Browsers are the lens through which we see the web. And considering how much the content of the web has changed over the past few decades — from low-quality images to constant streaming video — it sometimes feels a little strange that the browser experience has remained more or less the same.

How might browsers change in the next five years? While there’s no way to definitively predict the course of the future, Here are a few ideas of some ways the web could continue to evolve in the near future.

In the web’s current state, there are many ways to discover content. But the current experience is saturated with algorithmically-generated feeds and search results tied to very particular keywords.

There are many users for whom the entire web is just what they can see through these methods. But they aren’t the only way to discover web content — and as time has passed, some of the more interesting ways to discover web content have fallen away from the core.

Search engines ideally function as a set of paths that can connect us to topically specific parts of the web. But today, many people are at the mercy of a single search engine, whose programming decides which search results are the “best” results, usually featuring paid content and ads at the top.

We’ve all had the experience of searching for something with specific keywords, and then having something else pop up. Even though keywords can help us find a wide range of options, some search terms yield pages of unrelated, ambiguous or inaccurate content. This is probably going to be even more noticeable as the web continues to expand.

Webrings were a solution for this on the early web. Webrings were community-built groups which would organize web content, in ways which let users explore themed sections of the web without having to dig through results from the entire thing.

Social media platforms like Facebook have proven to be an imperfect replacement for web rings. They allow us to discover a constant stream of content, but it’s all recommended by algorithm rather than human interaction — not to mention that it is filled with ads.

Facebook also suffers by trying to be a service for everybody. The human experience is not supposed to be about ‘one size fits all’ content or profiles.

Webrings were most effective when they would focus on particular communities or interests. But a service which tries to be all things for all users is more likely to gravitate towards content for “the average user” — so that most of what users end up discovering will feel impersonal and non-customizable.

The browsers of the future might get more mileage out of webring-like systems instead of just using algorithm-driven search engines. This would help bring the human element back into the exploration of the web, while helping filter out unrelated search results. Webrings are already coming back through services like Neocities.

Rather than trying to build the perfect algorithm which can find the “best” content for the most people, why not let people recommend content to each other themselves?

Remember StumbleUpon? Anyone could click on a button and be taken to a surprise website, either from a set of all interesting websites, or within a favorite category. Imagine a service like StumbleUpon built into the browser itself, allowing people to discover new content by category instead of keywords.

This kind of social platform would encourage its community to compile and share in a browser-supported context, allowing a more vibrant experience than one built around re-posting algorithm-approved links.

Bookmarks are amazing helpers for keeping track of content on the web. At their best, bookmarks let us save content to explore later — and this system of re-discovery can be just as important as finding something for the first time.

Unfortunately, the default bookmark experience in browsers is kind of a mess. Bookmarks often feel more like a giant filing cabinet full of forgotten content than an endless library which is fun to dive back into. There aren’t easy ways to categorize or collect bookmarks, either.

A more customizable bookmark system would help different users match their browsing experience to their needs.

Someone writing a research paper might want a bookmark system which can mark key places in long articles. A visual artist might benefit from a focus on visual previews and color-coded bookmark grouping. Some people might appreciate the ability to randomly explore their own bookmarks, while others might prefer something tightly structured or easy to rearrange.

The best part about bookmarks is that they can affect a lot more than just the look and feel of the browser — they could even be used to form communities. For example, bookmarks could easily form the backbone for a webring-esqu platform of user-shared links like the one hinted at in the previous section

When you bookmark something, you could also have the option to make it part of a ring, to help others to discover it as well. After submitting enough bookmarks, you could be voted in by community members as a trusted node in the network, preventing spam submissions and improving overall quality.

Different networks could exist for education, research, art, current events, and smaller, more niche categories. Rather than trying to serve everyone with the same omnipresent Google-like service, these sub-networks would acknowledge that the web is different for everyone, and try to give the best possible exploration tools to each of its diverse audiences.

Delicious was a popular bookmarking service in the early 2000s, which allowed users to share marked sites with each other and the rest of the web. After being acquired by Yahoo in 2005, Delicious w