Gartner’s AADI Conference kicked off this morning at the Park Plaza in London with the keynote Gartner Keynote: Integrate the Past. Embrace the Present. Shape the Futuredelivered by Gartner’s Andy Kyte and David Mitchell Smith. Kyte and Smith talked about the Nexus of Forces (cloud, mobile, social and information) that is driving both change and uncertainty. They framed their conversation against the skills and assets that organizations bring from their past, but also about the technology and process “baggage.” To move forward, they said, “…organizations must not just integrate legacy but also do so in a way that minimizes dependencies on legacy thinking.”
What to Do with Legacy Systems
The single biggest problem most organizations have today is exactly what to do with legacy systems that are crucial for the business as both systems of record and as business process support. David Mitchell Smith started off the conversation by saying that the key to embracing the future is making a choice to shape it and not be passive about the change that’s coming.
Shaping the future, Smith said, is about understanding what got us to today. That requires understanding the intersection points of the Nexus of Forces. He said that while each element of the Nexus is strongly disruptive, in combination they set the stage for innovation to happen.
Ripples on a Pond
Smith used the analogy of ripples on a pond, saying that the key for organizations is to understand the effect of the pond’s ripples, but to pay particular attention to the points where the ripples intersect. Smith gave the example of what he called the mobile imperative, where mobile needs to fit into IT, but where IT also needs to change and not treat mobility the same way as previous technology waves. Smith said that mobile strategies should not work in a vacuum and that mobility should be part of every other strategy, including security, integration and development. Don’t outsource mobility, he said, because the skills required are going to be more important in the future. His mobility points were as follows:
- Mobile changes everything
- Mobility is about people, not devices
- No platform, vendor, device or form factor will win; many will co-exist
- Mobile is a key enabler in the Nexus of Forces
- Mobile strategies should not be in a vacuum
- Leverage existing skills and assets is a key goal
- Consumerization sets the agenda
Smith turned the stage over to Andy Kyte, who asked the audience to think about what’s happened in application development over the past five years and then compare that with expectation for the next five years. He offered three scenarios:
- Demand for application capability is going to decline because the work has been mostly done
- The level of demand will remain mostly the same, with organizations living with a backlog and working their way through it
- The demand for application capability is going to grow because user expectation is going to grow dramatically, driven by the Nexus of Forces
Growth of Demand
Growth in demand is going to be relentless. With IT budget cuts and with the huge amount of time and attention to managing the legacy systems, our ability to meet that growth in demand is extremely limited. There is a massive amount of unsatisfied demand.
Kyte said that even worse, technology vendors are coming to the business and saying, “We can satisfy your demand.” This is a problem, as the solutions being sold to the business complicate the job of IT, making life stressful for IT personnel.
Kyte said that the vast majority of organizations are struggling to cope with current demand and will be inundated by the demands that are coming down the road.
The Team, The Team, The Team
Kyte feels that IT is made up of warring teams that defend their applications and refuse to share artifacts across other teams. These teams see it as a battle for declining budget. On the opposite side, Kyte said that the most successful IT groups operate as a single team and share resources. The best IT is highly collaborative and open to external ideas. He challenged the audience to move from “toxic baronies” that exist in most organizations to teams of people who work cooperatively.
Right Things Right
The best organizations know what’s truly needed and can say, “This is what we should be working on now.” Kyte said that the challenged IT departments he sees are putting enormous effort into things that are “a great landing… wrong airport” because of siloed funding that goes to equally siloed business managers. The siloed managers are in competition for budget in a zero-sum game.
Kyte feels respect is usually delivered more often through very flat organizational structures. Organizational charts in the most challenged organizations read like “long scrolls,” with the people doing the actual work many levels down from the leadership. The smart and innovative people at the bottom aren’t listened to and have limited positive effect on the organization.
In the flat structured organizations, on the other hand, managers aren’t expected to deliver innovations and instead recognize the individuals who are thought leaders and structure teams around them. Time in grade has little to do with the influence and the organizations are true meritocracies.
Design for Life
The best organizations are focused on “design for life,” which is what Kyte calls the, “Shocking difference between the best and the rest.” Kyte said that, “Most IT organizations are ‘stuffed to the gunwales’ with project stuff,” and the culture is all about the success of the project alone. People in challenged organizations want to be attached to big projects that have longevity and less risk of near-term failure. Success in the poor environment is constituted by being on time, budget and quality, but each of those three “legs of the milking stool” is highly subjective and not necessarily aligned to the organization’s needs.
Designed for life, on the other hand, is focused on functionality, reliability, maintainability, usability and things that create systems for life rather than satisfying short-term need for “screens.” The ability to design for life satisfies the attributes for success today and into the future. The challenge? It takes greater design work done by “designers” and not necessarily by the ubiquitous architect.
Everything as a Service
These designers are focused on “everything as a service” rather than focusing on architectures that don’t satisfy future needs. Highly stable, highly predictable and persistent systems are developed within organizations that put capabilities under a layer of services. To do this, they’re using application architectures based on Greek mythology, and in particular, the Ship of Theseus. Without getting too historical, the Greeks had to constantly replace the planks of Theseus’ ship, a monument to his military victory. Eventually, the ship was no longer the original ship, but had been completely changed out.
For Kyte, the Ship of Theseus is a model for IT, where services can be constantly replaced and for transformation to take place constantly and gradually without having to bring in new architectures or throw away large portions of legacy systems.
The best organizations are focused on reuse. Kyte expressed this as an IT maturity model that looks like this:
- I reuse what I have written
- We reuse within the team
- Departmental reuse — we share between teams (advanced level)
- We reuse within the company
- We reuse what we find outside our organization
Traditional IT is stuck at levels 1 and 2, where the best IT organizations are able to consistently reuse at levels 3 and 4 while embracing the idea of level 5.
Selfish software is about writing services that are designed defensively so that they can be used in environments that are unknown to the designer. These services must be self-testing, self-monitoring and “selfish” in a very positive way. Invocation of these services is safe in many environments, including with third parties that want to reuse this “paranoid” work. Kyte pointed out that it costs more upfront, but these services can be reused much more readily and will survive much longer.
Kyte summarized with, “Services should not trust their environment.”
Kyte noted that leading organizations invest a significant amount in quality assurance. They see QA not as something that happens at the end of the development process, but instead throughout delivery. Kyte described this practice as QA being done through design reviews before coding. In Kyte’s opinion, 60% of QA should be done in design review with only 20% coming at the end of development.
This article first appeared on The TIBCO Blog and has been lightly edited.