Software Engineering

There are 43 posts filed in Software Engineering (this is page 1 of 5).

Should the software engineering profession pursue collective bargaining?

Greg Wilson, who has built a significant career in the field of software engineering, recently wrote a set of essays called “Sex and Drugs and Guns and Code” with the sub-title “a few things programmers should know about society to understand big tech, social media, and AI”. The essays cover a wide set of subjects, with this intent from the introduction: “The essays that follow will explore a few things I wish I had known earlier: where power comes from, how it is used, how its use is hidden, and how people have held the powerful accountable and made the world a fairer place.”. Greg also provides multiple references to other sources supporting the topics in the essays, which I greatly appreciate.

For today, I am going to point to one of the topics – collective bargaining. My experience in software engineering within the aerospace industry is that almost none of the engineers I worked with were part of a union. The only example I have heard of is an engineers union at Boeing, which is an affiliate of International Federation of Professional and Technical Engineers (IFPTE), AFL-CIO. In one of my previous companies, a co-worker made a comment about “we ought to start a union”. I said; “You say the word and I will call the AFL-CIO!”. When I checked into it though, we would have to get people to sign cards to indicate interest, then fight with management, etc. I told this to my friend, and he said “Sounds like too much work”, and that was the end of that.

Recently, I had lunch with a former colleague who is still working at one of my previous employers. His tales of office politics and lack of promotion opportunities made me glad that I was retired. I related the story I listed above, saying that it was an alternative to “stay or leave”. I am not sure if anything will come to it, but I felt the need to share.

I also told my friend about this series of essays – again, he did not express immediate interest, but I think I will send him a link. I think this paragraph from the essay “When The Algorithm Comes for You” is appropriate, given that AI is transforming the software engineering industry (bold text is my addition):

In 2023, the Writers Guild of America struck for five months over issues that included AI. When the strike ended, the WGA had won explicit contract language: AI cannot write or rewrite scripts, and scripts cannot be used to train AI systems. The Screen Actors Guild reached a parallel agreement that included restrictions on the digital replication of performers’ likenesses without ongoing consent [Kelly2022]. These victories established enforceable contractual limits on what employers could do with AI—limits that individual workers negotiating alone could never have secured. The lesson is not specific to Hollywood: wherever workers have collective bargaining rights, they can negotiate from a position of strength. Professional associations, open-source communities, and standards bodies can create analogous leverage in sectors where formal unions are absent or weak.

Since my retirement, I have had time to reflect about my career, the difficulties and successes, and how I could have approached things differently. I think I will add that topic to my “Future Essays” list….

Co/Recursive: Beautiful Code – Inside Greg Wilson’s Vision for Software Design –

Greg Wilson has been on a decades-long quest to transform how we teach and talk about software design. From getting rejections for using the term “beautiful code,” to empowering scientists through workshops on Python and Unix, Greg has pushed to bridge the gap between theory and practice.

Join us as Greg shares his failures and epiphanies along the way. You’ll hear how he revolutionized research computing by showing physicists the power of profilers. How he taught grad students the elegance of shell scripts. And how he’s crusaded to create a shared language to discuss software architecture with the nuance of true craftsmanship.

Greg’s captivating journey reveals that with perseverance and the right examples, we can elevate software design discussion to an art form. But that we’ve got a long way to go. You’ll come away enlightened and eager to level up your own understanding of software design.

Greg Wilson: Eleven Tips for Organizational Change “This was a proposal for the US Research Software Engineer Association 2023 conference, but was rejected…. I recognize that they are incomplete—in particular, that they are strongly biased toward what works in affluent, democratic societies—but if you’re tired of rolling heavy rocks up steep hills over and over again, maybe they will help.”

Booklist Browser app update

This weekend, I moved my cut at Booklist Browser app requirements from the Github issue to its own page. Next, I took a cut at defining the possible domains for the Booklist Browser app, using concepts as described in “How to Engineer Software” by Steve Tockey (https://www.construx.com/books/how-to-engineer-software/). My next step will be to define use cases based on the domain definition and requirements.

Paying for the development of software

I recently read a post on funding of software developed for academic/research use, the author stated that many grants are for new applications, and that funding agencies should consider supporting existing open source software that goes unfunded. There were links to papers such as the development of AstroPy, and a report by Nadia Eghbal on funding of Internet digital infrastructure development.  I have touched on this topic before in a recent post about the development/funding story of the Node Package Manager. All of these stories touch on some common problems:

  • People write software, put it out on the web, people use it or they don’t.
  • If a lot of people start using it, and start asking for features/fixes, how do those get done? How do the developers get reimbursed for their time and effort?
  • What if a business uses open source software at no direct cost? How can they depend on that software being supported and available?

Businesses that develop physical products that contain custom software pay for that software as part of the development cost of the product. This may be the best example of funding for software development.

To me, the idea of a subscription fee makes the most sense. The user paying a fee gets a specified level of support, for example, or early access to new versions. The developer gets a steadier stream of income than one-time payments (like for smartphone apps). There are other problems to address that I am not going to solve in this post, but people who develop applications should consider these topics before starting (they should “count the costs”).

 

 

Is good code boring?

Juraj Malenica writes saying “You should develop boring code”. His post highlights the following:

  • Make your code readable
  • Keep your code predictable
  • Your code is your documentation

If we take these as attributes of “boring code”, are these also attributes of “good code”? Let’s take a look.

Code readability (the ability to understand a source code function) is a good one. For most of my career, I have been working with existing code bases, and ability to understand what has been done before is very important. Working in the avionics industry, requirements and traceability go a long way towards helping understanding legacy software. If this is not available, avoiding the use of difficult-to-understand coding techniques can enhance readability.

In Juraj’s post, code predictability covers naming conventions for parameters/functions/files. For the projects I have worked on, coding standards help to define these conventions. Juraj mentioned a Python standard, there are many available.

Finally, code as documentation is a good goal. If there is no other project documentation, the source code should contain all the information necessary for another developer/engineer to pick up the code and add new features.

After this review, I would say these “boring” elements are also included in code that would be considered “good”. Is this everything needed? No (it is necessary, but not sufficent), but this is a good start.