A hub is a place of convergence where elements are integrated to form a coherent object. It is an enclosed, contiguous space wherein the included items or elements are unique; i.e. not duplicated.

Familiar physical examples are the hubs of wheels and in Information Technology (IT), devices, which, for example, allow several computers to use a single shared printer.

Your brain is a hub; your hub of consciousness, that reveals the key attributes of a hub emphasized by the bolds words above. Consider how quickly your brain would fail if any one of these key attributes or conditions of 'knowing' fail.

In information sharing and knowledge representation individuals who own data and who could contribute expertise may be distributed, but whatever is assembled, has to be a single object that also critically 'obeys' the key 'rules' in bold in the opening sentence above.

When information elements are shared in a single contiguous space the 'meaning' that a network of links establishes is not broken as similar meaning would be when an inter-database link in a distributed network (see below) is broken. Furthermore, many; virtual databases could be created within the same, contiguous information commons that can be owned, labeled and controlled to function indistinguishably from a separate, 'private' database. A familiar example of this can be found in Internet banking wherein we all control our own accounts without 'parking' our own computers and databases on the bank's premises.

In contrast to a hub, a 'ring' of separate databases has no hub. One of the consequences of this is that information items are distributed and may be duplicated, either as identical items or as an array of slight variants of each other or even 'false friends'. The basic process of assembling nodes is linking which has to be done first in order to assemble a coherent object of knowledge representation.

The 'meaning' of an set of inter-linked information items, is created by an external (meta) formula that 'gets & displays' the pattern while leaving the contributing items in place. When a link to any contributing database 'breaks' the assembled meaning is shattered. A 'web ring' like this is somewhat like a soap bubble that pops when a molecule in its surface is 'removed'.

The current, dominant information management strategy on the Internet is to link separate databases together (the dots in the last diagram above) in a way that would allow users to look for information elements in all the inter-linked databases using a single search procedure. Such networks of interconnected databases are interoperable without using a central hub.

A search question such as 'find species pollinated by bats' might return an answer / list as illustrated by the black lines inside the circle in the diagram on the right. The 'value' of the returned 'answer' depends entirely on the links between the separate databases being 'unbroken'. If a links fails, the comprehensiveness of the answer is reduced accordingly, and in many cases the meaning of the whole set may be completely lost.

Such distributed networks of discrete databases are inconceivable without a meta language of keywords, tags and other elements (collectively an 'ontology'), that is used to render the individual systems interoperable. These meta-languages need vocabularies that are large enough to attach an external label - a 'tag', on every possible factorial combination among elements/items contained in the separate databases. For example, interaction vectors such as:


1. Blue fly x red flower = pollination
2. Red fly x red flower = pollination
3. Blue fly x red flower = nectar feeding ... and so on...
... each requires a unique tag to distinguish between them:
1. [[Blue fly x red flower = pollination]Tag 1]
2. [[Red fly x red flower = pollination]Tag 2]
3. [[Blue fly x red flower = nectar feeding]Tag 3] and so on into infinity.

Obviously no two tags can be the same if they 'profile' different interactions, so the vocabulary of the meta language used, must be as large as the number of all possible combinations between all information items in all the interlinked databases. To reduce this impossible demand for an infinitely large 'new' language, the individual 'tag words' may be arranged and grouped into a hierarchical system of classification that vastly decreases the demand for infinite vocabularies; e.g. as postal codes are able to pinpoint a house anywhere in a large country. However, as Edelman & Tononi (2000): (Consciousness. How Matter Becomes Imagination) have pointed out, every conscious experience of the world generates its own, unique ontology. An ecologist seeing a sparrow attack a hawk, would consciously experience a unique event that would have to be uniquely tagged to make it 'findable' among all possibly similar events. Therefore, since each conscious experience of the world is unique and has a unique ontology, we're back to the requirement for a meta tagging system that has infinite discriminating capacity since there are an infinite number of ways of consciously experiencing the world.

It might be worth recalling at this point that if only 35 factors combine in a pattern (ab, ba, ac, ca, bc, cb, abc etc.' for 35 and not just 3 factors), this results in 235-1 possible unique connections - a number that is larger than the number of seconds that have passed since the Big Bang created the universe.

While distributed networks are very small and while they deal with only the simplest, first 2 or 3 items in the hierarchy integers > real numbers > limited alphanumeric data (e.g. specimen labels in museums) > food web interactions > community ecology > climate regulation, meta ontology systems could be useful to a limited extent. But when it comes to even simple ecology webs, database interoperability fails cognitively instead of technically; i.e. 'interoperability' is a social problem and not an IT one.

At a glance the diagrams below illustrate how much more 'applicable' meta ontologies are to taxonomy hierarchies than they are to trophic food webs:


Imagine what would be involved in tagging each and every unique interaction in even the simplest food web in a way that will assign a unique meta-tag to every possible interaction…

For these and other reasons, the CA.EcoPort website and service is a hub.