Behind the courtain of a paper: Interdisciplinary research from the idea to dissemination

Behind the courtain of a paperInterdisciplinary research from the idea to dissemination

Federico GobboFGobbouvanl

Dublin Institute of Technology 7 november 2015

1 of 40

Introduction

2 of 40

My main affiliation since Feb 2014

My official web page in Dutch at the UvA3 of 40

My research triangle

httpfedericogobboname4 of 40

My academic life in one slide

MA in Media Studies (1998) PhD in Computer Science (2009)

5 Universities (Italy the Netherlands China) in 10 years of(almost) full-time career

my first publication in 1998 was in language acquisition

in the last 10 years approx 100 publications in different fields

publications in different languages (English Italian Esperantomainly)

co-authored works with mathematicians philosophers economistsengineers

Interdisciplinarity is a practice

5 of 40

My academic life in one slide

MA in Media Studies (1998) PhD in Computer Science (2009)

5 Universities (Italy the Netherlands China) in 10 years of(almost) full-time career

my first publication in 1998 was in language acquisition

in the last 10 years approx 100 publications in different fields

publications in different languages (English Italian Esperantomainly)

co-authored works with mathematicians philosophers economistsengineers

Interdisciplinarity is a practice

5 of 40

Computer science hard or soft

Media Studies are part of the humanities so this is my academicheritage Where are the boundaries of the humanities Terencersquosmotto humani a me nihil alienum puto (nothing human is strangeto me)

Computer Science is a peculiar beast if you deal with algorithmscomputability theory computational logic (theoretical part) as wellas the hardware technicalities per se (eg embedded systems)measures of the quality of the software it is a hard science Hardsciences are driven by strongly motivated mathematical modelsHuman beings play no role

Whenever we consider the human-machine interface inComputer Science we are (also) inside the humanities

6 of 40

Computer science hard or soft

Media Studies are part of the humanities so this is my academicheritage Where are the boundaries of the humanities Terencersquosmotto humani a me nihil alienum puto (nothing human is strangeto me)

Computer Science is a peculiar beast if you deal with algorithmscomputability theory computational logic (theoretical part) as wellas the hardware technicalities per se (eg embedded systems)measures of the quality of the software it is a hard science Hardsciences are driven by strongly motivated mathematical modelsHuman beings play no role

Whenever we consider the human-machine interface inComputer Science we are (also) inside the humanities

6 of 40

Outline

1 the idea how to choose the good one

2 why to collaborate with others

3 how to collaborate with others effectively

4 dissemination of results copyright issues and social networks

We will see an example throughthe case study analysis of a published paper

7 of 40

Outline

1 the idea how to choose the good one

2 why to collaborate with others

3 how to collaborate with others effectively

4 dissemination of results copyright issues and social networks

We will see an example throughthe case study analysis of a published paper

7 of 40

The Minimal Levels of Abstraction in the History ofModern Computing

DOI httpdxdoiorg101007s13347-012-0097-0

Part one the idea

How to choose the good one

9 of 40

We are in the Information Overload Era

ccopyhttpwwwmakemarkcouk

Nothing new under the sun

In 1945 Vannevar Bush physician and mathematician already in theteam in the construction of ENIAC publishes a paper As we maythink where he addresses the problem of the externalization ofhuman thinking through the new technologies

Taxonomies and classification systems are artificial they get old verysoon an item can often be classified under two different branchesHuman minds do not classify they work by associations

Keeping trace of our associations is the only way to have an externalmemory which can be shared among different individuals

11 of 40

The Memex in use

source Bush Vannevar (1945) As you may think12 of 40

Lesson learnt n1

Check if someone had already published an

elaboration of your idea

if yes discard it

if no proceed and elaborate

What is a virtual machine

ccopy httpyvettetechiecom

Our case study

The idea started from an observation and a conversation betweenthe two co-authors

Observation we talk about virtualization and abstract machines(eg VMWare)

Conversation how to find a minimal number of levels ofabstraction in order to define a computer

Strangely enough we found that nobody had done it before

15 of 40

Our case study

The idea started from an observation and a conversation betweenthe two co-authors

Observation we talk about virtualization and abstract machines(eg VMWare)

Conversation how to find a minimal number of levels ofabstraction in order to define a computer

Strangely enough we found that nobody had done it before

15 of 40

From the abstract of our case study

From the advent of general purpose Turing-completemachines the relation between operators programmersand users with computers can be observed asinterconnected informational organisms (inforgs) henceforthanalysed with the method of levels of abstraction (LoAs) risenwithin the philosophy of information (PI) [ ] (my emphasis)

source Gobbo amp Benini (2014) Philos Technol

16 of 40

Part two

Why to collaborate with others

17 of 40

How NOT to do research nowadays

ccopy Richard Linderum

The opportunities of interdisciplinary research

researches with different backgrounds can offer new perspectivesto old problems

sharing ideas often helps to see them more clearly

people belonging to different academic communities can findpotential readers that shape the idea into a form suitable forpublication (maybe more than one)

it is more fun you never get bored of your research

19 of 40

Lesson learnt n2

Research results live inside (at least) one

academic community

if your idea is interesting for an academic

community go for it

if not try to reshape it in new terms

referring to the current literature in the field

From the abstract of our case study

From the advent of general purpose Turing-complete machinesthe relation between operators programmers and users withcomputers can be observed as interconnected informationalorganisms (inforgs) henceforth analysed with the methodof levels of abstraction (LoAs) risen within the philosophyof information (PI) [ ] (my emphasis)

source Gobbo amp Benini (2014) Philos Technol

21 of 40

Our case study key concepts for the academiccommunity

inforgs interconnected informational organisms

method of levels of abstraction (LoAs)

Philosophy of Information (PI)

All terms come from the British-Italian philosopherLuciano Floridi (Oxford)

22 of 40

Our case study re-definition of common terms

What does lsquocomputerrsquo mean We limited our analysis on standardmodern digital computers based on Von Neumann Machines Youshould always avoid ldquofuzzyrdquo terms so to prevent possiblearguments against your line of reasoning

A good example in the paper is the notion of observables which isphilosophical Another good example is the notion of category whichis used within a precise mathematical theory ie the category theory

23 of 40

Lesson learnt n3

Pay attention to the key concepts

link your key concepts to a renowned

academic tradition

redefine common terms in a inequivocable

way to prevent possible arguments

(You can also prepare the counterarguments in advancethat you will put in the paper this is a bit old-fashioned)

Lesson learnt n3

Pay attention to the key concepts

link your key concepts to a renowned

academic tradition

redefine common terms in a inequivocable

way to prevent possible arguments

(You can also prepare the counterarguments in advancethat you will put in the paper this is a bit old-fashioned)

Back to the father of AI and Computer Science

I propose to consider the question ldquoCan machines thinkrdquo Thisshould begin with definitions of the meaning of the termsldquomachinerdquo and ldquothinkrdquo The definitions might be framed so asto reflect so far as possible the normal use of the words but thisattitude is dangerous If the meaning of the words ldquomachinerdquoand ldquothinkrdquo are to be found by examining how they arecommonly used it is difficult to escape the conclusion that themeaning and the answer to the question ldquoCan machinesthinkrdquo is to be sought in a statistical survey such as a Galluppoll But this is absurd Instead of attempting such a definitionI shall replace the question by another which is closely relatedto it and is expressed in relatively unambiguous words

source Turing (1950) Computing Machinery and Intelligence

Part three

How to collaborate with others effectively

26 of 40

Avoid known pitfalls with co-authors

pitfall you need a shared domain language ndash sometimes thesame term is used in different ways according to the academiccommunities (think about words like lsquoevolutionrsquo lsquosystemrsquo orlsquolevelrsquo) If you do not have agreement on this it will be easy toattack you

solution prepare it well in advance especially if you do not knoweach other

pitfall you risk that down-to-earth researchers find your papertoo unconventional eccentric and they reject it because they donot understand all the technicalities

solution address the paper to a open-minded community Do notengage with such people Agree in advance with your co-authors

27 of 40

Avoid known pitfalls with co-authors

pitfall you need a shared domain language ndash sometimes thesame term is used in different ways according to the academiccommunities (think about words like lsquoevolutionrsquo lsquosystemrsquo orlsquolevelrsquo) If you do not have agreement on this it will be easy toattack you

solution prepare it well in advance especially if you do not knoweach other

pitfall you risk that down-to-earth researchers find your papertoo unconventional eccentric and they reject it because they donot understand all the technicalities

solution address the paper to a open-minded community Do notengage with such people Agree in advance with your co-authors

27 of 40

Avoid known pitfalls with co-authors

pitfall you need a shared domain language ndash sometimes thesame term is used in different ways according to the academiccommunities (think about words like lsquoevolutionrsquo lsquosystemrsquo orlsquolevelrsquo) If you do not have agreement on this it will be easy toattack you

solution prepare it well in advance especially if you do not knoweach other

pitfall you risk that down-to-earth researchers find your papertoo unconventional eccentric and they reject it because they donot understand all the technicalities

solution address the paper to a open-minded community Do notengage with such people Agree in advance with your co-authors

27 of 40

Avoid known pitfalls with co-authors

pitfall you need a shared domain language ndash sometimes thesame term is used in different ways according to the academiccommunities (think about words like lsquoevolutionrsquo lsquosystemrsquo orlsquolevelrsquo) If you do not have agreement on this it will be easy toattack you

solution prepare it well in advance especially if you do not knoweach other

pitfall you risk that down-to-earth researchers find your papertoo unconventional eccentric and they reject it because they donot understand all the technicalities

solution address the paper to a open-minded community Do notengage with such people Agree in advance with your co-authors

27 of 40

Two ways for paper co-writing

1 divide et impera (lsquodivide and rulersquo) each co-author writes a-prioridefined sections In other words each one is ldquokingrdquo of hisherpart

pros amp cons suitable for remote teams interaction anddiscussion are low risk an ldquoeditedrdquo paper can be weak Revisionof the other parts is needed You need to be well organized

2 pair writing you write the paper together two people with onekeyboard ndash possible two screens) Borrowed from PairProgramming ndash an eXtreme Programming technique pros amp conspersonally engaging you need a good relation with theco-author(s) if not instead of spare time you lose it in generalthe final quality is better

28 of 40

Two ways for paper co-writing

1 divide et impera (lsquodivide and rulersquo) each co-author writes a-prioridefined sections In other words each one is ldquokingrdquo of hisherpart pros amp cons suitable for remote teams interaction anddiscussion are low risk an ldquoeditedrdquo paper can be weak Revisionof the other parts is needed You need to be well organized

2 pair writing you write the paper together two people with onekeyboard ndash possible two screens) Borrowed from PairProgramming ndash an eXtreme Programming technique pros amp conspersonally engaging you need a good relation with theco-author(s) if not instead of spare time you lose it in generalthe final quality is better

28 of 40

Two ways for paper co-writing

1 divide et impera (lsquodivide and rulersquo) each co-author writes a-prioridefined sections In other words each one is ldquokingrdquo of hisherpart pros amp cons suitable for remote teams interaction anddiscussion are low risk an ldquoeditedrdquo paper can be weak Revisionof the other parts is needed You need to be well organized

2 pair writing you write the paper together two people with onekeyboard ndash possible two screens) Borrowed from PairProgramming ndash an eXtreme Programming technique

pros amp conspersonally engaging you need a good relation with theco-author(s) if not instead of spare time you lose it in generalthe final quality is better

28 of 40

Two ways for paper co-writing

1 divide et impera (lsquodivide and rulersquo) each co-author writes a-prioridefined sections In other words each one is ldquokingrdquo of hisherpart pros amp cons suitable for remote teams interaction anddiscussion are low risk an ldquoeditedrdquo paper can be weak Revisionof the other parts is needed You need to be well organized

2 pair writing you write the paper together two people with onekeyboard ndash possible two screens) Borrowed from PairProgramming ndash an eXtreme Programming technique pros amp conspersonally engaging you need a good relation with theco-author(s) if not instead of spare time you lose it in generalthe final quality is better

28 of 40

Part four dissemination

Publication of results copyright issues andsocial networks

29 of 40

We live in difficult times

source twitter account Mgm and Mrk

The copyright issue and the social networks

the big players of the academic publishing industry (CUP OUPSpringer IEEE Elsevier Taylor amp Francis Brill etc) have specificregulations on the publication of your work

some adopt some kind of Open Access policy (see below) othersnot

You are always right in publishing pre-prints if you followsome rules and use the convenient ldquomagic formulardquo

31 of 40

The ldquomagic formulardquo for pre-prints

1 take the final draft or the camera ready and cut off thestylesheet page numbers included If you submitted your paperwith LATEX simply use documentclassarticle

2 insert a cover page where you will indicate the final destination ofpublication

3 in the cover page insert a disclaimer too Beware disclaimerschange according to the venue If you do not find it on the netyou can ask to the board of the journal for it

32 of 40

An example the rules for Springer

An author may self-archive an author-created version of hisherarticle on hisher own website and hisher institutionrsquosrepository including hisher final version however heshe maynot use the publishers PDF version which is posted onwww springerlink com Furthermore the author may onlypost hisher version provided acknowledgement is given to theoriginal source of publication and a link is inserted to thepublished article on Springers website The link must beaccompanied by the following text ldquoThe original publication isavailable at www springerlink com rdquo

33 of 40

Case study from another paper of ours

ccopy Gobbo amp Benini (2013)

Use official DARs (Digital Academic Repositories)first

URL httpdareuvanl

The paradox of ldquosocialrdquo repositories

CCcopy K Fitzpatrick 26 oct 2015 Academia Not Edu Seehttpplannedobsolescencenet

The challenge of Open Access

Open access journals ndash mainly electronic ndash are available to theirreaders free of charge Access is open across the net The journals arefinanced by payments being made for the article to be publishednot by payment being made to access the article throughsubscriptions

There is an ldquoOpen Access rainbowrdquo (Bill Hubbard SHERPANottingham) policy spectrum where the author should get orientedbut not easily

A good resource is the white paper How open is it

URL httpswwwplosorgopen-accesshowopenisit

37 of 40

The colours of Open Access

green publishers that allow both the preprint and the postprint(authorrsquos final) to be archived

blue publishers that do not allow preprints to be archived but willallow postprints (either the authorrsquos final version or the publisherrsquosPDF)

yellow publishers that allow their authors to archive theirpreprints ndash draft uncorrected papers ndash but not their postprints(publisherrsquos PDFs)

white publishers which refuse to grant their authors any rights toarchive their work online

gold publishers ask to the author to pay the publication fee(directly or through a research grand or similar)

38 of 40

A simpler way to say that

CCcopy Bill Hubbard Repositories Support Project