Post

This made me remember that I need to re-read "living with complexity". Best practices as a concept has driven me nuts in the past because you would find contradicting ones and often there was little argument for the "best' part of their description.

Very informative video! It's likely the fact I'm not a UX designer, but I was initially very confused as to the pushback against "sensible" and I wonder if that's a terminology thing specific to UX design, as it seems from context that people are using it to mean "perfect" or "best"? Whereas I'd /

think of "sensible" as more "not awful". More specifically I guess two things: "Viable": I want the default layout to be viable for whatever I want to do, even if not optimal. Nothing I need should be eg hidden until I press a shortcut, it should be either visible or clear how I get there.

"Inviting": The software should not be overwhelming or scary. Failure here is kinda the inverse of failure in viability, where too much is immediately visible and I can't form a mental model easily. You could probably think of this in terms of "branching factor" when it comes to menus etc 🤔

Forgive me if I missed some aspect but to me, "default" implies that there are options. So before Blender added left select, right click wasn't the default, it was the only way. After, left was the *sensible default* since that's what most people expect.

And so basically I agree with you that there should be ways to customize the product, but you still should think of what the best initial value for a setting is. For a pilot chair maybe extended all the way is the best because then everyone can get in it before starting to adjust it :)

Blender did have an option to change it back then, so it was a default, but there was a much wider assumption that it was the 'sensible' default, so who or why would ever change it? You should indeed think of an initial value, but do not think of it as 'best'. That way lies ruin, and perscription.

I wonder if instead of sensible defaults, there could be sensible presets? maybe?

Especially if the presets are presets only for a part of the application. Like a subset

Really enjoyed the video, great job!

Most of all, I hope this finally helps convince more teams that their tools need customization of their UI, so that users can get the best workflows out of them. And for people to stop applying 2000's website UX/UI paradigms to professional daily use tools. Please stop, you can't blindly do that!

Web UI conventions sure cast a large shadow over the entire UI world, largely not in a good way. And that’s coming from me, a web frontend dev and UX designer who works on a complex web app.

I found your perspective quite interesting and it reminds me to be less afraid to make certain parts of our tools more readily accessible for our specialist users, even if it means there are lots of inputs visible at once.

I have received many nice comments on this video so far, thank you! One in particular stood out to me: "You are so correct in all of this, you can really tell you have a lot of experience!" But, my thoughts on Tool Design haven't changed, since about 2014. I was called nuts back then. Times change

It's really strange to have things I used to be ridiculed for, now be shown as examples instead I'm not going to let that happen to the next generation. It's one of the big reasons I always listen to arguments from students, juniors, and others who are just starting out. All we have to do is listen

Nice, this video is already doing quite a bit better than my previous ones. Thank you to everyone that subscribed!

A peek behind the curtain: These are the thumbnails and titles I have set up so far. None of them sound particularly outstanding to me, but it is helpful for them to pull folks in, because that raises the click-through-rate on youtube, which then shows the video to more people. A perpetual cycle.