It may seem like blasphemy, but actually the ability of Cloud Computing technologies to virtualize applications and gain resulting hardware efficiencies, and thus cost-savings, is not actually the most interesting part or compelling business reason for its adoption.
Sure a ton of money is spent on under-utilized servers and storage and of course the cost savings are there to be had, but these benefits pale in comparison to those that can be enjoyed through the ability of the Cloud to accelerate application software innovation.
Building better integrated, more user-centric applications faster, is the killer Cloud use case.
Open Government Innovation
Naturally to achieve this a new paradigm in software engineering is involved, but actually the bulk of the benefit will come from re-engineering organizations and SDLC processes, especially procurement and enterprise architecture of LOB (line of business) applications.
The reason for this is to unblock ‘Innovation Gridlock‘. The stickiness that builds up due to years of legacy applications layered atop one another.
For example in government, like most large organizations, they typically have a very slow-moving enterprise application deployment process. If the Social Welfare unit decides it needs a new case management system to improve how effectively they deliver care services, then a sequence of business analysis, RFP writing, supplier review and decision, purchase and project implementation is carried out and can take many months and years.
Add in that any number of these might be in effect at one time and the rate of “innovation enablement throughput”, the amount of new software being deployed to enable new innovations, grinds down to near if not entirely zero.
Et voila: Innovation Gridlock.
This is why similarly with developments like Open Data it’s not the technology that’s the most interesting part. Rather it’s the open source development model that is, the fact that the creation of new software capability for online government systems is happening outside of the government IT department.
By publishing open data sets the goverment is enabling “citizen developers” to create software for them, meaning the innovation enablement throughput rate is no longer constricted by their internal process.
It’s also an approach of lots of small code projects, rather than one huge enterprise effort. These developers are quickly publishing lots of useable services that provide citizen value. The innovation throughput rate is high.
As this blog from Microsofts Open Data expert highlights, it’s this combination of open source software and open data and the ability to publish to the Cloud that makes all this possible.
Open Linked Data
In the introduction I also said “better integrated” applications, highlighting the other critical value of this Open Data approach.
The other heavy price that the large enterprise pays for their monolithic application deployment approach is that they’re typically driven by their departmental hierarchies too, ie. Social Welfare wants their own case management system, on their own hardware and their own database. That’s how they end up with an estate of all this under-utilized hardware!
In business process terms it also results in a situation where customer information is duplicated across them. John Smith may receive Healthcare treatment as well as Social Welfare, but these systems have no knowledge of this. Each John Smith is entirely unique to them, as far as they’re concerned, and this is how government processes become disjointed.
The remedy to this will come in the form of Internet standards for identifying these attributes and sharing them across multiple systems, what is known as ‘Linked Data‘. As the Open Data approach matures so it will come to use these methods and thus become Open Linked Data.
Cloud Aware Applications will be better integrated, more user-centric and deployed much faster than the legacy enterprise app approach.