Christopher Oezbek, oezbekinf'fuberlin'de 1

1
Seminar Ausgewählte Beiträgezum Software
EngineeringPart II Outline/Paper/LaTeXChristop
her OezbekFreie Universität Berlin, Institut für
Informatikhttp//www.inf.fu-berlin.de/inst/ag-se/
2
Review of the Literature Search

• Share your experiences!
• What worked, what didn't?
• Where was progress easy and where was it hard?
• Which tips were useful and which useless?
• What would you do differently?
• How much time did you invest?
• Comments in general?

3
Review of the Summaries you have written

• Most of you did an okay job
• Still a lot of room for improvement
• All the typical mistakes present
• Please be aware that these mistakes will not be
  tolerated in your final version

4
Common Mistakes when writing a paper

• No clear train of thought
• No connections
• No separations
• No concept
• Too complicated sentences
• Lack of citations
• Proof missing
• Plagiarism
• Usage of imprecise words and fillers
• Spelling and commas
• Mixing up English and German

5
No train of thought - No Connections

• Train of though is the single most important
  property of a paper.
• Without it people will not be able to follow you.
• To achieve this make sure that you connect
  sentences and paragraphs using an explicit
  structure or connector words.

6
No train of thought - No Separations

• The converse of the above
• Sentences need to be separated using paragraphs,
  enumerations or separator-words (like "Next", "On
  the contrary", "Afterwards"), otherwise they can
  be very confusing to read.

7
No concept

• This is most important for the talk but also has
  implications for the paper.
• Make sure that you have a structure in your
  explanations
• There should be healthy mix of theory and
  examples.
• If you forget the examples then the paper is to
  abstract.
• If you forgo the theory your paper lacks depth.
• Combine complicated material and easy summeries.
• Not everybody can follow your explanations.
  Provide backup routes and key-frames to reset
  from.
• Concept and train of thought are the same issue
  on different levels.

8
Too complicated sentences

• Watch out for sentences that are nested too deep.
• You should read over them several times and
  detangle them.
• Suggestion Im ersten Teil des Textes werden
  Terminologie und folgende Gliederung für
  empirische Untersuchungen nach Fenton und
  Pfleeger (1997) vorgestellt ...

9
Lack of Citation - Plagiarism

• If you quote passages from other publications,
  then you need to mark them using quotes and a
  link into your reference section.
• Same holds for ideas (even if you paraphrase
  them).
• If you copy text from other authors without
  citing them, you risk losing your reputation as a
  scientist (and your certificate in this seminar).

10
Lack of Citation - Proof missing

• It is easy to jump to conclusions that you cannot
  back up with a logical argumentation or
  experimental results.
• Be aware that experienced readers will not
  tolerate remarks that don't stand on solid
  grounds.

11
Wrong use of Citation

• Don't overdo the use of literal citations.
• Just copying the words of important authors does
  not make the paper valuable.

12
Use of imprecise words and fillers

• Make sure that words like "maybe", "probably",
  "for sure", "certainly", don't weaken your
  argumentation.

13
Spelling and Commas

• Use a spellchecker (!)
• Review your comma-knowledge.

14
Mixing up English and German

• Use the following rule when adding English terms
  to a German paper.
• All terms should be used in German unless no
  appropriate German translation can be found
  (P2P).
• On the first occurrence you should list the
  English term in parenthesis.
• If you have to have the English term gt italics.
• Watch out that you don't translate to literally
  from English. There are a lot of "False Friends"
  out there.

15
LaTeX and Under- and Overfullboxes

• To help LaTeX with splitting words on line
  boundaries you have to add "\-" into words that
  are not in the dictionary at all positions where
  the word could be split.
• This gives hints to the LaTeX compiler where
  split.
• For instance Ver\-ständ\-nis\-feh\-ler

16
Here the same points with a positive spin

• Make sure that your paper has...
• Good structure
• Clear train of thought
• Good understandability
• Good readability
• Clean Layout
• Illustrating Examples and Pictures
• Correct Spelling and Grammar
• Logical sound argumentations and citations to
  back them up

17
LaTeX - Why?

• We want to use LaTeX for all the documents
  produced. This has several reasons
• LaTeX is the standard for scientific documents.
• You should have come in contact with this system
  so that you can contrast it to What You See Is
  What You Get solutions like Word.
• It is easy to enforce a unified look and feel for
  a set of documents produced by different authors.

18
LaTeX - How?

• What you should do (Windows instructions)
• Install Miktex - A windows distribution for
  LaTeX.
• Install TeXnicCenter - A powerful editor for
  LaTeX sources.
• Get the template LaTeX files from the
  Paper-Template.zip.
• It contains a little demonstration LaTeX file
  that has a lot of the important features that you
  will need to write your paper.
• The Not So Short Introduction to LaTeX2
• Lyx - Is an editor for LaTeX that is kind of a
  hybrid between LaTeX and Word. Lyx tries to
  display what it thinks LaTeX will produce, but
  uses LaTeX in the background. It works mainly on
  Linux.
• TeXmacs - GNU TeXmacs is a free scientific text
  editor, which was both inspired by TeX and GNU
  Emacs.

19
Outline

• The outline should present the vision of what you
  would like to present from the topic and
  structure it in a way that can be followed by the
  reader.
• Ask yourself the question "Which three things do
  I want people to remember from my talk?"
• The general structure of a paper is the
  following
• Introduction - Open the topic, place it inside
  the context of the seminar, motivation
• Fundamentals / Definitions - Explains
  requirements that are necessary for
  understanding.
• - n. Main aspects to be discussed
• n1. Conclusion, criticism, look-ahead (i.e.
  where to do research), review of used papers.

20
Assignment (II)

• Create an outline for your topic.
• Make an appointment with me before the next
  session (11.07.05).
• We will discuss the outline.
• Fill gaps in your conceptual understanding and
  find rationals/papers for missing logical links.
• References
• Markus Kalb, 2004. Anforderungen und Tipps.
• Hinweise zur Bearbeitung eines Seminarthemas
  (RWTH Aachen)

21
Thank you!