Mar. 2nd, 2025

I've been interested in web browser development for a while now. Trying to find a lightweight web browser than can be built independently from source and still supports the ability to view the average modern web page is extremely difficult. I feel like the main browser developers now have so much of a monopoly over web development that they managed to become the web standards committee, taking this job away from the W3C which originally maintained web standards. They control what standards become part of the web and even eliminate the need for certain products. Products like Flash and Silverlight became practically obsolete with the advent of HTML 5. While Flash and Silverlight weren't exactly a boon for Free and Open development, the current web standards are so difficult to develop for that independent commercial and Free/Open development alike are both at a loss to keep up.

Some of the browsers I was able to build myself included Netrider, D+ and lynx. Netrider ( https://sourceforge.net/projects/netrider/ ) is based on an older version of webkit. It has a FLTK front end, but was designed to easily allow for other front end development using other GUI libraries. I've used it with SDL and other options. I also ported it to work on Windows so it provided a cross-platform solution. D+ is an older fork of Dillo. Unlike Dillo it works on a variety of operating systems including FreeDOS. Lynx is a text based browser.

There were several text browsers at one point but now development seems to have stopped. Originally one could surf the web using just a text based browser but trying to do so now, is extremely difficult. Many web pages just don't render well in text alone. This is doubly troubling because it's removed the ability for many to use text based browsers and it's made it more difficult for those who are visually impaired to surf the web. I often used text browsers to make sure that the web pages I created rendered well and were easy to navigate. This was one test I used to ensure the pages were user friendly for the visually impaired and the browsers they often use. Some other older text browser projects include links, links2, elinks, w3m. Links was an interesting project as far as text browsers go because it allowed some viewing of graphics using the Linux framebuffer or SDL.

Fifth ( https://github.com/clbr/fifth ) is also an interesting web browser alternative. It's available as a package for TinyCore Linux. Like Netrider, it's based on an older version of webkit and provides an FLTK front end. Unlike Netrider, it's not as portable to other platforms such as Windows.

I looked at netsurf at one point because it's one of the few active independent web browser projects out there. I was unable to build it myself from scratch. It had several libraries, little documentation on building and I had difficulties finding out which version of which library worked with the others. So, I had no luck getting it to build independently as a cohesive project.

Some suckless.org developers solved the issue of getting a modern day web browser to build from source by adopting surf. Surf uses webkit and a GTK GUI front end. The Linux from Scratch project currently includes lynx, links and these browsers in their documentation: https://www.linuxfromscratch.org/blfs/view/svn/xsoft/graphweb.html

There are several interesting independent browser development projects but like surf, they typically use libraries such as webkit or gecko from a larger web development project.

Some independent operating systems are trying to find a way around the browser issue by developing their own. Ladybird from SerenityOS is an example of this: https://ladybird.org/

Some developers are trying to simplify browser development by creating an alternative to the HMTL 5 standard. This is where Gemini comes in. Unfortunately, Google has renamed their AI to Gemini as well which makes it difficult to search for information on the Gemini protocol. One can read more about Geminispace here: https://geminiprotocol.net/ There was an interesting critique of the Gemini protocol from the developer of curl: https://daniel.haxx.se/blog/2023/05/28/the-gemini-protocol-seen-by-this-http-client-person/ I'm not sure if any of his issues have been addressed but so far, Gemini hasn't hit the mainstream for replacing the World Wide Web and HTML 5. I have to say, I do like the looks of the Gemini based web sites. However, not all Gemini browsers can render all the sites. The protocol is still evolving and some functionality is experimental. So, it seems to share some similar issues with World Wide Web surfing.

The problem of being able to easily build a web browser from source without requiring a huge number of libraries has been around for a while now. It appears that it will continue to be an issue for independent operating systems and lower resource computers for a long time to come. It would be great to find solutions that would make web browser development less of a monopoly only attempted by major players. I'm surprised groups like the Free Software Foundation don't feel this is an area that should be considered for their High Priority Projects. I guess many believe that as long as there are Free, Libre, Open Source browsers out there that provide source code, it shouldn't matter that it requires a large organization just to build them while building from source is out of reach for the average developer. However, my personal philosophy is that one of the key advantages of Open Source is to be able to build and modify the code for yourself. If I'm unable to do that with a project, it loses a lot of its usefulness for me.

Here's a short but in no way all inclusive list of some web browser projects that avoid using major web libraries such as as webkit and gecko:
https://ladybird.org/index.html
https://www.netsurf-browser.org/
https://github.com/CobaltBSD/neosurf
https://dillo.org/
https://sourceforge.net/projects/dplus-browser/
https://github.com/textbrowser/dooble
https://lynx.invisible-island.net/

That's a quick rundown of the state of alternative browser development. If I've missed any other relevant FLOSS browser or Open protocol options, you can reach me on Mastodon and let me know.

April 2025

S M T W T F S
  12345
6789101112
13141516171819
20212223242526
27282930   

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated May. 23rd, 2025 12:21 pm
Powered by Dreamwidth Studios