Information Technology Reference
In-Depth Information
clumsily worded, does not require fifteen hundred words of amplifying explanation.
Some authors seem to like to make, and then explain, remarkably asinine observa-
tions, such as “Users today frequently use a search engine to answer queries”. But
this is not as bad as asserting, say, that “with the growth of data online, users have
given up trying to use Web search engines”, or that “the development of large RAM
means that there is no longer a need for offline storage”.
11
Nor is it as bad as basing a
program of research on a fallacy or delusion, such as “users download software ille-
gally as a form of protest against corporate business practices”, or “recent advances
in machine learning mean that soon it may no longer be necessary to write software”,
or “increases in the efficiency of processors are encouraging people to develop their
own database utilities, based on open-source toolkits”.
12
What these examples have in common is that they are dogmatic—and in some
instance, foolish—opinions expressed as fact. They embody a catastrophic misjudge-
ment of the likely readership, and possibly of reality.
Another display of ignorance is in relation to past literature, as noted above. All
too often, authors have either paid no attention to key past work, or have not troubled
to understand it properly and dismiss it in a few words. It is astonishing how often
authors don't seem to have searched for related papers; a common exercise of mine
is to paste a paper's title directly in to a search engine, because highly relevant
but uncited papers are often in the first page of results. When the background and
literature review are crushed into a few brief paragraphs, it is almost certain that the
author has nothing to contribute.
A “Writing Up” Checklist
Regarding
the scope of the work
,
In what forum, or kind of forum, do you plan to publish?
Is the scope of the work well defined?
Is there a single, clearly articulated research question or goal? Have you identified
which aspect of the work is of greatest impact, or of greatest interest?
What would success in the project look like? What would failure look like? Can
you anticipate the form of the outcomes in either case?
11
Some papers in the area of in-memory databases rest on a similar assertion, namely that with the
falling cost of memory technologies databases no longer need be maintained on disk. I'm intrigued
by the fact that these claims have now been made for well over twenty years, a period in which
memory and disk capacities have grown by a factor of at least 10,000. While there is some validity
to the assertion—it is undoubtably true that some things can be done in memory today that required
disk in the past—it does illustrate that such claims depend not just on technology, but also on the
context in which the technology is used.
12
If the gross logical failures of these statements are not obvious, pause here and take the time to
analyze them yourself.
