On the enterprise side, integrating the access and discovery layers gives a lot of advantages, especially around data governance. On the web, we give users tools to connect data sources, document them, and share/audit access to them. When a query comes through the endpoint, since we're implemented as a Postgres proxy, we can rewrite/filter/drop it in accordance with rules, or we can forward it along to the upstream data source(s) and/or join across them. If you use Splitfiles to generate versioned data, we can also provide data lineage/provenance and full reproducibility.
We've been working on this for ~3 years but are still pretty early. If anyone wants to help, we just raised a seed round and are hiring a remote team -- check my comment history for links.
[0] https://www.splitgraph.com
[1] https://www.splitgraph.com/docs/working-with-data/using-spli...
Discoverable
Addressable
Trustworthy and truthful
Self-describing semantics and syntax
Inter-operable and governed by global standards
Secure and governed by a global access control
A reminder that Thoughtworks was highly influential in pushing Microservices. This may be an elaborate mea-culpa ("oops, SOA was actually more sensible") without admiting 'culpa', rehashing SOA with a set of features (above) that look awfully like those highly elaborate SOA proposals with XML and all sorts of meta-data to 'couple' these "data products' (previously called Services).You don’t really build a data lake, you just end up with one.
That said, both lakes and lakehouses are very valid ideas and not buzzwords !