UPDATE: Proposal now being developed after fantastic community support. Check out satRdays on GitHub and contribute your opinions!

I had a contact from a very nice chap in Dallas a month ago about whether in the R world we do anything like SQLSaturdays.

The great thing about the SQLSaturdays he said was not that they’re free (well it helps!) but that they’re on his time. Developing his skills was something he couldn’t get signed off by his boss so he wanted to be able to do it by himself.

In answer to the question of whether there are local(ish) weekend conferences happening regularly for R, my answer was “not really” and it’s a shame because the R community is fantastic. I started thinking about why we don’t have them and what would be needed to change that.

Free / cheap regional small-medium conferences are a must for growing user knowledge and speakers in R.

What are SQLSaturdays?

SQLSaturdays are free, SQL Server related conferences held around the world. The SQLSaturdays focus on developing technical knowledge and skill with talks at a range of levels. The first SQLSaturday was in 2007, and in 2015 it’ll have grown to 115 events across the globe. That’s a fantastic achievement, especially as we’re generally seeing the SQL events expand into Business Intelligence and Analytics.

They’re run by local communities to grow support and attendance for their user groups, but they’re supported globally by PASS who provide the brand, the amazingly helpful Karla, and the digital infrastructure needed.

Who are PASS?

PASS was set up in 1999 by Microsoft and CA Technologies to help promote and educate users of SQL Server around the world. PASS run the PASS Summit, a much larger paid-for conference that acts as a fund-raising vehicle for the support of PASS’ infrastructure, SQLSaturdays, and the Virtual Chapters. The way PASS operate the events to help grow the user base and the quality of knowledge within the SQL community, provide a great example that the new R Consortium could learn from – especially given Microsoft are key partners of the group. PASS operate with some salaried employees but for the most part this massive organisation is staffed with volunteers – it’s not perfect but they do a great of supporting learning and the community.

What’s the problem for doing them for R?

If they’re so awesome, why aren’t we already doing them? There are a few main reasons:

  • There’s been not been a central association / organisation with this in their remit
  • The user base for R is much smaller than SQL Server was, even back in 2007, let alone today
  • A corollary of fewer users is that there are fewer speakers. This makes it harder to get 5 to 10 good quality speakers for a conference in Cardiff or Albuquerque for instance
  • The academic roots of R means that almost R conferences happen during the week* and potentially there could be a lot of pushback to weekend events
  • There may not be enough companies deriving benefits from the R language and who are willing to pay to grow the user base and brand awareness

What it would take to get them going for R?

In terms of what it would take to do these, the answer is “not much”. Doing a minimal replication would entail:

  1. A group of people who can produce a central platform
  2. Some pilot user groups out there willing to organise a conference
  3. A few good sponsors who can support the first conferences
  4. A group of people willing to travel to speak and/or attend
  5. A dab hand at graphic design to come up with a neat logo

What next?

I think R Saturdays could be a great way to make R more accessible. Community events don’t happen because of just one person though, it can literally be a cast of thousands when you include the attendees, so more folks are required to make it happen – so we’ll need some willing people to help tackle this.

It’s theoretically within the remit of the R Consortium to provide a central platform for these events. It’d additionally support the big useR! conferences too, so this could be an excellent proposal to put in front in of the Infrastructure Committee. I raised it at the Panel Discussion at EARL 2015 Boston and the Consortium are keen.

I intend to submit a proposal, but before I do, I’d really love to have some discussion on the concept and potential pitfalls people might see. Please either discuss below or tweet with #RSaturday to @SteffLocke.

 * I'm aware there's some "Citation Required" - if you have evidence to the contrary, please give a heads up!
SQLSaturdays but for R?
Tagged on:                         

15 thoughts on “SQLSaturdays but for R?

  • 04/11/2015 at 20:05


    This is the model that Security B-Sides —http://www.securitybsides.com/w/page/12194156/FrontPage — uses to help promote cybersecurity awareness and provide a forum for practitioners and researchers (and those with a morbid curiosity for cyber 😉 to come together, engage in some talks and rock the hallwaycon. I’m 100% for this and would be glad to help promote R in this fashion. Where do I sign up?!

    • 04/11/2015 at 20:18

      You’re already signed up FOREVER 😉 Thanks for the support and it’s great hearing about how other communities are doing it. Means we can incorporate lots of best practice!

      I’m looking up hallwaycons now – sounds awesome, and cheap

  • 04/11/2015 at 20:34

    This is a fantastic idea and something I’d love to organize in Hungary, or more generally in Eastern Europe, or in Los Angeles. Do you plan to submit an official proposal to the R Consortium? Please let me know how can I contribute — until then, I keep thinking about the next practical steps on who to contact for good talks, venue and some sponsorship in Hungary/Los Angeles.

    PS: I hope this post will appear on R-bloggers. If not, it should 🙂

    • 04/11/2015 at 20:58

      Yep – an R Consortium Proposal is the plan but I wanted to see whether folks liked the SQLSaturday model, what else is out there, and who wanted to help run with it. So keep thinking and I’ll be making the draft proposal on github when I get back from EARL.

      I’m not on r-bloggers at the mo. I submitted ages ago but never heard anything about it. I’ve resubmitted so fingers crossed. If you’re on r-bloggers, it’d be great if you could do a little post and refer people here.

      • 04/11/2015 at 21:55

        Cool, looking forward to follow/contribute to the GH repo when it’s ready.

        About R-bloggers: to avoid confusion, I do not want to cross-post, and I think Tal Galili will be happy to add your blog and this post to R-bloggers. Pinging him on Twitter, just in case 🙂

  • 04/11/2015 at 21:31

    Being far from your expertise level in R (until now I’m more of a cadet pirate), but I would throw my hat in to participate. I would always like to attend such events, have no clue how to organize one, but I might even submit to speak, once I reach that level; and I could make it to Cardiff almost any time, Albuquerque is another question… 😉

    First thought that came to my mind though: with Microsoft integrating R into the SQL Server environment, should you maybe start off by establishing R tracks into SQL Saturdays, and then maybe spinning them off, if and when the participation and feedback speaks for it?

    • 04/11/2015 at 21:47

      I considered trying to gel the two but there’s a lot of R that won’t happen in SQL Server and vice versa. I think there’s a lot of value in dual running, like some of the sqlsats & sharepoint Saturdays do. It’ll definitely help spring board them 🙂

  • 05/11/2015 at 13:44

    How is this different from the R User groups?

    • 05/11/2015 at 15:56

      Hi Adnan – good question!

      So user groups are typically evening affairs, during the week, with a few speakers. This means that it can be difficult to attend for people with long commutes, hectic family lives etc. Then there’s a large dimensionality issue with the audience – each person will have certain interests and a certain level of experience with R – this makes providing a balanced set of sessions across the year really difficult to achieve.

      User groups are still excellent things, but running a (small) conference gives the user group the facility to greatly increase the breadth of topics & levels that their attendees have access to. They also serve a really handy purpose of raising awareness about the user group – a big campaign about a conference is likely to reach more ears than the standard marketing around your user group.

      The SQLSats, and thus the proposed RSats, are a great way to give content but they’re also excellent opportunities to do a few other things that a standard user group typically won’t be able to facilitate:
      – Meet great vendors: it can be tough convincing someone like Revo or Plot.ly to come along to a user group but to a conference it’s much easier. This gives the UG members opportunities to talk to people who might be able to help with challenges they’re facing
      – Increase the speaker pool: getting speakers can be really tough, and with only a few at anyone time you really want them to be the best possible but we need more speakers and they need more opportunities to practice and develop
      – Possibly fundraise: some UGs add a “pre-con” day of paid training workshops that can contribute not only to the cost of running the RSat but also fund the UG through the rest of the year
      – Global recognition: some SQLSats are way more desirable than others to attend / present / sponsor and that’s because of how much awesome work the organisers put in. These events have a lot of excitement around them and can help make your UG renowned

      Hope this outlines some of the key differences from my perspective 🙂

  • 05/11/2015 at 16:16

    I think this is a fantastic idea!

    Question from my experience developing and teaching an R-flavored Software Carpentry: Do you plan this to be like a conference, where each speaker does his/her own thing, or more like a semi-standard workshop, where there is some core content an attendee can always count on?

    SWC is most definitely the latter, which has huge value for the speakers/instructors and the attendees. But deciding what’s in and what’s out presents a juicy opportunity for long email and GitHub threads that never really resolve. Debating whether to use the Hadleyverse or the pipe operator or base graphics etc etc etc. But you wouldn’t want all the talks to be about astronaut, bleeding edge stuff, if you’re really trying to reach the unconverted. So these basic questions of “how I use R” will come up.

    How to give attendees really useful and fundamental information … w/o getting sucked into the usual debates?

    Just gathering a few experts and helping people get an entire R stack installed and working, e.g. version control and a build environment on Windows, would be a great community service and nice component of an RSaturday.

    • 05/11/2015 at 16:42

      I hadn’t given too much thought to the specific content, but it’s an excellent point 😀 Most SQLSats run a relatively typical conference format but it’s non-mandatory and people like to shake it up.

      I LOVE the idea of a standard beginner workshop that could be dropped into the format though – as you say it’d add tremendous value for new people and if we had some sort of standard material it’d be a great resource not just for the confs but for the Consortium & community more widely. It’s valuable enough that I think it’d be a worthwhile project in it’s own right tbh 🙂

      • 05/11/2015 at 21:14

        OK got it re: standard conference format! I like your idea of basic content that could be dropped in. As you point out, the R user base is smaller than SQL and the experts are even thinner on the ground. But the number of people who are curious or just beginning is quite high. Would be good if RSaturday had sthg for both.

  • 05/11/2015 at 16:19

    Some comments received on twitter etc

  • 09/11/2015 at 11:45

    The proposal repo is now live: stephlocke/RSaturday-poposal

    I’ve added a bunch of questions in the Issues area, please opine and/or add more questions of your own. I see this working like the R Community – distributed, opinionated, & massively productive 🙂

    PS I also created a boilerplate repo so if you’d like to build your own proposal for the ISC, you can crib my structure based off the guidance provided on the ISC page. Be warned though, Hadley thinks it’s a little heavy in terms of info provided so you can strip it back a bit. I’d rather give plenty of thought to (potentially) asking for money from folks though but YMMV.


What do you think?