This week I had the opportunity to sit in on an M.Sc Thesis Defense at Guelph. In most of my courses so far at Guelph the profs have stressed that we should avoid weasel words at all costs in our Thesis and our defense.
The definition of a "weasel word" varies depending on the source but generally they are words which are misleading, make baseless claims or are exaggerations to make ideas sound better. Or like a weasel sucks the insides out of an egg, the weasel word sucks the value out of the document it appears in.
When I watched the defense I found out one reason why they should be avoided. The person defending used the word "understanding" in his thesis. One of the goals of his thesis was to "understand" the field and as a result, at the conclusion of his thesis he claimed an "understanding" of the field. However it is impossible to ever have a complete understanding of a topic. It is very easy to point to something related to his thesis that he doesn't fully understand and thus this type of language should be avoided. It may seem like a small detail but imagine how a paper would sound if it were full of this type of language. It would be like one big exaggeration leaving the reader doubting the real knowledge of the person who wrote it.
<center><img src="/uploads/2008/08/reynolds1preview-300x225.jpg" alt="Reynolds building, University of Guelph" /></center>
The reason I'm writing this entry is to compile a list of computer science related "weasel words" which similarly should be avoided.
* Understand / Understanding: As in I "understand" the field I am studying thoroughly and completely
* Optimal, Optimized: As in "My algorithm is the optimal solution to the problem" when you really mean it is a "better solution than currently exists". All someone has to do is find something slightly better or a small improvement to make your work invalid. At the very least this will require a mathematically proof to show your idea is truly optimal so if you don't like math it may be best to avoid this word.
* Best: See above for optimal
* Proved / Proven: Many people like to use this word after some experimental results agree with a hypothesis they have proposed. However a hypothesis is never really proven completely. It can only be verified. If something has been proven it is usually mathematically proven and as such there should be no need for an experiment or hypothesis at all.
* Theory: I hear people using theory all the time and think they are talking scientifically. In science a theory is a well verified, time tested hypothesis or group of hypotheses. In common english many people try to use "theory" when they really mean idea or guess. For instance on television the other day someone asked "whats your theory on this?" to another character on a show. If you use theory in your thesis make sure you use it correctly.
* Hypothesis: Similarly to theory, many lay-people use this word incorrectly. A simplified defintion of a hypothesis is an "explanation of a phenomenon". Like theory many people will use it when they really mean idea or guess.
* (more to come as I watch more defenses / come across more on the inter webs)
The following are some links I found on weasel words and other language to avoid when writing formal academic documents:
* <a title="Wikipedia: Avoid weasel words" href="http://en.wikipedia.org/wiki/Wikipedia:Avoid_weasel_words" target="_blank">Wikipedia:Avoid weasel words
* <a title="How to Write A Dissertation - Purdue" href="http://www.cs.purdue.edu/homes/dec/essay.dissertation.html" target="_blank">How to Write A Dissertation - Purdue