# Towards v2

OCA, as the name state is the overlay capture architecture. It brings data semantics as a first-class citizen and imposes organizational and structural focus on what so far was implicit or scattered. No clear picture introduces ambiguities that lead to errors. OCA, by introducing task specific overlays, promotes data enrichments so in essence adding metadata by defining overlays. In other words, these overlays enable adding metadata that in many cases is important, but was not captured properly along with the data.

Recent practical applications revealed that:
- overlay concept has limitations, especially when they co-exist without

the limitation of the approach that anything can be overlay, especially within flat structure, so where all overlays are considered from within the same level or layer. In fact, they become obsolete in favor of the `OCA Bundle` concept that aggregates them into one piece. Proper organization and segregation is also essential to practially approach a given problem. Organization and segregation lead to how people organize themselves and how they manage to distribute the responsibilities.

We propose a foundational restructure of the current approach to address the above concerns, so to include existence of various groups of interests that influence how in the end the data is captured, presented or used. While it's tempting to assume it's the responsibility of the people that work with the data to also take care of it, the reality is more complex. The final shape of what is the data, what it contains, how it's tailored and so on is the mixture of various influencing groups that come from different environments and have different expectations. The data is the common denominator among all these groups, because they all demand to benefit from it, but at the same time their benefits depend of how it's structured.

To reconcile the interests of these groups so to enable them to participate, both organizationally and technologically, we must evolve the current organizational paradigm. What currently exists under OCA umbrella, we must extract into distinct components, especially overlays whose existence is not determined by semantics, but other factors, that is, business/governance or presentation. Business/governance and presentation are two additional players, except semantics, that impact and inevitably constitute the data. This is the trio:


We have:
- semantics that encompasses the part of the metadata that is about data mechanics. The lowest level of metadata that brings raw bytes additional description, including data type, meta, encoding, potentially label and description for human reasoning. Potentially quantity of the information and standard, but solely for informational purposes.
- business/governance that covers the widest part of the data, the "end product" and how it looks like. Entries, formats, conditions, ranges, among other things, are all the consequence of rules, [the invariants](https://hackmd.io/Uocbyg88S_SNj-yaW1aO4w), that are imposed within the ecosystems. The origin of these rules is important, that is, the "external forces", namely businesses or other regulations. By imposing rules, they narrow their interest in particular data.
- presentation that enables expression of an intent, i.e., digital information consumable by human perception through human senses—a “translation” layer of something digital to something visual, auditory or tactile (as of 2024).

What it means for the existing ecosystem is releasing OCA from the business related burden that introduces unneccessary complexity into the semantics layer. The semantics layer becomes smaller, but more pure and cohesive. At this level, it is no longer known what are entries (entry/entry code overlay) or conditions (conditional overlay). They advanced into distinct, upper layer.