mateja

Subpages deserve details and polish too. Deposits page for @flashnet 👇

it's live, check it out: flashnet.xyz/products/deposi…

How do you know if a direction is worth pursuing? Not just exciting in the moment, or something that looks good in your head, but worth the time and craft it takes to make it real? Will it feel as good as I’m imagining it, or should I expand my range and explore a completely different approach? And how do I figure that out before I’ve already spent days and hours on the wrong one? These questions tend to be both really important and almost impossible to answer by just thinking about them. Personally, I have a lot of ideas, and if I’m being honest, most of them aren’t great for what the project actually needs. So instead of picking one and hoping, I use Wonder to generate 20 or 30 rough versions of something like a hero section first. None of them are polished, and most of them end up being absolutely wrong, but seeing them side by side makes it pretty clear which two or three are worth giving my time to. So this is how I usually do it: 1. I generate 20 to 25 rough directions in Wonder, for something like a hero section, and see which two or three are actually worth my time. 2. I take those into Figma and edit them by hand to give them the polish and craft they need. 3. I bring them back into Wonder, where they either go into production or through a few more rounds of iteration. Curious how everyone else does it, so drop your thoughts below :))

There’s this thing called unit bias, where we assume one unit of something is the right amount. One plate, one bag of chips, one episode of something... Casinos know this really well. That’s why you play with chips instead of cash, so a $100 chip just feels like one chip and is way easier to let go of. Product design does the same thing all the time. AI tools charge in credits instead of dollars, infinite scroll removes the idea of one page, and autoplay turns one episode into a whole season. Bet you didn’t even noticed those things Whoever designs what one unit looks like quietly decides how much people will use, spend, or scroll. That’s a lot of power and knowledge hiding in something as small as a default, and I think we don’t talk about it enough as designers.

Is it possible to use Dia without any AI features and essentially have just Arc?

I’ve heard that there is a joke in Silicon Valley that when people leave a normal tech job they say “onto the next adventure,” and when they leave an AI job they say “I’m retiring to write poetry, please spend time with your families.” To be honest, I can’t decide if it’s just a joke or if people closest to this stuff really are staring into the abyss

Instead of using localhost to see how your designs look live, you can preview them right on the canvas, next to the frame where you built them. If you're unsure about something or think it could be improved, you can reach out to the team and it will probably be fixed very soon. They are super responsive (one of the main reasons I switched from Paper 👀).