Preserving the Integrity of the Scholarly Record

Preserving The Integrity of The Scholarly Record

http://www.flickr.com/photos/shinez/5000985919/

Peter Burnhill, EDINA @ University of Edinburgh

NaAonal Library of Scotland George IV Bridge 5.30pm 16th February

Preserving The Integrity of The Scholarly Record

http://www.flickr.com/photos/shinez/5000985919/


NaAonal Library of Scotland George IV Bridge 5.30pm 16th February

Take Home Message: 1)  Archive Streams of Issued Content 2)  Avoid Reference Rot

The Scholarly Record & Serials … [a focus on the digital]

‘The Scholarly Record’ has a fuzzy edge

‘e-‐journals’

Websites, Databases, Repositories

‘Book-‐length work’


ConAnuing Resources, inc. Serials


‘e-‐journals’






Issued in Parts (Serials)

Content changes over Ame

(IntegraAng)

‘e-‐journals’






Other ‘resources needed

for scholarship’ Issued in Parts (Serials)

Content changes over Ame

(IntegraAng)

‘e-‐journals’



‘Gov Docs’

1.  What exactly is the scholarly record? •  What of that now ‘issued on the Web’?

•  And what if we limit focus to what could get an ISSN?

2.  Whose responsibility is it to act as steward? Each research library; library consorAa;

naAonal/state libraries/archives?

& is this a naAonal, or a trans-‐naAonal challenge?

The following quesAons are implicit:

An Article, once available in print on-shelf locally …

… is now online & accessed remotely,

‘anytime/anywhere’ => Improved Ease of Access J

But what of Continuity of Access? Will it be still be there tomorrow?

Libraries boast of ‘e-collections’, but maybe now they only have ‘e-connections’

Picture credit: hgp://somanybooksblog.com/2009/03/27/library-‐tour/

=> real & present danger for the integrity of what is published as scholarly record

10

This is a global challenge: trans-national action

%age of 132,806 ISSN issued for e-serials (December 2013)

US: 20% UK: 8.6%

Rest of World: 71%

Researchers (& libraries/publishers) in any one country are dependent upon content written and published as

serials in countries other than their own

So, who is offering digital shelving?

①  Web-scale not-for-profit archiving agencies:

②  National libraries …

③  Research libraries: consortia & specialist centres …

Ingesting content with archival intent …

National Science Library, Chinese Academy of Sciences

National Science Library, Chinese Academy of Sciences

Many archiving organisations a Good Thing

“Digital information is best preserved by replicating it at multiple archives run by autonomous organizations”

B. Cooper and H. Garcia-Molina (2002)

Some bad stuff will happen!

A Project to Pilot an E-‐journal PreservaAon Registry Service

Need to know who is looking after what & how?

ISSN Register

E-J Preservation Registry Service

E-Journal Preservation

Registry

user requirements

(a)

(b)

ISSN-‐L as kernel field

METADATA on extant e-serials

METADATA on preservation action

Digital Preservation Agencies

Pilot: CLOCKSS, Portico; BL, KB; UK LOCKSS Alliance



ISSN Register

E-J Preservation Registry Service

E-Journal Preservation

Registry

user requirements

(a)

(b)

ISSN-‐L as kernel field

METADATA on extant e-serials

METADATA on preservation action

Digital Preservation Agencies

Pilot: CLOCKSS, Portico; BL, KB; UK LOCKSS Alliance



The Keepers Registry

"Tales from the Keepers Registry"

Serials Review 39.1 (2013)

… to discover who is looking a5er what

thekeepers.org as Global Monitor

*New in 2014*

Library of Congress and Scholars Portal now reporAng in

e-‐journals should be easy – right?

the Keepers Registry recorded

In 2011, 16,558 Atles ‘ingested & archived’ by at least 1 ‘keeper’

in 2013, 21,557 in 2014, 26,195 now 26,712

9,731 'ingested & archived' by 3+

… more archiving & as more archives report into Registry !

Some signs of Progress:

Wrigen & produced by Julie Brown, 1989

“Are we there yet?” … “Don’t think so”

‘Ingest Ratio’ = titles being ingested by one or more Keeper / ‘online serials’ in ISSN Register

= 26,195 / 136,965 [in March 2014]

=> 19% (We do not know about 80% of all resources having ISSN)

‘KeepSafe Ratio’ = titles being ingested by 3+ Keepers / ‘online serials’ in ISSN Register

= 9,656 / 136,965

=> 7%

Evidence on what libraries care about

Using Title List Comparison tool in Members Area of Keepers Registry As reported in: P. Burnhill (2013) Tales from The Keepers Registry: Serial Issues About Archiving & the Web. Serials Review 39 (1), 3–20. hgp://www.sciencedirect.com/science/arAcle/pii/S0098791313000178, &hgps://www.era.lib.ed.ac.uk/handle/1842/6682

In 2011/12 three major research libraries in the USA (Columbia, Cornell & Duke)

checked archival status of serial Atles regarded as important

‘Ingest RaKo’ = 22% to 28%, ie about a quarter

=> fate of c.75% is unknown

very many ‘at risk’ e-‐journals from many small publishers

BIG publishers act early but incompletely

Priority: find economic way to archive content from …

… logs for the UK OpenURL Router*

•  8.5m full text requests in UK during 2012 => 53,311 online Atles requested

Analysis in 2013::

‘Ingest RaKo’ = 32% (16,985/53,311)

=> over two thirds 68% (36,326 Atles) held by none!

Evidence based on what Researchers Use

* As reported in Keepers Registry Blog, OpenURL Router passes ‘discovery’ requests to commercial OpenURL resolver services; developed & delivered by EDINA as part of Jisc support for UK universiAes & colleges

… logs for the UK OpenURL Router*

•  8.5m full text requests in UK during 2012 => 53,311 online Atles requested

Analysis in 2013::

‘Ingest RaKo’ = 32% (16,985/53,311)

=> over two thirds 68% (36,326 Atles) held by none!

Evidence based on what Researchers Use

* As reported in Keepers Registry Blog, OpenURL Router passes ‘discovery’ requests to commercial OpenURL resolver services; developed & delivered by EDINA as part of Jisc support for UK universiAes & colleges

“I believe we've … a problem here.” [John Swigert, Jr.]

Another threat to the integrity of the record

Language Technology Group Funded by the Andrew W. Mellon Foundation

‘Reference Rot’ When what was referenced & cited ceases to say the same thing, or ‘has ceased to be’

hJp://www.snorgtees.com/this-‐parrot-‐has-‐ceased-‐to-‐be

Reference Rot = Link Rot + Content Drift

“when links to web resources no longer point to what they once did”

Link Rot

‘Link Rot’

+ Content Drift: What is at end of URI has changed, or gone!

http://dl00.org 2000




(a) Dynamic content as values on webpage changes over Ame

(b) StaKc content but very different (o{en unrelated) web pages

Hiberlink: Time Travel for The Scholarly Web 1. Threat: Creating evidence on extent of ‘Reference Rot’

–  Main focus: references (& URIs) made in Journal Articles •  "Scholarly Context Not Found: One in Five Articles Suffers from Reference Rot"

–  PLOS One paper published on 26 December 2014.

•  Harvard Law Library & permaCC reference rot in Supreme Court judgments

•  http://www.newyorker.com/magazine/2015/01/26/cobweb

–  Also looked at Reference Rot & the e-Thesis, ETD2014

2.  Remedy: Opportunities for productive intervention –  Identify workflows: preparation, publication, ingest

–  Prototype tools to avoid or limit reference rot

–  Pro-active or ‘transactional’ archiving as remedy •  Embedding such ‘solutions’ in existing tools & infrastructure

•  Propose/test new infrastructure for temporal referencing –  supporting & using the Memento protocol


hgp://www.res|ulliving.com/wp-‐content/uploads/2013/12/Time-‐1024x861.jpg

Preserving the integrity of the scholarly

record

•  Robust Link - re-factor the HTML link that is returned

‘Infrastructure’ to Enable Remedy

<a href="http://www.bnf.fr">

Link to the BNF

</a>

b)  Augment Link with a set of Datetime & location pairs <a href="http://www.bnf.fr"

mset="2014-05-19,

http://archive.today/zdpAn 2014-05-15 memento">

Link to the BNF

</a>

a)  Take simple URI - to French National Library (say)

hgp://robustlinks.mementoweb.org/

Remedy for The Integrity of The Scholarly Record

Envisage the best opportuniAes for IntervenAon to make Remedy, to ‘flash-‐freeze’, either to avoid reference rot or to ‘stop the rot’. 3 basic workflows: ① Study: PreparaAon -‐> (Review) -‐> Submission ② PublicaAon: Editorial -‐> (Revision) -‐> Acceptance -‐> Issue ③ Post-‐PublicaAon: Deposit/Ingest -‐> Provide/Access -‐> Use

IdenPfy the Actors involved in: ① ComposiAon: author/creator ② Public Release: editor/referee/copy ③ CuraAon: librarian / repository manager / archivist

Hiberlink Plug-in: help authors & middle-folk do the right thing:

①  Triggers archiving of referenced web content when it is noted in:

– Zotero - used by authors to manage references

https://www.zotero.org/

–  Open Journal System (OJS) - used by OA publishers

https://pkp.sfu.ca/ojs/

②  Returns Datetime URI for archived content that can

be used in the citation

Two-step Remedy To Avoid Reference Rot

Time’s Up!

thekeepers.org hiberlink.org

•  See also •  thekeepers.blogs.edina.ac.uk •  safenet.blogs.edina.ac.uk/

HelpDesk: [email protected]

Education

Preserving the Integrity of the Scholarly Record