Selfish Scientists, Egocentric Engineers, Truculent Techies

1
Selfish Scientists,Egocentric Engineers,Truculen
t Techies

• Some Stories from the Trenches
• A Personal Perspective
• Carole Goble
• The University of Manchester, UK
• carole.goble_at_manchester.ac.uk

3rd Intl Conf on e-Social Science, Uni of
Michigan, Ann Arbor, USA 8th October 2007 After
Dinner Keynote
2
(No Transcript)
3
(No Transcript)
4
GSK
ID MURA_BACSU STANDARD PRT 429
AA. DE PROBABLE UDP-N-ACETYLGLUCOSAMINE
1-CARBOXYVINYLTRANSFERASE DE (EC 2.5.1.7)
(ENOYLPYRUVATE TRANSFERASE) (UDP-N-ACETYLGLUCOSAMI
NE DE ENOLPYRUVYL TRANSFERASE) (EPT). GN MURA
OR MURZ. OS BACILLUS SUBTILIS. OC BACTERIA
FIRMICUTES BACILLUS/CLOSTRIDIUM GROUP
BACILLACEAE OC BACILLUS. KW PEPTIDOGLYCAN
SYNTHESIS CELL WALL TRANSFERASE. FT ACT_SITE
116 116 BINDS PEP (BY SIMILARITY). FT
CONFLICT 374 374 S -gt A (IN REF.
3). SQ SEQUENCE 429 AA 46016 MW 02018C5C
CRC32 MEKLNIAGGD SLNGTVHISG AKNSAVALIP
ATILANSEVT IEGLPEISDI ETLRDLLKEI GGNVHFENGE
MVVDPTSMIS MPLPNGKVKK LRASYYLMGA MLGRFKQAVI
GLPGGCHLGP RPIDQHIKGF EALGAEVTNE QGAIYLRAER
LRGARIYLDV VSVGATINIM LAAVLAEGKT IIENAAKEPE
IIDVATLLTS MGAKIKGAGT NVIRIDGVKE LHGCKHTIIP
DRIEAGTFMI
5
Social Ecosystem
Reference Notes Techniques
Community collective intelligence
In silico experiment
Lab Book
A Scientist
A Scientist
6
e-Science is Systematic Support for
Collaborative, Accelerated, Innovative
Research.enabling-Scienceempowering-Scientists
e-Science is Science
7

• Collaborative content
• Resources data workflows ontologies tools
  protocols techniques research know-how
• Share Reuse
• Collaborative development
• Systems, data sets, ontologies
• Build Adopt Reuse

8
Structure prediction
Phylogeny
Software Engineers
Omics
Chemists
Bioinformaticians
Systems Biology
Functional genomics
BioMed
Theory
Computer Scientists
Mathematicians
Biologists
Practice
Crystallographers
Service Providers
System Administrators
Resource Providers
9
Bio-Tribes, Bio-Nations, Territorial Techies
10

• Collaborate and share with my colleagues and
  friends I trust.
• And people I dont and may never know. And
  rivals?
• Actually they dobecause
• They are compelled to.
• There is a culture.
• Its in their best interest.
• Good citizenship.
• Rewards.
• Open Source Community

11
The Selfish (or Self-interested) Scientist

• A biologist would rather share their toothbrush
  than their gene name

Mike Ashburner and others Professor Genetics,
University of Cambridge,\UK
Data mining my datas mine and your datas
mine
Rich Giordianos Counterfeit Sharing
12
The Seven Deadly Sins of Bioinformatics

• Parochialism and Insularity
• Exceptionalism
• Autonomy or death!
• Vanity Pride and Narcissism
• Monolith Meglomania
• Scientific method Sloth
• Instant Gratification
• 3115 views, so this hit a nerve.
• http//www.slideshare.net/dullhunk/the-seven-deadl
  y-sins-of-bioinformatics/

Reinvention!
13
Reuse is Really HardAnd that goes for software
too.

• Hell is other peoples stuff.
• Metadata is the key.
• And often an afterthought ?

14
Andy Laws First (Format) Law

• The first step in developing a new genetic
  analysis algorithm is to decide how to make the
  input data file format different from all
  pre-existing analysis data file formats.

female
male
0
1
crimap
1
0
Keightly
1
2
Knott and Haley
http//bioinformatics.roslin.ac.uk/lawslaws.html
15
Biologist exceptionalism
Im different. We are all individuals.

• I know there is already a gene name for that
  gene, but, I don't like it and it doesn't fit in
  with my schema.
• It would be better if I wrote the script I need
  so I know what it does, how it does it and how to
  modify it later because I havent specified what
  it was supposed to do in the first place.
• 20 formats for sequence?
• 250 pathway databases?

16
Art is IScience is we.

• Claude Bernard
• (1813 - 1878)

17
Art is IScience is I.Science is we when it
suits me.e-Science is me-Science

• Carole Goble
• (1961 - )

18
Rewards Scientists
Advance the frontiers of Science. Get on with
some SCIENCE.

• Competitive advantage.
• Be the first with the Nature paper.
• Credit, credibility, fame, acclaim, recognition,
  peer respect, reputation.
• More funding.
• Get my result/approach/technique/workflow/ontology
  adopted.

19
Fears Scientist

• Beaten by lab X or Professor Y.
• Protecting my turf.
• Being misinterpreted or misrepresented.
• Looking stupid.
• Losing control.
• Taking a risk
• Releasing results too early.
• Be distracted from my Science
• Getting left behind.
• Being out of fashion.

20
PhDs at Work
21
The Ontologists Tale.

• Lets build Ontologies!
• Ontologies are consensual developed and shared
  knowledge.
• Controlled vocabularies for linking up data and
  stuff.

22
Endurants, Perdurants, Being, Substance, Event
Philosophers
Spiritual guides
Aesthetics
Life Scientists Capulets
Knowledge Representation Montagues
OBO
Theoreticians
Pragmatists
A means to an end Content providers
The end Mechanism providers
Carole Goble and Chris Wroe The Montagues and
the Capulets, in Comparative and Functional
Genomics December 2004, vol. 5, no. 8, pp.
618-622(5)
23
W3C Semantic Web for Life Sciences mailing list
Why dont you biologists modularise your
ontologies properly?
Er, well, like how should we do it properly and
where are the tools to help us?
We dont know and we havent got any. But here
are some vague guidelines.
24
There are no proper ontologies in biology! We
have all this incredible stuff in our language
you arent using. Look at this example.
How do I handle my legacy? The data I need to
describe isnt mine and isnt neat and tidy. The
ontology is already used by thousands of people
every day.
But its only got 20 classes! I need 250,000! And
its a trivial made-up example. How would using
all this help me do my job? Who will train the
curators? Who will pay for the effort? Where are
the tools?
Tell them to start again and do it properly this
time.
If you learn some logic then you can use this
OWL-RDF editor thingy (that only scales to 20
classes)
25
Rewards Computer Scientists
Advance the frontiers of Science. Get on with
some SCIENCE.

• Competitive advantage.
• Publish.
• Credit, credibility, fame, acclaim, recognition,
  peer respect, reputation.
• More funding.
• Get my results/approach/system/design/algorithm/fo
  o-bar adopted
• Showing off how clever we are.

26
(No Transcript)
27
Rewards Software Engineers and Service Providers
To build useful, sustainable Software and
Services. Get on with making something.
Preferably with the coolest technology.

• Competitive advantage.
• Credit, credibility, fame, acclaim, recognition,
  peer respect, reputation. Showing how clever I
  am.
• More funding. Though the rewards are poor.
• To get my system/data set/foo-bar adopted.
• Add value to my service/software/ data set for
  low or no cost.

28
Mutual Dependency and Antagonism

• How can I get Scientists to use my software,
  which is clearly superior to what they already
  have and unaccountably like?
• How do I make them work with me.
• Why dont these computer people give me something
  I can actually use. Preferably now?
• Make my favourite desktop application faster.
  Make my dataset bigger. Get me their data set.
  Dont let them see my results until I say so.
  Give me something I couldnt get before.

29
Simple to use is not the same as simple.Stuff
you cant see must be easy stuff. Right? ?
30
The Integrationists Tale.

• Lets build systems that link all these data sets
  and tools together!
• myGrid Taverna Workflow system

31
myGrid Taverna Workflow
http//www.mygrid.org.uk
32
Trypanosomiasis in Cattle
A PhD student. Paul Fisher.

• Identified a pathway for which its correlating
  gene (Daxx) is believed to play a role in
  trypanosomiasis resistance.
• Systematic and comprehensive automation.
  Elimination of user bias.

Fisher P et al A systematic strategy for
large-scale analysis of genotypephenotype
correlations identification of candidate genes
involved in African trypanosomiasis, Nucleic
Acids Research, 2007, 19
33
Principles of Engagement

• Content is King data, workflows, services.
• Be Open to build Critical Mass.
• Keep adoption costs low.
• Fit into the scientific world as it is.
• Change by stealth. Track it. Predict it. React to
  it.
• Dont be prescriptive. Scientists control.
• Cooperate. Get Others to Add Value. Use Network
  effects. Think local Act Global.

34
Just enough, Just in timeJam today and Jam
tomorrow
Very BAD
Pain
Just right
Good, but Unlikely
Gain
35
Many of these are marketing points
caBIG User Advocate
36
Long tail
OReilly Book
37
A good User Experience outweighs Smart
Features.Eat Your Own Dog Food Innovation is
not necessarily Cleverer Infrastructure
Scientists are Naughty
38
Computer Scientists
Life Scientists
39
Mars vs Venus

• Not my problem Lets solve this other problem
  which isnt your problem but is fun and leads to
  interesting software. And papers.
• Over-complication Lets solve this harder problem
  than take the easier route that solves your
  problem.
• its simple myth My granny can write
  workflows...

40
Venus vs Mars

• Paternalism Repeating the same old mistakes
  despite our experiences.
• Short termism instant gratification It just
  about holds together to get the results for my
  paper. Lets hope the PhD student doesnt
  leave...Oh...
• Hackery. Simplifications, hackery and monolithes
  now stores up trouble down the road. Act in
  haste, repent later.
• Calling me a plumber.

41
Marco Roos
42
Standards are boring blue collar scienceBy not
making shareable reusable software, we can
publish every single monolithic software
solution. Hurrah!
43
Changing Scientific Method

• Dry people hypothesise
• Bench scientists validate.
• make sense of this data
• to
• does this result make sense?

Fisher P et al A systematic strategy for
large-scale analysis of genotypephenotype
correlations identification of candidate genes
involved in African trypanosomiasis, Nucleic
Acids Research, 2007, 19
44
Luddism? Surely not!
45
Recycling, Reuse, Repurposing

• Workflows are memes.
• Scientific commodities.
• Traded know-how.
• To be exchanged and traded and vetted and mashed.
  
• Acceleration.
• Quality.

46
(No Transcript)
47
The Social Networkers Tale.

• myExperiment
• Sharing content
• AND
• Sharing development!

48

• Screen shot

49
From me-Science to we-Science

• Tribal bonding and sharing
• Crossing Tribal Boundaries
• Across communities and disciplines (MIT)
• Intellectual Fusion Swarming breaking down
  silos
• Understanding outside my expertise. E.g. sources
  of error
• Metadata challenges.
• Social challenges.

50
Scientists mashing it up for themselves
51
This is fun!

• Incentives and inhibitors for sharing content
• Newbies.
• Projects amongst networks of friends.
• Across communities for intellectual swarming.
• Quality, Protection, Reputation. Market place.
• Incentives and inhibitors for sharing development
• Perpetual beta.
• Best friends.
• Mash ups.

52
Structure prediction
Phylogeny
Software Engineers
Omics
Chemists
Bioinformaticians
Systems Biology
Functional genomics
BioMed
Theory
Computer Scientists
Mathematicians
Social Scientists
Biologists
Practice
Crystallographers
Service Providers
System Administrators
Resource Providers
53
Rewards Social Scientists
Advance the frontiers of Social Science. Get
on with some SCIENCE.

• Competitive advantage.
• Be the first with the X paper.
• Credit, credibility, fame, acclaim, recognition,
  peer respect, reputation.
• More funding.
• Get my result/approach/technique adopted.

54
Challenge for e-Social Science for ME

• Timeliness
• Rapid Churn
• Participative activity
• Strangers participation
• Added value to the Scientists and the Engineers.

• Ignoring the Star Trek Prime Directive

55
So Just Do It.e-Social science. For what?Jam
for All.Thats a real challenge.
56
Acknowledgements

• With considerable thanks to
• David De Roure
• Robert Stevens
• The ontology, myGrid and myExperiment teams over
  the years.
• And all our long suffering users.
• My Systems Administrator Mr Cottam
• http//www.myexperiment.org
• http//www.mygrid.org.uk
• Funders EPSRC and JISC