Summary of Software Art Thou: Glenn Vanderburg - Real Software Engineering

This is an AI generated summary. There may be inaccuracies.
Summarize another video · Purchase summarize.tech Premium

00:00:00 - 01:00:00

The video discusses the history of software engineering, and how it has often been modeled after traditional engineering disciplines. However, the author argues that this is not an effective way to approach software engineering, and that a more document-based approach is needed in order to improve the discipline.

00:00:00 software engineering has gone wrong, and Glen Vanderburg believes that it can be fixed by returning to the principles of engineering. He starts by discussing how software development first began to be viewed as engineering, and how the first conference was characterized by debate and humility, while the second conference was characterized by certainty and some arrogance. Vanderburg argues that academic software engineering should aim to become more like engineering, and he provides a brief history of the field.
00:05:00 The author discusses the history of software development, focusing on the period from 1969-2001. He points out that while mainstream software engineering had a high failure rate, academic software engineering set a target of an idealized view of engineering, but they were unable to achieve success. In response, a number of people started trying to find new more appropriate disciplined ways of building software. The agile movement was one of these attempts, and it became popular in 2001 with the publication of the "Agile Manifesto."
00:10:00 The speaker discusses the history of academic software engineering, discussing how it went wrong from the beginning. He argues that software development is an engineering field, and that the rational model of decision-making is a caricature of engineering. He also discusses how objections were ignored from the beginning, and how this led to the current state of academic software engineering.
00:15:00 Glenn Vanderburg, lead editor of the first conference on software engineering, presents a memoir of his experiences in the field of software engineering. He argues that the rational model, which assumes that the designer derives their design from a statement of requirements, is unrealistic and that software engineering started with the assumption of engineering as applied science.
00:20:00 The video discusses the history of software engineering and how it went from being a discipline focused on designing documents to one focused on code. It also discusses how one of the problems with this shift is that code can't fill the role of documentation in a software engineering process. Finally, the speaker discusses how academic software engineering has disproportionally emphasized correctness over cost, which has had negative consequences on the project life cycle.
00:25:00 The video discusses the differences between engineering disciplines, with civil engineering and structural engineering being more towards the defined end of the spectrum. It also points out that software engineering falls more towards the empirical end of the spectrum, as it does not typically design artifacts but processes.
00:30:00 In this video, software engineer Glenn Vanderburg discusses the creative process behind engineering, emphasizing the importance of intuition and experimentation over formal methods. Vanderburg also discusses the importance of creativity in aerospace engineering and other engineering disciplines.
00:35:00 In this video, engineering student Glenn Vanderburg discusses the principles of real software engineering. He discusses how engineers use mathematics to prove that a software design will work, and how they add safety factors to account for inaccuracies in the model. He also discusses the Tacoma Narrows Bridge disaster, in which the bridge's thin deck was subjected to a forces that nobody had ever seen before.
00:40:00 The video discusses the history and work of Glenn Vanderburg, a bridge engineer who pioneered the use of reinforced concrete in bridge construction. He used models at multiple scales to establish the validity of his designs, and later, the math caught up and demonstrated their feasibility. His designs are still used today as standard parts of the bridge builders toolkit.
00:45:00 This video discusses the history of software engineering, and discusses how earlier software engineering processes were modeled after traditional engineering disciplines. The video also discusses how current software engineering processes use computer models to validate designs.
00:50:00 The video discusses how software engineering should be done, comparing it to real world engineering. The main points are that first, a design must be produced in document form, then it must be handed off to laborers who build the resulting structure. However, this analogy has never worked well, and the problem is that the second step, transferring the design to code, has never been done well enough. DevOps can help with this by providing automated unit testing and integration testing.
00:55:00 This video discusses the importance of documents in software engineering, and how they can help to improve the discipline and rigor of a software process. It features a discussion of tabular mathematical expressions, and how they can be used to specify complex logic and requirements for code. Finally, it discusses the five practices that are just "noise suppressors" when it comes to software engineering processes, and how they can be removed to create a more rigorous continuum.

01:00:00 - 01:05:00

Glenn Vanderburg discusses the benefits of using software engineering practices, such as pair programming and extreme programming, to reduce the cost and risk of developing software. He notes that these practices are effective at achieving this because they provide feedback and verification of every decision, which helps to improve the reliability and scalability of software.