I think this will be the major feature of Web3 browsers apart from helping you give granular access to your data.
That said, there are a number of people working on the next generation of the content layer of the Internet. I hesitate to call it the Web, because that's not really a good name. Mesh is a better name for it, and the one I use. In order to answer what this Mesh browser would look like I need to describe what the Mesh would look like.
First, would NFTs be needed? Don't know. They are probably not necessary, certainly not in their current form. Blockchains are a different story, through proof of algorithmic work is not going to work long term due to environment and other issues. The jury on proof of stake is still out, but looks promising.
The three big things about this Mesh are:
- You control your identity. No one is able to tell you who you are, because that is, at the end of the day, a fundamental inalienable right. You can have a government issued identity you use to interact with their services, but you get to choose if that is tied to your work identity, your leisure identity, etc.
- You retain control of your data. This is tricky, because once you share data it's going to be stealable in reality. I see some sort of punitive system for people who steal and misuse others' data, and a system for identifying that misuse.
- Everything is decentralized - This will do away with the platforms like Facebook and Twitter, and even Mastodon, though Mastodon is closer to what I am talking about.
- First, it will be more of an agent in the cloud plus a browser client than a client. You'll specify the things you are interested in, manage your vendor and personal relationships, and your browser will use those relationships to discover and queue content for you. Within that content you may find other relationship possibilities, and use the browser to request forming a digital relationship, under certain terms. The other party's agent might counter with different terms, or accept, and so on. Once the relationship is established their agent would publish a catalog of available content (think Atom feed on steroids) to your agent, who would then start queuing content matching your interests.
Your agent would also get content catalogs from the people and orgs you had relationships with, weighted in value based on your valuations of their past content and the relationship, as well as other metrics that might be determined adhoc (for example, emotionality of content, or ratings from others you trust).
Let's look at an examples, buying an airline ticket for a work trip.
Today, getting a ticket online means either going to an airline's web site and purchasing it or going to a travel site like Travelocity (first one that came to mind), and entering the origin/destination then filtering down to what you want.
Someday, getting a ticket online may mean having already entered your travel preferences for work travel and weighted their importance to you (wide seats +2, better food +1, layovers -1, etc.), then you use the browser to tell your agent (or your agent prompts you, based on your calendar entries) to find tickets using your work identity. Your agent goes to each of the airline vendors you or your company have a relationship with (and possibly offers a relationship to other airline vendors through your company's agent) and solicits bids on tickets. The agent narrows these bids down based on your criteria, and the browser presents you with a list of choices when you are on next (or immediately if you are still on).
To summarize, modern browsers are dumb clients, tomorrow's browsers will be smart web concierges and the means to present the information those concierges acquire.