FoRC Rm Saarbrucken

David Shotton's Notes: Day 1

 

Anita:  The Past

See slides: http://www.slideshare.net/anitawaard/introduction-of-the-past-for-future-of-research-communications

Don’t take anything for granted - be open to new views, don’t just push your old ideas.

 

“Life is what happens while you are waiting for something else” - long breaks in conference for talks

 

The Past

 

Douglas Englebart, 1968: Hypertext  “Content represents concepts, but there is also a relation between the content of concepts, their structure, and the structure of other domains of human thought, that is too complex to investigate in linear text”  (echoes of Ayn Rand)

http://sloan.stanford.edu/MouseSite/1968Demo.html±player11

Demo of collaborative editing across internet

 

Hypertext: modular content components, meaningful links and claims > evidence

 

Gives history, via ABCDE Format of modular computer science proceedings papers, ending in LiquidPub 2010 Structured Knowledge Objects, HCLS Rhet Doc - medium-grained core narrative components, and DoCO!

 

Argumentation relationships - Simon B-S work

 > claim evidence networks - Hyper, Nanopublications, SWAN

 

But almost no content has been created in this way.  Why?

 

Four periods:

Pre-web online databases, hypertext

Web 1990 - 2000 Web, preprint servers, standards

2000-2005 Semantic Web, open access

2005-2011 Social web

 

1990 - 2000 standards - XML, PDF, MathML, et

2000- RDF

 

What has stuck - Word and PIDF - coercion interests

XML html - shallow learning curve - high functionality

RDF - semantic message

social media - simple tools to express basic human need

 

Business models

Up to 1980: Publishers sold to libraries

1990 - ArXIV,

2000 BioMed Central, PLoS, Creative Commons 0 author-pay

2005 o open data movement

 

But commercial business model engrained

Societies and author-pays models also become publishers

Indignation drives Open Access - but also have a day job - just need to publish in Science and Nature

 

Research data

1990 - digital repositories, JSTOR, Cite seer

2000  Workflows and Grid - Taverna and myGrid

2005 - Dataverse, DataCite, MyExperiment, Data journals

Workflow tools not yet ubiquitous - why not?

 

Attribution and credit

Until 1980s - impact factor

Since 2000 H-Index and Google scholar, but not much has changed

 

Factors driving support

Commercial publishing

Community support  LaTEX

OPEN ACCESS

Ease of use - Google Scholar,

Academic credit depends up t - Grant proposal forms complex and non logical, yet “essential” so people do it

 

Simon Buckingham-Shum   Net-centric scholarly discourse?

 

We construct knowledge through language and discourse.  How to go beyond journals?

Vannevar Bush 1945 “ We may some day click off arguments on a machine with the same assurance that we now enter sales on a cash register”  Computable arguments

2001: ScholOnto Project “IN 2010 will we be publishing results primarily as prose papers?”

 

Scientific infrastructure would permit asking questions - what is the evidence for this claim?

Was this prediction accurate?

Who build on this idea?

Who challenges this idea, and with what arguments?

are there distinctive perspectives on this problem

 

Systems:

1665 throws a long shadow - Le Journal des Scavans, Phil trans Roy Soc > networks

 

Any community of enquiry is a complex adaptive system responding to a changing environment - physarum - absorb and reorganize in response to external disturbance

 

IS CURRENT SYSTEM NOT FIT FOR PURPOSE?

HCI and sense making -

 

see slides

Higher quality of conversation within scientific literature

Making formal and informal connections

Sense making is about plausibility, coherence and reasonableness

We construct meaning ins scientific discourse  there is rarely one master world view

 

Tools assume a wealth of documents, plus online tools and analytics, making meaningful interpretations and connections between information elements

Start to tell a story linking statements / publications

 

 

Empirical studies of users, inform user interface design.

 

Example of analyzing blog on nuclear power

 

Collaborative web infrastructure for structured deliberation

Cohere visualization of semantic annotations on publications

 

OpenEd Evidence Hub

 

Agnes Sandor, Xerox - human and machine annotation of discourse

Cohere + Utopia  - annotations in Utopia

 

David Rosenthal - Finance

 

Follow the money. Research publication is a multi-billion dollar industry

 

HighWire pioneered on-line publication

Libraries switched from purchase to leasing info online

Cancellation lost all access to past materials - post-cancellation access

Publisher goes out of business = Preserving the record

LOCSS project  Peer to peer network for preservation

 

NSF funding, Mellon Foundation additional funding

 

LOCSS went into production all five other archives failed

Negotiations time consuming and expensive

Concept resurrected by Phil Bowen as head of Mellon > Ithaka with $55m Mellon cash

Portico front end

 

Outsourcing library collections

Charging foo open access content 

Close connections with Elsevier

 got all Elsevier content

 

Portico became selected repository, but has failed to get sustainable funding

 

Libraries unwilling to pay enough this funded in with JSTOR

 

 

National libraries - legal and technological barriers have prevented them from fulfilling role, except Dutch, who have no legal mandate

 

Publishers set up own archive - dark archive of journals just in case > CC republication

CLOSS  2007 this has been the future of LOCSS

 

Driving costs out of system

Would not have survived if libraries has not found technology useful for preserving other stuff

 

Digital copyright deposition is not required.

 

Hits to your own site have value, even if from non-subscribers

 

Portico leaks content slowly to library = Ithaka will own all academic content which they can later monetize, as does JSTOR

 

Value of post-cancellation access dominates field - expensive content

But Elsevier is not going away

Stuff that is at risk is open access stuff from minor publishers - funding this preservation is much harder

 

Lessons: Funding for long-term preservation needs to be buried in something that delivers access in short term

Increasing returns of network effects so that quickly become monopoly - non-profit monopolies harder to dislodge

Everything we are talking about is copyright

Copyright is totally broken 0 but control systems that publish, preserve and manipulate contend broken

 

No open access publisher has sustainable business model - including PLoS

 

 

Paul Groth : Current status

11:10

Papers on paper, or online

Mendeley, Endnote

Find things on Google Scholar, digital libraries

Papers by email - PDFs not self citing

Twitter feed > URL > clack and download

 

Conferences - slides and talks, Lanyard.com, Slideshare

 

Databases - UniProt etc.

 

Informal - twitter, blogs, beyond the PDF discussion forums, mailing lists, e-mail

 

Measurement - PMetrics - through journal articles only

 

Phil Bourne: 

We have a share vision, but have we a shared goal.

“ A little history of the present”

 

Push a goal form Beyond the PDF re SMA - happening slowly - everyone doing what they have always done, not towards the shared goal

 

What can Phil do towards the shared goal

 

Some shifts: perception of what is scholarship is changing v slowly

Now a committee for how to measure scholarship at UCSD

But young faculty members far less focused on improving scholarship, and more on Nature papers

all about numbers and quality journals

 

SciVerse is a game changer - iPhone changed the way we do software apps all operate off same interface

 

SciVerse similar over shared content base - extend into data content world

PDB provides data without services, that are left to user, independently

Need to create app style model over data . easier to use

 

Data Journal?  Workflows critical? 

Wellcome and Highs and Max Planck announcement about a data journal - not so dumb

 

Must have more regards for publishing data - get a proper citation out of it - PROPER data citation

 

Institutional repositories - roach motel - data checks in but not easy to get data out!!!!  Not v useful

 

Data journal - database maintainers become publishers

 

Phil blogged this and got lots of response

 

Technically possible, but is it a good idea. 

 

Executable paper idea on top of this.

 

Use data journal to control a software environment

 

Workflows  Yolanda Gill as USC asked to take info

Tried to create a workflow - v hard to reconstruct

Need tools to create and record workflows as part of scholarly record

 

Publish this with the workflow.

 

Fiona Murphy Impediments to Data publication

 

Impediments to integrating data:

Technology: people collect data ad hoc and postdocs leave - no deposit policy

Provenance lacking

Interfaces unhelpful British Atmospheric Data Centre - interface prevents people doing what they want

Format issues - which will survive and which will create black holes - laserdiscs

Funders: mandates but not incentivizing, toothless demands for compliance

Researchers: Suspicious of sharing their IP - keep data for reuse

ClimateGate people v worried and have gone underground to avoid exposure

Researchers lack time to do non-essential data management - at end of project looking towards the next project

Building research team - data scientists don’t have career structure

Publishers want to help, but not clear where to go and what to do

Cannot see benefits of publishing data themselves

Tough time for library subscriptions

Used to deal with librarians, and have lost touch with end users

What do researchers need from publishers - need conversation

`subscription model brings in money - why change it?

What expertise is needed for new data products, what is business model

Not used to building partnerships with funders and policy makers, or even other publishers to provide industry-wide solutions

 

Who is responsible for data  where does it sit?  How to link to article?  Publishers don’t want to clutter sites with huge amounts of data

 

Michael Kurtz:  11:40

 

myADS - Collaboration with ArXiv at Cornell

Not refereed journal with the best paper in the field - for 10 years, 8000 subscribers

 

Number of articles per year 3.7% increase - not sustainable, so new tools for filtering, specializing

New publication models that don’t take this into account will fail

 

History:  1987 Astrophysics Data Systems workshop

 

“Urgent need for a master directory of all NASA observations linked up to ground based data

Still not achieved. Not easy to do. 

 

None of this to do with literature - all about data

Early work on own browser before Mosaic, query over two databases at same time

 

Goes through developments

Now integrates data and journal articles

 

Click through to one paper - can link to full text of paper or to data products, including archival data > list of sources of data for that paper. ‘

 

Click through to table of data

> raw telescope images

> dataset references to ‘reduced’ data

 

Can see what data used for in past

 

Steve Pettifer  Utopia

 

Can liberate date from tables, export to excel, etc.

Can extract images and browse through them

Play back video behind image

Overlaying on PDF, not embedded between, so further readers can see comments

Trans-membrane receptors re-visualized according to scheme, by pulling info from database, on the fly

 

Simple API

 

Ref linking to Science Direct

Will pull bibliographic info from Science Direct > open cited article in Utopia without having to download PDF

 

Reconstructions document using DoCO

Completely automatically no training

Useful for text mining app e.g. just of body text

 

Use this for PDF reader for visually impaired

 

Cohere:  Assertions in text seen in Utopia > Cohere graph of assertions

 

Mendeley and Open Citations

 

Open PDF and goes off to Mendeley and open citations - > related articles, pull back info from Mendeley, > related articles

 

On the fly generation of citation networks on the fly from open citations, and can link through to other articles on node

 

Todd Vision: Capturing the long tail of orphan data

 

Most researchers data not published in useful way that gets credit.

Policy effort among journals in ecology and evolutionary biology

 

technically unimpressive

 

No undue burden on authors or journal in terms of semantically rich metadata

 

Low hanging fruit - added value

 

Low barrier to submission - link with MSS processing system of journal - back passage message passing freeing researcher from metadata entry.

DOI assigned to data.

 

Could imaging additional layering on top of archived data

 

Compromise approach - red hearings: Researchers wont bother.  25% uptake when voluntary, 75% when mandatory

 

Free themselves of burden of data archiving and management

 

Storage needs not overwhelming so far - on average one big spreadsheet per paper , < 1 mbyte per paper

 

Spreadsheet will not be reused:  several dozen data packages downloaded 100s of times.  1000 data sets per year

 

Costs real and not overwhelming - up front payment $50 per paper for long-term storage

Trivial relative to supplement al material rates and cost of publication, and orders of magnitude less that cost of creating data in first place

 

Want data to be treated as first class objects, and be given credit

 

Want to support data citation, but mechanisms to track usage not yet created.

 

Cameron Neylon - Research communication is different from paper publication

The paper is not the research, but a (poor) representation of the research - we have been talking about making it a better representation

But can we communicate THE RESEARCH

Data does not spring fully formed from research = workflows required as well

In wet lab: mixing solutions to put into a machine in hope it will spit out a useful spectrum

Don’t forget to document the early research activities, materials,

Process of authoring and reviewing is v expensive

We could publish the research object itself - could be done at lower cost

 

Herbert van der Sompel

 

Need framework of thinking for future discussion

4 functions:

 

Registration - claim a finding - time stamping

Certification - peer review etc.,

Awareness  - discovery

Archiving - how do we hold on to all this stuff

 

Derived function that drives everything,

Reward - based on the impact factor

 

(Christine Borgman has written similar stuff more recently)

 

The above is an abstraction of the journal system - do they still apply to what we are planning to create?

 

 

“Looking at the future through the rear-view mirror”

 

Look at network effects that were not present before.

 

Judith: Have never talked about databases, that are the primary reference points for most people. Most people’s career are independent of journals except where necessary for their career.

 

Phil: We need some goals:

1  Intra - things we could do together, that might be unrecognized by the lab researcher

2 Inter - the Dryad and Utopia that can impact the external community in a huge way

3  The interface between these two - influencing policy that could change the status quo.

 

Looking at professional science, separate from society at large

Royal Society investigation is about connection to society

 

David de Roure:

 

Also bringing people together - building the social graph between scientists

 

Informal communication  - spectrum -  formal publications.

discussing, collaborating, reuse, reproduce

 

 

Requirements of scientific discourse that do not apply to normal discourse.

 

How to validate scientific information to society on the Web

 

Anna de

 

Hard science concentrates on databases

Other approach is scientific discourse - why is A citing B?

 

Proposals for working groups;

 

3 sessions: presentations on Thursday am

 

5 rooms Here, nothing

Karslruhe donwstairs 4

S003, 1

S004, 2

S105, 3

 

Requirements _ two reports from each group = a manifesto - recommendations to funders, government, publishers, libraries

 

vision - impediments - how to overcome

 

topics:  Research data and code - repositories, data papers, executable codes, workflows + New forms and tools -  nanopublication, text components, Utopia, chemistry tools, authoring tools, How to create value chain across all these

 

Assessment and impact - rewards, citations.  New ways to assess science

 

Business models - Ed and David - How to make sustainable systems

 

Social platform (citizen science and crowdsourcing) , outreach to society, and continuity

 

Tim Clark: how to build an ongoing ‘college’

 

Need to ground stuff and get good uptake from researchers

 

seek some funding for and construct a web community for continuing FoRC and friends - funding for travel for collective experiments

 

Todd wants to vertically integrate over all topics for a new intervention

 

 

Breakout group 1: topics:  Research data and code - repositories, data papers, executable codes, workflows + New forms and tools -  nanopublication, text components, Utopia, chemistry tools, authoring tools, How to create value chain across all these

 

Paolo: How can we use this meeting to start to work together and leverage one another’s work

 

Gully:  How do we effectively collaborate. Steve makes himself available and gets it to work 

 

Steve: Just blundering around

Phil: Terry on the practitioner side, while Steve does the technology - starting with lack of understanding of one another’s domain

 

 

Paolo: how to integrate across groups - Steve just “we’ll do it”

Problem is first understanding what the different groups do, how to recognize our own boundaries,

 

What we want:

Paolo: Integrate with Utopia both ways, Use gully’s model, use Open Citations, Citation of data.  Need citation of workflow. 

 

[Need API over Open Citations.]

 

Judith Blake:  Bring; works with publications Need to dig data out of publications.  Need markup to identify bits of paper>  Utopia could do that for her and enable markup.

 

Gully:  Do: NLP ontology systems,  want to reason over the data.  The act of scientific discovery is key p building a breakthrough machine.  Text mining in progress with Judith in progress

Wants to do : create collaborations in which his stuff gets used. 

Challenges - not invented here.  How to use other people’s stuff more easily. 

 

David: Data management tools and SPAR applications:

Need open citation information, expert markup for open research reports

 

Steve Pettifer: will collaborate with everyone round table by end of year

 

David de Roure - myExperiment - workflows

People want to work flow bundles into packs - research objects

Workflows4Ever - focus on provenance of research objects

 

Carole Goble: Research Objects lifecycle  not expecting all items to be in one place. - SysMO project

Other systems for methods, scripts, SOPs, ISA infrastructures,  store data spreadsheets, etc. and models.

Need curation at data creations time

Built metadata index to other peoples databases

 

Todd: Model for ingesting heterogeneous content.   has a way to test out stuff with collaborating scientists.   What would enable research if data were . . . .

 

Anne: Cohere tool for allowing collaborative annotation on documents - not following a particular ontology - How to do higher level reasoning on data, using ontological markup

How to support open annotation - understand collaborative dynamics of human and machine annotation

 

How to support this human_ machine collaboration and higher level reasoning about connections between concepts . . .

 

Cohere imports XML or RDF - wants to suck in annotations for analysis for cognitive reasoning.  How to improve visualization.

 

Phil:  Collaboration network embarrassingly small. and don’t produce anything anyone can use.  Have PDB!  Have literature = database integration in house.

 

Todd and DataBank both accept minimal metadata. 

 

Gully: what is purpose of it all - ontologies concerned with the correctness of the representation.

 

We are not creating domain ontologies - but rather generic ontologies that can be shared.

 

Link Open Citations with e-Journal club

F1000 is successful

 

Interactive journal club is a learning procedure

 

Paolo: some people don’t accept summarization of a paper

 

Carol: what are the bottlenecks in research:

 

Bottlenecks;  Information overload and choice of best papers

Todd uses many tools for this

 

?? citation numbers

 

Gully:  IN genetics, lots of infrastructure already in place

Neuroendocrinology  no databases - have to read the papers

Different domains have different levels of informatics support.

 

Also, everything is interdisciplinary, so domain specific is not enough

 

Read want to read the papers that are just at the intersections of two domains -

 

Judith - papers not correctly tagged in PubMed

 

How to crowd source annotations - but also issue of authority and choice of vocabularies - folksonomies must be open

 

Other bottlenecks

 

Carole: Crowdsourcing only works with specific tasks, and need funding and organization

 

Other bottleneck - knowing what is going on in my own lab

 

Preserving from a multi-site multi-community project all results, so can continue after the project.  Quite complex for multi-site - staff turnover. 

 

People happy to share point to point, or publish to the world, but not to share data to the world - no reward for this, and many potential penalties

Todd not using word “sharing” 

 

No policing and no rewards\\

 

80% of all clinical trials mandated by law to publish data do not - no one fires them!

 

Phil:  But things are changing.  Your papers have to be in PubMed central.

 

Carrot: et an extra citation

 

Prisoner’s dilemma - need everyone to do it for the common good

 

Bottleneck: Start-up time for new people in the lab - too much information to acquire - the time has not changed over time, but the pressure on productivity has increased, so hire a postdoc who will be more productive.

 

Research data management still mostly paper based and non=collaborative

 

EverNote would have helped 20 yrs. ago if useful thinks built on top

 

If Phil had something that could digest lab notes and suggest what new papers to read in the morning.

 

Tools that are too specific, never do what YOU want - cannot customize

 

How to bring all people together:  What is our connection diagram

 

Domain names

FoRC not so good

Force *** better

 

Future of Research Communication and E-scholarship

 

Force11.org free

ForceEleven.org and Force-Eleven.org  available

 

ScholarForce.org available

ScholarlyForce.org available

 

ForceInternational.org available

 

OpenForce taken

 

Day two

 

Domain names

FoRC not so good

Force *** better

 

Future of Research Communication and E-scholarship

 

Force11.org  free

ForceEleven.org and Force-Eleven.org  also available

 

ScholarForce.org available

 

ScholarlyForce.org available

 

ForceInternational.org available

 

 

Ed Hovy: The Future

Issues:  data deluge

 

, repeatability, citability, mashability

algorithmic assistance to researchers

Web of data and web of documents

Walled garden

 

[See slides]

 

Slide 12 additional:  Argument, terminology, claim and evidence

Your argument is part of broader discourse - need to cite others

 

Background Claim Citation of evidence is citation of other paper.  Cant get transitive closure of claim right back to evidence through papers

 

Results claim cites evidence in paper - figures and tables, methods

 

Must be cognitive of these patterns. 

 

Scientific argument is finding something de novo and fitting it to what is already known

 

Robert Dale:  What is it we mean by scientific communications - do blog posts may exclude evidence and just be argument!

 

Judith: Exploratory and descriptive projects (microscopy) now don’t get funded

 

Ed Hovy: What I’d like to get from Publications in the Future

 

A disruption algorithm: 

Vision to improve the communication of knowledge between scholars using new technology

Why do we publish: record new knowledge with time stamp, to communicate, to permit people to measure me

 

In future, need to provide better access to knowledge,

 

to a=permit better knowledge communication, and

 

better knowledge verification and extension

 

WHY DONT WE DO IT TODAY?

 

Why would our vision sick?

 

When PoF is easy to produce = learn and use tools quickly, capture all aspects of knowledge

Easy to consume - ingest results faster and deeper

Cheap

More effective - readers get more info, more accurately - ditto for reviewers and tenure evaluators

Cool

ONE REMOVE BLOCKADES AND GIVE BENEFITS, JUST DO IT

But we don’t have clear vision, clear statement of why, understanding of impediments and cost, understanding of procedures

Goals of meeting

List and organize methods

Study impact and effects - why beneficial

study financial implications

Work on how to make it happen

 

The Report

Improving vision

Vision 

Technology

 

Impacting our world

Social aspects

Coolness

 

Overcome obstacles  -

Financial

Get ball rolling

 

Benefits

PoF:  Contextualizing The Web of X

X = ideals, theories, papers, social network, research network

 

 

e.g. Stephen Wan’s work

Reflect

Active pdf Greenacre and Hastie

 

Put non-text info into text - data movies, images, data collection software

 

Who will maintain, update and extend

Who will do this - publishers are achieves

 

Different fields differ - must allow for these

construct paper to fit knowledge domain

 

Discourage free-form argumentation and bullshit

Understand intellectual framework specific for discipline

 

e.g. KEfED model for biomedical experiments - gully Burns

Input and output dependent and independent variables

 

If you can find pattern, can help people right

 

Ethics: balancing concerns - electricity and nuclear power = precautionary principle

 

Time: Evolution of music over history

Monteverdi > Bach > Beethoven

Evolution of understanding of harmony: horizontal melodies > vertical chords

Foundation: timeline

 

Impacts of PoFs Society

What are we paying for

Reviewing, distributing

Google is disruptive

Blogs, avoid journals

What next between producer and consumer?

What jobs remain, and who will pay for it?

 

In long-term, will code work, data grows in incompatible ways, formats go out of date

 

Someone has to do the work

It would be good if we know the individual costs, rather than bundled costs

 

Both consumer and producer needs to pay

 

Where to go from here

Paper light (free)and paper rich (semantically enhanced, costs $$)

Get publishers involved - show them roles

Get people involved

 

= = = ==

 

EXCELLENT!  Get slides.

 

Udo Hahn: no better system than peer review.

Simon: Paper discussing a different type of pattern: Evaluation of software system has these points  . . . methodology, users, etc.

Different disciplines use different paradigms of argument

 

Who can access communications - subscription barriers

How easy are they to use

What is happening - concentrating on journals, forgetting about books and monographs

 

Anita: Manifesto light and manifesto full!

 

Leslie Chan: More that just putting stuff on the Web - need to use the potential of the Web.

 

Ivan: Discussion concentrates on the journal.  But historians write books.  We do not have that in mind.

 

Ed: Model of book + CD + web site + iPad app - book by Al Gore on environment

 

Judith:  Put aside the idea of journal - the future is the internet

Argument structure and templates > web

eBooks can have versioning

Trip Advisor & Google Analytics - each book has ranking by community - measure of value

 

These give value without subscription - look at different financial models for journals??

 

David R:  Concentration on journals: this world is about to get disrupted.  University press books is much more disrupted - no one wants to pay for stuff humanities writes, since none wants to read it!

 

Need to understand what is wrong at moment - opportunities for disruption

 

Peer review working worse - increased volume and workload - wrong rewards.

 

so look at what is wrong now.

 

Robert Dale:  Need to decide what product category we want to describe

 

Gully: get away from “Paper”.

 

Carole: Concentrate on long term preservation

 

Can we read it in 40 years time - complicated on-line artifacts

 

Ed: users will convert to modern format if important enough to read

 

Majority of science not worth saving

 

Phil:  NY Times - smart people could not look beyond 5 years, building on what we know

Totally disruptive: Knowledge is a stream > river > sub-disciplines to permit usefulness

But now moving to stream - rolling flow of knowledge in on-line news, on-line newspaper, evolution of Wikipedia - not individual editions

 

David de Roure - The Execution of Dave 2.0

 

Coshh assessment of making tea - process

Taverna workflow for bioinformatics data analysis

Other types of workflows too - enable dealing with data deluge

 

Help to reproducibility, and description of process itself

 

Flickr, YouTube, Slideshare, myExperiment - social media

 

myExperiment pays attention to license, credit, attribution

Need to record not just workflows but inputs and outputs - > package research object

Describe relationships between components

 

R dimensions

 

Reusable

Repurposable

Repeatable

Reproducible

Replayable

Referenceable

Revealable

Respectful - provenance, IP, lineage

 

Yin Yang - data and method/process/execution

 

www.methodbox.org Iain Buchan

Executable thesis: Now throw new data at it and get new results

 

e.g. Dexy

 

Executable code within document

 

Spreadsheet is the same thing

 

Computable Document Format  CDF  from Wolfram

 

Executable Paper grand Challenge

 

Winner: Collage:  Cloud executable

 

The Executable Journal - new role for the scientific publisher??

 

Executable papers

 

then strayed into science fiction of automated creation of new scientific results - machines are users too!!

 

Gully:  How does this help knowledge synthesis - bringing stuff together into a coherent framework

 

DDR:  This is the biggest impediment

 

Cameron:  Cost of discovery of publications is major point -

 

Simon: People good at argumentation reasoning - let computers get on with the simple things

But people suspicious of automation - is that attitude changing?

 

Many people go straight to the data, rather than waiting for the publication

Much research is “mundane” discovery - hard and intellectually challenging precursor to the “ah ha” moment - intellectual slog

 

Tim: Want to be able to cite workflows as well as data

How to do this? 

 

Extension of CiTO / FaBiO    fabio:Workflow exists

PROVENANCE?  Leave that to the Provenance vocabulary

 

fabio:ResearchObject does not - we decided it was too fluid to be defined, and that ORE covers this

 

ISA hierarchy - useful, and defines SPOs too - all packaged

 

Tim Clark: ALSO ASK PEOPLE TO PUT IN THE CONCLUSION - AS IN MIIDI

 

Cameron: I want to play in the lab, capture that, then point others to the data - need provenance model in retrospect, not in advance

 

Gully: differentiate observations and interpretations

 

DMS: Conceptual models, presuppositions - need to be aware since these determine observations you make

But observations will always be true, interpretations will vary - need to distinguish, so that others can re-interpret

 

Steffan Decker FoRC

 

Changing to web of knowledge - science only part of this

Linked data happening right now

Services for search and collaboration and mining

Applications and commercialization

 

http://lod-cloud.net/

 

How does the real world interact with science

 

Where are the sweet spots between paper and web?

 

Qrate.me - a smart way to use your smart phone

 

Scientific posters at conference - read a URL from a QR code

Scan the poster, tweet it, Google etc.

 

Mendeley:  video

How to create a QR code

 

Provides universal interface for everything we are talking about

 

- slide presentations, etc.

 

SmallWorld demo UIST 2011 presentation

HiVE testbed

 

Creation of museum of world of warcraft !!  HTML links into 3D world and vv

 

?? CLAROS virtual museum

 

We need clear vision of where to go

More in society than research - research data only part

 

“Tourist attractions” along the route: e.g. 3D navigation and QR codes

 

Kinect rendering of 3d world

 

Phil:

Make solid models of virtual 3D things

 

Mike.rogers@ec.europa.eu  Toward Horizon 2020  Role of scientific data - a survey

 

EU has brought many benefits

“We threaten people with money!”

 

“Innovation” is the buzz word, including innovative ways of recycling stuff we have already and of collaboration

Include on every page of application

 

Horizon 2020:  Framework program for research and innovation

Incorporated European Institute for INNOATION AND TECHNOLOGUY EIT

 

INCREASE IN BUDGET 46% = Euro80 billion!!

 

Excellence in the Science Base

Research infrastructure that other people can use - lots of solutions to many problems

 

Will finance infrastructures, more money up front

 

By end of year, words in place - we need one line in place describing this area

 

New member states need to catch up

 

Lots on global cooperation , esp. developing countries, regardless of geographical location, synchronized calls with USA, etc.

 

Now in listening mode - now tall EU formally what comes out of this workshop

 

Open Scientific consultation - number of questions asked but want to hear what people wish to say

 

EU budget funded by VAT - so some questions designed for man in street and simplistic

 

Avoid taking the science out of life

 

“push for improved knowledge circulation in the European Research Area”

 

Willing to listen to international set of opinions

 

 

Now gathering opinions and formulating projects - getting language in place that defines legislation

 

European Parliament increasingly interested in Science and Technology

 

Keep your Euro MP in the look - they are very influential

 

Leibnitz who run Dagstuhl have office in Brussels

 

Need to show evidence of benefits of research - driven by bean counters

 

Keep in touch with project officer, can adapt plans accepted  - flexibility is OK

 

Lots of ways to fund individuals

IRSES international Research staff Exchange Scheme - funds for exchange visits - application

 

Fund 5K ph. ds per year

Research projects FET program

Marie Curie Actions are small and manageable for individuals

 

http://www.ec.europa.eu/research/csfri/