Your Ideas for Birds of a Feather Sessions?

To foster relationships among people doing similar work we’ve made time and arranged spaces for 9 total Birds of a Feather (BoF) sessions.They will be held during breakfast and lunch, so you can grab a meal and head to your preferred BoF room.

Some topics seem obvious to us. For example, we will definitely set aside rooms for Life Sciences, Financial Services, Education, and Business Operations Analytics BoFs (the latter will include sales & marketing, human resources, product science, customer support, IT and similar departmental analytics). However, BoFs don’t have to be aligned by industry or type of data.

What exactly is a Birds of a Feather Session? As we see it, a BoF is just a short unconference within a conference, organized or left un-organized (mostly) by participants! Topics may be narrow or broad. Some may have agendas and others may be purely for networking. At a minimum, each room will have a friendly RStudio proctor, chairs, a screen to present, and flipcharts for those who are inspired to create discussion sub-groups, share material broadly, or collaborate.

What Birds of a Feather sessions would you like to attend?
We'd like to encourage y'all to start proposing and upvoting in replies below!

Once the BoF session topics are decided, we’ll load them into our mobile app for the conference (under construction - stay tuned). This, along with, will allow for Pre-BoF discussions so you can hit the ground running in San Diego!.

Some folks I think will be interested:
@kmprioli, @thomas, @taras, @daattali

Thanks, @EconomiCurtis!

I propose outcomes research - my field is Health Economics and Outcomes Research (HEOR). I don't know if many people will be in the HEOR space but I suspect outcomes research may be popular.


I'd be interested in a session that is something along the lines of geospatial data analysis and (interactive) visualisation.


I've been thinking about it for a bit.
What I've heard so far is 3 different ways to categorize BOF sessions.

  • By industry (what your company actually does):

    • Transportation
    • IT
    • Humanitarian relief
    • Retail
      ...and so on
  • By profession / occupation (what you actually do for the company):

    • Analysts
    • IT
    • Management / Leadership
      ... and so on
  • By R interest:

    • Geospatial analysis
    • interactive viz
    • retrospective analytics
    • predictive analytics
    • migrating to R
      ...and so on.

Now, the latter group is already covered by the conference. I.e. you attend sessions that interest you.

So, we're left with either industry segmentation and occupation segmentation.

  • The latter one (occupation) is easier to organize as there is a finite amount of different occupations, but you are getting there at the expense of the industry insight.
  • The former one (industry) would be harder to segment and group properly (good luck: there is an infinite amount of industries). And once you group different specific industries into a larger group (say, put logistics, supply chain, transportation and moving into one group) you again lose that industry insight to a degree.

I guess, segmenting BOFs by occupation would be the most rational choice. IT folk in one group, analysts and data scientists in another group, team lead, management and senior leadership - yet another group, educators and consultants - one more group. And so on. Problems must be similar across industries and some cool stuff can materialize from these discussions.


I suggest a Civic/Public data BoaF session.


A session for Non-profit/NGO workers?


A session for data-as-art people, accidental art,

A session for "the bizarre, unexpected and un-publishable".


What I would like to see how the data in silos in different depts within a govt. agency or across agencies are/can be shared for decision making. Also, how are agencies handling large streams of data from connected devices and what are they doing to overcome sampling biases in such datasets.

1 Like

Medical/Health related?


Two BoF sessions I would love to see:

  • Software and package development with R and RStudio.
  • Data product development: web applications, static websites, report generation, web APIs, books, (etc) all created with R and RStudio.

Have you tried the heemod package? it looks pretty nice, though I have not taken on a project with it.

1 Like

I haven't, but it looks really interesting! I typically use TreeAge for decision analysis and Markov modeling, but would love the chance try these things in R - unfortunately, TreeAge is what my boss is comfortable with, so it would be an uphill battle for me to move away from it!

I have mostly used TreeAge, but I only do HEOR about once every 3 years or so, when a topic is really compelling (hard to publish HEOR in my field). Kinda annoying to have to re-up the TreeAge license. I like the (many) heemod vignettes, and the ability to customize graphic outputs with ggplot. It might be fun (if you can find the time) to do an analysis in TreeAge and in parallel in heemod. Antoine is very responsive to email and willing to help newbies to heemod.

I would like to see a BoaF session around the use of agile in data science

I really like these ideas. I think they're great examples of things that get touched upon in various conference sessions (not necessarily rstudio::conf), but are often in specific contexts (e.g. individual projects or domains). I think they're also areas that involve coordination within teams and across skill levels, and it could be really helpful to have a session with a diversity of POVs on the development/release/adoption/modification lifecycle.

Best practices for an integrated (well-centered) R system:

  • use of R packages to box project analysis
  • use of git / github (public and/or private) for version control, back-up and to allow the contribution (both from the rest of the world whe repository are public and from colleagues even in case of private repository)
  • use of Rmarkdown for documented and reproducible analyses
  • use of cloud services and related clients (ownCloud, Dropbox, iCloud, OneDrive, GoogleDrive, ...) to be free to work at work, at home, on PC, on laptop, on mac, on different system with the same domain account, ...
  • use of simple rules to organize folders both to manage different projects and to manage different aspects of the same project (analysis, reports, papers, presentations, bibliography, ...)

The main purpose is not to learn how to git / github, packages, Rmarkdown or cloud clients work but how to put together all these beautiful things to work and organise everything in a perfect R-world ... leaving your colleagues free to collaborate following your own rules and procedures but also allowing them to not follow those rules without brake all your work (to convince peers or boss to change their habits is not always possible or simple, although you can have the opportunity to organize work as you like)


I would be keen to attend and participate in a session discussing challenges and solutions around open data and reproducible workflows for the government sector. Could be a part of a 'sector' BoaF session rather than a stand-alone?


Current Challenges in Reproducible Workflows

Some example issues:

  • data versioning (e.g. how to capture source data without creating duplicates, extra challenges when data not open licenced)
  • software versioning (e.g. R version, package versions, OS etc)
  • best practices for documentation of finished projects (SessionInfo + package versions + data?)

How about a session for people who work in analyzing risk / integrity / fraud that's not necessarily of a financial nature?

Or one for people who run community meet ups and the good they can do for cities / local partners?