Warning: This is a highly fuzzy market; I welcome proposals for clarification!
Direct Interaction with Domain Names:
Users verbally or manually specify exact domain names (e.g., "www.facebook.com") to access a specific online content or service.
Typing, speaking, or manually selecting a full domain name in a search bar, address bar, or other input method to reach a specific site or platform.
Examples of YES resolution:
Users predominantly use browsers or similar tools where they enter domain names to access websites.
Websites and online platforms prominently display their domain names as a primary means for users to recognize or return to their content.
Examples of NO resolution:
The majority of web interactions are based on fuzzy requests where specific domain names are not referenced, like "show me what my friends are doing" or "play the latest Taylor Swift song."
Entities, brands, or content sources are accessed by name or description alone without the user needing to know, specify, or see a domain (e.g., "Take me to Facebook" without any ".com" or country-specific extension).
Domain names might still exist deep in the tech stack but are abstracted away to the extent that an average user doesn't see, interact with, or need to be aware of them.
Resolution Criteria:
The prediction will be deemed YES if, at any point in 2034, a significant portion of interactions with the web still involve directly using or seeing domain names, as described above.
The prediction will be deemed NO if the majority of interactions have moved away from direct domain name use as described in the 'NO' interactions.
If ambiguous, the market will be resolved based on
Industry survey if they exist; otherwise
Based on a poll. This will not resolve to PROB
1,000
1.00Note: I encourage anyone who feels like the description no longer reflects their intended bet to sell their position and reach out. I'll managram you any losses incurred. This offer is valid for 24h.
I'll treat web content (available on an HTML page regardless of other formats) as a part of the larger internet, which adds app-only activity like email, gaming, text/voice/video chat. For simplicity, I'll treat "apps" as including voice assistants and the other non-domain-name stuff you mentioned.
If you measure interactions by time spent on the internet, then the public already spends way more time in apps than in a browser. But say you want to measure "invocations" (tapping an app, saying "Alexa", following a link/URL) for only web content (which may have an app too). I would still expect that the whole world makes more of these invocations without seeing a domain than with one. Most popular websites also have native apps with the same content. People find those apps through app stores or voice search. And popular apps don't show very many domains or links, since they would kick out of the app.
I can only see your criteria being met if you restrict the measurement to something like:
First-time invocations (tap a new domain vs. download a new app)
Counting the raw number of available apps vs active domains
Focusing on power users (maybe weighted by complexity of interaction)
Counting users who have tapped or entered any URL in the past month
Basically, I think we've already transitioned to a post-web world. Sure, domains have stuck around under the hood. But so has Unix, and nobody is wondering whether >50% of computer interactions are displayed as Unix-based.
Hmmm. All great points.
I think the 50% criteria might end up being too much of a technicality and will take away from the spirit of the question.
Even if most interactions happen via apps, URLs still play a prominent role in how the web operates. One of the first things a start-up does is purchase a URL (and name their company according to cheap/available URLs.) Anyone using a laptop or desktop browses to website via the address bar or bookmarks.
@jgyou Thanks for the clarification. One way to estimate the importance of URLs in general could be to look at web traffic sources. According to this large 2020 study, about 27% of web traffic was "direct", meaning the request had no referral data. Even though this includes bookmarks, links in emails and other stuff, we should still expect this number to fall if people in the future are typing in fewer URLs directly.