Sharing our work on sensemaking data in science + connections to nanopubs
Ronen Tamari
Hey all,
Just wanted to share a preprint of ours that may be of interest to this community. We presented it last month at the Metascience conference.
TL;DR - we try to make the case that attention/sensemaking data (eg data related to what researchers are attending to and their assessments of content) are an important kind of nano-scientific knowledge that gets extracted by platforms (social media, publishers, science analytics companies) instead of helping to power content curation and discovery networks.
If you’re curious about what sensemaking data looks like, here are some examples from Twitter- this one contains inline annotations and this one a general assessment of an article. We think these these data belong on the “sensemaking layer” of science in an open and FAIR format, and not just on social media.
Anyway, I can see a lot of potential synergies with the nanopubs ecosystem, so I'd be happy to hear any related feedback, thought or comments!
Thanks!
Ronen
Barend Mons
Prof. Barend Mons
VODAN: https://www.go-fair.org/implementation-networks/overview/vodan/
Leiden University Medical Center
President of CODATA
GO FAIR international Support and coordination office
Mail: baren...@go-fair.org
ORCID: 0000-0003-3934-0072
sent from my IPhone
Op 28 mei 2023 om 23:43 heeft Ronen Tamari <ronen....@mail.huji.ac.il> het volgende geschreven:
--
You received this message because you are subscribed to the Google Groups "Nanopublications" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nanopub-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nanopub-users/2f8c6826-8870-479d-bb43-0bd9fdf1b5f4n%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Tobias Kuhn
Thank you for sharing this. This is very interesting, and I believe
there is a lot of overlap with what nanopublications can do. So, I would
find it very interesting to explore further how (part of) this approach
could work when implemented as nanopublications.
I have two observations.
First, it seems your approach doesn't aim to change the primary methods
of publishing research (red parts in Figures 1/2/3), which is of course
perfectly fine. But I think this approach becomes even more powerful
when these red parts become more structured and fine-grained too, so one
can refer to a particular hypothesis, equipment, interpretation, etc.
and not only to a scientific paper as a whole.
Second, you write in 4.2
> Importantly, sensemaking data should be self-
> sovereign, with users controlling the accessibility of their data.
but later in 7.5 you also clarify that
> our proposal is focused on the public FAIR sharing of sensemaking data
and I think that's a very important point. It's also a point in which
your approach seems to differ from nanopublications. Nanopublications,
in most of their use cases so far, are only about public data, with
"public" in a strong non-revocable sense.
Publishing in this sense means that the user no longer controls the
accessibility of the data. This release of control of the publishing
user, in turn, allows for others to depend and reliably build up it. So,
there are not only clear downsides but also clear advantages to *not*
allow the user to control the accessibility after publication.
But these are just observations, which I hope make sense, and not points
of criticism.
Regards,
Tobias
On 28.05.23 21:43, Ronen Tamari wrote:
> Hey all,
>
> Just wanted to share a preprint of ours that may be of interest to this
> community. We presented it last month at the Metascience conference
> Open Access science needs Open Science Sensemaking (OSSm): open
> infrastructure for sharing scientific sensemaking data
> TL;DR - we try to make the case that attention/sensemaking data (eg data
> related to what researchers are attending to and their assessments of
> content) are an important kind of nano-scientific knowledge that gets
> extracted by platforms (social media, publishers, science analytics
> companies) instead of helping to power content curation and discovery
> networks.
>
> If you’re curious about what sensemaking data looks like, here are some
> examples from Twitter- this one
> the “sensemaking layer” of science in an open and FAIR format, and not
> just on social media.
>
> Anyway, I can see a lot of potential synergies with the nanopubs
> ecosystem, so I'd be happy to hear any related feedback, thought or
> comments!
>
> Thanks!
>
> Ronen
>
> --
> You received this message because you are subscribed to the Google
> Groups "Nanopublications" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to nanopub-user...@googlegroups.com
Ronen Tamari
Thanks for the feedback Tobias!
Some comments regarding your observations:
> First, it seems your approach doesn't aim to change the primary methods of publishing research (red parts in Figures 1/2/3), which is of course perfectly fine. But I think this approach becomes even more powerful when these red parts become more structured and fine-grained too, so one can refer to a particular hypothesis, equipment, interpretation, etc.and not only to a scientific paper as a whole.
Yes, 100% agreed. The “finegraining” of scientific research wasn’t the focus of the paper, so we didn’t want to introduce the additional complexity of bringing in that framework. And our sensemaking markers (the blue nodes) are compatible (or even work better, as you mentioned) with more fine-grained red nodes: for example, a marker indicating my approval of a paper can refer specifically to the methods section, even if I didn’t like the literature review.
> It's also a point in which your approach seems to differ from nanopublications. Nanopublications, in most of their use cases so far, are only about public data, with "public" in a strong non-revocable sense.Publishing in this sense means that the user no longer controls the accessibility of the data. This release of control of the publishing user, in turn, allows for others to depend and reliably build up it. So, there are not only clear downsides but also clear advantages to not allow the user to control the accessibility after publication.
That is a great point and one we are still actively discussing. There is tension between the friction to publish and the irrevocablility of published data. E.g., there is less friction to publishing on Twitter where you know you can always go back and delete your Tweet. If you know it is irrevocable, you might share less information (which can be negative) though you also might think more about what you’re posting (which can be positive). It seems there might be room for a spectrum of sensemaking markers, where some are more “light” in the sense that they’re not irrevocable, but also not citeable (with DOI): for example, the fact that I read a specific paper. Some are more “heavy” and probably should have a DOI + be irrevocable- e.g., a review of mine on a particular paper. Perhaps a rule of thumb might be that for sensemaking marker types that are more likely to be built upon directly by scientific discourse (a review), they should be irrevocable. And for marker types that are less a part of the discourse (my having read a particular paper), they are flexible and revocable.
Anna Sofia Lippolis
Anna Sofia
Philipp von Essen
Thanks for your question and welcome to the nanopublications group.
Ronan will be presenting his work at the next Nano Session on 11 July
and it may be a good opportunity to discuss it there. All you need to
know about the meeting: https://nanopub.net/sessions
Have a nice week, Philipp
On 29.06.23 16:13, Anna Sofia Lippolis wrote:
> Hi,
>
> The article was truly an interesting read. As someone new to this
> field, I came across your definition of sensemaking data, which refers
> to implicit behavioral data generated through app usage. I'm currently
> trying to understand where meaning-making analysis/meta-analysis fits
> within this context. Specifically, I'm interested in the annotation of
> metaphorical/analogical usage and reasoning in texts, as it appears to
> provide valuable data for understanding science, theory
> conceptualization, and their evolution over time. Moreover, it could
> contribute to consolidated systems of knowledge representation for
> future researchers and offers points for discussion and exploration.
> Although it may not be directly related to app usage, do you think
> this type of analysis could still be relevant in the realm of
> "modeling collective behavior of scientific communities"? I'd
> appreciate your insights on this matter.
>
>
> Anna Sofia
>
>
> Il giorno domenica 28 maggio 2023 alle 23:43:49 UTC+2
> ronen....@mail.huji.ac.il ha scritto:
>
> Hey all,
>
> Just wanted to share a preprint of ours that may be of interest to
> this community. We presented it last month at the Metascience
> Open Access science needs Open Science Sensemaking (OSSm): open
> infrastructure for sharing scientific sensemaking data
> TL;DR - we try to make the case that attention/sensemaking data
> (eg data related to what researchers are attending to and their
> assessments of content) are an important kind of nano-scientific
> knowledge that gets extracted by platforms (social media,
> publishers, science analytics companies) instead of helping to
> power content curation and discovery networks.
>
> If you’re curious about what sensemaking data looks like, here are
> some examples from Twitter- this one
> on the “sensemaking layer” of science in an open and FAIR format,
> and not just on social media.
>
> Anyway, I can see a lot of potential synergies with the nanopubs
> ecosystem, so I'd be happy to hear any related feedback, thought
> or comments!
>
> Thanks!
>
> Ronen
>
> https://ronentk.github.io/
>
> You received this message because you are subscribed to the Google
> Groups "Nanopublications" group.
> To unsubscribe from this group and stop receiving emails from it, send
> <https://groups.google.com/d/msgid/nanopub-users/da49517a-c17f-4e35-93ce-bdf8652c7a8bn%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.
--
Philipp von Essen
Knowledge Pixels – publishing the meaning of science
Phone: +41 76 581 87 58
Zeughausstrasse 31
8004 Zürich
https://knowledgepixels.com/
Ronen Tamari
Hi Anna Sofia,
I’m happy the paper was interesting for you! Regarding your comments,
As someone new to this field, I came across your definition of sensemaking data, which refers to implicit behavioral data generated through app usage.
Sensemaking data by our definition includes implicit behavioral data but also explicit annotations (tags, votes, ratings, marginal comments) and commentaries made by researchers (see pages 1-2 for more discussion).
So yes, I think your interest in annotations made by scientists is quite related, particularly to questions around modelling collective behavior of scientific communities, as you mentioned. One difference perhaps is, if I understand correctly, that you are interested in annotations made by researchers explicitly for studying metaphorical/analogical usage and reasoning in texts. In our case, we are interested in the annotations made by researchers in the normal course of their work, this kind of data may or may not be useful for particular research questions such as the metaphorical usage question.
As Phillip mentioned, I’ll be presenting this work next week at the Nano Session, happy to discuss more there or elsewhere.
Best,
Ronen