Putting Open Science into Practice at the Barcamp Science 2.0 #s20bar

OSR_Sonderlogo_2016_s20bar_iTunesAs you might have recognized, we’ve attended the Barcamp Science 2.0 and the first day of the Science 2.0 Conference on Monday and Tuesday this week. Background was, that Konrad and me (or: the Open Science Radio) has been invited/asked to participate and report about the Barcamp Science 2.0 (and if possible the following conference) in Cologne this year.

Both events are organized under the umbrella of the Leibniz Research Alliance Science 2.0, which also has approached us if we could imagine to accompany mainly the Barcamp with the Open Science Radio. Indeed we could since this year’s barcamp topic was „Putting Science 2.0 and Open Science into practice!“, hence our interest.

After a bit of brainstorming we figured that an introductory podcast before the event and a couple of smaller episodes with a few session hosts directly from the event may be a good way to do it. So we did an introductory podcast with the main organizer (and research coordinator of the Leibniz Research Alliance Science 2.0), Guido Scherp, ahead of the barcamp and 10 episodes (both in English and German) with session moderators directly from the event. We are also still planning to wrap-up the whole barcamp-related episodes with another one with Guido to look back at the event and draw some conclusions.

If you are interested, please head over to the website of the Open Science Radio and give it a listen. The rest of this blogpost will be a short review of the Barcamp Science 2.0 from my very own perspective.

The Barcamp

Before really getting started, let me give you one last preliminary remark: I myself am not a scientist. Quite some years back I have worked for about 5 years within one of the big German research association, specifically within research communication (call it research PR). However, science has always been one of the subjects I was and am interested in. When I was still working within the academia I always asked myself why communication is mostly closed, even though there is some fascinating stuff happening. Over the years I more and more got the impression that not only communication should be more open, but pretty much anything could and should in most cases be more open. That’s pretty much my turn to the whole open science issue(s) put into a few sentences. Anyway, even though I left the academic system and started working in the industry (or private sector, free economy, or whatever you like to call it), my fascination for science and the practices of open science in particular did decrease. It grew even stronger, I got more interested and also more involved within the open science community – up to the point that I launched this blog and the Open Science Radio nearly 4 years ago. Even in my current job, which is somewhat still related to academia (especially with regards to learning), I am increasingly experiencing the advantages of open science. In my case that mainly is being able to access scientific literature freely. To cut a long story short, my views are often a mix of both worlds – my insights I gained from within science, but growingly also my insights from a company that might benefit from a growing openness of science. Maybe this explains my argumentation at some point. 😉

If you are not familiar with the event format of the barcamp you might want to read a short article about it. But in general it is an unconference format, which means that the content (presentations, discussions, etc.) is planned mainly in the beginning of the event itself.

Unfortunately I was only able to attend a few sessions since we also had to concentrate on recording, editing and publishing of the Open Science Radio episodes, but I actually was able to attend a few sessions and others have probably been equally filled with intense discussions, nice chats and useful insights. Because that’s what it has mainly been for me – the diverse crowd coming from various scientific disciplines, various scientific stakeholders and even from non-scientific institutions (and companies as in my case) has provided an interesting mix of views, interests and objectives. That absolutely fostered interesting discussions and made the whole barcamp a generally very enjoyable experience.

Hence, the offered sessions have been equally diverse within the open science context. Luckily the barcamp organizers had also set up a pad for each session so that you can have a look inside and see what has been discusses in each session. Here’s the sessions with their respective pads linked:

11:30 – 12:15
Session 1: Incentives for Open Science
Session 2: Preregistration for publications
Session 3: Jupyter Notebooks

12:30 – 13:15
Session 1: Is Open Science bad for Science?
Session 2: Peer Review
Session 3: Wikipedia & Wikidata as a workbench

14:15 – 15:00
Session 1: Tools for Open Science
Session 2: Infrastructure for Open Science
Session 3: Package Management for research projects
15:15 – 16:00
Session 1: Data formats for Open Science
Session 2: Teaching Open Science
Session 3: SciHub good or bad

16:30 – 17:15
Session 1: Practicalities of data sharing
Session 2: Analyzing scholarly tweets
Session 3: Structuring research and publications

I don’t really want to give a summary of what has been discussed within the session I’ve attended. On the one hand I won’t be able to recall all the strands of argumentation and on the other hand the session pads can provide this much better. But I want to discuss a few things – a few aspects from sessions, as well as a few general remarks about the barcamp, the topic or related stuff.

First of all a general remark. Indeed I wished that there would be more interest of the private sector (e.g. of those so highly-valued SME’s all the funding calls refer to). I think the perspective of the „business“ on a topic such as open science can be a valuable addition to such an event. Let me explain. I think one argument of open science, and a drive for science in general, has always been that an increasing openness will foster an increasing knowledge transfer from science into the economy, causing a higher innovation potential and finally leading to more innovation. Or seen from the other side (and from my own experience): a more open academia will increasingly become a more reliable and used source of either a) information to help building businesses and products (journal articles, reports, etc.), b) data to build products (open data, etc.), or c) talents (HR). But apart from that rather abstract level, there is a growing number of businesses build around scientific infrastructure and targeted to the scientific world, offering services to ease quite different problems within science on many levels – starting from (collaborative) article writing, to the issue of metrics, to „workbench-related“ services such as web-based lab notebooks and workflow description services. At this year’s barcamp Xpansa was a nice example. Roman of Xpansa used the opportunity of a barcamp session to introduce their new product Sci.AI to the audience and collect some feedback. Perfect match. I am not saying that we should turn the barcamp into an advertising event, but I think that the business side has also something to offer in such events.

What more? This year’s ignition talk (some what say keynote, but for me such a presentation should ignite some discussion, hence the name) was given by Felix Schönbrodt, post-doc at the LMU Munich and also involved in a „taskforce open science“ (my formulation) at the university. From the program:

Putting Open Science into practice: Simple but powerful steps for researchers, universities, and journals
Even if I embrace the values of research transparency – where should I begin? How can I move my institution a little towards openness? This short talk highlights some recent developments and practical examples that enable researchers and institutions to put Open Science into practice.

As I’ve already said in the first Open Science Radio episode regarding the barcamp, I really liked that Felix brought „simple“ into the title of his talk, because that is what we need to provide. Open science is quite a complex area of different topics along the research process. This raises the entry barrier, so if we wanna convince people to inhale and put open science into practice we need to lower the barrier and show them simple steps first. Simple, but as Felix also said, powerful. Felix also stressed that we need to leave our open science bubble. Of course, once you’re interested in a certain topic, we tend to surround ourselves with like-minded people. This is necessary to find and form a community. But it holds also the danger to stay within it and get stuck in a „preaching to the converted“ state. This is for sure nice, but it will hardly change something. As Felix said, we need to get out of the bubble, turn to those who haven’t met open scientists yet, who haven’t much encountered this fascinating subject. We need to address those people at their events in order to get the idea spread.

One of the interesting question to Felix was what price he paid for his openness. If I understood him correctly, his answer was tow-fold. On the one hand he invests quite some time to do his research openly, so due to this „sacrificed“ time he might not be as effective as he could be if he would use all his time only for the classical approach of doing it. That might of course result in e.g. something like publishing less papers in the long run. But the more interesting part of the answer refers to the indirect concept of the career path. Felix stated that he didn’t just try to „master the game“ to make progress on the academic career ladder and that maybe his openness has cost him one or the other career chance. Actually this „mastering the game“ aspect he mentioned is something that I find absolutely intriguing. I don’t wanna go into too much detail at this point because in this respect I am lacking experience from insight this system. But I very much hope (and Konrad too) that we might get the chance to speak to Felix again and in more detail in another OSR episode.

I’ve also attended two sessions regarding the more infrastructural aspect of open science. With plans for an European Open Science Cloud (that are still rather unclear to most extent), this is a hot topic at the moment that was also mainly addressed at the Science 2.0 conference. However we constantly came back to the same points in our discussions: if we want to publish and openly share scientific data we need to have a infrastructure that is trustworthy and long-term available, but also fundable. And the modes of funding are not fixed yet. There are different options on the table, but there is not yet a common agreement. However it seems that the position of a dedicated research data manager seems to be increasingly important. Of course this is something that libraries could partly take over, but data and methods are so diverse and different that data managers might need to be implemented on lower levels (disciplines, research lines or even smaller). Furthermore, publishing data is not enough. To foster reproducibility, re-use, etc. we need to publish the tools for data analysis, methods and workflows. This needs to be regarded when planning a research infrastructure and some of these aspects still have a major question mark over them. Listeners of the Open Science Radio already know it, but infrastructure is such a broad topic that we usually pretty much fail to shed light on it entirely. But we hope to have a guest in one of the upcoming shows who can provide an overview.

One of the other sessions I attended was the session on teaching for open science, moderated by Andreas Leimbach. This session was quite interesting because most (or even all?) of the participants actually see a need for a thorough open science education for either fellow researchers, or as early on as students. There are quite some approaches how to do it, but it seems that nobody has actually a clue for a „golden way“, if something like that exists. This seems to be particularly difficult with students who are hard to motivate besides forcing them with marks. It actually came down to a question that also came up in other sessions: what incentives to do open science can we offer? It’s the old „What’s innit for me?“ question. This question can be answered on thousand different ways depending on the individual situation and it can even be answered to a certain extent on a general level. But we still seem to lack a systematic answer to it; an answer that we can formulate as a motivation for those who we want (and need) to convince to do open science. Especially since many of the incentives are rather in the long-term and even a bit abstract. Many of the incentives (advantages of open science that is) will only be fully effective once many researchers turn to open science. Because only with a critical mass of researchers the positive network effects and increased efficiency of science (e.g. through re-use of existing data) will be significant. So how do we motivate researchers (and future researchers) NOW to do open science where most of the system is not yet recognizing these efforts, of even worse punishing them by disregarding open scientists in career decisions.

Introducing open science early on in university with students certainly is a way we need to go. But even there it is hard to give it enough room. A single or even two lectures are probably not enough to introduce this wide topic and to connect all the various facetted aspects of it. And how do we teach it best? Andreas brought up Software Carpentry as a good example. Yes it is, but with SC the case is a bit different. It focuses on teaching how and with what means can I do computing within my specific subject. Hence, it rather a specific approach aiming to solve a certain problem in a rather hands-on manner. With teaching open science the case will be a bit different in my opinion. Of course, there are hands-on aspects and tools are a major aspect of it too. But there are also quite a number of rather complex non-technical issues such as legal aspects, economic aspects, maybe even systems theoretical aspects. Maybe an open science basic education needs to be much more of a mix of theory, best practices, hands-on tools, etc. This definitely needs more thought and much more community effort – in this respect, Software Carpentry is a good example. By the way – we had an interview episode with Greg Wilson (co-founder) of the Software Carpentry in episode 38 ofd the Open Science Radio.

There is one last aspect I want to mention, actually coming back to the general barcamp format. This year’s Barcamp Science 2.0 was the 2nd edition (Barcamp 2.0 2015) and it probably won’t be the last one. One of the drawback is, that often with recurring events with changing audience (at least a bit) one is stuck in a topic loop. Of course, projects evolve, new technologies arise, stories happen (as with Sci-Hub this year). But there is always the danger that we discuss the same general aspects, the same questions over and over again. So maybe there is a way how we could turn this and also evolve in our discussion through paying attention to last year’s findings? Without referring to something like a „barcamp proceedings“ or something like that, but maybe there is a way to build are discussions on the foundations of last year’s.

And connected to this: such events will only have an effect if discussions and arguments turn into activites. So it would be great to see some activities arise from the barcamp, or from ideas we’ve discussed there.

The Role of the Open Science Radio

I want to come back to the beginning of this lengthy blogpost and finish it by referring again to the Open Science Radio. For us as the Open Science Radio this was an experiment since we did not accompany any event in such a manner. The Open Science radio is a fully voluntary project, we have no commercial interest. All donations by listeners through Flattr are usually re-donated (last year we donated equally to Wikimedia and Open Knowledge). Hence we also did not charge any remuneration for reporting from the barcamp (disclaimer: our travel and accommodation costs have been taken over by the research alliance).

I think, all in all, this was a successfull experiment with a number of episodes. Our aim was to provide a quick insight into the event while it still takes place. Hopefully our episodes together with the tweets provide exactly that – an informative overview of the 2016 Barcamp 2.0. However we quickly recognized that accompanying an event of this type is not an easy thing to do if you want to a) follow the activtities and discussion so you can relate to them and b) participate in the sessions and actively take part in discussions. This „dual role“ certainly made it more difficult for us. It would have been much easier if we could juts concentrate on one, but honestly we are just too interested in the topic to not go full on. 😉 Also the relative short time between session (15min were planned , but with a few delays this time shrinked, you know the reality) made it quite difficult to speak to session moderators or participants without holding them back too long from the next session. By the way, as mentioned in the beginning, we also participated in the first day of the Science 2.0 Conference (follow the tweets of #sci20conf). Just in case you wonder why there are no episodes from the conference: well, a conference is something totally different. Since it is even more stringent so that it wouldn’t have been possible to get interviews after presentations in the same way as at the barcamp. But we will soon do an episode on the conference to share our experience.

For the very last, I want to do 2 things. First of all, I want to thank the Leibniz Research Alliance (especially Guido Scherp) for inviting us and giving us the opportunity to meet and talk to so many great people! It was our pleasure! And finally I want to ask a few short questions in particular (but not exclusive) to our regular Open Science Radio listeners. It would be great if you could share your opinion in the comments:

  1. How did you like this experiment and publishing the event-related episodes within the normal Open Science Radio feed?
  2. Did you think that these episodes have been labelled clearly enough (with the OSR logo version including the #s20bar hashtag and a special episode image)?
  3. Do you think we should repeat reporting from events (I’m actually not thinking about 10 times a year, but rather 1 or 2 times)?

Thank you very much for your attention! You’re invited to discuss or share any insights in the comments!

Cheers,

Matthias