I find that a lot of talk about “virtual ethnography” misses this and instead attempt to explore Internet relationships and behaviour as if they are completely different and unrelated to their informants’ in-person lives.

What I found in my fieldwork is that doing online ethnography is little different from other flavours of ethnography in that you have to examine not just a single aspect of your informants’ lives in order to be able to appreciate their practices and motivations online. This is equally true of everybody else online: Social ties are immensely strengthened by in-person meetings. As Gabriella Coleman has argued, online sociality augments offline sociality, rather than the other way around. In a similar vein, Brigitte Jordan labels this mixing of physical and digital fieldwork “hybrid ethnography” and argues that “the blurring of boundaries and the fusion of the real and the virtual in hybrid settings may require rethinking conventional ethnographic methods in the future.”

I don’t know exactly how ethnographic methods may require rethinking, I can only point to a description of how I combined different research methods, online and in-person in my fieldwork. If you’re curious, you can read my reflections on being in a digital field and my experiences there in my field report [pdf], which I’ve just uploaded for the first time (shame on me for putting it off for so long).

Having said that, I don’t think that online ethnography on its own is without merit. There is plenty of potential to learn from people online from behind the computer screen. But there is one other central issue here: It is incredibly easy to just observe others and not participate online. They can’t see you so there’s no social awkwardness associated with lurking. Not only is it unethical to some extent (just because it’s public doesn’t mean you shouldn’t let people be aware of your presence), but it is also a bad way to do research.

Actively sharing yourself, participating on equal terms is the cornerstone of participant observation, giving you the best possible opportunity to experience what your informants are experiencing. And it is the central way to build trust with people online. Actions do speak louder than words online. Much louder. And it is perfectly possible to do online participant observation. A great example of this is Michael Wesch‘s fascinating study of the Youtube community. Both Wesch and his students shared themselves through videos of their own in a way that garnered both respect and interest in their project. Video is a much more personal and credible way to interact than text online, and it is well worth the time to check out Wesch’s presentation of their study (available on Youtube, of course).

I’d love to hear about others’ experiences doing fieldwork online. So please do share.

Opening the source

Now that I’ve officially finished my fieldwork, and with all the talk going on about Open Access Anthropology, I thought I’d try my own little Open Access experiment. I’ve decided to publish the question guide I’ve used for my fieldwork under the GPL. I’ve even indented and commented them in proper code fashion (or, at least, as far as I’ve been capable of emulating it).

Also, at suggestion of one of my informants, I’ve answered my own questions to offer my informants and other interested parties a bit more background in my own interest in computers, Ubuntu and the F/OSS world.

One of my hopes in doing this is that more people in the community will find interest in looking at the questions, and possibly even writing up their own answers. If that were to be the case, I would love to see them and incorporate them into my thesis. So send me any answers you feel like writing! 🙂

On Free Software Conferences

When I tell people that I do fieldwork among Free Software developers, I often try to relate it to more traditional anthropological ventures as a way to make it clearer to people what it is I do. Traditionally, anthropologists travelled to the part of the world that used to be colonized and lived among the conquered natives. Seeking to understand their social structures, their values and their rituals – basically their way of life.

So I compare Free Software projects to these native tribes that anthropologists usually study. For it is a community that is built around a common interest in computing and shared values around that interest. The difference is that it is a tribe that is defined not by its association to any specific place but rather by its use of a technology.

For most of the time, the interaction within Free Software projects are shaped by the technology they use – mailing lists, IRC channels, web forums, even VOIP phone calls, but once in a while they gather at conferences to create those real human face-to-face connections that add a vital, physical dimension to the social life of a project.

These ‘tribal gatherings’ have been described as ‘the quintessential hacker vacation’ and reminds me most of all of a (Boys’) Summer Camps. They are intense festivals celebrating all things hackish where the developers gather to wear themselves down with sleep deprivation, cumulative hangover, shared passion for technology and constant social interaction. The conferences offer excellent opportunity to revitalize and energize the developers’ interest and belief in the project.

Conflicts are resolved, plans are laid out, specifications are written, unexpected meetings happen and friends are made. A different degree of collaboration is made possible by the conference as the tribe convenes and for a week or two actually is a temporary village of its own.

This week such a village has been erected by the Ubuntu community at the Google Headquarters in Mountain View, California. More than 160 Ubuntu developers, community members and upstream developers have convened from around the world in order to review and plan the next release of Ubuntu, due in April 2007.

The Ubuntu community differs from many other Free Software projects in their close relationship to Canonical – the company that employs a number of the core Ubuntu developers, drives the tight time based release schedule, and organizes the very focused and professionally planned Ubuntu Developer Summits to which they sponsor a good number of the community members and upstream developers from projects such as KDE, LTSP and GNOME whose work is central to Ubuntu.

This means that a snapshot of the Free Software world as seen from an Ubuntu perspective gathers every six months to collaborate and get to know each other in person – something that often proves to be invaluable when it is augmented through the digital means of communication where a previous awareness of a person’s personality and physical presence can make the difference between understanding and conflict.

Since conferences are such an integral part of Free Software development, the location and organization of the conference is essential, and this got me thinking. With Google hosting this conference, the facilities are generally in very good order. Unfortunately, the hotel where most of the developers are staying are 15 minutes away by bus and the logistic problems of ferrying people back and forth are not insubstantial.

Other conferences in other locations have other problems(such as bad food, bad bandwidth and bad conference facilities) so my thought is: Why not make a hotel specialized for the needs of free software projects? Based on my observations, I’ve drafted up an idealised list of requirements:

– Located close to an easily accessible

What Bikeshed?

Mark Shuttleworth’s recent post on the new gaudy desktop prettiness of Ubuntu has received a good deal of interest and discussion (more than 130 comments and counting).

Pretty much all of that discussion was summed up in one of those comments:

# Murray Cumming Says:
October 25th, 2006 at 7:31 pm

The bikeshed is brown.

The bikeshed in question is this one, and the idea behind it has proven to be one of the most powerful ways of explaining why so much energy is spent on inconsequential nitpicking in F/OSS projects. Especially when it comes to artwork and usability.

Why we have anthropologists

Native speakers can rarely explain the grammatical rules of their own language. In the same way, those who are most ‘fluent’ in the rituals, customs and traditions of a particular culture generally lack the detachment necessary to explain the ‘grammar’ of these practices in an intelligible manner. This is why we have anthropologists.

Kate Fox

I came across this quote in a little book by British popular anthropologist Kate Fox, and though it is a simplistic statement, I kind of like the way it presents anthropologists – it does makes us sound very important, doesn’t it?

Meanwhile, in my field of study, all of the computer people I interview prove to be very reflective and often discuss community structure and governance issues with a fair bit of detachment. So I often end up discussing these topics with them on equal terms. I wonder whether I can still call it anthropology… 🙂

I’m in Oslo this week, interviewing another couple of Ubuntu hackers, and learning even more about how it is to participate in Free Software development. I’m getting to the point where I have so much data that I’m really looking forward to beginning the thesis writing proper and see what sense I can make of all of this.

Ubuntu governance discussions

It didn’t take long for my specification on community governance best practices to be superseded by an avalanche of community and governance-related topics that are already approved for the upcoming Ubuntu Summit. Clearly, it is something the governing bodies have been meaning to put on the agenda for some time. And basically, it looks like these topics will open up for a thorough evaluation and possibly a complete reworking of the growing community structures. It will be very interesting to see how these discussions pan out:

2006 Community Council Nominations

2006 Technical Board Nominations

Ubuntu Membership Management

New Developer Approval Process

MOTU (Universe Maintainer) Organization

Developer-only mailing list

Ubuntu Forums governance

Ubuntu Local Community Team Organization

Coordinating Local Community Teams with regards to Countries and Translations

Opening up for so many discussions at once would be unthinkable in most other F/OSS projects, but there is such a professional energy and direction within Ubuntu that this kind of massive reworking does seem feasible. As one of my informants (that’s anthropological slang for persons whom I’ve interviewed in relation with my fieldwork) said, “I don’t know if there is a ‘usually’ in anything we do yet.”

Maybe this will turn out to be an occasion to set down such traditions and practices.

Going to San Francisco!

After some deliberation, I decided to blow all of my remaining grant money plus a little extra to buy a ticket to the Ubuntu Developers’ Summit at the Google HQ in early November.

This is quite a big step, since I only just came back from three weeks of fieldwork in Ireland, England and Scotland and I will barely have time to reorganise my notes and get everything settled before I jet off again. Going to the summit will also mark the end of my fieldwork in the Ubuntu community, and when I get back from the US, it will time to enter a state of intense thesis-writing.

For the Ubuntu Summit in Paris, I had added a couple of specifications on my own proposed research which didn’t really attract that much attention. This time, I’ve tried to write specifications on some of the issues I’ve come across in my time in the Ubuntu community, so as to be a bit more constructive. I’ve written a couple which I would love to get some feedback on:

Improving the participate page and team pages

Scaling the community governance

Documentation reorganization for topic-based help

.. I actually had a few others, including an idea for an Ubuntu usability team to coordinate between the KDE and GNOME usability upstreams and the Ubuntu and Kubuntu desktop teams and all of the various usability wishlist bugs that come in. I don’t know how useful such a team actually would be, but it does seem that there is a gap there at the moment.

Oh, and finally. Coming to San Francisco and Mountainview, I have no idea where I can stay for the summit. Any locals who would care to give me some hints? 🙂

KDE signatures

One of the frustrations with being on the road doing my fieldwork is that I can’t be on-line often enough to keep a solid presence in the Ubuntu community or even on my blog.

The birthday consumption of my bottle of Danish herb snaps on my last night at the aKademy
resulted in seven KDE contributors signing a pound coin with hopes of creating a cult item to be auctioned on e-bay:

Signed pound coin detail

Since there wasn’t a whole lot of room for signatures on such a pound coin, we added a Certificate of Authenticity to go with it:

Certificate of Authenticity

For those not in the know, the wantonness is a KDE in-joke, and the signatures belong to

Adriaan de Groot, KDE Quality Czar
Allen Sandfeld Jensen, KHTML Wiz
Ellen Reitmayr, affectionately known as ‘Usability-Ellen’.
Dirk Müller, one of the old Konquerors.
Florian Grässle, Master of usability and snaps-swigging both.. 😉
Robert Knight, young champion of the Konsole.
Tobias Klein, KDE PIMpster extraordinaire.

Now, I don’t actually like e-bay very much, nor do I have a usable account there, so there’s no auction yet before I get back to Denmark and have better time for arranging it. But I can say as much as that all the proceeds will go directly to furthering the cause of KDE. 🙂

Konference for Developers in ?ire

I have arrived safely in Dublin’s fair city, and have found it harder than expected to find decent lodgings, due to the huge crowds drawn to some sporting event taking place outside of town. But I’ve found a modest hostel dorm bed, and are now waiting for the registration to begin for the aKademy conference – the annual gathering of developers, designers and other contributors working on the K Desktop Environment.

KDE was the first open source desktop environment, hoping to contain a complete desktop use experience with web surfing, email, word processing and more. Announced in 1996, the name is a typically hackish joke on the then popular proprietary Unix desktop environment named CDE (the Common Desktop Environment).

KDE is based on the Qt toolkit which at first was not entirely free software which caused a fair few flamewars, and eventually the creation of the rival GNOME project in 1997. Since then Trolltech, the Norwegian company behind the toolkit, has made their work available under a fully acceptable Free Software license, and that issue has ceased to be a concern, and instead the main issue is ensuring peaceful collaboration and co-existence between the two development communities – something that can be more difficult than you might expect.

Not only are there deep differences in the way the two communities approach the basic aspects of the development having chosen differing programming languages for the basic architecture and differing design philosophies for the user configurations. But even simple considerations on the look and feel of the desktop itself can easily reach almost religious levels.

Currently, KDE development is focusing on the release of the KDE4 which seeks to refine KDE even further based on the experiences of the past years Open Source desktop work – not only with regards to technical infrastructure but also with regards to graphical design, look and feel and usability.

I’ll be here in Dublin for the conference, hoping to learn more about the role of Kubuntu, the KDE version of Ubuntu, within the KDE community, and how the relationship between the distribution and the upstream developers is working. Oh, and maybe drink a few pints of Guinness, as well.

Part of the tribe

Yesterday, I was approved for Ubuntu membership and am now an official member of the Ubuntu community. Becoming a member is just about the most formal procedure in the Ubuntu community, and it is still very, very relaxed.

New member candidates are approved at the Ubuntu Community Council meetings which are held every two weeks on the ubuntu-meeting IRC channel. Members write a short introduction of themselves and the work they’ve done in the Ubuntu community and add themselves to the meeting agenda in the Ubuntu community wiki. And based on that, the four members of the Community Council make their decision to either approve membership or ask for further involvement or testimonials from other community members who can vouch for the candidates.

I, too, did a wiki page presenting myself and the based on the work I described there, I was asked questions by the members of the community council which I guess is the closest Ubuntu comes to having tribal elders. As all the Ubuntu IRC channels are logged, you can easily find and read the full transcript of the meeting with the review of, and interview with, all of the candidates.

It was a quite strange experience, sitting in a room full of Ubuntu developers as the meeting unfolded in complete silence, everybody working and only a few of them taking the time to follow the meeting on IRC.

So what does being a Ubuntu member mean? Well, officially it means that you’re allowed to vote on community-wide issues which has been put up for a vote. But so far, in the almost 2 years of active Ubuntu community, there has not been a single occasion for a vote.
What else? Well, I am now allowed to make my own Ubuntu business cards, have my own email address and get my blog on the Planet Ubuntu aggregated community blog.

.. hey, wait. Maybe I ought to do that.