I was introduce to the term 'slop dev' by a recent acquintance named 'Mauve'. It was inspiring and made me think more deeply about the term.
You aren't vibecoding, you are producing slop.
Vibecoding is a misnomer
I think many developers and even people today have heard of the term 'Vibe coding' the term was coined by Andrej Karpathy, a well known AI researcher and influencer in the field. I think that vibe coding does not capture that well the result of what is produced, it is also singular to code where as slop deving or slop development is more accurate. You want to just churn out effectively garbage content. A major thing that is missing from vibecoding is that you are not even coding at all. In fact, it isn't vibe coding if you are coding. Though the term has recently evovled to almost encompass any major usage of AI in writing code. The original usage of the term meant the wholesale abandonment of the code. Prompting the AI is all you need. We need a better term that captures this, because honestly its as bad as the term 'serverless'.
Instead of vibecoding, you are engineering stuff to get thrown away or turned into gold. I think that it also elucidates the harm of generating a ton of code during slop dev. The more you produce from slop dev the worse it is. Keep it minimal. Encourage less rewrites. Keep asynchronous development narrow. Work on one little bit at a time. A little slop to go is fine, a lot is just harmful. Slop also seems great for validating ideas, but it also can harm 'brand' images (if you dislike the term branding that is understandable, it is a marketing term, feel free to replace that with reputation or street cred if that makes you feel less icky). If all you read is stuff produced by slop you get sick and your brain stops working well.
Thought Pollution
Slop dev is bad for the soul.
I have a hypothesis that using artificial intelligence to avoid thinking is something that many people have started to do. I think its something especially present in education where students are already demotivated to learn about a particular topic and teachers are demotivated to mark. This hypodthesis is supported by some scientific paper that I haven't read yet apparently but will get to eventually. If you are interested in reading what it has to say you can read that here or if you are the type that needs someone to watch half read a paper or abstract you can watch that here. I am not going to pretend that I have read something or that my view is particularly more scientific than a hunch. It is just a hunch that seems somewhat prevalent in some other people (see Theo's video about it (There is no need to read the whole video just about 5-10mins past this time stamp)).
In short, Slop dev is bad for the soul. It is bad for your own growth and development as a person and it erodes the confidence that you have in your own independent thinking. Yet that isn't all, there is a community erosion and pollution aspect to doing slop dev that is also a huge problematic issue. I think that it gives people another excuse not to help each other.
To some extent, early tools like Google did this as well. For example, often people in discussion have arguments about certain topics like the definition of a word. Usually, I think that arguments about definitions are usually not arguments about what the common definition of a term is. Sometimes it truly and so dictionaries are useful, however, other times definition discussions are actually about the differing ideas that people have that they are associating with the words. They have a disagreement not merely with the terms but a deeper conversation that is being squashed by an overbearing colloquial use of the terms. For example, when discussing 'intimacy' with a partner, I may describe it in a particular way and give a couple of examples of it. My partner might say, "but isn't intimacy defined this way?" and they may look up the definition on Google and we might close the discussion there. When a much deeper concept was being unraveled that was merely being associated with the word.
Why we embrace slop
Despite the amount that slop ruins our minds and our engagement with others, we embrace it because we feel we need things to get done as soon as possible. A culture of immediacy and convenience at the detriment of many other things has led us to embrace generate a ton of trash in a completely unsustainable way. Forget the impact that AI has on the energy grid with fossil fuels. This embrace of slop will bring great detriment to improving skills our skills. If your skills degrade then you find greater and greater trouble describing what you need to the slop generators. As more slop is produced, it makes it harder for the machines to generate non-slop. The signal is lost in the noise. Eventually all you are left is slop.
Rapidity and velocity in development has made developers suffer at long-term planning for projects. Many scoff at think deeply about any system: "But we have to validate the idea first" isn't a reason we can't take some time to think about what idea we are even validating in the first place! What is even stranger is that this race to the bottom for velocity is usually followed by a much longer period of abandonning the maintenance of the software produced by it. Why do we hate maintaining software? I think a large reason is because our development pipeline rarely takes the time to think about maintaining it longterm. Even though that maintenance will be where most of the cost truly lies.
