[closed] more info about Relations: group, collection, cluster

I'm seeking more insight & understanding regarding relations. Specifically more detailed, complete and/or expansive information about the Relation types: group, collection and cluster. Are they even used? Or should they even be used? Under what circumstances and for what purposes? What are the differences between them?

I know that the 'group' relation type is used to associate/link the Great Lakes of North America. Apparently because it can contain other relations, in fact it seems to be able to contain every type of object/data.

Let me explain some of the reasons WHY I have interest in this topic. This all relates to a particular interest I have in mapping hydrological systems and basins; catchment and drainage areas and so on. I also have interest in various practical applications of such data in regards to developing more accurate systems for measuring precipitation and estimating flooding and the time for water levels to subside in rural and developing areas. As well as integrating other information regarding agriculture, mining and other industries in order to include some information about contamination of these waters with various toxins/wastes. Where they are likely to end up? At what concentrations? How long they will remains, persist and pose threat? And how they could interact. We take for granted in developed nations that our governments are dealing with these concerns. But those in poor, developing areas (often rife with corruption) are generally left in the dark to fend for themselves either because the information is not disseminated (whether that is due to the lack of infrastructure or intentional withholding/cover-ups) or the information doesn't even exist. Such powerful and important information which can be crucial to health or even survival should be available to anyone who cares. It seems likely that even in the West we are likely not as well-informed or protected as we might expect/hope/believe. This is specially likely in regards to the aftermath of natural & man-made disasters resulting from industrial mistakes/failures or (potentially secret) gov't research/test gone wrong.

This can also potentially apply to improved & more detailed map data facilitating more efficient & effective searches, and superior routing through the use of more detailed and comprehensive relations mapping. Including indoor maps of shopping facilities, schools, national parks, recreational and tourist sites & facilities.

There are other potential, interesting applications for small & subsistence farmers; and much more.

Any thoughts, feedback, insight, warnings and other information or sources of information will be sincerely appreciated!