a web browser is a program that displays a HTML document, in the same way that a text editor is a program that displays a txt file, or a video player displays MP4s, AVIs, etc. HTML has been around for a while, first appearing around 1990.
HTML alone is enough to create a website, but almost all sites bring in CSS as well, which allows you to customise the style of a webpage, such as choosing a font or setting a background image¹. javascript (JS) is also used for many sites, allowing for interactivity, such as hiding and showing content or making web games.
together, HTML, CSS, and JS make up the foundations of the modern web. a browser that aims to be compatible with as many existing websites as possible must therefore implement all three of them effectively. some websites (twitter, ebay) will fall back to a “legacy” mode if javascript support isn’t present, while others (mastodon, google maps) won’t work at all.
implementing javascript and making sure it works with HTML is a particularly difficult challenge, which sets the bar for creating a new browser from scratch staggeringly high. this is partly why there are so few browsers that aren’t based on firefox or chrome² that are capable of handling these lofty requirements. microsoft has decided to switch edge over to a chrome-based project, opera switched to the chrome engine years ago, and so on, because of the difficulty of keeping up. that’s not to say there aren’t any browsers that aren’t based on firefox or chrome – safari isn’t based on either of them³ – but the list is certainly sparse.
due to the difficulty of creating and maintaining a web browser, and keeping up with the ever evolving standards, bugs and oddities appear quite frequently. they often manifest in weird and obscure cases, and occasionally get reported on if they’re major enough. an old version internet explorer rather famously had a bug that caused it to render certain elements of websites completely incorrectly. by the time it was fixed, some websites were already relying on it. microsoft decided to implement a “quirks mode” feature that, when enabled, would simulate the old, buggy behaviour in order to get the websites to work right. most modern browsers also implement a similar feature. this just adds yet another layer of difficulty to creating a browser.
the complexity of creating a web browser combined with the pre-existing market share domination of google chrome makes creating a new web browser difficult. this has had the effect of further consolidating chrome’s market share. chrome is currently sitting about about 71.5% market share⁴, with opera (which is based on chrome) adding a further 2.4%. this means that google has a lot of say over the direction the web is headed in. google can create and implement new ways of doing things and force others to either adopt or disappear. monopolies are never a good thing, especially not over something as fundamental and universal as a web browser. at its peak, internet explorer had over 90% market share. some outdated websites still require internet explorer, which is one of the main reasons why windows 10 still includes it, despite also having edge.
footnotes:
- before CSS, styling was done directly through HTML itself. while this way of doing things is considered outdated and deprecated, both firefox and chrome still support it for legacy compatibility. having to support deprecated standards is yet another hurdle in creating a browser.
- chrome is the non-free (as in freedom) version of the open source browser “chromium”, also created by google. compared to chromium, chrome adds some proprietary features like adobe flash and MP3 playback.
- the engine safari uses is called webkit. chrome used to use this engine, but switched to a new engine based on webkit called blink. safari therefore shares at least some code with chrome.
- based on http://gs.statcounter.com/browser-market-share/desktop/worldwide. note that it’s impossible to perfectly measure browser market share, but this is good for getting an estimate.