Recently a thread of discussion arose in an open source software users list to which I wanted to reply. I’m not going to mention the specific project (I’ll call it TheProject here) because I don’t have anything nice to say about it, and I’ve chosen not to make my reply in that community. In fact, I unsubscribed from the email list after seeing this exchange. Nonetheless, I think it’s worth presenting my reasons for abandoning purpose-made open source software.
For context, the string of responses to a question about requirements for using the software included a response from a frustrated user who didn’t have the technical resources to achieve the desired ends. One of the responses, to which my reply is directed, had the nerve to suggest that the frustrated party had not tried hard enough, which was the proverbial straw that broke the camel’s back. I’m increasingly irked by this attitude from software developers, who seem to assume that anyone can overcome these technical barriers.
News flash: some people simply don’t have the background, skills, or motivation to learn how to code and fix software themselves, and these people are just as much captives of open source software developers as they would be captives of proprietary software developers. You can assign blame as you like, but those are the facts. The only real difference is that in open source, you can complain to your captors, and sometimes it might make a difference. Of course, most of the time, complaining will just earn you an “RTFM” response.
Here’s my reaction to seeing a frustrated user told that he hadn’t tried hard enough:
Unfortunately, I’m afraid that we are in the same boat as X_. The general lack of support has made it too difficult for us to properly maintain or customize our installation, and like X_, my very few requests for help have been completely ignored. Even asking specific people for help has not provided any progress. We are not trying to do anything fancy with customization, but even seemingly simple things have been out of reach. I have seen wonderful implementations, so I know it can do great things, but apparently not in my hands.
When I attempted to deal with these problems, the documentation routinely had placeholders instead of useful content when it came to addressing the issues I was trying to solve. For what it’s worth, a wiki is often a very poor solution to software documentation, as it usually lacks the necessary structure and completeness for supporting end users. With TheProject, I can’t even RTFM because there’s just no help for me there, so in the end, I can’t solve my own problems or even begin to untangle them. Since I cannot get assistance, it’s simply not a sustainable solution for us, nor is a hosted or paid service an available option. Some other open source projects of this general type do manage to provide outstanding support, although I wouldn’t attempt to speculate why; just as an example, myGrid’s Taverna team does a brilliant job with supporting a complex software product – they are reliably responsive and helpful, as is the user community.
This is not a matter of “giving up too easily”, which accusation is profoundly insulting for a semi-skilled end user to hear over and over. That is a good way to alienate users who genuinely want to learn and can make other valuable contributions, e.g. in user support, documentation, and usability, all of which TheProject could use in healthy measure. That unfortunately pervasive attitude is enough to make me abandon the software, because it signals a fundamental disconnect between developers and users that I know I can’t overcome and I’m no longer willing to tolerate. After making an earnest effort for months, I got nowhere with TheProject. It makes me feel bad that I can’t contribute code or get anything to work the way I know it could, but it makes me very angry that anyone would assume that I haven’t made enough effort. In any case, software that makes me feel this way isn’t worth my time. By contrast, working with Taverna is liberating because it is enabling, and I have learned quite a few new technical skills as a result of the community support and encouragement.
We are very shortly migrating out of TheProject and into Drupal using multiple modules. We find that for our needs, this works far better out of the box, is easier to manage, sufficiently flexible, and fairly straightforward to customize. More importantly, the Drupal community is also very supportive, particularly of those who wish to improve their skills but may need some mentoring, or even just a little assistance here and there. I inherited this installation rather than selecting it, and the choice to move to Drupal is more strategic than just answering our problems with TheProject. With Drupal, we are able to address several needs, and will be replacing 3 open source softwares with just one, which is much more sustainable in a resource-poor environment.
My notes from this year’s OSS conference were blogged on the FLOSShub site that I’ve been developing for the last year, so I’m just providing links to the specific posts of interest here – consume what you like, and leave what you don’t.
This was my second time attending the OSS conference, and while both were great experiences, this year was definitely an improvement over last year. It wasn’t about venue, either, since last year the conference was in Milan, Italy, and this year it was in Skövde, Sweden – although admittedly, the Swedish hospitality seemed much warmer than the Italian venue. The keynote speakers were fantastic, and I got to meet both of them, each of whom is a genuine, good person, and a notable leader in OSS. The doctoral consortium was an impressively high quality affair; the reviewers had clearly put a lot of thought into their responses to our work, and it was refreshing to talk with other students who “get it” in terms of shared context for research.
Of course, there were also paper sessions and the IFIP 2.13 business meeting, for which I’ve also posted notes. The papers were generally very good, and I think overall quality of the work and the discussion was better than last year. This seems to be an excellent venue for work on open source, in terms of networking opportunities, developmental support, and quality of the research and community. I look forward to attending next year, when the conference will be held in exciting South Bend, Indiana.
Obviously enough, I haven’t blogged in awhile, although not because I have nothing to say. I’m hoping that I’ve now found the thread that this blog has been missing, a theme to make it worth writing and reading. And it was obvious: the social life of information.
When I talk about the social life of information, I mean many things. The interactions of people and society with information, from creation through use and preservation, are myriad and fascinating. How do we do it? It is at once trivial and yet complex; there is more than a lifetime of research to be done in researching social information behavior.
Clearly, that alone cannot be a focus. My focus as a researcher is a good start, though, and I’ve been working on refining that for a couple of years now. My current research focus is on the interaction of organizational design, work design, and information technologies in distributed work. More specifically, I’m interested in coordination of distributed work, and even more specifically, in certain kinds of work – voluntary work and scientific work. Even those qualifications don’t narrow things down as much as it might appear, but it’s a good start.
Moving forward, I’m going to try to focus on just a few things. First, I’ll discuss conferences and other collaborative efforts, because spreading the memes is important. Second, I’ll write a little about my research, because sharing research results with non-academics can make a difference. In addition, I may occasionally discuss some of the technical geekery that I engage in to do my work, because blogs that do this have been enormously helpful for me, and I certainly have some specialized knowledge that can be shared in kind. Finally, I plan to dig into my views on open movements, particularly open source, open data, and open science.
So that’s the new agenda for my blog; we’ll see how well it goes. Comments are welcome, if you have ideas of things I ought to be talking about.
Last week, I attended the ACM Conference on Human Factors in Computing in Boston, acting for a fourth year as the official conference photographer. The role has its pros and cons; the downsides are that I don’t get to see everything I want, and I spend a very long week (approximately 100 hours) to do the job “right”. On the plus side, it has allowed me access to many leaders in the CHI community while performing a highly visible service, and that level of comfort and familiarity with the “who’s who” of a research community is hard to acquire. This was my last year as a student volunteer and conference photographer, however, and they presented me with a fantastic digital photo frame in appreciation. It’s wifi-enabled and pulls an RSS feed off my Flickr photos.
Boston was great fun, too; I walked over 35 miles and took about 3200 photos in the week I spent there. I particularly enjoyed seeing a Shepherd Fairey exhibit at the ICA, and the Gehry building on the MIT campus. Both are more outlandish and visionary than pictures can capture, and of course, I didn’t actually take any photos of Fairey’s work, though it was really a wonderful show.
—-
Monday, April 6
Opening plenary – Judy Olson – “Even small distance matters: Social ergonomics in collocated and remote teams.”
Four big “important topics” warranting further research: How to design a large activity, generalize across field studies, human robot interaction, and social ergonomics. Social ergonomics is “understanding the ways people interpret social interaction so as to inform design and use.” Major categories include the physics of space and reciprocity, conversation, impression management, and time. We don’t typically notice these things until they’re broken – e.g. broken physics of space, where others can see you without you necessarily seeing them, and because you can be seen and heard, they will assume you are nearby in time and space. I actually have (most of, but hopefully soon all of) the slides from this talk if you’re interested in more detail. It was really excellent.
–
Panel – “Creativity, Challenges and Opportunities in Social Computing”
Gerhard Fischer, Pamela Jennings, Mary Lou Maher, Mitchel Resnick, Ben Shneiderman
The focus was primarily on theories of creativity; not much attention to or exploration of creativity and social computing in group or organizational contexts.
Mary Lou Maher discussed collective intelligence and contrasted this to socially intelligent computing, which brings together people and computers for new forms of collaboration, communication and emergent intelligence that weren’t previously achievable through human or computer alone. Examples include Internet-based collaborative diagnosis of Thalium poisoning, AstroAlert for astronomy hobbyists, Goldcorp Challenge, NASA ClickWorkers, and the ESP Game. Key research challenges: What are effective methods for studying emergence of “human-computer creativity”? How do methods of study scale from small through large numbers of entities in these systems? What computational, cognitive and social substrates and abstractions enable and facilitate the design of systems with emergent human-computer creativity properties? What design techniques and technical characteristics enable open systems for the fullest breadth of social creativity?
Mitchel Resnick demonstrated “Scratch” which seems to fit the “collab” model of creative peer production (from a CSCW 2008 paper). This seems to have interesting properties to contrast with other forms of mass collaboration and social production (FLOSS, Wikipedia) as the nature of the work is primarily creative. Very interesting observations of children independently realizing the value of open sourcing their own creative content (for remixing, which is a primary activity in Scratch) and capitalizing on their abilities to create a niche role within the community.
Pamela Jennings’ discussion was more abstract – about enabling creativity through computing. Ben Shneiderman discussed a model for motivating technology-mediated social participation, “from reader to leader”.
–
Paper session – Education and Science
“Friend or Foe? Examining CAS Use in Mathematics Research”
Andrea Bunt, Michael Terry, Edward Lank – Uni Waterloo
CAS stands for computer algebra system. Found uneven uptake in mathematicians, leaving aside issues of collaborative work; they used CAS only for specific purposes and had strong preferences for paper. Some reasons included the affordances of paper, the desire to “stay sharp” on computational skills, and a general mistrust of the CAS because of the lack of internal logic transparency. I talked with the first author later at a break and gave her two more reasons that mathematicians may prefer paper to CAS: 1) mathematicians are trained on paper and chalkboards; 2) CAS operate at a very low level of abstraction, while most substantial mathematical research requires a very high level of abstraction.
“Pathfinder: An Online Collaboration Environment for Citizen Scientists”
Kurt Luther, Georgia Institute of Technology; Scott Counts, Kristin B. Stecher, Aaron Hoff, & Paul Johns, Microsoft Research
I met Kurt last year at CSCW, where he presented a great paper comparing FLOSS to Wikipedia and “collabs” for collaborative animation creation. The paper created a new system for citizen science data contribution and interaction. The effort was focused primarily on system development for a custom in-house deployment of a system that seeded and then recorded additional contributed data related to commuting and air quality in Seattle, where they were specifically supporting participant discussion and analysis of their data, providing participants with tools to take a more active role in inquiry than is typically seen in citizen science projects. In this respect, Kurt saw potential for an unintentional “dark side” of data contribution, in that participants seemed likely to contribute additional (potentially skewed) data that supported their views when replying to ongoing discussion of an issue.
—-
Tuesday, April 7
Panel – “Mobile technologies for the world’s children”
Allison Druin chaired the panel, which included panelists from UNICEF, OLPC, LeapFrog Enterprises, and Sesame Workshop. The session started with a video of a boy from Indonesia, very seriously posing the question to the panel, as to whether teleportation is realistically achievable. Yes, teleportation. Well, it would be a mobile technology, wouldn’t it? I didn’t stick around much longer than that.
Paper + Panel – Scientometric Analysis of the CHI Proceedings
Paper: Christoph Bartneck, Jun Hu – TU/E
Panelists: danah boyd, Gilbert Cockton, Robert Kraut
Among the trends pulled out in the paper, the shift in organizational representation has moved from more industry to more academic; Apple used to publish in CHI but no more. Best paper winners don’t seem to acquire more citations than other papers; there are some extreme outliers in both the “normal” and “best paper nominees” categories, but none in “winners”. Bob Kraut criticized some of the analytic methods; danah boyd said that there’s more to impact than publication.
–
Paper session – Software Developers and Programmers
“Comparing the Use of Tangible and Graphical Programming Languages for Informal Science Education”
Michael S. Horn, Erin Treacy Solovey, R. Jordan Crouser, & Robert J.K. Jacob, Tufts University
Implemented a museum display on coding robots: collaborative coding opportunities using code blocks, either in a GUI or a tangible block, to assemble programs that are then executed by (modified Roomba) robots. Obviously, two experimental conditions; neither seemed better than the other in terms of evidence of learning outcomes, but the tangible blocks were more engaging for children. In particular, girls were far more likely than any other demographic (boys, adult men, adult women) to engage with the exhibit in the tangible blocks condition, and to lead the activity. Interestingly, adult males were the only demographic that preferred the mouse-based GUI input condition.
“Designers Wanted: Participation and the User Experience in Open Source Software Development”
Paula M. Bach, John M. Carroll, Pennsylvania State University & Robert DeLine, Microsoft Corporation
An extended version of the paper that Paula presented at the iConference, and a better delivery with more background to the work.
“Understanding How and Why Open Source Contributors Use Diagrams in the Development of Ubuntu”
Koji Yatan & Khai N. Truong, University of Toronto; Eunyoung Chung & Carlos Jensen, Oregon State University
Initial assumptions about FLOSS and development work were clearly predicated on a different model of production than community projects that interest me. Found something better to do.
–
Design Vignettes
This was a session incorporating a wide variety of hands-on multimedia demos, from fanciful to purposeful. It’s rather hard to describe these design examples, as they included incredible industrial design (especially from TU/E-Philips and NYU) combined with an amazing array of technological novelty. They literally made me gasp in surprise and delight. My favorite was the computer program that could determine and help appropriately design the structural integrity qualities for laser-cut paper irregular polyhedra lamp shades. They were beautiful.
There was also a “breathing” lamp shade, T-shirts with an LCD display that reacted to the colors of objects in the physical environment, a Jackson Pollock-style Wiimote-controlled light painting system complete with a (literal) light paint-bucket color mixer, and spinning ceramic plates (think variety show) used to spin tracks for DJing. I tried it, and it worked! I didn’t even break any plates.
—-
Wednesday, April 8
Paper session – Studying Wikipedia
“Coordinating Tasks on the Commons: Designing for Personal Goals, Expertise and Serendipity”
Michel Krieger, Emily Margarete Stark, & Scott R. Klemmer, Stanford University
This was a neat plug-in (WikiTasks?) for Wikipedia that allowed people to track tasks that needed to be done on Wikipedia, but not just for themselves – also general tasks for the collective. Users could then adopt these tasks and keep track of their completion. More people created tasks that other people created than those that they did themselves. Enthusiastic user response.
“Coordination in Collective Intelligence: The Role of Team Structure and Task Interdependence”
Aniket Kittur, Bryant Lee, & Robert E. Kraut, Carnegie Mellon University
Very cool extension of work I saw presented at CSCW. Involved looking at the coordination costs of having more editors and the level of centralization of the work. Basically, more editors isn’t helpful after a point, at which juncture centralized efforts make a positive difference.
–
Paper session – Photos and Life Logging
Baby Steps: Evaluation of a System to Support Record-Keeping for Parents of Young Children
Julie A. Kientz, University of Washington; Rosa I. Arriaga & Gregory D. Abowd, Georgia Institute of Technology
Digital baby book for recording developmental progression to support well-baby visits more effectively. Built in functionality for sharing newsletters about baby’s progress, as well as other thoughtfully targeted features, plus recording usual baby book types of memories. Interesting effects on physicians and families perceptions of each other – overall physicians had more respect for the families when they could see evidence of the parents’ attentive monitoring of developmental progress. On the flip side, one couple lost respect for the physician.
Making History: Intentional Capture of Future Memories
Daniela Petrelli & Steve Whittaker, University of Sheffield; Elise van den Hoven, Eindhoven University of Technology
Had families create time capsules, documented and archived the contents, asked the families why they included each thing, then returned them to the families for opening in 25 years. Families were instructed to include both physical and digital artifacts. Interesting analysis of what they did include – most often things directly related to people, as well as a lot of images and records of the mundane, e.g. “day in the life” diaries and grocery store receipts. With respect to digital objects, 3 families included none because they didn’t know how well they could rely on the technology’s later usefulness. 3 families included playback devices (iPod, CD player, etc.) with memory media because they felt this was the only way to ensure that they would later be able to access digital content; 4 families included only memory media with the explicit assumption that means of dealing with archival media would be worked out in 25 years. (If only!) Did a nice classification of object types with analysis of participants’ reasonings for including objects such as the symbolic types which often had no explicit index, e.g. a roll of toilet paper because “we’re a nappy-free family now” or mismatched socks because the household was chaotic. Families felt that they would enjoy decoding the clues that they had left for themselves, and therefore an explanation of symbolic items was not needed.
—-
Thursday, April 9
Student Research Competition – Finalist presentations for 3 graduate and 3 undergraduate students. The topics were: usability in industrial control systems, privacy labeling design, cognitive consequences of social search, computer game immersion, a kinetic typography engine, and a multi-touch interface to control multiple mobile robots.
–
Student Design Competition – Finalist presentations for top 4 student design teams; theme was designing technologies to support local rather than global consumption. (2010’s theme: encouraging people to take a walk!)
The University of Michigan teams took first and second place! First place, TreasureHunter is a social media system for supporting reuse through thrift stores. Second place, MIFresh, is a kiosk system to support consumption of local produce in Detroit grocery stores. Third place was University of Iowa’s CropConnect, a site design to support a CSA. Fourth place went to WattBot, with a design for an iPhone app interface for monitoring household electricity consumption.
–
Closing Plenary – Kees Overbeeke – Dreaming of the Impossible
Kees Overbeeke of TU/E’s Industrial Design program presented on his vision for the integration of design, science, and engineering. He presented a very lovely, very European, and very inspiring view on design and technology. His abstract is a statement of beliefs about design that structured an intriguing look at his school’s approach to developing talented industrial designers through an approach that emphasizes learning skills through skills:
* Design is about people. It is about our lives, our hopes and dreams, our loneliness and joy, our sense of beauty and justice, about the social and the good. It is about being in the world.
* There is a primacy of action. In accordance with Merleau-Ponty’s and others’ approaches to epistemology, I strongly believe that meaning cannot be detached from action. Meaning is in (inter)action. There is a primacy of embodiment.
* A design theory consequently must be a theory of action and the embodied in the first place, and of meaning in the second, and not the other way around. Reflection on action is the source of knowledge.
* The methods used must be rooted in design practice, in the socio-cultural and multi-cultural environment, invigorated by experimental and technological methods from other disciplines.
* Intuition and common sense should be high on the agenda. They should be exploited to the maximum. Le sens commun n’est pas si commun, as Voltaire said.
* Design practice and research are powerful generators of knowledge. They are a way of looking at the world and transforming it. Just as classical disciplines are. Design teaching and research should therefore be interwoven.
Right now, I should be writing something else. I ought to be working on writing two grant applications, a journal article revision, and another journal article draft. Today I even asked for leniency on portfolio writing deadlines in favor of these other more pressing writing tasks. So I shouldn’t be blogging. I should be writing for money or publication.
I’m just too frustrated to do that writing, though, and not writing about what frustrates me is standing in the way of real writing productivity. What a mess. So what’s frustrating? It’s not just having too much on my plate, though I wish I had the gall to tip the dish and watch things slide off into the garbage. For the moment, I’m caught up in stupid graduate student angst over defining my disciplinary home. Again.
I’ve more or less accepted that I’ll have to claim Information Systems as my field for my end-of-coursework exams. It’s a pragmatic and strategic choice; pragmatic because IS is an “acceptable” field of study in my context (don’t even get me started on that; suffice it to say that choice is illusory), and strategic because it’s my advisor’s field, and therefore presumably there are benefits to be conferred from his accumulated social capital. All of this makes good logical sense and appeals to my functionalist sensibilities.
However, the problem arises when after four years of graduate school, I finally get a chance to take an IS course. After reading IS literature for the last 11 weeks, I am finding few links to my interests and the reading is generally pretty dull. There has been a relatively small subset of the literature that has been of any interest, and the more I read, the less I want to.
So the net effect is that I don’t want to read, and I don’t want to write, because nearly all of the reading and writing I’m doing is on topics that I don’t know or care enough about to make it seem worthwhile. It’s completely unrewarding and just makes me unhappy, and I’m really tired of hearing that it’s good for my character. The more time I have to spend this way, the more I question my suitability for the profession. I’m trying to be patient, since there is only a month left until I finish my coursework requirements, but that’s never been one of my strengths.
Hopefully I’ll make a more optimistic post in another month when the semester is over. In the meantime, I’ll just keep working, because there’s not much else for me to do.
