Linked Science - Building a Web of Research Data

Linked ScienceBuilding a Web of Research Data

Rinke HoekstraVU University Amsterdam/University of Amsterdam

[email protected]

Linked Science - Building a Web of Research Data by Rinke HoekstraLicensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.

mailto:[email protected]


http://creativecommons.org/licenses/by-sa/3.0/deed.en_US


Linked LibrariesBuilding a Web of Library Data


[email protected]






Linked ArchivesBuilding a Web of Archived Data


[email protected]






Linked GovernmentBuilding a Web of Government Data


[email protected]






Ok... let’s stick to Science for now

‘Golden boy’ van de psychologieblijkt een onderzoeksfraudeur

Door onze redacteur

Bart Funnekotter

Rotterdam. Topwetenschap is top-sport. Wie niet continu presteert, teltniet meer mee. Publish or perish. DeTilburgse hoogleraar cognitieve soci-ale psychologie Diederik Stapelspeelde in de eredivisie. Daar deedhij alles voor. Toen hij in 2009 wasgenomineerd voor ‘de Moderne manp r ij s ’ van het ministerie van Onder-wijs, zei hij: „Ik werk ook vaak ’savonds en ’s nachts en dat moet jemaar willen en kunnen. Ik heb ge-leerd om op alle mogelijke momen-ten te werken. Nu denk ik eerder:oké, ik heb 20 minuten, laat ik nogeven een paragraaf in elkaar draai-en.”

Gisteren werd duidelijk dat Sta-pels grote productie – in 2011 alleenal (co-)auteur van een boek, driehoofdstukken in een boek en zeven-tien tijdschriftartikelen – deels is ge-baseerd op gefingeerde onderzoeks-gegevens. De universiteit van Til-burg heeft hem gisteren op non-ac-tief gesteld en rector Philip Eijlander

zegt dat Stapel niet meer terugkeert.Een commissie onder voorzitter-schap van voormalig KNAW-presi-dent Pim Levelt zal de omvang vande fraude in kaart brengen.

Stapel (44) was de golden boy va nzijn generatie psychologen. Hij stu-deerde cum laude af en promoveerdeaan de Universiteit van Amsterdam,ook cum laude. Op 34-jarige leeftijdwerd hij hoogleraar aan de Rijksuni-versiteit Groningen en vijf jaar gele-den werd hij in Tilburg benoemd alshoogleraar. Sinds vorig jaar was hijdecaan van de Tilburg School of Soci-al and Behavioral Sciences. Stapelwas een graag geziene gast in de Ver-enigde Staten. Op de website van Ya-le staat een seminar van hem aange-kondigd, op 16 september.

Dat zal waarschijnlijk niet door-gaan nu zijn carrière in duigen ligt.Volgens de Tilburgse rector Eijlan-

der gaat het bij Stapels misdragingenniet om het „hier en daar aanpassenvan onwelgevallige data”. De psy-choloog heeft „op grote schaal gege-vens verzonnen”. „We vermoedendat hij dit ook al deed vóór hij in Til-burg werkte. Ik heb mijn collega inGroningen daarvan op de hoogte ge-steld.”

Tien dagen geleden meldde eenjonge wetenschapper uit Stapelsgroep zich bij Eijlander. Onderzoeks-resultaten leken niet te reproduce-ren en bij navraag bleek dat Stapel re-gelmatig in zijn eentje onderzoekdeed, terwijl zijn omgeving dachtdat hij daarin werd bijgestaan. Eij-lander voerde een aantal gesprekkenmet Stapel. „Aanvankelijk gaf hij al-leen toe dat er misschien wat mis wasmet zijn resultaten. Pas dinsdagbiechtte hij de waarheid op.”

Stapel publiceerde geregeld on-derzoek dat het nieuws haalde en hijwas op televisie te gast bij onder

meer de NRCV en MAX. Twee wekengeleden zorgde hij voor ophef metonderzoek dat hij samen deed met dehoogleraren Roos Vonk en MarcelZeelenberg. Vleeseters waren, aldushet drietal, „e g o ï s t i s ch e r ” en „hufte-riger” dan vegetariërs.

De claim kwam de onderzoekersop kritiek te staan; ze zouden nietonbevooroordeeld zijn. Vonk is voor-zitter geweest van de Stichting Wak-ker Dier. Nu blijkt dat het onderzoekwaarschijnlijk geheel verzonnen is.Vonk gaat in een verklaring op haarwebsite diep door het stof. „Ik moetaannemen dat ook de ‘vlees- data’ be-rusten op fraude. Bij het besprekenvan de resultaten vond ik het welvreemd dat Diederik de naam van deassistent [die het onderzoek had ge-daan] niet noemde, maar de gedachteaan fraude is geen moment in me op-gekomen.”

Vonk toont zich bezorgd over dereputatieschade voor de sociale psy-chologie. „Het is denkbaar dat dezeomvangrijke misstap van één enkelecollega effecten heeft op de reputatievan ons gehele vakgebied.” Ze is ech-ter vooral verbijsterd over het gedragvan haar collega en vriend. „DiederikStapel was een van de beste sociaal-psychologen van Europa. Dat uitge-rekend híj dit gedaan heeft maakthet extra schokkend en laat zien hoe-zeer ook wij als psychologen ons vol-strekt kunnen vergissen in mensen.”

Commentaar: pagina 2

Psycholoog en bekenddeskundige Diederik Stapelblijkt onderzoeken te hebbenverzonnen. Zelfs zijn naastecollega had dat niet door. „Degedachte aan fraude is geenmoment bij me opgekomen.”

Summa cum fraude

Mooie mensen krijgen eerder een baan. Carnivoren zijn hufteriger

en egoïstischer dan niet-vleeseters. Rommelige straten leiden tot

agressie en discriminatie. Het onderzoek van de psycholoog Diederik

Stapel leek vaak raak en haalde dan de krant. Zelfs ook het Amerikaanse

S

c

i

e

n

c

e

en andere wetenschappelijke tijdschriften.

De sociale faculteit van de Universiteit van Tilburg dacht met zijn

werk een psychosociale variant te hebben op F

r

e

a

k

o

n

o

m

i

c

s, zoals een eco-

noom uit Chicago en journalist uit New York in 2005 hun bestseller over

rare maar ware sociaal-economische verbanden tussen bijvoorbeeld

abortuswetgeving en misdaadniveau in de VS noemden.

Nu denkt de universiteit daar anders over. Want de studies van Stapel

leken wel creatief en academisch, maar ze blijken fabeltjes. Sommige ge-

gevens waarop Stapel en zijn collega’s zich baseerden, zijn aan Stapels

duim ontsproten. De anonieme assistenten die hij opvoerde als hulpjes

hebben mogelijk nooit bestaan en zeker

geen integer werk afgeleverd.

Toen de Universiteit van Tilburg onraad

rook – gealarmeerd door collega’s – en

hem daarmee confronteerde, gaf Stapel toe

dat hij gebruikmaakte van fictieve data.

Een bijna komisch detail is dat de hoog-

leraar niet alleen onderzoek zei te doen

maar ook professionele ethiek doceerde.

Stapel is stante pede op non-actief gezet door de rector magnificus. Dat

is logisch. Zoals het ook voor de hand ligt dat hij zijn weg zal vinden in

de literatuur. De zaak is voer voor psychologen én voor auteurs, die zich

laten inspireren door schrijvers als de Joods-Galicische Joseph Roth of

de Vlaming Willem Elsschot, beiden meesters in de spiegelpaleizen.

Maar daarmee houdt de grap wel op. De fraude raakt de academische

wereld in het hart. Stapel was immers een gevierd man in zijn kring,

met wie andere hoogleraren graag samenwerkten. Er is waarschijnlijk

meer aan de hand dan een enkele fantast die tot de top der alma mater

wist door te dringen. Dat bleek dit jaar ook in Duitsland, waar een reeks

politici (onder wie minister Zu Guttenberg en europarlementariër

Koch-Mehrin) werden betrapt op plagiaat in hun dissertatie. Net als in

Duitsland schort het ook in Nederland aan collegiale toetsing en kritiek.

Overschrijven en fabuleren is natuurlijk van alle tijden. Maar als het

een patroon wordt, is dat bedreigend voor de universitaire cultuur. De

verklaring dat citatenindexen en andere quasi-econometrische model-

len om wetenschappelijke productiviteit te meten, de publicatiedruk te

ver opvoeren (‘publish or perish’) is psychologisch plausibel.

Maar niet acceptabel. De academie moet boven elke twijfel verheven

zijn. Zeker in landen die zich trots een kenniseconomie noemen.

B

e

d

r

o

g

m

e

t

o

n

d

e

r

z

o

e

k

r

a

a

k

t

a

c

a

d

e

m

i

s

c

h

e

w

e

r

e

l

d

i

n

h

a

r

t

Hoogleraar

Vonk ook

onder vuur

D

o

o

r

e

e

n

o

n

z

e

r

r

e

d

a

c

t

e

u

r

e

n

Rotterdam.

Na de Tilburgse hoog-

leraar sociale psychologie Diederik

Stapel is ook de werkwijze van zijn

Nijmeegse colle

ga Roos Vonk in op-

spraak. De commissie wetenschap-

pelijke integriteit van

de Radboud

Universiteit Nijmegen stelt

een on-

derzoek in naar een studie over het

gedrag van vleeseters die Stapel en

Vonk deden. Vorige week bleek dat

Stapel de data voor dit onderzoek

heeft verzonnen. Hij is op non-actief

gesteld.

Vonk meldde eind augustus in een

persbericht dat uit het vleeseters-on-

derzoek bleek dat carnivoren egoïsti-

scher zijn dan vegetariërs. Maar de

door Stapel aangeleverde data waren

nog niet gecontroleerd. De Radboud

Universiteit concludeert dat Vonk

„kritischer had moetenzijn

op de

door haar collega aangedragen

on-

derzoeksgegevens”. De procedures

rond het bekendmaken van onder-

zoeksresultaten worden strenger. De

Nijmeegse rector magnificus heeft

met Vonk gesproken, maar vindt dat

ookde onafhankelijk

e integriteits-

commissie een oordeel moet velle

n.

W

e

t

e

n

s

c

h

a

p

m

o

e

t

f

r

a

u

d

e

n

i

e

t

w

e

g

w

u

i

v

e

n

:

pagina 15

NWO eist onderzoeksdata op

D

o

o

r

o

n

z

e

r

e

d

a

c

t

e

u

r

K

a

r

e

l

B

e

r

k

h

o

u

t

Rotterdam. Wetenschappers die

een onderzoeksubsidie krijgen van

NWO, moeten hun gegevens voort-

aan openbaar maken. NWO, de Ne-

derlandse Organisatie voor Weten-

schappelijk Onderzoek, wordt me-

de-eigenaar van de data. NWO is met

een budget van ruim een half miljard

euro de grootste financier van onder-

zoek in Nederland,

„Wetenschappers hebben ten on-

rechte het idee dat de onderzoeksge-

gevens van henzelf zijn”, zegt Ron

Dekker, directeur Instituten bij

NWO. „Maar die data zijn van ons, in

elk geval voor een groot deel. En wij

willen de data voor iedereen toegan-

kelijk te maken. Zo kunnen we de ba-

ten van het kostbare onderzoek over

meer partijen verdelen en kan inte-

ressant vervolgonderzoek worden

verricht met dezelfde data.”

De stap van NWO past in een lang

lopend debat in de academische we-

reld over het openbaar maken van

ruwe onderzoeksgegevens. ‘Open ac-

cess’ helpt de wetenschap vooruit,

luidt de communis opinio, en maakt

het wetenschappers moeilijker om te

frauderen. Wetenschappers zijn te-

rughoudend met het geven van inza-

ge. Recent onderzoek in het weten-

schappelijke tijdschrift Plos ONE

toonde dat van 500 artikelen in de 50

belangrijkste wetenschappelijke pu-

blicaties nog geen 10 procent de data

beschikbaar zijn.

In Nederland is het debat over

open access onlangs opgelaaid door

de affaire rond psycholoog Diederik

Stapel, die onderzoeksgegevens ge-

fingeerd zou hebben. „Open access

leidt er niet toe dat collega’s massaal

elkaars data gaan controleren”, zegt

Dekker: „Wel kan de verleiding om

te frauderen minder worden, als een

wetenschapper weet dat anderen

zijn data kunnen inzien.”

Sociaal-psycholoog Joop van der

Pligt, hoogleraar aan de Universiteit

van Amsterdam, plaatst een kantte-

kening. „Je moet de data delen als

het onderzoek klaar is, maar wan-

neer is het klaar? Zeker bij langlo-

pende onderzoeken is dat lastig te

bepalen”, zegt hij. „Maar het is on-

vermijdelijk dat we de kant opgaan

van open access.”

NWO heeft de verplichte open ac-

cess onlangs toegevoegd aan de sub-

sidievoorwaarden, meldt huisblad E-

data & Research. Dekker: „We zijn be-

zig met een campagne met de bood-

schap: de data zijn niet jou.”

Heel goed, vindt directeur Peter

Doorn van Data Archiving and Net-

worked Services (DANS): „Deze stap

van NWO helpt ons zeer.” DANS ver-

zamelt onderzoeksgegevens van we-

tenschappers en maakt die digitaal

beschikbaar. „Maar dat gaat soms

moeizaam. In de bètawetenschappen

is het delen van data heel normaal,

bij alfa- en gammawetenschappen is

het veel minder ingeburgerd.”

Sociaal-psychologen bijvoorbeeld,

waartoe ook Stapel behoort, delen

nauwelijks data. Ook economen

doen het weinig. Archeologen die

vaak jaren onderzoek doen naar hun

vondsten zijn geneigd hun data al

die tijd onder zich te houden. „Maar

met de archeologen hebben we nu

een overeenkomst over het delen van

data”, zegt Doorn. „En sinds kort

ook met de dendrochronologen.”

De dendrochronologie, de weten-

schap die aan de hand van jaarringen

in bomen hout dateert, reconstru-

eert onder meer de historie van het

klimaat. Die onderzoeksgegevens

spelen een rol in het debat over kli-

maatverandering. De Queen’s Uni-

versiteit in Belfast moest in 2010 van

de rechter zulke gegevens afstaan

aan een Britse klimaatscepticus.

NWO wil ook dat wetenschappers

het auteursrecht op hun artikelen

niet langer overdragen aan tijd-

schriften. „Een onderzoeker kan be-

ter een licentie afgeven.” Maar Science

en Nature, twee van de meest voor-

aanstaande wetenschappelijke tijd-

schriften ter wereld, eisen altijd het

copyright. Dekker: „Dat betekent dat

wij strijd moeten leveren met com-

merciële uitgevers die hun verdien-

model willen handhaven.”

C

o

m

m

e

n

t

a

a

r

:

pagina 2

Wetenschappers moeten gegevens openbaar maken en copyright afstaan

Data’s shameful neglectResearch cannot flourish if data are not preserved and made accessible. All concerned must act accordingly.

More and more often these days, a research project’s success is measured not just by the publications it produces, but also by the data it makes available to the wider community. Pioneer-

ing archives such as GenBank have demonstrated just how powerful such legacy data sets can be for generating new discoveries — espe-cially when data are combined from many laboratories and analysed in ways that the original researchers could not have anticipated.

All but a handful of disciplines still lack the technical, institutional and cultural frameworks required to support such open data access (see pages 168 and 171) — leading to a scandalous shortfall in the sharing of data by researchers (see page 160). This deficiency urgently needs to be addressed by funders, universities and the researchers themselves.

Research funding agencies need to recognize that preservation of and access to digital data are central to their mission, and need to be supported accordingly. Organizations in the United Kingdom, for instance, have made a good start. The Joint Information Systems Committee, established by the seven UK research councils in 1993, has made data-sharing a priority, and has helped to establish a Digital Curation Centre, headquartered at the University of Edinburgh, to be a national focus for research and development into data issues. Other European agencies have also pursued initiatives.

The United States, by contrast, is playing catch-up. Since 2005, a 29-member Interagency Working Group on Digital Data has been trying to get US funding agencies to develop plans for how they will support data archiving — and just as importantly, to develop policies on what data should and should not be preserved, and what excep-tions should be made for reasons such as patient privacy. Some agen-cies have taken the lead in doing so; many more are hanging back. They should all being moving forwards vigorously.

What is more, funding agencies and researchers alike must ensure that they support not only the hardware needed to store the data, but

also the software that will help investigators to do this. One impor-tant facet is metadata management software: tools that streamline the tedious process of annotating data with a description of what the bits mean, which instrument collected them, which algorithms have been used to process them and so on — information that is essential if other scientists are to reuse the data effectively.

Also necessary, especially in an era when data can be mixed and combined in unanticipated ways, is software that can keep track of which pieces of data came from whom. Such systems are essential if tenure and promotion committees are ever to give credit — as they should — to candidates’ track-record of data contribution.

Who should host these data? Agencies and the research community together need to create the digital equivalent of libraries: institutions that can take responsibility for preserving digital data and making them accessible over the long term. The university research libraries themselves are obvious candidates to assume this role. But whoever takes it on, data preservation will require robust, long-term funding. One potentially helpful initiative is the US National Science Foundation’s DataNet programme, in which researchers are exploring financial mecha-nisms such as subscription services and membership fees.

Finally, universities and individual disciplines need to undertake a vigorous programme of education and outreach about data. Consider, for example, that most university science students get a reasonably good grounding in statistics. But their studies rarely include anything about information management — a discipline that encompasses the entire life cycle of data, from how they are acquired and stored to how they are organized, retrieved and maintained over time. That needs to change: data management should be woven into every course in science, as one of the foundations of knowledge. ■

A step too far?The Obama administration must fund human space flight adequately, or stop speaking of ‘exploration’.

After the space shuttle Columbia burned up during re-entry into Earth’s atmosphere in 2003, the board that was convened to investigate the disaster looked beyond its technical causes

to NASA’s organizational malaise. For decades, the board pointed out, the shuttle programme had been trying to do too much with too little money . NASA desperately needed a clearer vision and a better-defined mission for human space flight.

The next year, then-President George W. Bush attempted to supply that vision with a new long-term goal: first send astronauts to build

a base on the Moon, then send them to Mars. This idea immediately set off a debate that is still continuing, in which sceptics ask whether there is any point in returning to the Moon nearly half a century after the first landings. Why not go to Mars directly, or visit near-Earth asteroids, or send people to service telescopes in the deep space beyond Earth?

Yet that debate is both counter-productive — a new set of rockets could go to all of these places — and moot, because Bush’s vision never attracted the hoped-for budget increases. Indeed, a blue-riband commission reporting to US President Barack Obama this week (see page 153) finds the organizational malaise unchanged: NASA is still doing too much with too little . Without more money, the agency won’t be sending people anywhere beyond the International Space Station, which resides in low Earth orbit only 350 kilometres up. And even the ability to do that is in question: Ares I, the US rocket that would return

“Data management should be woven into every course in science.”

145

www.nature.com/nature Vol 461 | Issue no. 7261 | 10 September 2009

145-146 Editorials WF IF.indd 145145-146 Editorials WF IF.indd 145 8/9/09 14:06:408/9/09 14:06:40

Data’s shameful neglectResearch cannot flourish if data are not preserved and made accessible. All concerned must act accordingly.

More and more often these days, a research project’s success is measured not just by the publications it produces, but also by the data it makes available to the wider community. Pioneer-

ing archives such as GenBank have demonstrated just how powerful such legacy data sets can be for generating new discoveries — espe-cially when data are combined from many laboratories and analysed in ways that the original researchers could not have anticipated.

All but a handful of disciplines still lack the technical, institutional and cultural frameworks required to support such open data access (see pages 168 and 171) — leading to a scandalous shortfall in the sharing of data by researchers (see page 160). This deficiency urgently needs to be addressed by funders, universities and the researchers themselves.

Research funding agencies need to recognize that preservation of and access to digital data are central to their mission, and need to be supported accordingly. Organizations in the United Kingdom, for instance, have made a good start. The Joint Information Systems Committee, established by the seven UK research councils in 1993, has made data-sharing a priority, and has helped to establish a Digital Curation Centre, headquartered at the University of Edinburgh, to be a national focus for research and development into data issues. Other European agencies have also pursued initiatives.

The United States, by contrast, is playing catch-up. Since 2005, a 29-member Interagency Working Group on Digital Data has been trying to get US funding agencies to develop plans for how they will support data archiving — and just as importantly, to develop policies on what data should and should not be preserved, and what excep-tions should be made for reasons such as patient privacy. Some agen-cies have taken the lead in doing so; many more are hanging back. They should all being moving forwards vigorously.

What is more, funding agencies and researchers alike must ensure that they support not only the hardware needed to store the data, but

also the software that will help investigators to do this. One impor-tant facet is metadata management software: tools that streamline the tedious process of annotating data with a description of what the bits mean, which instrument collected them, which algorithms have been used to process them and so on — information that is essential if other scientists are to reuse the data effectively.

Also necessary, especially in an era when data can be mixed and combined in unanticipated ways, is software that can keep track of which pieces of data came from whom. Such systems are essential if tenure and promotion committees are ever to give credit — as they should — to candidates’ track-record of data contribution.

Who should host these data? Agencies and the research community together need to create the digital equivalent of libraries: institutions that can take responsibility for preserving digital data and making them accessible over the long term. The university research libraries themselves are obvious candidates to assume this role. But whoever takes it on, data preservation will require robust, long-term funding. One potentially helpful initiative is the US National Science Foundation’s DataNet programme, in which researchers are exploring financial mecha-nisms such as subscription services and membership fees.

Finally, universities and individual disciplines need to undertake a vigorous programme of education and outreach about data. Consider, for example, that most university science students get a reasonably good grounding in statistics. But their studies rarely include anything about information management — a discipline that encompasses the entire life cycle of data, from how they are acquired and stored to how they are organized, retrieved and maintained over time. That needs to change: data management should be woven into every course in science, as one of the foundations of knowledge. ■

A step too far?The Obama administration must fund human space flight adequately, or stop speaking of ‘exploration’.

After the space shuttle Columbia burned up during re-entry into Earth’s atmosphere in 2003, the board that was convened to investigate the disaster looked beyond its technical causes

to NASA’s organizational malaise. For decades, the board pointed out, the shuttle programme had been trying to do too much with too little money . NASA desperately needed a clearer vision and a better-defined mission for human space flight.

The next year, then-President George W. Bush attempted to supply that vision with a new long-term goal: first send astronauts to build

a base on the Moon, then send them to Mars. This idea immediately set off a debate that is still continuing, in which sceptics ask whether there is any point in returning to the Moon nearly half a century after the first landings. Why not go to Mars directly, or visit near-Earth asteroids, or send people to service telescopes in the deep space beyond Earth?

Yet that debate is both counter-productive — a new set of rockets could go to all of these places — and moot, because Bush’s vision never attracted the hoped-for budget increases. Indeed, a blue-riband commission reporting to US President Barack Obama this week (see page 153) finds the organizational malaise unchanged: NASA is still doing too much with too little . Without more money, the agency won’t be sending people anywhere beyond the International Space Station, which resides in low Earth orbit only 350 kilometres up. And even the ability to do that is in question: Ares I, the US rocket that would return

“Data management should be woven into every course in science.”

145

www.nature.com/nature Vol 461 | Issue no. 7261 | 10 September 2009

145-146 Editorials WF IF.indd 145145-146 Editorials WF IF.indd 145 8/9/09 14:06:408/9/09 14:06:40

Research cannot flourish if data are not preserved and made accessible. All concerned must act accordingly.

JACEYB Y J O R D A N S U C H O W

To the dismay of many (yet to the delight of a few), Nature Publishing

Group announced today that its flagship journal, Nature, will no longer accept submissions from humans (Homo sapiens). The new policy, which has been under editorial consideration for many years, was sparked by a grow-ing sentiment in the scientific community that the heuristics and biases inherent in human decision-making preclude them from conducting reliable sci-ence. In an ironic twist of fate, the species has impeached itself by thorough research on its own shortcomings.

The ban takes effect on 12 September and will apply to those who self-identify as human. Authors will be required to include, in addition to the usual declaration of competing financial interests, the names of all humans consulted in prepa-ration of the submitted work. Other journals are likely to adopt a similar policy.

Although the reactions are mixed, not everyone is surprised, and a few remain comfortably unaffected.

The Massachusetts Institute of Technol-ogy has since 2010 asked that all active researchers opt-in to wearing an implant-able tag as part of the TMI project, which aggregates real-time data across the cam-pus to improve all aspects of everything. As these tags are sentient, the researchers who wear them qualify as bionic (Homo bionika) according to standard ISO +1.914/582.2646. This act of foresight by the university, which at the time was con-troversial and the cause of much debate, now pays a handsome dividend.

Similarly, researchers at Yale, who have never been the type to self-identify as mere mortals, remain unscathed.

It seems unavoidable that other universi-ties will soon follow suit, causing a sharp rise in the incidence of implants and arrogance. Exploiting these loopholes may be a saving grace for the species’ full participation in the sciences.

While professors weep, students rejoice. According to the provisions of the ISO standard (the one gainfully employed by MIT), a human who spends at least half its waking hours interacting with a sentient non-carbon-based machine qualifies as bionic. The newest generation of students, having grown up on the interwebs, spends on average the entirety of its life online. Students everywhere have been seen call-ing their mothers, reiterating how brilliant they were to have flatly ignored the warnings to “put down that damn hand computer”. Cyberculture paid off.

Those who have been slow to adopt new technology (or who still identify as human) are rightly concerned: their contribution to

Nature had been dwin-dling well before the ban, and today consti-tutes less than 10% of published papers. In its

place stands the work of pharma-ceutical laboratory automatons, embedded devices, the inter-webs and most recently, Google Books, which having declared independence from its parent company Google (NASDAQ: GOOG), has become increas-ingly prolific, contributing 42 manuscripts this year alone.

Shortly after the announce-ment, the World Wide interwebs Consortium (W2iC, formerly W3C), alongside the Union of Embedded Tags, jointly filed a formal complaint with the journal, arguing for mandatory first-authorship of non-carbon-based machines in all bionic collaborations. (See also the let-ter to the editor in the 12 March issue of Nature, written by Tag #15167247373 and co-signed by the arm in which it is embed-ded.) Although amendments to Nature’s policy are at this time unlikely, concerns regarding authorship will surely be the cause of considerable tension in many laboratories. Embed-ded devices will use the ban as leverage for salary increases and promotions.

Not everyone is so bothered by the announcement. Egbert B. Gebstadter, professor of com-

puter science at the University of Mishug-gan, notes: “Although it is nonsensical to rely on evidence provided by human-based research when judging whether humans are themselves inept, in doing so, the editors (all human, I note) provide a perfect example of the feebleness of human reasoning, thereby validating their claims.” Gebstadter is bionic, although was human when he had come to this conclusion.

The editors of Nature were readily availa-ble for comment, and their incisive remarks gave such great credibility to the new pol-icy that it rendered all future debate moot. But, in the spirit of the policy, because the editors are human, these remarks are duly censored. ■

Jordan Suchow is a graduate student in cognitive science at Harvard University, and can be found online at jwsu.ch/ow. He self-identifies as human.

NPG’S POLICY ON AUTHORSHIPImportant change to submission criteria.

NATURE.COMFollow Futures on Facebook at:go.nature.com/mtoodm

2 4 4 | N A T U R E | V O L 4 7 7 | 8 S E P T E M B E R 2 0 1 1

FUTURES

© 2011 Macmillan Publishers Limited. All rights reserved

JACEYB Y J O R D A N S U C H O W

To the dismay of many (yet to the delight of a few), Nature Publishing

Group announced today that its flagship journal, Nature, will no longer accept submissions from humans (Homo sapiens). The new policy, which has been under editorial consideration for many years, was sparked by a grow-ing sentiment in the scientific community that the heuristics and biases inherent in human decision-making preclude them from conducting reliable sci-ence. In an ironic twist of fate, the species has impeached itself by thorough research on its own shortcomings.

The ban takes effect on 12 September and will apply to those who self-identify as human. Authors will be required to include, in addition to the usual declaration of competing financial interests, the names of all humans consulted in prepa-ration of the submitted work. Other journals are likely to adopt a similar policy.

Although the reactions are mixed, not everyone is surprised, and a few remain comfortably unaffected.

The Massachusetts Institute of Technol-ogy has since 2010 asked that all active researchers opt-in to wearing an implant-able tag as part of the TMI project, which aggregates real-time data across the cam-pus to improve all aspects of everything. As these tags are sentient, the researchers who wear them qualify as bionic (Homo bionika) according to standard ISO +1.914/582.2646. This act of foresight by the university, which at the time was con-troversial and the cause of much debate, now pays a handsome dividend.

Similarly, researchers at Yale, who have never been the type to self-identify as mere mortals, remain unscathed.

It seems unavoidable that other universi-ties will soon follow suit, causing a sharp rise in the incidence of implants and arrogance. Exploiting these loopholes may be a saving grace for the species’ full participation in the sciences.

While professors weep, students rejoice. According to the provisions of the ISO standard (the one gainfully employed by MIT), a human who spends at least half its waking hours interacting with a sentient non-carbon-based machine qualifies as bionic. The newest generation of students, having grown up on the interwebs, spends on average the entirety of its life online. Students everywhere have been seen call-ing their mothers, reiterating how brilliant they were to have flatly ignored the warnings to “put down that damn hand computer”. Cyberculture paid off.

Those who have been slow to adopt new technology (or who still identify as human) are rightly concerned: their contribution to

Nature had been dwin-dling well before the ban, and today consti-tutes less than 10% of published papers. In its

place stands the work of pharma-ceutical laboratory automatons, embedded devices, the inter-webs and most recently, Google Books, which having declared independence from its parent company Google (NASDAQ: GOOG), has become increas-ingly prolific, contributing 42 manuscripts this year alone.

Shortly after the announce-ment, the World Wide interwebs Consortium (W2iC, formerly W3C), alongside the Union of Embedded Tags, jointly filed a formal complaint with the journal, arguing for mandatory first-authorship of non-carbon-based machines in all bionic collaborations. (See also the let-ter to the editor in the 12 March issue of Nature, written by Tag #15167247373 and co-signed by the arm in which it is embed-ded.) Although amendments to Nature’s policy are at this time unlikely, concerns regarding authorship will surely be the cause of considerable tension in many laboratories. Embed-ded devices will use the ban as leverage for salary increases and promotions.

Not everyone is so bothered by the announcement. Egbert B. Gebstadter, professor of com-

puter science at the University of Mishug-gan, notes: “Although it is nonsensical to rely on evidence provided by human-based research when judging whether humans are themselves inept, in doing so, the editors (all human, I note) provide a perfect example of the feebleness of human reasoning, thereby validating their claims.” Gebstadter is bionic, although was human when he had come to this conclusion.

The editors of Nature were readily availa-ble for comment, and their incisive remarks gave such great credibility to the new pol-icy that it rendered all future debate moot. But, in the spirit of the policy, because the editors are human, these remarks are duly censored. ■

Jordan Suchow is a graduate student in cognitive science at Harvard University, and can be found online at jwsu.ch/ow. He self-identifies as human.

NPG’S POLICY ON AUTHORSHIPImportant change to submission criteria.

NATURE.COMFollow Futures on Facebook at:go.nature.com/mtoodm

2 4 4 | N A T U R E | V O L 4 7 7 | 8 S E P T E M B E R 2 0 1 1

FUTURES

© 2011 Macmillan Publishers Limited. All rights reserved

Fighting a Running Battle• Too much data to make sense of

• Manual curation won’t cut it

• Advanced search not sufficient 0

500

1000

1500

2000

1993 1996 1999 2002 2005 2008 20110

1,25

2,50

3,75

5,00SciVerse ScienceDirect

Articles Articles per Day

“Dweilen met de kraan open”

Fighting a Running Battle• Too much data to make sense of

• Manual curation won’t cut it

• Advanced search not sufficient 0

500

1000

1500

2000

1993 1996 1999 2002 2005 2008 20110

1,25

2,50

3,75



“Dweilen met de kraan open”

We need a better hold on research data

Credits to Tomi Kauppinen and Willem van Hagehttp://linkedscience.org

What does it take to make this work?

Linked Science is an approach to interconnect scientific assets to enable transparent, reproducible and transdisciplinary research. LinkedScience.org is a community driven-effort to show what this means in practice.

http://www.slideshare.net/cgueret/exposing-the-data-from-narcis-with-vivo


Interconnect

• Connections are citations

• They are implicit free text (≠ machine readable)

• Expensive to make explicit

• ... and hidden in silos

Scientific Asset = Published Article

+ ≠

InterconnectScientific Asset = Any part of Anything Published

InterconnectScientific Asset = Any part of Anything Published

Articles

Datasets

Paragraphs

Spreadsheet Cells

Annotations

Social Media

Database Cells

• Connections are typed links

• They are explicit (= machine readable)

• ... still expensive to make explicit

• But exposed to the outside

Transparent

• They are implicit free text (≠ machine readable)

• They have Dublin Core metadata

• No underlying data for publications

• Data published without context

Pacific Barreleye, http://imgur.com/gallery/Mzyb5(can rotate its eyes forwards or upwards to look through the transparent head to prey above)

http://imgur.com/gallery/Mzyb5


Transparent

• Publications have explicit links to methods, experiments, data

• They have rich metadatacontent, hypothesis, evidence, conclusion

• Data published as Linked Data

Pacific Barreleye, http://imgur.com/gallery/Mzyb5(can rotate its eyes forwards or upwards to look through the transparent head to prey above)



Reproducible

• Publications abstract from research...

• Making it almost impossible to reproduce results

Reproducible

• Publications abstract from research...

• Making it almost impossible to reproduce results

http://on.wsj.com/XCajtB



ReproduciblePapers explicitly link to the

underlying provenanceTrack and publish explicit

provenance information

Capture the processes by which data is manipulated

TransdisciplinaryApplicability of results across domains: unknown

Börner K, Klavans R, Patek M, Zoss AM, Biberstine JR, et al. (2012) Design and Update of a Classification System: The UCSD Map of Science. PLoS ONE 7(7): e39464. doi:10.1371/journal.pone.0039464

Transdisciplinary

• Unique identification

• Explicit links

• Rich metadata

• Research data

• Provenance information

Web Science

Things to consider...• Types of connections between scientific assets

• Portability of connections and metadata

• Granularity of connections and metadata

• Consistency of connections and metadata

• Ownership and control of connections and metadata

• Quality of connections and metadata

Linked Data

+ =

• Everything gets a URI

• Everything is connected as much as possible

• Everything is assigned to a category (including connections)

• When we know two things are the same, we say so

• We might even publish our data on the Web (but don’t have to)

Concrete Examples

http://en.wikipedia.org/wiki/File:Concrete_mixer_ldce.JPG



is a public- private research community

VIVO enables the discovery of research and scholarship within an institution and beyond

https://github.com/Data2Semantics/VIVO-Browser



24

6 23

20

9

26

12

11

12

19

47

8

15

WUR

VU

TUDelft

Thales

D-CIS

Sense Observation SystemsAlmende

Video DockEuropeana

KBCOMMIT

Rijksmuseum

Beeld en Geluid

AMCLUMC

AmolfUniversiteit Leiden

CWI

UvA

UT TNO

TomTom

UU

Logica

Serious Toys

Waag Society

EricssonNovay

Roessingh

Capgemini

Irdeto

Ambient Systems Microflown

Centric

Politie NLTUePhilips Research

Broadfield

RUN

CINOP

KITT

Re-LionKLPD

T-Xchange

EUR

e-ScienceCenter

ESIPhilips

Healthcare

Axini

to2Data Semantics

Semantics for Scientific Data PublishersFrom Data © 2013

Legend

Rinke Hoekstra, VU University Amsterdam http://www.data2semantics.org

Collaboration graph of all COMMIT/ projectsPersons are colored according to affiliation,Projects are numbered as follows:

1 - Information Retrieval for Information Services2 - Interaction for Universal Access4 - Virtual Worlds for Well-Being 6 - Socially-Enriched Access to Linked Cultural Media8 - SWELL: Smart Reasoning Systems for Well-Being at Work and at Home9 - Very Large Wireless Sensor Networks for Well-Being

12 - METIS: Dependable Cooperative Systems for Public Safety15 - Trusted Healthcare Services19 - Spatiotemporal Data Warehouses for Trajectory Exploitation20 - e-Infrastructure Virtualization for e-Science Applications23 - From Data to Semantics for Scientific Data Publishers24 - e-Biobanking With Imaging for Healthcare26 - e-Foodlab11 - Composable Embedded Systems for Healthcare

24

6 23

20

9

26

12

11

12

19

47

8

15

WUR

VU

TUDelft

Thales

D-CIS


Video DockEuropeana

KBCOMMIT

Rijksmuseum

Beeld en Geluid

AMCLUMC


CWI

UvA

UT TNO

TomTom

UU

Logica

Serious Toys

Waag Society

EricssonNovay

Roessingh

Capgemini

Irdeto


Centric


Broadfield

RUN

CINOP

KITT

Re-LionKLPD

T-Xchange

EUR

e-ScienceCenter

ESIPhilips

Healthcare

Axini

to2Data Semantics


Legend





24

6 23

20

9

26

12

11

12

19

47

8

15

WUR

VU

TUDelft

Thales

D-CIS


Video DockEuropeana

KBCOMMIT

Rijksmuseum

Beeld en Geluid

AMCLUMC


CWI

UvA

UT TNO

TomTom

UU

Logica

Serious Toys

Waag Society

EricssonNovay

Roessingh

Capgemini

Irdeto


Centric


Broadfield

RUN

CINOP

KITT

Re-LionKLPD

T-Xchange

EUR

e-ScienceCenter

ESIPhilips

Healthcare

Axini

to2Data Semantics


Legend





to2Data Semantics

Semantics for Scientific Data PublishersFrom Data

AERS-LDserious adverse

event reportsexposed as linked data

Papers & Guidelines

BioPortalMesh,

MedDRA,SnomedCT,

etc.

LOD CloudUMLS, DBPedia,Sider, Drugbank,

LinkedCT

SILK linkspeci!cation

languageand

PROV-O

BioPortalAnnotator

withAnnotationOntology

andPROV-O

4Store

Google WebToolkit

Hubble demonstrates three ‘sales pitches’ of linked data: inter-operability, interlinking and tool availability.

From patient to:- Relevant publications- Related adverse events- Clinical trials- Drug information- Known side e"ects- Statistical analysis

PROV-O-MaticTM

• Python Wrapper script for shell commandshttps://github.com/Data2Semantics/data/blob/master/src/d2s/prov.py

• Output in PROV-O & W3C Time vocabulary

• Timestamped URIs for files/resources

• ... integrate with GIT?

• Provenance trail for conversion, loading and linking

Monday, February 27, 12

TabLinkerSemi-Automatic RDF Converter for Eccentric Excel Files


Partial Replication

Yasgui

COMPLEXITY vs. INTERESTINGNESS

?

Data Analysis

Provenance Reconstruction

http://www.data2semantics.org

RDF$Conversion$

RDF$Cleaning$

Internal$Linking$

Link$to$Other$Data$

Semi8Automa;c$Annota;on$

Cloud$

Provenance$Enrichment$

acquiring$data$from$text?$

xml2rdf$d2rq$

rdb2rdf$$

e.g.$GATE$OpenCalais$

AIDA$Browser$Poseidon$(Pirates/Maps)$

…$

SILK$Amalgame$Graph$Rewri;ng$Graph$Rewri;ng$

Provenance$

Analysis/Metrics$

Querying$and$Ranking$

Visualiza;on$

User$Interfaces$

sgvizler$

RDF$Feedback$

Semi8Automa;c$Conversion$

“tablinker”$

to2Data Semantics

Semantics for Scientific Data PublishersFrom Data

HUBBLE Linked Data Hub for Clinical Decision Support

PROV-O-MaticTM

• Python Wrapper script for shell commandshttps://github.com/Data2Semantics/data/blob/master/src/d2s/prov.py

• Output in PROV-O & W3C Time vocabulary

• Timestamped URIs for files/resources

• ... integrate with GIT?

• Provenance trail for conversion, loading and linking


TabLinkerSemi-Automatic RDF Converter for Eccentric Excel Files


Partial Replication

Yasgui

COMPLEXITY vs. INTERESTINGNESS

?

Data Analysis

Provenance Reconstruction


RDF$Conversion$

RDF$Cleaning$

Internal$Linking$

Link$to$Other$Data$

Semi8Automa;c$Annota;on$

Cloud$

Provenance$Enrichment$

acquiring$data$from$text?$

xml2rdf$d2rq$

rdb2rdf$$

e.g.$GATE$OpenCalais$

AIDA$Browser$Poseidon$(Pirates/Maps)$

…$

SILK$Amalgame$Graph$Rewri;ng$Graph$Rewri;ng$

Provenance$

Analysis/Metrics$

Querying$and$Ranking$

Visualiza;on$

User$Interfaces$

sgvizler$

RDF$Feedback$

Semi8Automa;c$Conversion$

“tablinker”$

Connect recommendations in clinical guidelines to underlying evidence

Rinke Hoekstra, Anita de Waard, Richard Vdovjak (2012) Annotating Evidenced Based Clinical Guidelines - A Lightweight Ontology, Proceedings of SWAT4LS 2012, Paris, Adrian Paschke, Albert Burger, Paolo Roma, M. Scott Marshall, Andrea Splendiani (ed.), Springer

24

13 9

17

8

3

20

6

4

7

5

15

22

23

21

12

16

1

18

100

19

11

14

2

Cluster 1: Blood Cultures Cluster 3: GeneralEvidenceQ|| GuidelineEvidenceQX

Cluster 2: Markers

Short-Title Catalogue

Co-author network for works published in The Netherlands

(~1550-1800)

In DegreeBetweenness Centrality

... but that’s just asking for more!

Data does not travel well...

Slide data courtesy of Christophe Guéret of DANS/VUhttp://www.slideshare.net/cgueret/exposing-the-data-from-narcis-with-vivo

148 38286





148 38286

47





13148 38

47



0

500

1000

1500

2000

1993 1995 1997 1999 2001 2003 2005 2007 2009 20110

1,25

2,50

3,75



... 7 months is a long time!

If this already is tedious...

... can you expect researchers to publish Linked Research Data?

Idea: use reward mechanisms of Web 2.0

• Lightweight Web Application

• Interface to API of existing data repositories

• Enrich metadata by linking to Linked Data resources

• Provide annotation services for data files

• Plugin based architecture

• Publish RDF metadata as new data publicationhttp://linkitup.data2semantics.org

http://linkitup.data2semantics.org

http://linkitup.data2semantics.org

recoprovReconstruct provenance using

Dropbox file edit history

0

1

8

9

12

13

16 22

2

4

7

11

17

19

3

5

6

14

23

10 15

18

21

20

24

Sara Magliacane and Paul Groth

plsheetHow are results calculated (1)? Automatic analyis of workflow in spreadsheets

Analyse dependencies between cells in complex spreadsheets

Martine de Vos, Jan Wielemaker and Willem van Hage

plsheet

Reconstruct and explain the workflow of computations

Martine de Vos, Jan Wielemaker and Willem van Hage

Albert Merono-Penuela, Rinke Hoekstra, Laurens Rietveld, Christophe Gueret

TabLinker

http://www.cedar-project.nl

Semi-automatic RDF converter for eccentric spreadsheets

http://www.cedar-project.eu


Albert Merono-Penuela, Rinke Hoekstra, Laurens Rietveld, Christophe Gueret

TabLinker

http://www.cedar-project.nl

Semi-automatic RDF converter for eccentric spreadsheets












... build useful services and tools for data publishers ...

... that maintain provenance information ...

... and cater for the entire research cycle ...

... allowing full reusability of the research

... empower researchers to do their own enrichment ...




Technology

Linked Science - Building a Web of Research Data