# -*- mode: org -*-
#+TITLE:     Computational Documents
#+DATE: June, 2018
#+STARTUP: overview indent
#+OPTIONS: num:nil toc:t
#+PROPERTY: header-args :eval never-export

This document gathers most references mentioned in the videos. This is
a rather crude list.

* Table of Contents                                                     :TOC:
- [[#m2-s0-computational-documents][M2-S0: Computational Documents]]
- [[#m2-s1-a-few-recent-controversial-studies][M2-S1: A few Recent Controversial Studies]]
  - [[#economy-austerity-in-fiscal-policy][Economy: Austerity in Fiscal Policy]]
  - [[#functional-mri][Functional MRI]]
  - [[#incorrect-protein-structures][Incorrect Protein Structures]]
  - [[#other-domains][Other domains]]
- [[#m2-s2-why-is-this-so-difficult][M2-S2: Why is This so Difficult?]]
  - [[#mistakes][Mistakes]]
  - [[#going-public][Going Public?]]
- [[#m2-s3-computational-documents-principles][M2-S3: Computational Documents: Principles]]
- [[#m2-s5-collaborating][M2-S5: Collaborating]]
  - [[#preparing-a-document-for-a-journal-or-a-conference][Preparing a Document for a Journal or a Conference]]
- [[#m2-s6-comparative-study][M2-S6: Comparative Study]]
  - [[#exemples-of-org-mode-documents][Exemples of Org-Mode documents]]

* M2-S0: Computational Documents
* M2-S1: A few Recent Controversial Studies
** Economy: Austerity in Fiscal Policy
- [[https://en.wikipedia.org/wiki/Growth_in_a_Time_of_Debt][Reinhart et Rogoff]]: /Growth in a Time of Debt/
- Herndon, Ash et Pollin
- Wray: combining data across centuries, exchange rate regimes,
** Functional MRI
- 2010: [[https://www.researchgate.net/publication/255651552_Neural_correlates_of_interspecies_perspective_taking_in_the_post-mortem_Atlantic_Salmon_an_argument_for_multiple_comparisons_correction][Bennett et al. and the dead salmon]] \smiley
- 2016: [[http://www.pnas.org/content/113/28/7900.abstract][Eklund, Nichols, and Knutsson]]. [[http://www.sciencealert.com/a-bug-in-fmri-software-could-invalidate-decades-of-brain-research-scientists-discover][A bug in fmri software could
  invalidate 15 years of brain research]] (/40 000 articles/)
- 2016: But it's [[https://www.cogneurosociety.org/debunking-the-myth-that-fmri-studies-are-invalid/][more subtle than it looks like]]. [[http://blogs.warwick.ac.uk/nichols/entry/bibliometrics_of_cluster/][Nichols]]. /\approx
  3 600 impacted studies/
** Incorrect Protein Structures
[[https://people.ligo-wa.caltech.edu/~michael.landry/calibration/S5/getsignright.pdf][A "buggy software"]].
** Other domains
- [[http://www.nature.com/nrd/journal/v10/n9/full/nrd3439-c1.html?foxtrotcallback=true][Oncology]]: "/half of published studies, even in prestigious journals,
  can't be reproduced in industrial labs/"
- [[http://theconversation.com/we-found-only-one-third-of-published-psychology-research-is-reliable-now-what-46596][Psychology]]: "/attempting to reproduce 100 previously published
  findings/, /only one-third of published psychology research was found
  to be reliable/"
* M2-S2: Why is This so Difficult?
** Mistakes
- [[http://theconversation.com/how-computers-broke-science-and-what-we-can-do-to-fix-it-49938][Computers broke science]]
- [[https://qz.com/768334/years-of-genomics-research-is-riddled-with-errors-thanks-to-a-bunch-of-botched-excel-spreadsheets/][Programming and data manipulation mistakes in Genomics]]
** Going Public?
- Someone may [[http://www.nature.com/news/the-top-100-papers-1.16224][benefit from my hard work]]
* M2-S3: Computational Documents: Principles
* M2-S5: Collaborating
** Preparing a Document for a Journal or a Conference
Requirements for producing a *pdf*:
- Internally, /pandoc/, /knitr/ or /emacs/org-mode/
- /LaTeX/ should be installed

  - http://blog.juliusschulz.de/blog/ultimate-ipython-notebook
    - https://github.com/kirbs-/hide_code is a must-have
  - http://svmiller.com/blog/2016/02/svm-r-markdown-manuscript/
  - https://github.com/balouf/Kleinberg/blob/master/KleinbergsGridSimulator.ipynb
* M2-S6: Comparative Study
** [[file:../ressources/orgmode_examples/README.org][Exemples of Org-Mode documents]]
# ** Bonus: expériences vécues 

# Éléments clés lors du choix:
# - Simplicité de prise en main vs. vrai éditeur
# - Où sont fait les calculs 
# - Multi-langage
#   - http://carreau.github.io/posts/23-Cross-Language-Integration.html
# - Gestion des langages compilés
# - Notions de caches et d'état

# Les principaux outils actuels:
# - jupyter
# - rstudio
# - org-mode

# Limitations:
# - Longs calculs
# - Grands documents
# - Solutions wysiwyg pour jupyter

# Historique/diff un peu compliqué pour jupyter