1. How to read scientific papers
Introduction - Pervasive Computing Course 2011
IT University of Copenhagen - Aurélien Tabard
mercredi 31 août 2011-w 1
Hi, I am Aurélien Tabard
I’m a post-doc in the SDG group working with Jakob and Morten here. And I spend quite
some time in the pitlab.
I’ll give you a short intro to something you may already be familiar with.
But it’s always good to start the year with some clear outlines.
Technical writing is like programming, you need to read a lot to get better at reading. There is
no magic potion. But reading it a good way to see how people are doing.
For the paper you’ll have to write at the end of the project, you should really get a paper you
like and use it as a model.
2. How to read scientific papers
Why so many papers?
Overview of paper reading.
How to go through your reading assignments?
1st, 2nd and 3rd pass.
mercredi 31 août 2011-w 2
3. Why so many papers?
Get you to know and understand the current state of the art
in the Pervasive and Ubiquitous computing field:
.: cover more than 20 years of research and exciting technology
.: present current questions within the field
.: create interesting new projects
mercredi 31 août 2011-w 3
4. How to Read a Paper
S. Keshav
David R. Cheriton School of Computer Science, University of Waterloo
Waterloo, ON, Canada
keshav@uwaterloo.ca
ABSTRACT 4. Glance over the references, mentally ticking off the
Researchers spend a great deal of time reading research pa- ones you’ve already read
pers. However, this skill is rarely taught, leading to much At the end of the first pass, you should be able to answer
wasted effort. This article outlines a practical and efficient the five Cs:
three-pass method for reading research papers. I also de-
scribe how to use this method to do a literature survey. 1. Category: What type of paper is this? A measure-
Categories and Subject Descriptors: A.1 [Introductory ment paper? An analysis of an existing system? A
and Survey] description of a research prototype?
General Terms: Documentation.
2. Context: Which other papers is it related to? Which
Keywords: Paper, Reading, Hints. theoretical bases were used to analyze the problem?
1. INTRODUCTION 3. Correctness: Do the assumptions appear to be valid?
Researchers must read papers for several reasons: to re- 4. Contributions: What are the paper’s main contribu-
view them for a conference or a class, to keep current in tions?
their field, or for a literature survey of a new field. A typi-
cal researcher will likely spend hundreds of hours every year 5. Clarity: Is the paper well written?
reading papers.
Learning to efficiently read a paper is a critical but rarely Using this information, you may choose not to read fur-
taught skill. Beginning graduate students, therefore, must ther. This could be because the paper doesn’t interest you,
learn on their own using trial and error. Students waste or you don’t know enough about the area to understand the
much effort in the process and are frequently driven to frus- paper, or that the authors make invalid assumptions. The
tration. first pass is adequate for papers that aren’t in your research
For many years I have used a simple approach to efficiently area, but may someday prove relevant.
read papers. This paper describes the ‘three-pass’ approach Incidentally, when you write a paper, you can expect most
and its use in doing a literature survey. reviewers (and readers) to make only one pass over it. Take
care to choose coherent section and sub-section titles and
2. THE THREE-PASS APPROACH to write concise and comprehensive abstracts. If a reviewer
cannot understand the gist after one pass, the paper will
The key idea is that you should read the paper in up to likely be rejected; if a reader cannot understand the high-
three passes, instead of starting at the beginning and plow- lights of the paper after five minutes, the paper will likely
ing your way to the end. Each pass accomplishes specific never be read.
goals and builds upon the previous pass: The f irst pass
gives you a general idea about the paper. The second pass 2.2 The second pass
lets you grasp the paper’s content, but not its details. The
In the second pass, read the paper with greater care, but
third pass helps you understand the paper in depth.
ignore details such as proofs. It helps to jot down the key
2.1 The first pass points, or to make comments in the margins, as you read.
The first pass is a quick scan to get a bird’s-eye view of
1. Look carefully at the figures, diagrams and other illus-
the paper. You can also decide whether you need to do any
trations in the paper. Pay special attention to graphs.
more passes. This pass should take about five to ten minutes
Are the axes properly labeled? Are results shown with
and consists of the following steps:
error bars, so that conclusions are statistically sig-
1. Carefully read the title, abstract, and introduction nificant? Common mistakes like these will separate
rushed, shoddy work from the truly excellent.
2. Read the section and sub-section headings, but ignore
everything else 2. Remember to mark relevant unread references for fur-
ther reading (this is a good way to learn more about
3. Read the conclusions the background of the paper).
mercredi 31 août 2011-w 4
• For instance before preparing this I read a bit :)
He are two references you can go to if you want to know more about reading.
the first reference, How to read a paper is linked on the blog, and it’s only 2 pages.
the second is a book from a French Professor of La Sorbonne who explains how to talk about
books you haven’t read. Which he does regularly.
But that guy has a trick: he already read a lot and knows a lot about litterature...
5. Disclaimer
You need to know the domain first!
...hence read
mercredi 31 août 2011-w 5
If you didn’t read a specific piece of Shakespeare but read 5 of them and essays about
Shakespeare writings then you have some idea about shakespeare’s plays, the plot, the style,
when it came up in his writing and why he was addressing this subject.
Why read?
because you have to build your own Universal Library so that you can situate one
paper within it.
Umberto Eco call the universal library the library containing all knowledge.
6. How to read scientific papers
Navigate between:
Skimming Careful reading
mercredi 31 août 2011-w 6
7. How to read scientific papers
Idea of the subject Good understanding
pass 1 pass 2 pass 3
mercredi 31 août 2011-w 7
8. Active reading
.: Get rid of distractions
.: Get a pen
.: Jump around, re-read, go backwards - These are not novels
.: Talk to others
mercredi 31 août 2011-w 8
• Reading requires attention, skipping some parts may hinter the overall
understanding
• Reading is active, get a pen,
• Go in whatever order you want.
Authors will (or should) tell you who is the killer from the beginning.
• Reading is a social activity, we share interesting papers we read, talk to people for
advices on papers...
9. Papers’ structures
Textbook
Vision / Overview
Technical paper
mercredi 31 août 2011-w 9
Before skimming or reading in details, it’s good to know that scientific paper share often a
very similar structure, that helps readers to browse articles.
Textbook:
- Stable knowledge, a few years old. vs. Current/ongoing
- overview vs. Precise (data, experimental process)
-
TODO:
SHOW A REAL PAPER SCREENSHOT
10. GridOrbit – An Infrastructure Awareness System for
Increasing Contribution in Volunteer Computing
Juan David Hincapi´ -Ramos, Aur´ lien Tabard, Jakob E. Bardram
e e
IT University of Copenhagen
DK-2300 Copenhagen, Denmark
{jdhr,auta,bardram}@itu.dk
ABSTRACT which seek to gain computational power by enlisting end-
.: Title
The success of a volunteer computing infrastructure depends user computers like PCs and game consoles.
on the contributions of its users. An example of such an
infrastructure is the Mini-Grid, a local peer-to-peer system Volunteer computing infrastructures however, like other in-
used for computational analysis of DNA. The speed of ana- frastructures, are often invisible to their end-users [16, 22],
and this invisibility poses a fundamental obstacle to their
.: Authors
lysis increases as more users join the Mini-Grid. However,
the invisible nature of such an infrastructure hinders adop- adoption [5]. A core question is how end-users can become
tion, as it is difficult for users to participate in an infrastruc- aware of such an invisible infrastructure, and start partici-
ture they are not aware of. This paper introduces GridOrbit, pating and contributing to it. Existing volunteer computing
a system designed to increase user awareness, fostering con- projects rely on individual and social motivations, leading to
tributions to this infrastructure. We designed GridOrbit us-
ing a participatory design process with biologists, and sub-
efforts in building communities, setting up competitions, and
rewarding users who participate [18]. Thus, there is a sub-
stantial overhead associated with recruiting users who will
.: Abstract
sequently deployed it for use in a biology laboratory. Our
results indicate that the number of contributors to the Mini- donate CPU cycles ‘for free’, and this in turn becomes a core
Grid increased with the use of awareness technologies. In
addition, our analysis presents their motives and behaviors.
challenge for the scientists using the infrastructure.
In this paper, we explore the use of awareness technologies
.: Intro
Finally, a characterization of user interaction with GridOrbit
to recruit contributors to a volunteer computing infrastruc-
emerged, which enabled us to understand how awareness
ture in a molecular biology research laboratory. Molecular
systems can be better designed. We see GridOrbit as an ex-
ample of a broader class of technologies designed to create
‘Infrastructure Awareness’ as a means to increase the contri-
biologists use the infrastructure to execute bioinformatics al-
gorithms for analyzing DNA/RNA sequences of millions of
.: Related work
bytes. This infrastructure uses peer-to-peer (P2P) technol-
butions to technological infrastructures.
ogy for distributing tasks to computers within the organi-
Author Keywords
Volunteer Computing, Infrastructures, Infrastructure Aware-
zation. This implies that the infrastructure requires many
users to participate. Thus, one central challenge is to mo- .: System description
tivate users to contribute despite the fact that only a minor
ness, Public Displays, Ambient Displays. part of them have the actual need of submitting tasks.
ACM Classification Keywords
H.5.3 Collaborative computing
To facilitate recruitment, we aimed at increasing the visi-
bility of the P2P grid. We engaged in a user-centered design
.: Evaluation
process with biologists, which resulted in the design of a sys-
.: Results
General Terms tem named GridOrbit. GridOrbit displays an interactive vi-
Experimentation sualization of the underlying activity in the infrastructure on
public screens (see figure 3), and provides users with feed-
INTRODUCTION back about their contributions through notifications on their
personal computers (see figure 4).
.: Discussion
Volunteer computing is a powerful way to conduct compu-
tation-intensive data processing by harnessing computing
GridOrbit was designed with the hypothesis that increas-
resources from large numbers of geographically distributed
ing awareness of a local resource sharing infrastructure will
individuals. The most prominent examples of volunteer com-
lead to broader participation and increased contribution. We
puting initiatives are SETI@Home and Folding@Home1 ,
1
http://setiathome.ssl.berkeley.edu/ http://folding.stanford.edu/
tested this hypothesis in a one-month deployment of the sys-
tem. Results show that public display visualizations and per-
sonal notifications can be used for creating an awareness of
.: Conclusion
Permission to make digital or hard copies of all or part of this work for an otherwise invisible infrastructure. We further show that
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, or
awareness of the activity of an infrastructure through both
public and personal displays supports the recruitment of new
contributors. Based on this case, we expand the notion of In-
.: References
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee. frastructure Awareness and discuss how it can improve the
CHI 2011, May 7–12, 2011, Vancouver, BC, Canada. adoption and thereby the value of voluntary infrastructures.
Copyright 2011 ACM 978-1-4503-0267-8/11/05...$10.00.
mercredi 31 août 2011-w 10
11. Questions to keep in mind
.: What questions does the paper address?
.: Is the problem relevant?
.: What are the main conclusions of the paper?
.: What evidence supports those conclusions?
.: Do the data actually support the conclusions?
.: What is the quality of the evidence?
.: Why are the conclusions important?
mercredi 31 août 2011-w 11
12. Pass 1 - Overview
Get a general idea
Situate the article within the Universal Library.
mercredi 31 août 2011-w 12
13. Pass 1 - Overview
Title + Authors
.: What is this about + Where does it come from
Abstract
.: What was done, what is the contribution
Sub-sections titles, figures
.: What is the paper structure / contribution
Medium (textbook, major journal/conference, workshop)
.: Who is the audience?
References
.: Is it a serious paper
mercredi 31 août 2011-w 13
This can be done almost out of the digital library or google scholar, no need to open or print
the pdf...
.: Who is talking?
.: What is the message?
.: What is the support / who is the audience?
.: Textbook - students (and teachers)
.: Major conference publication - researchers students in the field
.: Work in progress, workshop - peers working on a specific topic
Did you already read many references, are they citing the main references within the field.
14. Pass 2 - Get the authors message
Introduction + Conclusion
.: What is the problem (look for However)
.: How are the authors solving it
.: What are the authors’ contributions
Discussion
.: What are the insights, the limitations, the relevance of the work.
mercredi 31 août 2011-w 14
15. Pass 3 - careful reading
From beginning to end ...
... or backwards
.: Read actively, use pens
.: Talk to other students
.: Get help from related work papers if it gets too complicated
.: Come back to it later
mercredi 31 août 2011-w 15
If it gets too complicated move to another one that may be more
general
16. Living with it
Research papers summarize months or years of work in a few pages
Come back, re-read, question, re-discover.
mercredi 31 août 2011-w 16
If the paper is core to what you are doing.
Particularly relevant to your course project or masters’ thesis.
17. Plenty of other ways of reading...
Objectives
.: Literature review
.: Survey
.: Reviewing articles for colleagues, journal or conferences
Other sources:
.: Books
.: Patents
.: Web content
mercredi 31 août 2011-w 17