So this week, SAP, Microsoft and Adobe announced plans for an Open Data initiative. “The idea is to have multiple business applications using a common data model within Microsoft’s Azure cloud, a Microsoft spokesman told CNBC. Microsoft’s Dynamics 365, SAP’s C/4HANA and Adobe’s Experience Cloud.” Great! About time!
This addresses a key issue for anyone migrating to Cloud or Cloud infrastructure. First, the onerous task of shoehorning existing data models into Cloud and having a consistent enterprise approach to that. Second, once in the Cloud, promoting interoperability and mitigating the possibility of lock-in.
Microsoft CEO Satya Nadella said: “Think of all data as a continuous renewable resource.” For Incorvus, this lies at the heart of what is necessary for future computing – a primal shift from applications to data.
Data centricity views data as the life blood of corporate ecosystems. As Frank Herbert famously wrote: “The spice must flow.” Data that is sat in a corner – static, uncared for, and getting dirtier by the minute – should be increasingly a thing of the past because the value of data is realised when the data is an active asset. This is why the SAP-Microsoft-Adobe ODI initiative is so important to ecosystem design. (Having said that, if they get the ODI off the ground – there is no roadmap at the moment – it will cause difficulties for those forced to play catchup.) We believe it aims to reduce the data oncology nightmare by establishing common standards for metadata – across their individual platforms and applications, and jointly.
So ODI is good news for those moving to Cloud but there are three really thorny problems ahead: locational data, unstructured data and data governance:
- Locational (more correctly, spatial) data may include rasterised images (and has its own peculiar challenges).
- The information architecture of handling unstructured data, particularly large BLOBs, is another thing entirely as it will need to address how objects are referenced and workflowed rather than stored in the Cloud, using up storage space and bandwidth.
- Governance will be challenging as this will be impacted by:
- multiple applicable overlapping and interlocking regulations;
- multiple jurisdictions;
- the need to be compliant when moving data across borders – into Cloud or from Cloud to Cloud;
- and the need for the standardised data model to address the differing requirements of those different regulations and jurisdictions. Can they really achieve a one-size fits all on this? And what are the implications for ODI if they can’t?
ODI sounds like a great idea whose time is definitely overdue but unless it is to be more than simply a great piece of PR, Adobe, Microsoft and SAP will need to address locational data, unstructured data and data governance in order to really serve up what their clients will need. Even a brief look at how GDPR has been received in the US (websites not available to UK/EU users) shows how nervous US companies are of the legal pitfalls and penalties of GDPR. If the ODI model chooses not to address governance metadata, then really, how useful will it be to their prospective clients?