Holistic mojo from ojo, baby

The Computer Weekly Developer Network talks to Nathan Baranowski, director at ‘business, technology and change’ consultancy ojo solutions in line with a mini-series of posts related to the rise of what we have called the ‘holistic’ application.

Not perhaps a formally defined technology industry term, the holistic application is an app that has arisen in the age of cloud with containerisation, microservices and compartmentalisation of discrete components at its core.

Asked to comment on the move from monolithic silo’d applications to the new era of holistic, Baranowski says that monolithic applications still have a place, but there’s a far greater need for holistic applications for standard business and consumer requirements.

He says that when his firm is talking to clients dealing with sensitive data, they have to be certain that it is contained. In this case, monolithic would be the tool of choice – contained within one single application with no access or visibility from other applications.

But he points out, for e-commerce sites, bringing all customer decisions to one location, holistic applications with several APIs would be the logical option. It has to be tied to real usage needs.

CWDN: What else does holistic software mean to you?

Baranowski: Any person wanting to work with applications expects applications to talk to each other, to integrate… and to provide them with a seamless experience all in one place. This is what holistic is trying to be, but an app is, at the end of the day, just an app. Holistic provides a single point of view – it’s a window into data.  Customers want to develop ‘abc’ in order to meet their ‘xyz’ goal – they don’t talk about holistic views on anything – they want a useful service.

CWDN: Can we make monolithic applications holistic? Can we put the holistic into legacy?

Baranowski: I think they can if they’ve been built with a very good underlying core. However, very often apps are built in a certain way – so you end up with more of a mess… things are bolted on and bolted on. It becomes a Franken-process.

They can be made holistic, but it’s easier to build something from the ground up. It’s about understanding how we use data and where that data comes from. Can we pull data from everywhere and tell systems about everything, for example? It’s a big ask, but we could… until certain un-named regulations make it harder. Cognitive tools mean this data can be accessed without necessarily holding that data.

CWDN: Where does security and authentication fit into this?

Baranowski: Often the way apps have been built, from a monolithic perspective, is that they’re containers. Access is far more limited, so the security element is more at the server application level. Holistic means you need to think about maintaining that security, as well as how the data accessed is secured without that security looking like a colander. Cambridge Analytica is a prime example of misuse of APIs. The firm shouldn’t have been able to do what it did, but obviously, the people there did.

CWDN: What label should be attached to the new application breed?

Baranowski: Monolithic app development – it’s like building a house. You build it and then you live in it. You’re not going to change the structure… but might give it a makeover. For all intents and purposes, you’re stuck with what you’ve got.

With holistic, it’s about building with a principle in mind. Why should that end-goal exist and how can it be achieved? It’s forever evolving, and so should be able to move, develop and evolve quickly.

If monolithic is old school bricks and mortar, then holistic is applications without walls. It’s the ability to build things that can do everything – the potential to do anything that needs to be done without being contained. It’s essentially Lego or Play-Doh (sorry, copyright lawyers). You can take it apart, build it up again, dismantle it and re-shape it. There’s no set structure.

< class="wp-caption-text">Baranowski: “Think holistic from the ground up – avoid the Franken-process.”

Data Center
Data Management