HTML and CSS Reference
In-Depth Information
History of backwards compatibility
HTML and JavaScript both have a tremendous track record when it comes to backwards
compatibility. Web sites written in the late 1990's, relying on CGI scripts and early versions
of HTML, still work largely unchanged. They typically do not look any better than they did
in the 1990's, but they do still work.
Browser vendors take amazing care with backwards compatibility. They are aware that their
users will hold them accountable if they ruin their browsing experience on a popular web
site. As such, browser vendors invest huge resources in compatibility and regression testing
every time a new release of a browser occurs.
This process is not perfect. As browser vendors have moved to faster and faster release
cycles bugs do occasionally appear in existing web applications. Although this is an annoy-
ance, at least it is possible for web application vendors to provide patches to their applica-
tions without requiring users to perform any upgrade steps.
Although bugs do sometimes appear therefore, the underlying APIs and feature sets remain
remarkably stable once they are implemented in all major browsers.
As a result, web standards tend to be a process of extension rather than change, and this
process looks set to continue. As mentioned earlier, HTML5 is considered a living standard,
and will gradually change over time, without ever being
replaced in a wholesale manner.
