|Table of Contents|
|Table of Contents|
The modularisation process
In table 5.2, we have given an indication of the relative sizes of the modules and the sections. In the modularised articles, the Interpretation is by far the largest module: it takes up about half of the article A05. In the original version, the section Discussion (roughly speaking the counterpart of this module) also forms the largest section in the original version. The collection of the sections 6.Comparison of the measurements and the calculations, 7.Discussion and 8.Rotational coupling in A08, which can be grouped under a higher level heading of Discussion, is less dominant, but it still forms a large part of the article.
The main reason why Interpretation modules are comparatively large is because they play a central role in the article: in the Interpretation the experimental line of the article and the theoretical line are brought together. Moreover, the Results modules only seem to be very small, but that is artificial: firstly, only the text of the treated results is regarded in these size considerations, whereas the figures are the most important part of the treated results, as well as the most voluminous; secondly, as we saw, in the modularised version raw data could have been included, but they are unavailable in this case. Furthermore, the Interpretation modules are relatively large, because the Methods modules of the articles are kept relatively small, as a lot of information concerning the methods is represented at the mesoscopic level.
The Interpretation is not only an large module, but also a difficult one: it contains a lot
of information that has to be dealt with in a clear manner within the module, and it is also rather entangled
with the modules Theoretical methods and the Results.
The boundaries of the module: entanglement
The entanglement of the interpretation, the theoretical methods and the results is illustrated by the fact that part of the Interpretation is derived from sections that also contain information that is represented in the Theoretical methods or the Results. For example, a small part of the section 3.Potential curves is presented in A08-m5bi. The first part of the section 4.Calculations is given in A08-m3cii, whereas the second part was supposed to deal with the results of these calculations. However, most of that part is given in the Interpretation, in particular in A08-m5a.
The theory or theories and technique(s) used for the interpretation are described and discussed (with respect to their reliability and applicability) in the Theoretical methods module m3c. The usage of these theoretical methods, however, in the process of interpreting the results, is included in the Interpretation, with the description and discussion of the resulting interpretation(s). This dependency of the Interpretation on the Theoretical methods is expressed by means of links. In the modularised versions of the analysed articles, most of these theoretical details are applicable to more than one article and therefore presented in the mesoscopic module MESO-m3c. For efficiency, most links in the Interpretation module are directly targeted at the mesoscopic modules. However, we have included at least one link to the module Theoretical methods of the article under consideration, in order to express the coherence of the article.
In the modularised articles, a Results module could be separated from an Interpretation module. In the domain of experimental molecular dynamics, interpreting the results obtained in terms of scattering implies explaining them in terms of reaction mechanisms specified in a molecular dynamics model. In the Qualitative interpretation module A05-m5a, for example, the main features of the curve of the experimental cross section are described in the light of a simple atom-atom model for molecular collisions. In the Quantitative interpretation, theoretical cross sections are calculated using that model and compared with the experimental ones.5.18
Separating the results from their interpretation calls for overlap between the Results and the Interpretation module. In the first place, for the interpretation to make any sense, the core of the results has to be repeated in the Interpretation. In particular, the key figures representing the results that are to be interpreted are copied to the Interpretation module.
The description in words of the special features of the (graphically presented) results is included in the Interpretation module, rather than in the Treated results. For example, a description of the curves presented in A05-m4bi is included in the Qualitative interpretation module A05-m5ai. The description of the curves leans towards the interpretation of the curve: the authors highlight features that they consider relevant for the solution of the central problem. A more practical reason is that the discussion of those features in the Interpretation can be performed more smoothly if the features are described in the same module.
The second reason for overlap between the Results module and the Interpretation is that intermediary results generated in the course of the interpretation are copied to the Results module in order to allow the reader to locate and retrieve them as results.
Within the module: internal structure
Once the author has decided what to represent in the Interpretation, the representation of that information in the module does not follow automatically: the module itself is rather complicated as well. It is a large module, which may contain different parts. It is also argumentative in nature. The Interpretation module contains argumentation on the reliability of the interpretation(s), which is based firstly on the reliability of the results in the module Results (which in its turn depends on the reliability of the methods used to generate them in module Methods), and secondly on the reliability and applicability of the chosen interpretation theory.
. Many of the external references to other work appear
in the Interpretation modules. The function of the references is, for example, the comparison with the
interpretations given by others for similar results, or input of factual information copied from another
article into a calculation.
The resulting module
The Interpretation module is a very important one. In this domain, the interaction between theory and experiment is emphasised. The Interpretation can be seen as the `core' of the argumentation in the article. The argumentation in the Interpretation is `supported' by the arguments and detailed reports given in other modules: The Results module gives the details behind the salient features of the results that are explained in the Interpretation module, the Methods give the background of how those results were obtained and the details of the theoretical model taken from the theoretical toolbox to interpret the experimental results, the module Positioning announces what the author aims to do and why. The Outcome then summarises what has been achieved. Thus, the module Interpretation is strongly related to many other modules. For the sake of the coherence of the article, these relations, in particular the dependency relations, are expressed in explicit in hyperlinks.
Because the Interpretation is generally complicated, it is difficult to endow it with a clear structure. But it is also very important to do so in order to allow the reader to understand and accept the information. The importance of a careful explanation and a clear structure immediately becomes clear when the interpretation is consulted in the original articles: it is often difficult to keep track of the line of reasoning.
The act of interpretation may involve one or more sub-problem solution patterns in which different methods are used to interpret the results. The results can, for example, be explained qualitatively with respect to their general features. The results can also (additionally to the qualitative interpretation or not) be inserted as input into calculations based on chosen theories considered in parallel. These calculations can consist of a series of separate steps (each featuring a sub-problem-solution pattern), in which the output of each step is required as input in the next one. Or different types of results can be obtained independently, and then compared in the interpretation. Thus, there can be more than one complementary interpretation (qualitative and quantitative) that are merged into a complete interpretation. There can also be more than one explicit candidate interpretation before one is chosen. The guidelines for this domain allow for the distinction by the conceptual function of two constituent modules within the Interpretation: Qualitative interpretation and a Quantitative interpretation.
In A08-m5, two candidate interpretations are described and discussed. The measured differential cross sections are first interpreted qualitatively, using a simple semi-classical model that takes into account the Landau-Zener coupling (see A08-m5a). This qualitative interpretation is quite satisfactory. Then the authors proceed with a quantitative interpretation (in A08-m5b), in which they first determine the potential parameters (A08-m5bi). Subsequently, using those parameters as input, the differential cross sections are calculated in A08-m5bii1 based on the same semi-classical model. This quantitative interpretation, however, is not satisfactory. Therefore, the same experimental results are interpreted again in A08-m5bii2, this time taking rotational coupling into account as well. This last interpretation is singled out as the more successful one, because it provides an accurate, quantitative explanation of the reaction dynamics.
In these big and involved Interpretation modules the `module summaries' in the complex modules, such as A05-m5, A05-m5a and A05-m5b, are particularly important. The module summaries of the `constituent complex modules' A05-m5a and A05-m5b are almost literal copies of the highest level module summary. However, by making these aggregates explicit, and giving explicit `module summaries' in each of them, the clarity of the module is clearly enhanced. The reader who consults a separate module needs the information to learn of the context and he does not encounter all copies. The reader who consults the article in its entirety or the Interpretation module as a whole following the complete sequential path does encounter multiple copies of the same text. The essay-type path avoids most module summaries, but in cases of large and complicated modules, the readers is more helped than hindered by a reminder of the overall structure of the module.