An introduction to a series of articles derived from an AI’s outline.
We’ve published a series of articles about the pitfalls of the design process, and it’s finally been pointed out that I never wrote an introductory article explaining what all of this is about. If you’ll excuse me going about it in this backwards fashion, I’ll rectify that now.
The series initially came about in the way that many such things do; casting around for ideas for articles. We had recently decided to write articles about the huge variety of things we have encountered in our careers as software and hardware designers, which sounds great until you actually have to think of something to write. We had also been playing around with AI, seeing what we could make it do and the sorts of things it was actually helpful with. We had rejected getting an AI to write articles for us (I have standards and AI output often doesn’t meet them), but perhaps it could be used to generate ideas. Probably not anything earth-shatteringly novel — an LLM is after all the product of its training data — but at the very least it should show up what people think is interesting.
So we dutifully sat down and typed “Outline an article on the pitfalls of electronic product development” at the prompt. Some tweaking later we had a four page document outlining a series of five articles, each concentrating on a different step in the design process. Badly formatted and terribly superficial admittedly, but perfectly adequate for an outline. It even had appropriate alliterative titles for each article (and understood that “Conceptual” and “Quicksand” do alliterate), which pleased me far more than it should have.
I’m going to leave an analysis of the AI’s output and how it bears on public perceptions of electronic product design for another article. There’s quite a lot to say, and reading the articles first will give you more of an understanding of where we are coming from in our conclusions. What I will say at this point is that we didn’t really consider scale when we prompted for the outline. It’s interesting to see that the outline gravitates to concerns that have more effect on high volume manufacturing projects.
There are places in the articles where we point out this difference explicitly. This is most obvious when we get to the manufacturing stage, as the number of units you are intending to build clearly makes a very big difference to your priorities, but it does show up elsewhere. For example in the very first article we skipped over the question of whether or not there is a market for the device you are designing. Large companies can afford to do considerable market research into how the proposed device will be received, what are the “must have” features and so on. Failing to do this can mean that you have a solution in search of a problem, as the AI put it. In our experience the opposite is also problematic; believing wholeheartedly that your research has given you the One True Product can lead to attempting to build the impossible or designing something too expensive for the market.
Smaller product-based companies often don’t really have the resources to do more than informal market research. Doing some is still advisable — even just asking around your friends can reveal that your proposed device will not fill the niche you thought it would — but such companies often come about because of a perceived hole in the market. They are a lot more vulnerable to the overenthusiasm side of the coin, resulting in a proposed product they cannot afford to manufacture. Acorn were joking when they put a literal functioning kitchen sink in the “rocket ship” demonstration Risc PC back in 1996 (sorry, I can’t find any pictures), but it is an issue small companies can struggle with.
Scale matters in so many of the issues that can bedevil product development, and you should bear that in mind as you read the other articles. A process-heavy solution to a problem can kill a small company’s ability to function. A process-light solution can kill a large company’s reputation. The answer is always a balance, and finding the right balance for a particular product from a particular creator at a particular time is a skill that only comes with experience. We hope that in this series we can offer you a little of our experience and help you find that balance.