Information is selection
from a "domain" which is a common set of possibilities. Language vocabulary is only one possible domain - the domain can be precisely adapted to the application and defined online. Any piece of information then can be digitally transported as number sequence or "Domain Vector" (DV) which selects in the domain or "Domain Space" (DS). A DS is a domain which is a metric space. Every DV is unambiguously identified by a UL which is an efficient link to the DS definition. The DVs of a DS are accessible to similarity search with optional range restriction.
NumericSearch consists of 2 steps:
1. Selection of the DS directly
or by text search
2. Similarity and/or bounded search of DVs within this DS.
You can test it and contribute own DVs and also define own DSs. You can verify that DV based description and search has
much higher resolution and range
than word based description and search.
Therefore it is plausible that the principle will become standard on the web.