PJ Onori @pjonori.com · 16hI didn't know I needed to carve out a day to disable AI features from macOS 27. 151
PJ Onori @pjonori.com · 28/09/2026After many, many speaker talking about how much he meant to them, he said his final words. He spent the entire time thanking everyone else. He didn't utter a single word about himself. He's the epitome of a class act. He's given so much—and during his retirement, he chooses to thank everyone else. 061
PJ Onori @pjonori.com · 28/09/2026Mike Krukow has been a fixture of San Francisco Giants' radio and TV for 26 years. He's a future Hall of Famer. Today was his last day. If you're not familiar with the Giants, it's impossible to know how much he's meant to fans. And on his last day, he got a great sendoff. 282
Reposted by PJ OnoriDesign system news @designsystems.news · 27/09/2026Week of September 27, 2026 designsystems.news/#week-2026-9-27 001
PJ Onori @pjonori.com · 26/09/2026Is anyone surprised by this? youtu.be/uh7s2jr0b48?...youtu.beWe Investigated the Christian "Scam" Fronted by Tim Tebow | PTFOYouTube video by PABLO TORRE FINDS OUT 010
PJ Onori @pjonori.com · 21/09/2026I don’t say this with a sliver of hyperbole. I will never wear a helmet to use software. 110
Reposted by PJ OnoriDesign system news @designsystems.news · 20/09/2026Week of September 20, 2026 designsystems.news/#week-2026-9-20 102
PJ Onori @pjonori.com · 20/09/2026designsystemdocspec.orgdesignsystemdocspec.orgDesign system doc spec 0.21.0A machine-readable format for design system documentation, what it means for a document to conform to the spec, and every rule the spec enforces. 060
PJ Onori @pjonori.com · 20/09/2026DSDS 0.21.0 is out. - A new style guide for how to structure/order the schema - traits' traitType now provides more explicit callout to variants vs states via traitType - sections can now carry tags for easier categorization - A new apiId allows for mapping prop/attribute names to traits and combos 151
PJ Onori @pjonori.com · 17/09/2026Aren’t those just glasses? And if so, why the hell would meta be in the business of selling just glasses? That has absolutely nothing to do with spying on children. 000
PJ Onori @pjonori.com · 17/09/2026Nate has forgotten more about color than most of us will ever know about it. Interested to hear what he has to say. 130
Reposted by PJ OnoriDonnie D'Amato @donnie.damato.design · 17/09/2026Yesterday on Wireframe LIVE, @natebaldwin.bsky.social came on the show to talk about data in design. We chatted about our thoughts on the scalability problem of curating modes that are influenced by multiple dimensions. ☝️ If that made sense to you then watch below 👇 www.youtube.com/watch?v=cfi_... 122
PJ Onori @pjonori.com · 17/09/2026I didn’t say that at all. I said there was a simple answer. But there’s also a reason why it hasn’t happened. 000
PJ Onori @pjonori.com · 17/09/2026Hey, just a quick note: I’m constantly on the hunt for design systems content for Designsystems.news. Specifically: * Not written by AI * Not entirely about AI * Not published on some paywalled/subwalled platform (includes Medium/Substack) Send me your links—would love to read them.designsystems.newsDesign systems newsDesign systems news is a weekly design systems editorial by PJ Onori. 5163
PJ Onori @pjonori.com · 17/09/2026There's a simple answer (but admittedly hard for some)—stop using these platforms. 110
PJ Onori @pjonori.com · 16/09/2026The biggest takeaway of all of this for me is how problematic it is to choose a technology based on how much it's currently used. Because the next best thing is always right around the corner. 040
PJ Onori @pjonori.com · 15/09/2026I’m not sure if we‘re in the hardest time for design systems, but we’re definitely in the weirdest. 071
PJ Onori @pjonori.com · 14/09/2026It can be—potentially. It can also create a false sense of understanding. Measuring adoption is a descent into madness. 120
PJ Onori @pjonori.com · 14/09/2026I don’t know. Atoms are sometimes used for tokens. Sometimes for components. That alone causes ambiguity in definitions. And what‘s division between molecule and organism? Seems fuzzy to me. 000
PJ Onori @pjonori.com · 14/09/2026I don’t think all adoption is created equal. Using 5 Box components is way different than 5 navigation sidebar components. But the only way that kind of measurement can occur is by assigning a tier value to each component. 110
PJ Onori @pjonori.com · 14/09/2026I think if it was just a rename, the exercise would be pretty useless. I’m trying to come up with a more repeatable rule that can assign a component‘s level based on its structure. It’s the subjective assignment that I’m against. And, honestly, we could call each tier atoms, molecules, etc. 110
PJ Onori @pjonori.com · 14/09/2026But aren't those different things? Designers want categorization and engineers what inheritance. I'm viewing this from the lens of measuring adoption—and I think both are important, but for different reasons. 120
PJ Onori @pjonori.com · 14/09/2026Yeah, I hear you. This may be something that can't be objectively assigned. The current reality seems too squishy though. 110
PJ Onori @pjonori.com · 14/09/2026In 2026, you choose your CMS based in part by how much drama you can tolerate. 030
PJ Onori @pjonori.com · 14/09/2026Also worth mentioning—I just applied that approach to our own design system and had some pretty hard disagreements where some components ended up. So, the thinking needs some work. 😂 020
PJ Onori @pjonori.com · 14/09/2026Damned good question. This came up when I started thinking about measuring usage. I don't think all components are created equal—usage of some have more weight than others. But how do we weigh that? 120
PJ Onori @pjonori.com · 14/09/2026Perfect? No. But something that's repeatable and _should_ be more universally applicable across systems. Thoughts? 400
PJ Onori @pjonori.com · 14/09/2026I've been thinking about a more objective way to assign "tiers" to design system components. I'm wondering if something as simple as this: - Level 0: Components made from no other components - Level 1: Components made from Level 0 components - Level n: Components made from Level n-1 components. 210
PJ Onori @pjonori.com · 14/09/2026I'll never dismiss what Atomic Design has done to create an approachable model for design system hierarchies. That said, I find the divisions subjective—which in turn leads to wildly different interpretations. 2110
Reposted by PJ OnoriDesign system news @designsystems.news · 13/09/2026Week of September 13, 2026 designsystems.news/#week-2026-9-13 012
PJ Onori @pjonori.com · 13/09/2026People can and should fall on whatever end of the spectrum they wish regarding Ferrari. My main interest is how the underlying critique applies to design systems. I don't think it's an empty accusation. 000
PJ Onori @pjonori.com · 13/09/2026I think the tricky thing is that software companies became so big that individual effort became unwieldy. I still believe that software design within a large org has to have some way to keep cohesion—typically done with a design system. But to your point, there's another solution—stay small. 040
PJ Onori @pjonori.com · 13/09/2026Yeah. And to be clear, I'm not packing my bags and leaving design systems. But I also don't want to intentionally ignore its downsides. 020
PJ Onori @pjonori.com · 13/09/2026It’s hard to watch this as a design system practitioner and not have a nagging, “are we the baddies?” moment. m.youtube.com/watch?v=HgjX...m.youtube.comWhy Nothing Looks Cool AnymoreYouTube video by Design Theory 55816
Reposted by PJ OnoriDesign system news @designsystems.news · 12/09/2026Week of September 6, 2026 designsystems.news/#week-2026-9-6 003