Proposal: Deep Web
The definition of "Deep Web" according to Wikipedia is:
The Deep Web, also called the Deep Net, Invisible Web, or Hidden Web, is the portion of content on the World Wide Web that is not indexed by standard search engines.
Examples for this is content behind logins (with or without paywall) dynamically generated content, file formats not readable for search engines or URLs which can not be found by search engine spiders because they are not linked anywhere.
However, the current example questions seem to be about a different definition of the term which I have noticed to be used in certain circles: websites served through anonymization systems like TOR, Freenet or I2P.
These two definitions are contradicting: The TOR network, for example, can be accessed by anyone who installs the TOR software. Also, there is no technical reason why it can not be spidered by search engines like Google. The only reason why you don't see .onion search results from normal search engines is because they believe that those results are likely not relevant for most users because most users don't use TOR. In fact, there are search engines which spider TOR hidden services exclusively (which are usually hidden services themselves).
Which one of these two definitions is this proposal about?