New post: "You had a story" https://nolanlawson.com/2026/02/18/you-had-a-story/
I'm still exploring this topic, but this time through the lens of fiction (kinda).
Discussion
Loading...
@nolan This one hits hard, because I recognize the tendency toward elitist gatekeeping in our industry, and I know I'm not immune. So I see the fact that coding agents are now allowing people to bypass that as kind of a good thing, and also an indictment of our industry for not doing enough to lower the barrier of entry before.
Only kind of a good thing, though, because especially when you write apps that are exposed to the Internet, the cost of innocent beginner mistakes can be very high.
@matt Thank you, yes, I'm not immune to it either. When I reflect on a lot of my blog posts, they are written for experts to deepen their expertise (not always a bad thing).
For your second point I worry that it's not going to matter because software is becoming disposable. But for now yes there are really reckless experiments with things like OpenClaw that I think are going to do a lot of damage.
@nolan "they are written for experts to deepen their expertise ..."
I reflect on this a lot in accessibility. The majority of content being pumped out into the world on that topic is aimed squarely at beginners, with a subset by people who are trying to deepen their own knowledge through experimentation in a public way.
Accessibility content that is written specifically for and by experts is relatively rare, and it makes me wonder what that says about the industry. Is there no audience for it? Are experts just quietly doing their own thing to expand their knowledge, or sharing targeted expertise mostly with paid clients? Are the bulk of the people working in accessibility plateauing at lower levels?
@jscholes @nolan I'm embarrassed to admit that I haven't done much sharing of whatever expertise I've built on accessibility. On a few occasions, when someone has asked me a question, I've answered at length. But I haven't blogged or otherwise publicly shared what I know. To some extent that might have been because, when I was working on a third-party Windows screen reader (before I went to Microsoft), I felt pressure to keep the "dark art" proprietary. But it could also be just laziness.
@matt @jscholes Accessibility feels like a bit of an odd duck to me, because a lot of the expert content is "such-and-such works in NVDA, such-and-such is broken in JAWS," etc (or swap them around; it doesn't matter). It kind of reminds me of a lot of my performance benchmarking posts, where the results are outdated as soon as I write them, because Safari implements some optimization, Firefox makes a different tradeoff, etc. Maybe since it's a bit tedious/prone to staleness experts avoid it.
@nolan From my point of view at least, I was thinking more about experiential design and adjacent topics.
How do you make a good collaborative document editing experience? What can you do better for braille users? How do you allow someone to digitally explore a 3D object?
One big problem in web accessibility is that there's usually too much focus on the ARIA attributes, markup, and other technical details. A frequent opening gambit in conversations with clients is: "We're thinking about building X; what do you think about `aria-y` here and `aria-z` there?"
And if they're engineers, this makes some sense. But what I usually think is that they need to take a step back and think about the user experience, and then have the implementation details follow. And also that if it's reached the engineering stage with only the visual UX being worked out for users with no accessibility needs, we're having this conversation a bit late.
When the UX has been worked out, I will talk and test ARIA and HTML all day long.
I wrote a bit about this last year, and would like to publish more on it at some point:
https://jamesscholes.com/2025/10/31/on-aria-and-experiential-design/
@nolan I don't see anything wrong with experts deepening our expertise. That's how we produce truly outstanding software, or "hit the high notes" as Joel Spolsky put it.
But I wish we wouldn't present it to people outside our field as mystical incantations, or magic in the cloud. And having to grind Leetcode to pass an interview is also a questionable rite of passage.
@nolan On that point about the backend being presented as the magical thing in the cloud that makes your app work, I don't know, maybe of all the things that kids are forced to learn in school (especially middle and high school), one or two of them could be retired in favor of learning more about how computers actually work, including the backend.
@matt Yeah I was trying to sketch out a character; I've never done LeetCode for example. I agree it's not gatekeeping to be an expert, but I think a lot of us got comfortable with having control over things that, themselves, control other people and make them feel a bit powerless. Oddly this kind of goes back to a blog post I recall us talking a lot about years ago: https://nolanlawson.com/2018/05/16/should-computers-serve-humans-or-should-humans-serve-computers/