Abstract view

The zoomed-out, abstract views and the zoomed-in, detailed views must be clearly related. If you write, and we use the kwik-kache approach to cache information on the client, the link between this abstract intent and the code must be quite clear for it to survive the first few rounds of maintenance by new programmers. 

The relationship between an abstraction and a refinement is the refinement relationship. It is an assertion that one description of a configuration of objects and actions is a more abstract view of another (see Figure 6.15). 

Gain flexibility by using orthogonal abstract views of different parts of the problem, but retain traceability to the straightforward domain terms. 

Do not lose sight of the domain terms even in your code. Use refinement to maintain trace-ability even as your code becomes more decoupled and reusable. Wherever possible, recast the problem domain descriptions themselves using these orthogonal and more abstract views remember, you are actively constructing a model of reality and not passively discovering it. Use frameworks (see Chapter 9, Model Frameworks and Template Packages) to explicitly document the mapping from domain terms to terms and roles in the abstract problem descriptions. 

The detailed tasks for Fault Management and Network Expansion are now rewritten to use these tools (see Figure 7.12). The old network expansion tasks are replaced by new procedures, but both the old and the new achieve the same high-level aims the new tools provide a new refinement in which portions of the high-level tasks are performed by software tools. This is a good case for separating the detailed and abstract views of the business. We can then easily deal with the different realizations of the business processes, including those that involve software. 

Robert Bridges, was not a mathematician but a medical man as well as a classical scholar and poet, and took a more earthy, or less abstract, view. In the first book of his Testament of Beauty he has a striking metaphor for scientific research and scholarly endeavor. (The unusual words are not misprints, but are made in accord with his personal theory of orthography (Bridges, 1924).). 

There is nothing more difficult than a good definition, for it is scarcely possible to express, in a few words, the abstracted view of an infinite variety of facts. 

We therefore put forward a more abstract view of what a template is and consider a template as the sum of all connections between the species reacting with each other which are involved in geometrically controlling the reactivity in the desired way. It is the array of interactions and their spatial arrangement that count. 

Abstract view of part of the metabolic network of the bacterium E. coli. From http //www.genome.ad.jp/kegg/ kegg.html... 

Figure 7.3. An example of a PubMed record in Abstract format as returned through Entrez. This Abstract view for the first reference shown in Figure 7.2. This view provides links to Related Articles, Books, LinkOut, and the actual, full-text journal paper. See text for details.

Krii ma G, Khalilieh S, Ezzet F, Marino M, Kantesaria B, Lim J, Batra V. Effect of cimeti-dine on the i iarmacokinetics of deslciatadine. AAPS PharmSci (2001) 3, 3. Available at htq) //www.aapspharmaceutica.cQm/seaich/abstract view.asp id=824 ct=01 Abstracts (accessed 20/08/07). 

Identify subjects through examining objects and their structures to obtain this more complex abstract view of the issue. Each classification structure and each assembly structure will comprise one subject. 

Design Model A simplified and abstract view of the development process. 

As a more abstract use case one might consider to model only the launch use case that includes by definition programming, launch and initiation. This is also implemented in some systems. But this abstract view does not allow modelling the failure modes of non initiation and unintended initiation. Therefore such an abstract use case view, which only takes the intended... 

There is great diversity in the way in which fibres fracture, depending on the fibre type and the mode of application of stress. The subject can be studied at three levels the macroscopic applied forces the microscopic level, which shows the path of fibre breakage and the response of the molecules. Attempts at theoretical interpretations of the gross experimental observations in terms of molecular effects can be misleading, because abstract statements about the forces involved are related to abstract views of imperfectly understood structures. The intermediate level, for which the scanning electron microscope (SEM) provides concrete evidence, must be taken into account in order to understand the mechanisms of failure. 

In the early requirements elicitation phase, a detailed component model of the system with interfaces and interconnection may not be available. But for the initial specification of MTBF a very abstract view of the system and of its subsystems is already enough. This abstract representation of the system is called the product breakdown structure (PBS). In the following, we use a very simple ontology, defined along the lines of [1]. 

However, to say that representation is secondary and that all that matters is the more abstract periodic law is to express a philosophical prejudice. This is an issue on which writers such as Pierre Duhem, Norman Campbell and more recently Mary Hesse have all written very eloquently (Duhem, Physical Theory Campbell, Science Hesse, Models). The question is whether the scientist should focus on models or, instead, on the more abstract mathematical relationships or laws. According to some of these authors British scientists, at least at the turn of the twentieth century, were more predisposed to building and thinking about models whereas continental scientists were more likely to dwell in just in the abstract laws and formalism. The development of quantum mechanics on the European continent would appear to be an example of the triumph of the abstract view of physical theories over the desire to regsnd the subatomic world realistically. ... 

---