Past, Present and Future of Open Science (Emergent session): Advocating open science is easy for data consumers

[jsheunis]

Advocating open science is easy for data consumers

By Simon Eickhoff, Forschungszentrum Jülich & Heinrich-Heine Universität Düsseldorf

• Theme: Past, Present and Future of Open Science
• Format: Emergent session

Abstract

Some labs/individuals support and practice open science, others do not. While much of this may be due to individual socialization and believes, there is a deeper pattern

• Traditionally, computational/data scientists, i.e., those that benefit from available datasets tend to be overrepresented in the "open science space"
• In turn, clinicians who spend months if not years recruiting and assessing patients are often much less enthusiastic about open sharing.

The goal of this session is to discuss, wow can this gap be bridged, acknowledging that the advantages of open practices are already well recognized. Education about open science and appeal to moral responsibility seem to have limited effects, though, given the structural imbalance of open data, as well as the difference in incentive structure and career goals among the different populations.

Useful Links

Public Mattermost channel for discussions prior to, during and after the session.

Tagging @SBEickhoff @dr-xenia @pimpul

[derekbeaton]

Is this session going to be recorded/available after?

[demianw]

@jsheunis will the video be available?
Thanks!

[Starborn]

Ha ha- yes, provocative important topic. My quick contribution to this discussion (because my one brain is already on fire) Open Science mandates that the data which support the scientific result in a publication /research is made available for scrutiny evaluation (many mistakes are not catched by reviewers but by others who try to repeat experiments for example) this does not have any negative implication for the clinical researcher, other that they have to show their data Do you think clinical researchers which have nothing to hide should not show what data they have used? We need to transcend duality (ie the polarity of these two extremes examples that you posit) A clinical researcher, needs to accept to disclose the data when they publish research based on it, to enable some level of verification, to enable reproducibility. They need to show consistency (ie that the conclusion follows the result) We see a lot of arbitrary conclusions that do not follow the results, because ultimately the results are not really shown, only the interpreted/cleaned up results are shown. The rest of the data, after clinical scientists have published their work, should be made open access when the researcher has no more use for it - (if not earlier of course) because it would be wasted otherwise Nobody is obliged to embrace OS, only those who feel it is beneficial for their research, for others. Look forward to this discussion PDM

…

On Thu, Jun 25, 2020 at 4:44 PM Demian Wassermann ***@***.***> wrote: @jsheunis <https://github.com/jsheunis> will the video be available? Thanks! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#62 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACFKUCOXMJ43SVIFENQDRJTRYME7HANCNFSM4N2EPY6Q> .

[SBEickhoff]

Dear PDM Ha ha- yes, provocative important topic. My quick contribution to this

discussion (because my one brain is already on fire)

In my view, it's a critical topic. We can talk about infrastructure all day long, if we can't get commitment from people that actually have the data, it won't help. And in my view the only way forward is a give-and-take, not pressure

Open Science mandates that the data which support the scientific result in a publication /research is made available for scrutiny evaluation (many mistakes are not catched by reviewers but by others who try to repeat experiments for example) this does not have any negative implication for the clinical researcher, other that they have to show their data. Do you think clinical researchers which have nothing to hide should not show what data they have used?

There is a big difference between making data available for scrutiny and sharing it for re-use, even though both aspects are often conflated in the current discussions

A clinical researcher, needs to accept to disclose the data when they publish research based on it, to enable some level of verification, to enable reproducibility. They need to show consistency (ie that the conclusion follows the result) We see a lot of arbitrary conclusions that do not follow the results, because ultimately the results are not really shown, only the interpreted/cleaned up results are shown.

I would argue that this is in no way a specific problem of clinical research. Not at all. And, just as a side-note, does not require access to the raw data.

The rest of the data, after clinical scientists have published their work, should be made open access when the researcher has no more use for it - (if not earlier of course) because it would be wasted otherwise

We can all agree that this is desirable, but it brings us back to the starting point of just requiring something from data providers (should) for the benefit of others (data consumers), i.e., a one-way street of benefit.

Nobody is obliged to embrace OS, only those who feel it is beneficial for their research, for others.

I would agree here, though I would see it as highly desirable if more data were to be available through sharing Cheers Simon

…

On Thu, Jun 25, 2020 at 4:44 PM Demian Wassermann ***@***.***> > wrote: > @jsheunis <https://github.com/jsheunis> will the video be available? > Thanks! > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#62 (comment)>, or > unsubscribe > < https://github.com/notifications/unsubscribe-auth/ACFKUCOXMJ43SVIFENQDRJTRYME7HANCNFSM4N2EPY6Q > > . > — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#62 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AE2Y32WX7XIK4Q5VDIURKA3RYMJV5ANCNFSM4N2EPY6Q> .

-- =================================== Univ.-Prof. Dr. med. Simon B. Eickhoff Director, Institute of Systems Neuroscience Heinrich Heine University Düsseldorf 40225 Düsseldorf, Germany Tel: +49 211 586729-127 <0211%20586729127> eMail: [email protected] http://www.uniklinik-duesseldorf.de/systemneurowiss and Director, Institute of Neuroscience and Medicine (INM-7: Brain and Behaviour ) Research Centre Jülich 52425 Jülich, Germany Tel: +49 2461 61 1791 <02461%20611791> eMail: [email protected] http://www.fz-juelich.de/inm/inm-7/EN/Home/home_node.html

[Starborn]

Thank you Simon we seem to agree - we need more depth of discussion in open science topics Data sharing means showing the underlying datasets when research is published, should be mandatory for that only- but it is often presented as something else so that newcomers back away from openness and ebcome intimidated There are lobbies trying to mud the scientific research waters for reasons that we know all too well For example the issue of consent is presented as a massive roadblock to open science It is also important but as someone said, it is used as a blanket objection to transparent practices Consent is important and we need to get it right, but if we handle the matter of consent properly, it may not be an obstacle to open data Same for other related issues. Open Data/Science needs to be better understood, and to some extent it is necessary to overcome the manipulation of research results for whatever overt purposes. Let's continue to talk about this topic! Thanks for opening the thread PDM

…

On Thu, Jun 25, 2020 at 6:08 PM SBEickhoff ***@***.***> wrote: Dear PDM Ha ha- yes, provocative important topic. My quick contribution to this > discussion (because my one brain is already on fire) > In my view, it's a critical topic. We can talk about infrastructure all day long, if we can't get commitment from people that actually have the data, it won't help. And in my view the only way forward is a give-and-take, not pressure > Open Science mandates that the data which support the scientific result > in a publication /research is made available for scrutiny evaluation (many > mistakes are not catched by reviewers but by others who try to repeat > experiments for example) this does not have any negative implication for > the clinical researcher, other that they have to show their data. Do you > think clinical researchers which have nothing to hide should not show what > data they have used? > There is a big difference between making data available for scrutiny and sharing it for re-use, even though both aspects are often conflated in the current discussions > A clinical researcher, needs to accept to disclose the data when > they publish research based on it, to enable some level of verification, to > enable reproducibility. They need to show consistency (ie that the > conclusion follows the result) We see a lot of arbitrary conclusions that > do not follow the results, because ultimately the results are not really > shown, only the interpreted/cleaned up results are shown. > I would argue that this is in no way a specific problem of clinical research. Not at all. And, just as a side-note, does not require access to the raw data. > The rest of the data, after clinical scientists have published their > work, should be made open access when the researcher has no more use for it > - (if not earlier of course) because it would be wasted otherwise > We can all agree that this is desirable, but it brings us back to the starting point of just requiring something from data providers (should) for the benefit of others (data consumers), i.e., a one-way street of benefit. > Nobody is obliged to embrace OS, only those who feel it is beneficial > for their research, for others. > I would agree here, though I would see it as highly desirable if more data were to be available through sharing Cheers Simon On Thu, Jun 25, 2020 at 4:44 PM Demian Wassermann < ***@***.***> > wrote: > > > @jsheunis <https://github.com/jsheunis> will the video be available? > > Thanks! > > > > — > > You are receiving this because you are subscribed to this thread. > > Reply to this email directly, view it on GitHub > > <#62 (comment)>, or > > unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/ACFKUCOXMJ43SVIFENQDRJTRYME7HANCNFSM4N2EPY6Q > > > > . > > > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub > <#62 (comment)>, or > unsubscribe > < https://github.com/notifications/unsubscribe-auth/AE2Y32WX7XIK4Q5VDIURKA3RYMJV5ANCNFSM4N2EPY6Q > > . > -- =================================== Univ.-Prof. Dr. med. Simon B. Eickhoff Director, Institute of Systems Neuroscience Heinrich Heine University Düsseldorf 40225 Düsseldorf, Germany Tel: +49 211 586729-127 <0211%20586729127> eMail: ***@***.*** http://www.uniklinik-duesseldorf.de/systemneurowiss and Director, Institute of Neuroscience and Medicine (INM-7: Brain and Behaviour ) Research Centre Jülich 52425 Jülich, Germany Tel: +49 2461 61 1791 <02461%20611791> eMail: ***@***.*** http://www.fz-juelich.de/inm/inm-7/EN/Home/home_node.html — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#62 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACFKUCPJXYSAHYZSJSPDRFTRYMOZPANCNFSM4N2EPY6Q> .

[derekbeaton]

I've found the video on Crowd Cast! Now my next question is where can we pick up on this discussion? 4am was too early for me and this is a topic I'm very much interested in and involved with (both as a producer and consumer of data)

[jsheunis]

@derekbeaton @demianw yes the recording is available through crowdcast. If you register here for free you will receive the links and passwords to join. You can also join the continued discussion via text-chat on this public channel on the Brainhack Mattermost workspace.