发信人: craman (江山), 信区: PostDoctor
标 题: how to write a technical paper(整理）
发信站: BBS 水木清华站 (Wed Jan 23 20:54:09 2002)
A technical paper will usually have four sections. The purpose of
each of these sections is as follows:
Section I: Introduction. The Introduction should do the following:
1. Open up the subject. (The subject will be electromagnetic
fields in cylindrical dielectric geometrics, adaptive arrays in
packet radio, or whatever.)
2. Survey past work relevant to this paper.
3. Describe the problem addressed in this paper, and show how this
work relates to, or augments, previous work.
4. Describe the assumptions made in general terms and state what
results have been obtained. (This gives the reader an initial
overview of what problem is addressed in the paper and what has
packet radio, or whatever.)
been achieved.)
5. Overview the contents of the paper. ("Section II contains our
formulation of the problem. Section III contains the experimental
data, ...")
Section II: Formulation of the Problem. This section should do three
things:
1. Define the problem to be considered in detail. Typically this
section might begin with something like: "Consider a packet radio
system consisting of a single central repeater surrounded by user
terminals. Each user transmits packets to the central repeater
using a slotted ALOHA protocol. The transmissions from all users
are assumed to be on the same frequency." the discussion should
proceed in this way until the problem is completely defined.
2. Define all terminology and notation used. Usually the terminology
and notation are defined along with the problem itself.
3. Develop the equations on which your results will be based and/or
describe any experimental systems.
section might begin with something like: "Consider a packet radio
Section III: Results. This section presents the detailed results you
have obtained. If the paper is theoretical, you will probably show
curves obtained from your equations. If the paper is experimental, you
will be presenting curves showing the measurement results. In order to
choose the proper curves to present, you must first be clear what
point you are trying to convey to the reader. The curves can then be chosen
to illustrate this point. Whether your paper is theoretical or
experimental, you must provide a careful interpretation of what your
results mean and why they behave as they do.
Section IV: Conclusion. This section should summarize what has been
accomplished in the paper. Many readers will read only the Introduction
and Conclusion of your paper. The Conclusion should be written so they
can be understood by someone who has not read the main work of the
accomplished in the paper. Many readers will read only the Introduction
paper.
This is the most common format for an engineering paper. Of course,
the names of the sections may differ slightly from those above, but the
purpose of each section will usually be as described. Some papers
include additional sections or differ from the above outline in one
way or another. However, the outline just presented is a good starting
point for writing a technical paper.
To write your paper, you should proceed as follows:
Step 1: Start by writing a complete first draft of your paper, except
for the Introduction and Conclusion. (It is easiest to leave the
Introduction and Conclusion until after the main body of the paper is
written.) In writing your paper, keep the following in mind:
1. You must always present the big picture first and then work towards
the details. The other way around will not work. This is especially
true in the beginning of Section II, where you are explaining the
problem you are studying.
2. If you get stuck and cannot figure out how to explain something, a
useful trick is to imagine that you are telling a very good friend
what you are working on: just put down the words as you would say them to
your friend.
In writing your first draft, do not worry if the wording is not
the details. The other way around will not work. This is especially
perfect. Polishing the document comes later. When you are finished
with your first draft, put it away and do not look at it for a couple of
days before you begin Step 2.
Step 2: Make sure the ideas in the paper are in the right order. If
not, move blocks of the paper around with your text editor until they
are. Ask yourself: "Can the reader understand every passage strictly
from the material up to that point?" if not, add material or move
ideas around. Make sure there not gaps in your logical arguments, and make
sure you are not implicitly assuming that the reader understands
something needed to follow your arguments, even though you have not
stated it. The reader probably understands less than you think.
Step 3: Work on the transitions between ideas. Make sure that at each
stage the reader has a roadmap of where he or she is going. The reader
must be able to see the big picture. At the beginning of each section,
make clear to the reader in advance what the purpose of that section
Step 3: Work on the transitions between ideas. Make sure that at each
will be and how that section relates to the preceding material. At the
end of each section, you may also want to remind the reader that you
have now completed what you set out to do in that section. Then, point
out what the purpose of the next section will be, and so forth. These
connection statements are called transitions the reader must always be
able to see where you are going and why and how far you have
progressed.
Step 4: Check each paragraph for unity. Each paragraph should have one
main point. Usually the central point of each paragraph is stated in a
topical sentence at the beginning of the paragraph, but not always.
You should not mix different ideas together in the same paragraph. If you
are having trouble getting a certain section of your paper to sound
right, go through that section one paragraph at a time and ask
yourself what the main point of each paragraph is. Fogy writing is often due
to mixed-up paragraphs.
Step 5: Work on the sentences to reduce the fog index. The Fog Index F
is defined as F = 0.4 (L + P), where L is the average number of words
per sentence and P is the average number of polysyllables per 100
words of text. (A polysyllable is a word with three or more syllables.) To
evaluate the Fog Index for your paper, count the number of words per
evaluate the Fog Index for your paper, count the number of words per
sentence and the number of polysyllables per 100 words for a
representative portion of your paper five or six hundred words long.
Ideally, you should strive for fog index less than 10. In technical
writing, it is sometimes difficult to get the Fog Index below 10, but
a Fog Index above 15 is a warning that your material will be very hard
for a reader to follow. Consider the following examples taken from
typical office memos. Note that both memos say the same thing. To
reduce the fog index, you must do two things: (1) reduce the length of
your sentences (by breaking long sentences into shorter ones), and (2)
get rid of as many complicated words as possible (by using simpler
words instead).
Step 6: Get rid of as many passive verbs as possible. Always check
your paper for passive verbs. ("The data were measured and the results were
correlated.") Change as many verbs as possible into the active form.
("We measured the data and correlated the results.") Using too many
passive verbs makes your writing boring. Editor's comment: At their
worst, passive verbs can make writing incomprehensible. The reader
always needs to know who is doing what to whom, and a passive verb
often obscures both the agent and the recipient of an action. In Ted's
example, the passive form obscures who it was that measured the data
and correlated the results.
Step 7: Use verbs more than nouns. Do not bury the main action of your
sentences in nouns or adjectives. Instead, let the verbs carry the
and correlated the results.
action. Consider these examples: Example: " The annual report produced
a disappointed reaction from the sponsor." Revision: " The annual
report disappointed the sponsor." Example: " It is our expectation
that we will see radiation pattern improvement when the antenna is
elevated." Revision: " We expect to improve the radiation pattern by
elevating the antenna." Your writing will sound better if you move the
action into the verbs.
Step 8: Get rid of as many abstract words as possible. Your paper will
practically always sound better if you use specific words instead of
abstract words. For example, instead of " We determined the conditions
for performance improvement." say " We measured the noise variance
necessary to increase the bit error probability by 5 percent." The
problem is that abstract words often convey a different meaning to the
reader than the one you intended.
Step 9: Check for consistent use of verb tense. Many technical
professionals change verb tense frequently between the future and
present tense for no reason. You should check that your verb tense is
consistent throughout the paper. Usually it is simplest just to stick
to the present tense: " We present our results in Section III,"
instead of " We shall present our results in Section III." (Note that the
correct future tense for the first person (I or we) is "shall", not
"will.")
Step 10: Do not use "this" as a pronoun. Avoid sentences such as "
This is ..." and " This gives ..." When "this" is used as a pronoun, its
antecedent is often missing or poorly defined, and the resulting
writing usually looks amateurish. A typical example in engineering
writing is, " By increasing the impedance, the radiation level is
increased and the electric field becomes stronger. This means that ...
"
What specifically does "this" refer to? (Moreover, who did the
"increasing"?) The solution to this problem is to change "this" to an
adjective by inserting a suitable noun. For example, use "This result
is ...," "This difficulty is due to ...," and so forth.
Step 11: Check your entire document for subtle grammatical mistakes.
We are not talking here about elementary grammar problems. Presumably you
What specifically does "this" refer to? (Moreover, who did the
do not write egregious sentences such as " He don't got no potatoes."
Rather, the idea is to watch out for more subtle problems, which are
very common in engineering writing.
Step 12: Polish and polish. Check that the sentence rhythm and timing
are pleasing, and that the ideas flow clearly and simply. If some
section does not quite sound right, work on it some more. Eventually
you will get the right wording. You are looking for subtle changes you
will get the right wording. You are looking for subtle changes that
will improve the way the paper "flows." The main tools you should use
for this step are: (1) reducing the Fog Index by shortening sentences
and eliminating big words, as discussed in Step 5, and (2) checking
each paragraph to make sure it has one main point, as described in
Step 4. You will get better at this step over time, as you develop a "feel"
for clear writing.
Step 13: Write the Conclusion. The Conclusion section should simply
summarize for the reader what has been presented in the paper.
Step 14: Write the Introduction. The Introduction is frequently the
hardest part of the paper to write. It must be smoothly written. The
Introduction should address each of the items mentioned in the outline
on Page 1. Moreover, the points discussed in Steps 1-13 should all be
used to polish your Introduction until it is as smooth as possible.
