Widely Available Web Platform Features I'd Like To Learn
In my previous post on launching into the new year, I mentioned that I've not been great about keeping up with the latest advancements in the web platform. That said, I like to learn about things that I can actually use in my work. As such, I've decided to limit my active learning to web platform features that are considered "widely available" according to Google's Baseline project. That is, features that have been available in all major browsers for at least 30-months. As with all things in my life, I operate best within some constraints.
ChatGPT Deep Research Totally Biffed It
Part of my hope for 2026 is to become better about reaching for AI when I think it might help. This topic felt like an obvious opportunity. So I went into ChatGPT's "Deep Research" feature and I asked it to give me a bulleted list with all of the CSS, JavaScript, and DOM APIs features that have achieved "widely available" status in the last 2 years according to the Baseline project.
ChatGPT went off and churned for about 25-minutes and then came back with a list that included about 14 features (4 of which were font related). None of which was <dialog>, which is one of the "widely available" feature I've been very keen to start using.
W3C WebDX Community Group
After ChatGPT failed me so completely, I did some Googling and came across a set of "WebDX" (Developer Experience) repositories that are managed by the W3C (World Wide Web Consortium). Among them is a Web Features Explorer repository that hosts a set of release notes broken down by year and month. This list includes both "newly available" and "widely available" features.
Ummmm, Jackpot! This is exactly what I was looking for!
Looking through the last few years, here's the list of web platform features that caught my eye (many I skipped over or already know). This list contains both features that I haven't used before as well as features that I have used but don't yet feel professionally confident in my understanding.
2026
animation-composition- Array by copy methods (ex,
.toSorted())
2025
- Color functions (mildly curious)
:nth-child() of <selector>inertattribute- Media query range syntax (ex,
(400px < width < 1000px)) - Container queries
- Dynamic viewport units
:modalpseudo classoffsetmotion animationoverflow: clip- Array
.findLast()and.findLastIndex() translate,rotate,scaleas independent transforms
2024
- Array
.at() ::backdroppseudo elementBroadcastChannelAPI@layerCSS cascade layers<dialog>element:focus-visiblepseudo classObject.hasOwn()scroll-behavior- String
.at() structuredClone()for object cloningcolor-schemeenterkeyhintfit-contentaspect-ratioerror.causefor nested errorsblockandinlineCSS logical properties
2023
::file-selector-buttonpseudo-elementgapfor Flexboxclip-path:is(),:where(),:not()for list of selectorstext-underline-offset&&=and||=logical assignment??and??=nullish coalescingPromise.any()Promise.allSettled()- String
.replaceAll() - Numeric separators (ex,
_) in number literals queueMicrotask():read-onlypseudo classResizeObserverAPI
2022
- Array
.flat()and.flatMap() - Async iterators (
for await...of) - Custom elements (ie, web components)
<details>elementdocument.elementFromPoint():focus-withinobject-fitandobject-position::placholderand:placeholder-shown- Scroll snapping
- String
.matchAll()
Lots of Exciting Stuff
I feel so energized by this list. It holds the promise of kinetic potential about to be dropped. Additionally, I need to go back and freshen-up on all the native Object, Array, RegEx, and String methods that have been added to JavaScript in years prior. I know all the basics; but, haven't integrated many of the newer methods.
A Final Thought On Serendipity
When I asked ChatGPT to do the deep research, I had limited the time frame to 2-years because I just assumed that doing the research would be hard. And, I was trying to make it easier for ChatGPT to give me coherent results (which it turns out, it couldn't do). It wasn't until I tried to actually do my own research that I came across this list of release notes. Which, in turn, motivated me to explore several additional years back in the release timeline.
This is the kind of "Oh, interesting" serendipitous learning that I'm so afraid of losing if I lean too heavily into the AI agents. A huge part of learning for me, historically, has been reading and then shooting off on side-tangents.
I'm not saying that AI will make learning harder. I'm only saying I've always found value in going 17-pages into a Google search to find all the weird stuff I didn't know I needed to know about.
Of course, if you've never been past the first page of a Google search, you may have no idea what I'm even talking about.
Reader Comments
Parallel to your point about the pros of doing your own research rather than leaning on AI, I think this way about teleportation. If it were a real thing, I'd miss out on those small but magical moments created during the process of traveling. Sure, most of it is mind-numbingly uneventful...but sometimes you have to go through those moments to get to the good stuff.
@Chris,
Plus, it's a great time to listen to podcasts (or, music if you're one of those normal persons).
@Ben Nadel — You get me 🫶🏼
Post A Comment — ❤️ I'd Love To Hear From You! ❤️
Post a Comment →