drjwbaker/Transforming Research Through Digital Scholarship

## Transforming Research Through Digital Scholarship
**Transforming Research Through Digital Scholarship, British Library, 11 November 2013**
========================================================================================

*Andrew Prescott, How Arts and Humanities is Being Transformed through Digital Scholarship*
Some of transformation unclear ... offer some hints on what is going on.
Collaboration with institutions at the heart of the digitrans theme's development.
BL site of transformative moments: text, sounds recording, post, biblical concordances.
The latter were revolutionary, considered heretical
> in essence cut up the bible and rearranged it into a different order.
> is it a tool? act of philosophy? intellectual engagement?
Early humanities computing: numbering, counting, indexing approach.
Such work integrated, in dialogue with pioneering work of librarians: such as English Short Title Catalogue.
We are now very used to digital editions, settled approach...
Raises the question, why digital transformations? and why now?

Well, we're not just making more digital stuff now...
 - we are recycling, reusing, linking.
 - we are being more experimental, more ah-hoc.
 - we now have born digital data.
 - but above all, we can't read everything we need to read or want to read.
  > could conceive to doing so for Gladstone's 150k+ letters, but not GW Bush's 200 million emails.
  > where do we event start with that?

We need new methods to deal with size, and different types of data.
> more metadata in a tweet than there are characters.

Linking data now normal.
Reusing well structured data (such as Old Bailey) now normal.
Considering artistics designed databases as a way in is now normal.

Links with world of art help us think better about the materiality of data


*Mahendra Mahey & Ben O'Steen, Labs One Year One*
Developing data.bl.uk
Data exhibitions
Mechanical Curator: images we have no metadata for, that we've surfaced for the first time, and that we intend to generate data for.


*Pieter Francois, Sample Generator for Digitised Texts*
Sample Generator.
BL collections as a curse...
Narrowing down body of knowledge needed.
Maybe once we just picked through this in an ad-hoc way...
- Now epresentativity of data more important.
- And humanities changed.

Key points:
- citable
- documentation
- comfort with how digitised data links to total published text.

Part of the Labs, part of a team project.

Basic facts of Sample Generator:
- digital tool
- looks like a catalogue
- creates samples using metadata
- universally applicable, but focus on 19th Century Books

Helps you visualise the fit between a digital collection and total holings.
Sample generated mimics the distribution of the total physical library holdings.

Unbiased, exportable, citable, replicable.
BUT cannot overcome any problems with the metadata (typos, double entries, noise).

Shift of bias: you take away the bias of the digitisation process.

Hypothesis testing tool.

Ben: Sample Generator does term vector analysis; it is a way of applying a search engine to a different way in which we are used to it.

How do I sell this? In a way the Sample Generator foregrounds problems that are normal in searching catalogues.

*Dan Norton, Mixing the Library*
A DJ uses an information flow: selects from an archive, presents through an authoring process between data.
Fluid interactions with information, at the intersection between information.
Functionality, an annotation of links between data, then publisher as a history complete with annotations.

*Adam Farquhar, Labs launch*
Interested in moving beyond using the digital as the physical
> to supporting a new range of digital research style, methods, strategies.

*Sharon Howard, The Digital Panopticon*
Goal to link together currently digital dataset with new tools on a scale not tried before.
Understand criminal punishment in the 19th century
AND develop new methods for digital research.
Link between Old Bailey and Founders and Survivors data, plus anything relevant we can get hold of.
Bring together commercial providers, public institutions, researchers, family historians.

Ethics and digital history > this project can democratise history.
BUT also maybe we'll have to reevaluate our own practice to making this kind of digital data so freely available.

*Randolph Donahue, Fragmented Heritage*
Archaeological sciences.
Deal with cost of exploring large sites: how to compare large numbers of objects?
Engagement with citizen science.

*Tim Crawford, Transforming Musicology*
Musicology: deep human understanding of music.
Emerging area of Music Information Retrieval: computational analysis and matching.
Sematic Framework: global exchange of cultural concepts, process of scholarship.
> this project seeks to connect these three together.
Aim to identify things that don't work as much as do: to seek value from failure.

*Bill Thompson, Technologies that will impact digital scholarship in the next decade*
Quick answer: all of them.
Good stuff...
Embedded digital scholarship.
Big data a management term: becoming meaningless and misleading.
Enormous amount of grappling with catalogues and metadata.
Caveats...
I don't like digital, I don't like content.
 > need a more nuanced vocabulary.
 > maybe we need some jargon, some barriers: too keep people out, so we/they can them overcome them...
Doesn't like predicting ten years: too far, too tricky.
Some things are pretty clear (if we don't blow it as human civilisation):
 - faster and more pervasive networks.
 - more powerful processors.
 - overwhelming cascades of data.
 - embedded systems.
 - intuative interfaces.
 - additive manufacturing.
 - personal data, ethics, in some direction at least...

What is the challenge for scholarship?
 - how do we collect, preserve, and get access (and finding out we'll have got it terribly wrong).
 - making decisions over deaccessioning: new media, new challenges.
 - what is scholarship? Adaptation and incorporating new media into assessment.
 - the book works and we like some of its affordances: but will it all do this and will something else be able to do it better? (and as sustainably?) ... factors combine to create a conservative atmosphere.
 - new forms of collaboration, better than being there, seamless skyping et cetera.
 - new things to be scholarly about... you can tell which AutoCAD was used just by looking at building plans - digital affordances shaping the built environment ... Game worlds.

BUT scholarship has always been in crisis, always changed, always failed to capture everything.
> real challenge to describe the word, build tools so we gain clarity, then rebuild new tools, et cetera.
> is this a new eco-system of scholarship rather than a new paradigm?

***This work is licensed under a Creative Commons Attribution 3.0 Unported License.***
<a rel="license" href="http://creativecommons.org/licenses/by/3.0/"><img alt="Creative Commons License" style="border-width:0" src="http://i.creativecommons.org/l/by/3.0/88x31.png" /></a><br />This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by/3.0/">Creative Commons Attribution 3.0 Unported License</a>
	Transforming Research Through Digital Scholarship, British Library, 11 November 2013
	========================================================================================

	Andrew Prescott, How Arts and Humanities is Being Transformed through Digital Scholarship
	Some of transformation unclear ... offer some hints on what is going on.
	Collaboration with institutions at the heart of the digitrans theme's development.
	BL site of transformative moments: text, sounds recording, post, biblical concordances.
	The latter were revolutionary, considered heretical
	> in essence cut up the bible and rearranged it into a different order.
	> is it a tool? act of philosophy? intellectual engagement?
	Early humanities computing: numbering, counting, indexing approach.
	Such work integrated, in dialogue with pioneering work of librarians: such as English Short Title Catalogue.
	We are now very used to digital editions, settled approach...
	Raises the question, why digital transformations? and why now?

	Well, we're not just making more digital stuff now...
	- we are recycling, reusing, linking.
	- we are being more experimental, more ah-hoc.
	- we now have born digital data.
	- but above all, we can't read everything we need to read or want to read.
	> could conceive to doing so for Gladstone's 150k+ letters, but not GW Bush's 200 million emails.
	> where do we event start with that?

	We need new methods to deal with size, and different types of data.
	> more metadata in a tweet than there are characters.

	Linking data now normal.
	Reusing well structured data (such as Old Bailey) now normal.
	Considering artistics designed databases as a way in is now normal.

	Links with world of art help us think better about the materiality of data


	Mahendra Mahey & Ben O'Steen, Labs One Year One
	Developing data.bl.uk
	Data exhibitions
	Mechanical Curator: images we have no metadata for, that we've surfaced for the first time, and that we intend to generate data for.


	Pieter Francois, Sample Generator for Digitised Texts
	Sample Generator.
	BL collections as a curse...
	Narrowing down body of knowledge needed.
	Maybe once we just picked through this in an ad-hoc way...
	- Now epresentativity of data more important.
	- And humanities changed.

	Key points:
	- citable
	- documentation
	- comfort with how digitised data links to total published text.

	Part of the Labs, part of a team project.

	Basic facts of Sample Generator:
	- digital tool
	- looks like a catalogue
	- creates samples using metadata
	- universally applicable, but focus on 19th Century Books

	Helps you visualise the fit between a digital collection and total holings.
	Sample generated mimics the distribution of the total physical library holdings.

	Unbiased, exportable, citable, replicable.
	BUT cannot overcome any problems with the metadata (typos, double entries, noise).

	Shift of bias: you take away the bias of the digitisation process.

	Hypothesis testing tool.

	Ben: Sample Generator does term vector analysis; it is a way of applying a search engine to a different way in which we are used to it.

	How do I sell this? In a way the Sample Generator foregrounds problems that are normal in searching catalogues.

	Dan Norton, Mixing the Library
	A DJ uses an information flow: selects from an archive, presents through an authoring process between data.
	Fluid interactions with information, at the intersection between information.
	Functionality, an annotation of links between data, then publisher as a history complete with annotations.

	Adam Farquhar, Labs launch
	Interested in moving beyond using the digital as the physical
	> to supporting a new range of digital research style, methods, strategies.

	Sharon Howard, The Digital Panopticon
	Goal to link together currently digital dataset with new tools on a scale not tried before.
	Understand criminal punishment in the 19th century
	AND develop new methods for digital research.
	Link between Old Bailey and Founders and Survivors data, plus anything relevant we can get hold of.
	Bring together commercial providers, public institutions, researchers, family historians.

	Ethics and digital history > this project can democratise history.
	BUT also maybe we'll have to reevaluate our own practice to making this kind of digital data so freely available.

	Randolph Donahue, Fragmented Heritage
	Archaeological sciences.
	Deal with cost of exploring large sites: how to compare large numbers of objects?
	Engagement with citizen science.

	Tim Crawford, Transforming Musicology
	Musicology: deep human understanding of music.
	Emerging area of Music Information Retrieval: computational analysis and matching.
	Sematic Framework: global exchange of cultural concepts, process of scholarship.
	> this project seeks to connect these three together.
	Aim to identify things that don't work as much as do: to seek value from failure.

	Bill Thompson, Technologies that will impact digital scholarship in the next decade
	Quick answer: all of them.
	Good stuff...
	Embedded digital scholarship.
	Big data a management term: becoming meaningless and misleading.
	Enormous amount of grappling with catalogues and metadata.
	Caveats...
	I don't like digital, I don't like content.
	> need a more nuanced vocabulary.
	> maybe we need some jargon, some barriers: too keep people out, so we/they can them overcome them...
	Doesn't like predicting ten years: too far, too tricky.
	Some things are pretty clear (if we don't blow it as human civilisation):
	- faster and more pervasive networks.
	- more powerful processors.
	- overwhelming cascades of data.
	- embedded systems.
	- intuative interfaces.
	- additive manufacturing.
	- personal data, ethics, in some direction at least...

	What is the challenge for scholarship?
	- how do we collect, preserve, and get access (and finding out we'll have got it terribly wrong).
	- making decisions over deaccessioning: new media, new challenges.
	- what is scholarship? Adaptation and incorporating new media into assessment.
	- the book works and we like some of its affordances: but will it all do this and will something else be able to do it better? (and as sustainably?) ... factors combine to create a conservative atmosphere.
	- new forms of collaboration, better than being there, seamless skyping et cetera.
	- new things to be scholarly about... you can tell which AutoCAD was used just by looking at building plans - digital affordances shaping the built environment ... Game worlds.

	BUT scholarship has always been in crisis, always changed, always failed to capture everything.
	> real challenge to describe the word, build tools so we gain clarity, then rebuild new tools, et cetera.
	> is this a new eco-system of scholarship rather than a new paradigm?

	*This work is licensed under a Creative Commons Attribution 3.0 Unported License.*
	<a rel="license" href="http://creativecommons.org/licenses/by/3.0/"><img alt="Creative Commons License" style="border-width:0" src="http://i.creativecommons.org/l/by/3.0/88x31.png" /></a><br />This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by/3.0/">Creative Commons Attribution 3.0 Unported License</a>