I feel that this sentence is awkward,

The nature of convergences of topological relations of plane adjacencies, allowing for the reconstruction of roof corner geometries with preserved topology, can be derived from cycles in roof topology graphs.

If I explain the meaning what I am trying to explain in a simple way: If we think a roof corner (or any corner in a cub for example), we can assume it is made by intersecting adjacent roof planes. By knowing the property topology of adjacent planes, we can undestand the nature how that corner is made and how another corner is made (as they are not the same). This convergence of adjacent planes or lets say convergence of topological relations can be derived from cycles in a topology graph. At the same time, Correct recognition of nature of convergence of topological relations allow to reconstruct roof corners in a correct way (by preserving the correct topology).

SO, I want mention these all ideas in a sentence or two. Can anyone suggest me the best way to enhance this sentence even at least by breaking this.

  • Like many of your previous questions, this has a lot of highly technical content which I, and I think most other users here, don't understand. It would help if you explained in detail what exactly this means. Jan 25, 2014 at 12:28
  • 1
    @StoneyB: I have explained the behind idea of that sentence just below to my question. thanks. Hope you can help me now. As I am writing a technical paper and this sentence is written in the abstract of my paper, it is good if you help me to write this to be suited for that.
    – gnp
    Jan 25, 2014 at 13:10
  • 1
    Based on your explanation, here is one way to rephrase it (in two sentences): The convergence of topological relations of adjacent planes can be derived from cycles in a topology graph. Recognizing the nature of such convergence allows us to reconstruct roof corners correctly by preserving the correct topology. --I personally prefer a paper that explains things straightforwardly, concisely, and clearly, though such a paper is not the easiest thing to find. Jan 25, 2014 at 13:53

2 Answers 2


Let's start with the stylistic awkwardnesses.

  • Horror aequi   That’s a term from linguistics meaning “dread of the same thing”. Your opening stretch uses the same construction three times in a row: NP of NP of NP of NP, and the middle stretch uses it again: reconstruction of roof corner topologies. Repeating the same structure even once is unpleasant to readers and hearers, and multiple repetitions are confusing.

    Write for your ear, listen to what you are writing, and you will spot this very easily.

  • Hypernominalization   That’s a joke term which exhibits the very literary fault it names: turning verbs into nouns. Examples in your passage are convergence, relations, reconstruction. For a couple of centuries this has been a chronic vice among scientific and technical writers (who seem to regard ‘things’ as somehow more real than ’actions‘), and it is made even worse by the modern reliance on noun-heavy phrases like roof corner topologies and roof corner graphs. Using this many nouns makes it very difficult for a reader to untangle the relationships between them.

    Wherever something happens or somebody does something, express it as a verb. That will be stronger and more easily understood, and will help you avoid those ‘horrible’ of phrases.

  • Depersonalization   That means straining unnecessarily to keep yourself out of your narrative by a) overusing the passive voice (e.g., are derived from) and b) writing as if things which happen do so as a result of some abstract quality in the entities you are talking about rather than as result of your own actions (e.g. allowing reconstruction). This, again, is a very old vice in technical writing; but over the past couple of generations it has gradually faded, and today only very insecure writers make a fetish of impersonality.

    You may use the editorial we if it makes you more comfortable, but if you are talking about something you do or have done, say so. You are telling your readers how they may do something; make it easy for them to distinguish what they are supposed to do from the entities they are supposed to do it to.

There is also a very severe structural problem. What you are describing has three stages:

  • Inputs: roof topology graphs
  • Process: derive “the nature of convergences &c”
  • Outputs: roof corner geometries, with preserved topologies

Note what DamkerngT. does in his admirable first-order rewrite: He puts the outputs last!—whereas you put the inputs last.

I'm still not entirely clear what exactly you do, step by step, or exactly how the entities you name are related; in particular, I have no idea how geometries differ from topologies, whether cycles are entities in the graphs or a process you apply, and whether convergence is a characteristic you observe or a mathematical action you perform. But here's a second-order rewrite that at least puts the major pieces in the right order

We [do something with] cycles in the roof topology graphs to derive [?intersecting?] adjacent planes; these are the basis on which we reconstruct roof corner geometries while preserving the correct topology.

One final note: the phrase you begin with, the nature of, sends up a red flag for me. Nature is what in the trade we call a waffle word or weasel word. It is so vague that it is essentially meaningless; what it usually means is “something interesting which I ’m not going to take the trouble to define”. If you catch yourself using nature it’s usually a signal that you’re leaving out something your reader needs to know.


As StoneyB's comments says, the technical content in your question would be understood by only a few users of this site. That aside, I am going to make an attempt. A general principle to follow is that clarity should be preferred to conciseness.

As you have pointed out, long sentences are harder to understand because it forces the user to either remember a lot of detail until they reach the end, or repeatedly read it back and forth till they "get it".

In case of this sentence, you could probably reorder the clause as below.

If we allow for reconstruction of roof corner geometries with preserved topology, then the nature of convergences of topological relations of plane adjacencies can be derived from cycles in roof topology graphs.

The if ... then ... structure provides the reader with a logical "pause" after the if part. This lets their mind to "take a deep breath" (if you will) before moving on to the then part. However, in this case, the then part is still too long, and you could further break it up, using multiple sentences. One such example is as follows:

Suppose we wish to derive the nature of convergences of topological relations of plane adjacencies. If we allow for reconstruction of roof corner geometries with preserved topology, then these could be derived from cycles in roof topology graphs.

In your thesis, you need to consider what's the best way to break up the sentence based on your target audience as well as the specific context, but the above is one approach you could follow. Of course, you could keep repeating this until each such piece is "small enough", but then you run the risk of making your text unnecessarily verbose, so there's some tradeoff there you need to consider.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .