Home » Blog

Seeking advice and resources on Open Notebook Science

13 December 2007 12 Comments

The following comment was posted to the ‘About‘ page by Sharon Sonenblum from Georgia Tech. Rather than leave it there where people might not see it I thought I would bring it to the front for everyone’s attention.

‘I’m looking for some resources or direction for diving into open notebook science. I have been interested in the concept for quite some time and recently began following this blog and a few others. I am excited to see that ONS is real and growing, but I’m not sure the best places to start. I want to find out what other folks are doing, what software they are using and what has and has not worked. I also would love to chat with anyone doing research with human subjects to figure out how IRB restrictions play out in ONS.’

Hi Sharon, great to see people interested in ONS! I am sure others will offer comments and suggestions but I will put my tuppence in first. My main suggestion would be to dive in and see what works for you, within the limitations of what you can do. Depending on the kind of work you are doing and how you are already recording it there are a range of options. As I mentioned in yesterday’s post there are as many different approaches to ONS as there are people doing it. We are definitely at the stage of exploring what is possible, what works, and there is plenty of discussion and indeed disagreement over what the best approach is.

There are really two places you could start. The easiest, and possibly the safest way to dip your toes into the water, is to start up a blog that discusses your lab work in general. There are good examples of this kind of approach with Rosie Redfield’s lab being one of the main proponents (see also Michael Barton’s blog). This can be, but is not necessarily, Open Notebook Science as defined by Jean-Claude Bradley. From what you say there may be real issues with you making your primary data available. If it involves human subjects then I would imagine it will be very difficult, if not impossible, to make the raw data available due to ethical considerations. Certainly I would expect that any review board would require that any data that was released was anonymised and that subjects understood exactly what the release conditions would be. I am no expert in ethics and we don’t (as far as I know) have anyone in the ONS community who is dealing with either human or animal subjects. This is an area that I think is important and that we have yet to explore in detail; if we believe that some science (say chemistry) should be fully open but that some (e.g small scale drug trials) cannot be then can we draw clear boundaries? I don’t know the answer but clearly some care is required with this.

If you can get clearance to go fully to ONS then there are a range of options. I would say it depends a lot on what sort of data you are dealing with. Take a look at your existing lab book and see what it looks like. Is it an electronic document already? Could you simply put that online? Is it an index to a set of data files, spreadsheets, graphs, analysis? If so a Wiki may be the best approach and using a free hosted service, either Wikispaces as used by UsefulChem, or OpenWetWare, could be a good option. Here you can add data files and then add pages that describe, and index them, as well as pages for analysing and discussing the results. Is your lab book more of a journal? Then a Blog may be the best approach, although you need to be careful here about date stamps as many blog engines allow you to change the datestamp. We use an in house developed blog at Southampton that gets around some of these problems but this is definitely an alpha to beta stage product.

Finally, make sure you discuss it with the people around you. Many scientists are deeply uncomfortable with the whole idea of making the lab notebook available. Be sure that you understand and take into account any concerns. In some cases they may not be valid concerns but as with anything there are real risks with the open notebook approach. Take the opportunity to understand any concerns and be prepared to argue where you think they are unjustified, but in a constructive way. Hopefully you can find good discussion points on this blog, at UsefulChem, Open Reading Frames (see also Bill’s excellent three part series at 3 Quark’s Daily), petermr’s blog, Jeremiah Faith’s blog, Michael Barton’s blog, What you’re doing is rather desperate, Public Ramblings, BBGM…who have I missed?

Good luck and keep us updated! The best thing about ONS is the conversations that can get started.


  • Cameron you make all the key points I think. The bottom line is to just start with something simple and ethically possible. I like to recommend the Wikispaces/Blogger system because it uses completely free and hosted solutions. For details about how these interconnect watch this recent talk I gave about Open Notebook Science and I would be happy to help scientists set up.

  • Cameron you make all the key points I think. The bottom line is to just start with something simple and ethically possible. I like to recommend the Wikispaces/Blogger system because it uses completely free and hosted solutions. For details about how these interconnect watch this recent talk I gave about Open Notebook Science and I would be happy to help scientists set up.

  • Yes, I think you’ve pretty much covered it. I’d just put in a plug for the articles I think are the best summaries thus far of what we’re trying to do and how we’re trying to define it =>

    Bill’s three articles (which you mention above):
    http://3quarksdaily.blogs.com/3quarksdaily/2006/10/the_future_of_s_1.html
    http://3quarksdaily.blogs.com/3quarksdaily/2006/11/the_future_of_s.html
    http://3quarksdaily.blogs.com/3quarksdaily/2007/01/the_future_of_s.html

    and I also think Jean-Claude’s recent post is an important clarification of ONS:
    http://usefulchem.blogspot.com/2007/10/science-is-about-mistrust.html

    As far as human subjects go, I should be working with mouse and human data starting in March, so I could provide some information after that. But if Sharon goes there first, I’ll watch and learn.

  • Yes, I think you’ve pretty much covered it. I’d just put in a plug for the articles I think are the best summaries thus far of what we’re trying to do and how we’re trying to define it =>

    Bill’s three articles (which you mention above):
    http://3quarksdaily.blogs.com/3quarksdaily/2006/10/the_future_of_s_1.html
    http://3quarksdaily.blogs.com/3quarksdaily/2006/11/the_future_of_s.html
    http://3quarksdaily.blogs.com/3quarksdaily/2007/01/the_future_of_s.html

    and I also think Jean-Claude’s recent post is an important clarification of ONS:
    http://usefulchem.blogspot.com/2007/10/science-is-about-mistrust.html

    As far as human subjects go, I should be working with mouse and human data starting in March, so I could provide some information after that. But if Sharon goes there first, I’ll watch and learn.

  • Yup, that about covers it. The only thing I’d add is that you don’t have to do *everything* completely in the open. You can have a regular lab for the most part, and just make a subset of your work Open — at least initially.

  • Yup, that about covers it. The only thing I’d add is that you don’t have to do *everything* completely in the open. You can have a regular lab for the most part, and just make a subset of your work Open — at least initially.

  • Yes, just to reinforce Jean-Claude and Bill’s comment. Starting with something simple is a very good idea. Wikispaces/Blogger is a good place to start. Just having a play to see what the limitations are for you and how you work is very valuable. Once you’ve identified how any given system limits you it is time to think about what the best system for you would look like and to think about choosing from the available options.

    And please blog/record the experience of getting these things set up. I know there are some social scientists interested in following these processes so having a record is really good. For instance if you follow back through the different recorded talks that Jean-Claude has put up over the last few years you can see how different issues have come to the fore at different times. J-C, is there an index of all the talks somewhere? Perhaps we should put one up somewhere?

  • Yes, just to reinforce Jean-Claude and Bill’s comment. Starting with something simple is a very good idea. Wikispaces/Blogger is a good place to start. Just having a play to see what the limitations are for you and how you work is very valuable. Once you’ve identified how any given system limits you it is time to think about what the best system for you would look like and to think about choosing from the available options.

    And please blog/record the experience of getting these things set up. I know there are some social scientists interested in following these processes so having a record is really good. For instance if you follow back through the different recorded talks that Jean-Claude has put up over the last few years you can see how different issues have come to the fore at different times. J-C, is there an index of all the talks somewhere? Perhaps we should put one up somewhere?

  • Be wary of your university’s policy on intellectual property though. I’ve run into a few problems where I’ve been asked if I own the research or does the university. This leads to questions about what I am doing with material that is not mine.

    I’m still putting my research online, but I try not to rub it the face of people who wouldn’t approve. Guerilla open notebook science?

    Finally, one other point, you can’t patent anything you disclose online, so this is worth keeping in mind if you discover something which you think might have have commercial value.

    Apart from that I think Cameron has covered everything really well.

  • Be wary of your university’s policy on intellectual property though. I’ve run into a few problems where I’ve been asked if I own the research or does the university. This leads to questions about what I am doing with material that is not mine.

    I’m still putting my research online, but I try not to rub it the face of people who wouldn’t approve. Guerilla open notebook science?

    Finally, one other point, you can’t patent anything you disclose online, so this is worth keeping in mind if you discover something which you think might have have commercial value.

    Apart from that I think Cameron has covered everything really well.

  • Cameron, as for a list of all the talk see my podcast blog:
    http://drexel-coas-talks-mp3-podcast.blogspot.com/

    The last talk I gave at Swarthmore is probably good enough to see the history of how UsefulChem evolved. There are also talks on Open Education in there but many are on Open Notebook Science. Your talk at Drexel is there too.

    There are transcripts for many of the talks – sometimes it is nice to just scan the content of a talk instead of listening to the whole thing. Cameron – the transcript to yours should be done shortly.

  • Cameron, as for a list of all the talk see my podcast blog:
    http://drexel-coas-talks-mp3-podcast.blogspot.com/

    The last talk I gave at Swarthmore is probably good enough to see the history of how UsefulChem evolved. There are also talks on Open Education in there but many are on Open Notebook Science. Your talk at Drexel is there too.

    There are transcripts for many of the talks – sometimes it is nice to just scan the content of a talk instead of listening to the whole thing. Cameron – the transcript to yours should be done shortly.