So the question is why? and if it is useful what happened with the 2.14.x release? or if not should we delete them and may be cut down on the number of times google takes people to an outdated version?
Either way the link to the downloads is broken on that page
Ian we do not have control of this and we cannot remove pages.
We are trying to obtain the domain so we can make use of github pages (or ask osgeo for a server like for geotools.org).
I have an answer as to what happend with 2.14.x onwards (which predates our move to S3 for docs storage) - the process of adding the link to the docs involved manual hand-editing of the HTML of that page. Around when we moved to S3 for the docs, not everyone had the requisite level of access to do this step, so it was frequently skipped (I’ll also note that it was actually skipped fairly frequently before that as well, we just usually updated a whole bunch of links in a batch every few releases). Now that we have even less access (upload-only, basically, and only via Jenkins), changes aren’t really viable at the moment.
While I don’t know the original reasoning, I’ll take a stab at answering why we were hosting the old versions of the docs in the first place - many (probably the majority) geoserver users are on old versions, and it is convenient to be able to reference the docs for the exact version of GeoServer one is running. Most other projects do something similar (although they tend to stick to only major versions, which is a bit more manageable).
So the question is why? and if it is useful what happened with the 2.14.x release? or if not should we delete them and may be cut down on the number of times google takes people to an outdated version?
Either way the link to the downloads is broken on that page