I've noticed for some time now that Google (as amazing of a search engine as it is) and Amazon, as well as other large search platforms are still having a difficult time producing relevant results when searching for version specific information. Why does this seem to still be such a difficult feat to achieve?
For example let's say I'd like to find out how to install rvm on ubuntu. Here's the search I might perform:
"installing rvm ubuntu".
I find it odd that - by default, the first several results are all for Ubuntu 16 or 14 ... articles that are clearly written several years ago; especially considering I'm searching from an Ubuntu 19.10 machine. It seems that would be taken into account.
I've also noticed oddities when searching version specific issues. Notably Apache, MySQL, Ruby, PHP (PHP searches are horrible with Google) and Python is pretty bad as well.
I've even had a few times where I entered a non-version question and received a fairly dated result set; I then decided to clarity by adding version numbers, and ended up with an even more dated result set.
That, combined with the fact that Google should know my searches pretty well by now. After years of development related searches, shouldn't it be fairly well up-to-date with what versions I'm currently using and what I'm looking for?
Amazon also has a notably frustrating quirk; returning the exact model numbers your searching for seems to be something it has a very hard time performing. For instance, an Amazon search for
"256gb ssd" 90% of the first page results are exactly NOT that. If this this something they've purposefully implemented, then it's doing exactly not what it's there for. The only thing I can think of is that there is some technical hurdle that has yet to be fully overcome.
I'm quite curious as to why this seems to still be an issue if it's not, in fact, an intended "feature".
I get that google is complex, and it's probably well beyond my capability to grasp exactly what it's doing behind the scenes. In no way am I knocking it. However, given that such odd results seem to still be so abundant and so similar across the megalithic search and query systems, I suspect then that it's probably a technical hurdle that we've still not figured out collectively.
What exactly is that hurdle? Why do these search technologies appear, from the user's end, to still have such a difficult time with seemingly basic search requirements?