Hackathon Yackathon Reflections / Notes

IMG_0012_2

The hackathon is an interesting phenomenon in the way that people have organized around technology development and social issues. Closely related to the concept of open data, the sort of collaboration that goes on at a hackathon is often structured around themes in the public interest, whether that’s open government, disaster relief, or community-minded apps and technology solutions. Our Hackathon Yackathon was designed to invite hackathon organizers, participants and those interested in such events to look at them with a critical eye, asking the sort of meta-questions that would impede hacakthon’s goals if raised during such events.  We provided a means for people to submit topics and came up with a rough agenda beforehand, but the more intimate nature of this hackathon (focused around conceptual work and ideation) with a smaller crowd meant conversation flowed more organically through the entire group – instead of splitting into our breakout sessions as planned, we started with smaller groups that came together for our first “panel,” which flowed into the next and focused on two main themes: Hackathon methodologies, tools and outcomes, and the tension between civic hacking and data activism. We were fortunate to have a diverse set of perspectives and motivations at the event – VJ Um Amel and Willow Brugh joined us via GoogleHangout and we had a fairly casual conversation with well-thought out input from all the participants.

For me, a key outcome was the sense of the hackathon as a method that has many different iterations and is not beholden to a particular ethos or inherent quality – it focuses on not merely volunteerism, but invested participation, which can be cooped by commercial interests as “cheap labor.” This was something we could all agree on, however, the ideological divide between the hackathon as a model of cooperation versus competition was something that also reflected schisms between hackathons organized to contribute to municipal and governmental goals, versus those which are more activist in spirit and intent.

Another interesting point which I wish we could have devoted more time to is the divide between big data and observable community results. Hackathons draw from a spectrum of skill sets by encouraging participation from people who know how to do more than just code. However, they are not always inclusive – for the activist and even the civic mode, it can be difficult to involve people who can’t just build something on a computer in a day or two. In addition, there’s a need to address the agenda and viability of data (is it alive or dead, and who creates it for what purpose?). One possible answer we thought of was to incorporate some of the methods from citizen science, and to organize hacakthons in a series, where data collection methods are planed and tools are created, then people are trained and can go out to build the dataset themselves, and return to work on it at future events. This way, people of varying skill levels have greater agency, participation and investment in the project as it develops.

This is also important, because as VJ said, “People in the West fetishize data to the detriment of its content” – it may be convenient for us to build tools that are practical and feasible, but do they really solve any questions we need answers to? Are we building things that people need, or just making a wasted effort for the sake of the exercise?

On a final note, there needs to be a set of reusable tools that can be adapted from one hackathon to the next. There may not be a universal set of “best practices” for hacakthons across the board, but there are some outputs (experience) that can be applied, and documenting them is difficult. The “case study” and working groups approach OccupyData has taken is useful, but hackathons in general suffer from a lack of institutional memory and high frequency that dilutes their effectiveness. I’d argue that the most important output of a hackathon is the community that develops as people experiment with each other from one event to the next, but if a hackathon is a platform for building a concrete product rather than concepts, there needs to be more structure to their organization. Unfortunately, this diminishes the agency and representation of participants… the hackathon is a tool itself, a way of organizing, that is contextual and suited towards unique purposes depending on the organizers and the participants. The variety of options between those purposes and motivations reveal the tensions that exist when we try to think of the hackathon as having a singular format and structure, and organizers must try to align the way the event is held with their motivational values.

(reposted from MrLiterati.com)

—— Rough Notes

  • Intros
    • Bryan
    • Max – Superstorm Research Lab.org
      • government vs grassroots
      • space time differences
      • catagorized, short term model for governmental work vs the long term less categorized approach of the community
      • long term picture, more open idea of space, less categorization, better able to infiltrate space
      • jenny – hackathon as data and as a physical engagement
    • Meghan – OccupyRhetoric
    • Evan – Grad student at Cuny
      • really interested in the power that measuring things can have
      • data mining, data hype – criminal justice
      • democratizing capital and information skills
      • information workers who were laid off and never work again
      • no good system to retrain people and apply those skills
    • Christo reviewing occupypops
      • data as having a life of its own and how to foster spaces like this and improve the grounds for outcomes that support action and activists
    • Nitin – breaking the walls between academia and activists
      • hackathon as a mode of participatory research
    • Suzanne – data scientist, using data to improve outcomes
      • data has potential to do that in many contexts
      • too many hackathons
      • nothing substantial comes out of a one or two day project
      • occupydata – ongoing working groups
      • hackathon as a test bed –
      • digitizing street media manual
      • streetmedichandbook.net
      • kicking corporate booty – indiegogo – create some models
    • Ashley – Code for America
      • brigade – work on code, write white papers, organizing technical volunteers or developers – building an open source community
      • (volunteers vs participants) – hackathon models problem, creates a noncooperative privatized project – against the goals of civic technology community
      • don’t treat software as dialogue when you’re competing with others – moving towards a working group or an assistive community
    • Varied Comments
      • openhatch – free software foundation boston – Deb Nicholson – help developers make their first commits to a public, open source project
      • StackOverflow – 50 page doc on how to ask a question
      • Rhelp R assholes
      • structure of incentives that are problematic, new york is a “startup” city, economic development linked to big apps, want to create private products that make jobs – civic infrastructure and digital infrastructure has no resources to maintain it and is out of date, yet they’re throwing tons of money have people compete on using
      • if you can’t fix a hackathon, whats a better method?
      • Ricardo Dominguez
      • FreeCultureFoundation -complications of
      • Hardt & Negri – Declarations
      • Delivering something to activist groups is something that we haven’t always done successful
      • restricting the benefits projects have to groups you’re not fond of, or make sure that its incredibly useful for activists, campaigners, grassroots types in general, know that their aims and methods might be very different than what you are expecting
      • fad at the moment – new way of visualizing media
      • headlines and articles in traditional media re easy to show the volume of, can measure more easily with fewer sources, but social media means you also have to deal with interactions between reporting an event, tenor of the coverage, and trying to identify what really drove the story
      • tension between manual curation and automated “big data” stuff
      • Natural Language Processing for curating some of it
      • not wanting to document clicktivists
      • seeing how the story fragments vs seeing how it forms
      • problems of the question vs what you can do to address it
      • “unconferences are for people who don’t have tech skills”
      • competition vs cooperation
      • publiclab
      • what you measure is a form of politics
      • opposite MOOC Massively Offline Education something
  •  —— PANEL STARTS
    • VJ talking about hackathons with Sasha at MIT
    • How can we use these tools in academia and scholarship – when do you use something for pedagogy or for civic engagement
    • Suzanne talking about twitter and clustering and visual patterns – DataAnywhere
    •  – Need for people to start collecting data (not creating it but deciding what to collect) – 
    • we assuming that the public dataset is complete – bottom up vs top down data
    • changing the data model to include fields that aren’t part of legislative concerns
    • ability to affect the data determines whether or not it’s actually open
    • socrata – being able to see problems with differences between public data sets over the same information
    • data regulation vs public data
    • Chris: canvasing and data collection as activism itself
    • structurally government doesn’t care bout certain types of data – making those trends and its relevance apparent to authority
    • Ashley:what is a reusable tool that can come out of a hackathon?
    • Chris: should be low-tech – paper – framework for thinking, rather than a tool
    • VJ: outputs of hackathons are not tools, but visualizations and reports. – hi fi lowfi, people need to know skills themselves, resources and stuff, and that comes out of a hackathon as well – too many hackathons, not enough (institutional memory)
    • Christo: having a teach in inside a hacakthon is a good thing
    • tension between leadership, organizing and participation/representation
    • Questions: Data has an agenda
    • taarifa.org
    • max – research group formed organically
    • aggressive ground truthing threatened the group –
    • community based vs non community based – on behalf of a community or with multiple stakeholders who are mutually exclusive
    • VJ – Perspective of Egyptians/Arab/Syrians – People in the west fetishize data to the detriment of its content 
    • online data is superficial and detached from historical and cultural analysis
    • how can we incorporate that information in any analysis
    • Ashley: this is a homogenous group – mutual aid a lot of people here are of like mind, but trying to leverage populations to create “great things” unilaterally create great things – models of cooperation aren’t going to work as well
    • outside of an academic sphere you see more competitive models because you can map people’s goal generation to the model
    • government big data darpa stuff, or a civic urgency
    • Wilow the only way that government can interact with citizens is through the competitive model
    • Nitin question of scale?
    • Willow / Ashley competition doesn’t scale… both problematic
    • Willow Locality is important
    • Max you can’t unilateral leverage a heterogenous group – thats why advertising exists. So everything gets scaled
    • Nitin do hackathons function better when there’re homogenous or if it’s heterogenous
    • VJ heterogenous skill sets but homogenous culture
    • nitin agreement on the ethos, but disagreement on some of the perspectives – made it interesting
    • carl ensuring that the output of that existing group is useful and voluminous, we don’t have to do all things to all people, just produce a template that others with similar “spiritual” goals of horizontalness and stuff for others
    • Ashley what are those methods
    • Carl having fewer lines that you can draw around who’s in charge (?)
    • nitin can data or knowledge be the output – repositories?
    • VJ having an open repository seems basic and important, but what if it gets googlesque.. and how do you get people together to share it and explain it
    • nitin best thing we’ve been able to do is case studies, then teach ins in future hackathons
    • VJ whose your audience – people who are afraid of technology, (low skill sets?) activism or tech turns them away.
    • making the archive accessible – journalists
    • Ashley – the hackathons that just want to show people something one person thinks is cool drives others away… not the lack of work, but a level of separation that decreases productivity… community building is great, but making something is most important
    • problems of having an idea and just getting people there to work on it… again, agency
    • Willow original hackathons were about a sprint of work… now you have education, collaboration, problems come from where people shove all these things into one space
    • need a translator, etiquette
    • need to be more people in the space who can discuss it
    • importance of “baby steps” people trying things differently seeing in different modes and such
    • chris – totally different creatures depending on size
    • people who came and just said – “tell me what to do” (dependent on the ethos)
    • nitin – hackathons aren’t one singular beast, they are platforms for different action agendas
    • ethics of censorship in the commons
    • VJ eastern sense of using your body, vs western idea of anonymity and fear
    • willow – always question why you are gathering data, because that itself is a risk – surveying also makes people alienated and quiet – trust networks, never give anyone a password to a database you don’t trust, and you probably can’t do it at a hackathon
    • hackdaymanifesto.com
    •  need for education is huge / education is really hard tho
    • agenda for what we as a group wanted to do in the long term
    • AmplifyGood – enabling small businesses to get social marketing PR stuff – learn and do days
    • making a data commons superstormresearch
    • “narcism of small difference makes so many silos”
    • align goals and share resources – is that a totalizing approach?
    • rethink the framework of hacakthons from a feminist perspective and how we know each other and see each other via the body
    • VJ Mm Emma –  VJUMEM.com Laila r-shief.org