Unknown Unknowns

Reinventing the Research Article -Seven Questions on Science Publishing

Anita de WaardResearcher Disruptive Technologies,

Elsevier LabsNWO - Casimir Grantee,

Utrecht University

ELPUB 2008

http://www.elsevier.com/

http://www.elsevier.com/

http://www.cs.uu.nl/

http://www.cs.uu.nl/

Seven ’known knowns’ in online science publishing:


1. The internet has caused an information overload.



2. Science papers contain facts.




3. The narrative research article is outdated and needs to be replaced.





4. Since words contain meaning,






5. And words (and logic) contain scientific fact,







6. We just need to model them with xml + rdf;







6. We just need to model them with xml + rdf;

7. And the publishers should stop making all these papers.

1. The internet has caused an information overload

- My own experience (as a researcher):



- Easy: find what I know exists




- OK: Finding things I expect hope exist





- Hard: making sure I haven’t missed anything






- However, none of these make me feel overwhelmed.







- Infuriating:







- Infuriating:

- Trying to respond to people who ask me something







- Infuriating:


- Managing three email accounts on 4 computers







- Infuriating:



- Following up on plans and projects







- Infuriating:



- Following up on plans and projects

- However, we can improve the delivery of science content online.




- Pick (carve out) a first set of user needs, e.g.:



- Locate



- Locate

- Understand



- Locate

- Understand

- Believe (Be convinced)



- Locate

- Understand


- Explore



- Locate

- Understand


- Explore

- But this does not address WHAT you want to Locate, Understand, ..



- Locate

- Understand


- Explore


- Semantic network in pharmacology: ‘Grey out what I already know’



- Locate

- Understand


- Explore



1. How can we model a user’s interest?

1. The internet has caused an information overload1. The internet has caused an information overload


- Locate

- Understand


- Explore




2. Science papers contain facts

- With FEBS Letters Editorial Office in Heidelberg/MINT Database in Rome



- Structured Digital Abstract [Gerstein et. al]: ‘machine-readable XML summary of pertinent facts’




- For FEBS: provide proteins, methods, protein-protein interactions, as given in MINT:





- 2008: authors provide, editors check






- 2009: Word Plug-in tool suggests, authors (and editors) check

































- Issue: authors cannot be curators!



- Fact is not claim, but created by consensus post-hoc




- How do we model the process of consenses-building, of disagreement, of fact creation, of mistrust and doubt?





2. Can we create (tools for) an ontology of doubt?





2. Can we create (tools for) an ontology of doubt?

2. Science papers contain facts 2. Science papers contain facts

3. The narrative RA should be replaced

3. The narrative RA should be replaced Aristotle QuintilianQuintilianQuintilian Cell APA Style Guide

prooimion Introduction exordiumThe introduction of a speech, where one announces the subject and purpose

of the discourse, and where one usually employs the persuasive appeal of ethos in order to establish credibility with the audience.

Introduction Introduction

prothesis Statement of Facts narratio

The second part of a classical oration, following the introduction or exordium. The speaker here provides a narrative account of what has happened and

generally explains the nature of the case. Quintilian adds that the narratio is followed by the propositio, a kind of summary of the issues or a statement of

the charge.


Summary propostitioComing between the narratio and the partitio of a classical oration, the

propositio provides a brief summary of what one is about to speak on, or concisely puts forth the charges or accusation.

Abstract Abstract

Division/outline partitio

Following the statement of facts, or narratio, comes the partitio or divisio. In this section of the oration, the speaker outlines what will follow, in accordance with what's been stated as the status, or point at issue in the case. Quintilian suggests the partitio is blended with the propositio and also assists memory.

Table of Contents Article Outline

pistis Proof confirmatioFollowing the division / outline or partitio comes the main body of the speech

where one offers logical arguments as proof. The appeal to logos is emphasized here.

Results Methods, Results

Refutation refutatioFollowing the the confirmatio or section on proof in a classical oration, comes the refutation. As the name connotes, this section of a speech was devoted to

answering the counterarguments of one's opponent.Discussion Discussion

epilogos peroratioFollowing the refutatio and concluding the classical oration, the peroratio conventionally employed appeals through pathos, and often included a

summing up (see the figures of summary, below).Discussion Discussion

http://rhetoric.byu.edu/Canons/Arrangement/Exordium.htm

http://rhetoric.byu.edu/Canons/Arrangement/Exordium.htm

http://rhetoric.byu.edu/Canons/Arrangement/Narratio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Narratio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Partitio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Partitio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Confirmatio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Confirmatio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Refutatio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Refutatio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Peroratio.htm

http://rhetoric.byu.edu/Canons/Arrangement/Peroratio.htm

3. The narrative RA should be replaced Aristotle QuintilianQuintilianQuintilian Cell APA Style Guide

prooimion Introduction exordiumThe introduction of a speech, where one announces the subject and purpose

of the discourse, and where one usually employs the persuasive appeal of ethos in order to establish credibility with the audience.


prothesis Statement of Facts narratio

The second part of a classical oration, following the introduction or exordium. The speaker here provides a narrative account of what has happened and

generally explains the nature of the case. Quintilian adds that the narratio is followed by the propositio, a kind of summary of the issues or a statement of

the charge.


Summary propostitioComing between the narratio and the partitio of a classical oration, the

propositio provides a brief summary of what one is about to speak on, or concisely puts forth the charges or accusation.

Abstract Abstract

Division/outline partitio

Following the statement of facts, or narratio, comes the partitio or divisio. In this section of the oration, the speaker outlines what will follow, in accordance with what's been stated as the status, or point at issue in the case. Quintilian suggests the partitio is blended with the propositio and also assists memory.

Table of Contents Article Outline

pistis Proof confirmatioFollowing the division / outline or partitio comes the main body of the speech

where one offers logical arguments as proof. The appeal to logos is emphasized here.

Results Methods, Results

Refutation refutatioFollowing the the confirmatio or section on proof in a classical oration, comes the refutation. As the name connotes, this section of a speech was devoted to

answering the counterarguments of one's opponent.Discussion Discussion

epilogos peroratioFollowing the refutatio and concluding the classical oration, the peroratio conventionally employed appeals through pathos, and often included a

summing up (see the figures of summary, below).Discussion Discussion

The Story of Goldilocks and the Three Bears

Story Grammar Paper The AXH Domain of Ataxin-1 Mediates Neurodegeneration through Its Interaction with Gfi-1/Senseless Proteins

Once upon a time Time Setting Background The mechanisms mediating SCA1 pathogenesis are still not fully understood, but some general principles have emerged.

a little girl named Goldilocks Characters

Setting

Objects of study the Drosophila Atx-1 homolog (dAtx-1) which lacks a polyQ tract,

She went for a walk in the forest. Pretty soon, she came upon a house.

Location

Setting

Experimental setup

studied and compared in vivo effects and interactions to those of the human protein

She knocked and, when no one answered,

Goal Theme Researchgoal

Gain insight into how Atx-1's function contributes to SCA1 pathogenesis. How these interactions might contribute to the disease process and how they might cause toxicity in only a subset of neurons in SCA1 is not fully understood.

she walked right in. Attempt

Theme

Hypothesis Atx-1 may play a role in the regulation of gene expression

At the table in the kitchen, there were three bowls of porridge.

Name Episode 1 Name dAtX-1 and hAtx-1 Induce Similar Phenotypes When Overexpressed in Files

Goldilocks was hungry. Subgoal

Episode 1

Subgoal test the function of the AXH domain

She tasted the porridge from the first bowl.

Attempt

Episode 1

Method overexpressed dAtx-1 in flies using the GAL4/UAS system (Brand and Perrimon, 1993) and compared its effects to those of hAtx-1.

This porridge is too hot! she exclaimed.

Outcome

Episode 1

Results Overexpression of dAtx-1 by Rhodopsin1(Rh1)-GAL4, which drives expression in the differentiated R1-R6 photoreceptor cells (Mollereau et al., 2000 and O'Tousa et al., 1985), results in neurodegeneration in the eye, as does overexpression of hAtx-1[82Q]. Although at 2 days after eclosion, overexpression of either Atx-1 does not show obvious morphological changes in the photoreceptor cellsSo, she tasted the porridge

from the second bowl.

Episode 1

Data (data not shown),

This porridge is too cold, she said

Outcome

Episode 1

Results both genotypes show many large holes and loss of cell integrity at 28 days

So, she tasted the last bowl of porridge.

Episode 1

Data (Figures 1B-1D).

Ahhh, this porridge is just right, she said happily and

Outcome

Episode 1

Results Overexpression of dAtx-1 using the GMR-GAL4 driver also induces eye abnormalities. The external structures of the eyes that overexpress dAtx-1 show disorganized ommatidia and loss of interommatidial bristles

she ate it all up.

Episode 1

Data (Figure 1F),


Discourse Segments:


Discourse Segments:

- “A text is made up of Discourse Segments and the relations between them” - Grosz and Sidner, Mann-Thomson, Marcu, Swales


Discourse Segments:


- Discourse Segment Purpose: element that has a consistent rhetorical/pragmatic goal.


Discourse Segments:



- Define for Biological Research Article:


Discourse Segments:



- Define for Biological Research Article:<EXPERIMENTS> <Experiment> <Header header="h1">p53-Independent Initiation of G1 Arrest Induced by IR</Header> <Fact fact="fa1" factref="br26">Since the transcriptional response by p53 is a relatively slow process,</Fact> <Problem problem="p1">we asked whether initiation of a G1 arrest following genotoxic stress requires p53. <Problem><Method method="m1">We generated an MCF-7 derivative </Method> <Fact fact="fa2" factref="br24">that expresses the HPV16 E6 protein, which mediates degradation of p53(<Bibref bib="br24">[24]</Bibref>).</Fact><Result result="r1">In the presence of E6, p53 stabilization in response to IR was almost completely prevented in MCF-7 cells (<Figref figref="agami1.gif">Figure 1A).</Figref></Result><Result result="r2">Consistent with this, no induction of p21cip1 by IR was seen in the E6-expressing MCF-7 cells







- Narrative is how stories are told; ‘the truth can only be told in stories’....



- Scientific rhetoric is contained within the narrative




- Main goal of article is to persuade: ‘ The author is a medium that enables the article to get itself published (a la selfish gene/meme)’





- Science happens in language - science is done by creating successful persuasive texts IN ENGLISH! (empowerment rests on mastery of this genre)






- How to disentangle good science from good writing?






- How to disentangle good science from good writing?

3. How can we better represent online narrative? ...


PHC Growth arrestundergo



Paper A:implication

results

method

goal

fact

fact



Paper A:implication

results

method

goal

fact

fact

data 1

data 2 data 3



Paper A:implication

results

method

goal

fact

fact

Paper B:

data 4

data 5 data 6

implication

results

method

goal

fact

fact

data 1

data 2 data 3



Paper A:implication

results

method

goal

fact

fact

Paper B:

data 4

data 5 data 6

implication

results

method

goal

fact

fact

data 1

data 2 data 3



Paper A:implication

results

method

goal

fact

fact

Paper B:

data 4

data 5 data 6

implication

results

method

goal

fact

fact

data 1

data 2 data 3



Paper A:implication

results

method

goal

fact

fact

Paper B:

data 4

data 5 data 6

implication

results

method

goal

fact

fact

underpinning

data 1

data 2 data 3



Paper A:implication

results

method

goal

fact

fact

Paper B:

data 4

data 5 data 6

implication

results

method

goal

fact

fact

data 1

data 2 data 3



Paper A:implication

results

method

goal

fact

fact

Paper B:

data 4

data 5 data 6

implication

results

method

goal

fact

fact

data 1

data 2 data 3

method link



- How to develop systems that ‘reconstruct the salami’



- Claim-evidence networks: identify nr. of experiments supporting a claim, vs. nr. of papers containing two words in a sentence?




3. How can we better represent collections of online narratives?




3. How can we better represent collections of online narratives?

3. The narrative RA should be replaced 3. The narrative RA should be replaced

4. Words contain meaning


Sicilian?


Sicilian?


Sicilian?


Sicilian?


Sicilian?



- ‘A word is worth a thousand pictures’ (Don Loritz)



- The meaning of words occurs in context and is dependent on knowledge and experience



- The meaning of words occurs in context and is dependent on knowledge and experience

- This is even more so in science:PSA = Prostate-Specific Antigen or Pot Smokers Association of America?



- Cognitive linguistics: language and cognition cannot be separated - language acts are cognitive acts



- Lakoff, metaphor: ‘anger is heat’

http://cogsci.berkeley.edu/lakoff/metaphors/





- Meaning is created in the mind:a word is not (only) a ‘particle’ but (also) a ‘wave’:Hearing/reading is not unpacking a package, but resonating at a specific frequency - context is its medium - context-free language does not exist!







4. How do we model cognitive context?











5. Words (and logic) contain scientific fact


• “[Y]ou can transform a fact into fiction or a fiction into fact just by adding or subtracting references [and data]” – Bruno Latour, ‘Science in Action’,1987


“We generated an MCF-7 derivative that expresses the

HPV16 E6 protein, which mediates degradation of p53

([24]).”


http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WSN-414G7R9-8&_coverDate=07%252F07%252F2000&_alid=446007213&_rdoc=1&_fmt=&_orig=search&_qd=1&_cdi=7051&_sort=d&view=c&_acct=C000021878&_version=1&_urlVersion=0&_userid=3021463&md5=dd7bec7d85402ab499415bae0d0bac51





([24]).”

24. M. Scheffner, B.A. Werness, J.M. Huibregtse, A.J. Levine and P.M. Howley, The E6 oncoprotein encoded by human

papillomavirus types 16 and 18 promotes the degradation of p53. Cell 63 (1990), pp. 1129–1136. SummaryPlus | Full Text + Links | PDF (1728 K) | Abstract + References in Scopus |

Cited By in Scopus






http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WSN-4C593X2-17&_user=3021463&_coverDate=12%252F21%252F1990&_fmt=summary&_orig=search&_qd=1&_cdi=7051&view=c&_acct=C000021878&_version=1&_urlVersion=0&_userid=3021463&md5=beb9836c94f2a0bc765d2307fa3d5545&ref=full


http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WSN-4C593X2-17&_user=3021463&_coverDate=12%252F21%252F1990&_fmt=full&_orig=search&_qd=1&_cdi=7051&view=c&_acct=C000021878&_version=1&_urlVersion=0&_userid=3021463&md5=a806d17e92032086fd901a8d19b79405&ref=full




http://www.sciencedirect.com/science?_ob=MiamiImageURL&_imagekey=B6WSN-4C593X2-17-1&_cdi=7051&_user=3021463&_check=y&_orig=search&_coverDate=12%252F21%252F1990&_qd=1&view=c&wchp=dGLbVlb-zSkzV&md5=44cf20ddf4ccf370c9afe76a89e3628b&ie=/sdarticle.pdf


http://www.sciencedirect.com/science?_ob=RedirectURL&_method=outwardLink&_partnerName=655&_targetURL=http%253A%252F%252Fwww.scopus.com%252Fscopus%252Finward%252Frecord.url%253Feid%253D2-s1.0-498R-WBT0-TWC1-C1F4-00000-00%2526partnerID%253D10%2526rel%253DR3.0.0&_acct=C000021878&_version=1&_userid=3021463&md5=9348e4e4f957c12f6c1c7d47be554a5d


http://www.sciencedirect.com/science?_ob=RedirectURL&_method=outwardLink&_partnerName=656&_targetURL=http%253A%252F%252Fwww.scopus.com%252Fscopus%252Finward%252Fcitedby.url%253Feid%253D2-s1.0-498R-WBT0-TWC1-C1F4-00000-00%2526partnerID%253D10%2526rel%253DR3.0.0&_acct=C000021878&_version=1&_userid=3021463&md5=a1edc38ee5ced720550167a4bbff75d3



“In the presence of E6, p53 stabilization in response to IR

was almost completely prevented in MCF-7 cells

(Figure 1A).”



([24]).”



Cited By in Scopus





















Figure 1. Initiation and Maintenance of G1 Arrest Induced by IR(A) Stable MCF-7 clones containing either pCDNA3.1 (Neo) or pCDNA3.1-E6 were irradiated (20 Gy), and cellular protein extracts were made 2 hr later, separated on 10% SDS PAGE, and immunoblotted to detect p53 and cyclin D1 proteins.

“In the presence of E6, p53 stabilization in response to IR

was almost completely prevented in MCF-7 cells

(Figure 1A).”



([24]).”



Cited By in Scopus






















- Essential persuasive elements are non-textual



- Open Data, how to incorporate into ‘text mining’?




- Bioimage consortium (Shotton, Oxford): access biology images across a variety of sources (PLoS, Nature, Elsevier...) and create common metadata format





- SPIDER: Allowing shared access to epidemiology data (meta-epidemiology)






- Tie in to Open Data initiative, generalise, get buy in, sustainability:







5. How do we represent (and access) non-textual elements?







5. How do we represent (and access) non-textual elements?


6. Just model the facts with xml + rdf


- Content in XML - but what about overlapping tags?



- Versioning in DTDs/Schemas? Principle of hierarchical trees - not always best model of a content set




- First pass at relations in RDF (Resource Description Framework:





- Cohere: Open University - (open!) system of creating and linking claims
















- More experiments with RDF:







- DOPE: Semantic access to heterogeneous data in pharmacology








- OKKAM: Entity-centric web (EU-funded)








- OKKAM: Entity-centric web (EU-funded)

8

1. DOPE (2003)deduplicate thesaurus term

select co-occurrence terms

see results set + link to full-text

visualise overlap results


6. Just model the facts with xml + rdf- Yes, but:


- In practice: ScienceDirect does not use our XML... (shhh....)



- At Elsevier: Project Harpoon: ‘stab’ the document with metadata, asynchronous, linked in (XPath/XQuery), distributed




- Not solved in XML - how to access a phrase inside an article:





- access inside a PDF by coordinates? Format, content changes






- add IDs to every single element? Format, content, version changes?







- How to represent relations, even if we know where they link?








6. How can we better model discourse elements (and relations)?

6. Just model the facts with xml + rdf6. Just model the facts with xml + rdf- Yes, but:







6. How can we better model discourse elements (and relations)?

7. And publishers should stop making all those papers.


- 6 uses of a RA:


- 6 uses of a RA:

- job application


- 6 uses of a RA:

- job application

- report card


- 6 uses of a RA:

- job application

- report card

- thesis


- 6 uses of a RA:

- job application

- report card

- thesis

- conference tickets


- 6 uses of a RA:

- job application

- report card

- thesis


- research assessment


- 6 uses of a RA:

- job application

- report card

- thesis



- and yes, by the way, reporting on scientific work.


- 6 uses of a RA:

- job application

- report card

- thesis




- Scientists are evaluated largely based on publications: this enables their production to be evaluated by non-specialists


- 6 uses of a RA:

- job application

- report card

- thesis




- Scientists are evaluated largely based on publications: this enables their production to be evaluated by non-specialists

- This places an undue stress on quantity, conformity (for risk of being rejected), publishing for its own sake.


The real challenge:


The real challenge:

- in Holland, chemistry departments are dwindling


The real challenge:


- in large companies, nr. of PhDs is inversely proportional to power


The real challenge:



- direction of scientific research determined by managers for adolescents


The real challenge:




For science to survive, we need:


The real challenge:





- ‘Hanny’, who found a Voorwerp on GalaxyZoo.org


The real challenge:







The real challenge:






- Prof. Twalib Ngoma, Professor of Oncology from Dar-Es-Salaam, Nigeria


The real challenge:








The real challenge:







- Prof. Zimitri Erasmus, Sociologist from Cape Town


The real challenge:









The real challenge:








How can we access their science?


The real challenge:









7. How can we disentangle communication and evaluation (‘metric of attribution’ - virtual RFID)?


The real challenge:









7. How can we disentangle communication and evaluation (‘metric of attribution’ - virtual RFID)?

7. And publishers should stop making all those papers. 7. And publishers should stop making all those papers.

Seven ‘Known Unknowns’ in Online Science Publishing





2. Can we create an ontology of doubt?




3. How can we better represent collections of online narrative?











5. How do we represent and access non-textual elements?







6. How can we better model discourse elements and relations?







6. How can we better model discourse elements and relations?

7. How can we disentangle communication and evaluation?

Technology

Unknown Unknowns