Skip to content

Latest commit

 

History

History
64 lines (47 loc) · 2.48 KB

foreword.md

File metadata and controls

64 lines (47 loc) · 2.48 KB
Raw

Foreword

On the very beginning of the lectures, authors explain clearly what is their view on these lectures regarding naming them a part of "computer science" curricula. It is not in fact science what is done here, and it is neither related to computers only.

It is in fact about two things:

  1. How we define "what is true" - declarative knowledge
  2. How we define "how to" do something - imperative knowledge

Programming itself they see as a way to express our knowledge about things and give an imperative explanation of how to deal with that knowledge. What are the procedures we have to define in a programming language that will direct a process which is going to be executed by computer in order to do something useful.

On the other side there is a fact that software is inherently complex and that problems that software engineering is to solve are becoming more complex over time. Therefore, it is a task of "computer science" to give answers on how to manage that complexity and provide some means integrated in the programming language that can easily deal with it.

Techniques for managing complexity

Two basic notions we deal with are:

  • Data
  • Procedures which operate on the data

All knowledge we have is represented by these, and all the techniques are to be expresses by using data or/and procedures.

Technique 1: Black-box abstraction

Being able to define and name new pieces of knowledge and expose them as a black box entity in the language which can be used without knowing the inner details is crucial thing in managing complexity. That way, we create levels of abstractions and minimize number of items that we have to deal with on certain level.

Technique 2: Generalization through conventional interfaces

If we would define addition as operation in a language wouldn't be nice if it could add two numbers as well as two electrical signals. But without having to restructure all the code but only the operands which will certainly be different in these cases. Defining conventional interfaces is a way to make abstracting of operations away from the details of operands possible.

Some approaches to this problem are:

  • generic operations
  • modularization
  • object-oriented programming
  • operations on aggregates

Technique 3: Meta-linguistic abstraction

This is in fact about creating new programming languages that are more suitable for capturing knowledge present in some domain. It is about creating your own tool for the purpose before using a tool meant for something else.