Skip to main content

Difference betwen computer science & software engineering

From Dr. Dobbs: Software Engineering ≠ Computer Science

The author puts forward the thesis:
Software engineering will never be a rigorous discipline with proven
results, because it involves human activity.

He makes a good case. There are numerous formal methods that are being used to eat away at the fringes of the unprovable in software engineering through formal processes - but it usually requires defining the problem without a human element.

Indeed, look at some of the highly regarded practices for engineering today - with phrases such as "code smells", "team cohesion" - are we really expecting formal proofs?

At this point in the analysis, I'm shaking my head since I don't want to understand why a degree in Computer Science (like I have) might have been a waste... but looking at the diagram and reading through to the end, he makes the relationship between the two clear and makes the world (as I see it) whole again.

This is from the point of view of someone who gained a B.S. in C.S. from a school with a heavy emphasis on engineering disciplines, rather than pure science or mathematics. This was after a senior year transfer from a prior C.S. program at a school geared towards pure science and math. I understood as I entered my senior year that I simply wouldn't be employable except as a mathematician. I have been grateful for the opportunity to see the engineering side of world academically.
Labels:

Comments

Post a Comment

Popular posts from this blog

Let's Not Mess Around with Security on our Personal Systems Either!

Essential Security Practices for Your Personal Systems Ensuring a minimal level of cybersecurity, privacy, and availability on your personal systems means you need to manage the following essential practices. This is a brief overview of recommendations from sources like CISA, NSA, etc., focused on personal laptop, phone, and other systems' security. Anti-virus  I've found you'll get the best anti-virus protection and usability from a paid product - I've always had good luck with Norton labeled products. If you are looking for current vendor offerings see:  https://www.pcmag.com/picks/the-best-antivirus-protection Regardless of whether you choose to use a commercial product or open-source anti-virus tool, it is absolutely something you need to use. This is the minimally needed level of system security. Once installed, ideally, it should be invisible until there's a security problem it can't prevent or solve.   Backups You need to have at least a minimal level of ...
Post a Comment
Read more

RACI, Cybersecurity and NICE Framework

The NICE framework from a RACI point of view The NICE framework ( NIST SP 800-181 rev. 1) established a standard approach for describing cybersecurity work, in order to help stakeholders share a common language and ideally improve how to identify, recruit, develop and retain talent. It breaks down cybersecurity work role categories into: Oversight and Governance; Design and Development; Implementation and Operation; Protection and Defense; Investigation.  Which is very cybersecurity-centric and not related to common tools for project management within companies. Especially smaller enterprises that do not have dedicated people to mange and coordinate cybersecurity needs. A  RACI chart  is   a project management tool used to define and clarify roles and responsibilities within a project team.   It stands for Responsible, Accountable, Consulted, and Informed, and visually represents who is responsible for what, who is accountable for the outcome, who needs to be c...
Post a Comment
Read more

Typescript - It might not be easier, but but it's surely different

Typescript is a statically typed language, that is a superset of JavaScript. I've had the discussions and debates about that aspect of the language. I am all for static typing. Any way my tools can help me be better is alright by me. So I avoid the ' any ' type designation and make sure I have guards on ' unknown ' types, as much as I can.  Any  does not carry any useful type information, while unknown does, and allows it to enforce type checking.  Anything can be assigned to a variable of type unknown , but an unknown value cannot be assigned to variables of other types without explicit type assertion or narrowing. Similarly, no operations are permitted on an unknown value until its type is refined. This behavior ensures type safety and prevents runtime errors. (Refined with help from google). I bring this up because I was arguing with the compiler recently because I'd assumed both made no use on any type information in any circumstance - because I haven't ...
Post a Comment
Read more