Content Copyright © 2018 Bloor. All Rights Reserved.
NextStep 2018 in Amsterdam was the OutSystems conference for its low-code digital transformation developers, and for the launch of OutSystems 11, in Europe. OutSystems 11 addresses Large Application Portfolios (LAPs) – i.e., it promises large enterprises that it deals with Architectural Agility; Monitoring and Troubleshooting; and Governance – at scale.
I am still surprised that “everyone” isn’t flocking to “low code” solutions. I can see no other way of coping with the speed of business change in the emerging “mutable businesses”, although one of the characteristics of IT change is that there is a very long tail of stuff that has been working well-enough for ages, and delivered huge value to the company over years of use, and just isn’t going to be changed – modernised – any time soon.
This is why one of OutSystems’ sweet spots is providing custom add-ons to SAP systems, that provide up-to-the-minute services without any need to customise the background SAP system itself. Customising packaged systems is, as one speaker at NextStep said, “the worst form of development ever”.
Eventually, one will need to modernise legacy (or, at least have that option available in the cost/benefit discussion) and, in my opinion, low-code approaches, using BizDevOps for close alignment with the business needs in something approaching real-time, are going to be the most feasible modernisation option.
BizDevOps is the removal of business silos as well as Dev and Ops silos. It institutionalises larger feedback loops, from business outcomes to business requirements analysis and design (and service procurement), as well as the DevOps feedback loop from software construction to software delivery.
So, what is the barrier? Well, everybody knows that senior management buy-in is essential for driving change and modernisation in the mutable business. What everybody doesn’t seem to know is that buy- in by the minions that actually do the work and meet change issues day-to-day, is also vital. This means the encouragement of collaborative communities and mentoring etc., as well as training, and this needs buy-in from managers at all levels, with appropriate resources – an actual change budget – for managing change where needed. And vendors of Agile tools can help, by setting up developer conferences (just one example) – which is why I was pleased at the numbers attending NextStep, and the “buzz” at the conference (which is, apparently, bigger than OutSystems’ US conference).
I was talking around some of these issues with Rijk Zwaan, an international and innovative Dutch family business claiming a people-oriented company culture, often a good basis for Agile development. It develops new vegetable varieties and sells the seeds to growers around the world. I was told that has an unusually “bottom up” culture. It is effectively (as far as I can see) managing the change involved in introducing OutSystems for its Scrum team, and in implementing DevOps; but its Agile projects have to co-exist with more “Waterfall” managed projects, largely in conjunction with external suppliers. I imagine this will be true for many mutable businesses involved in fixed price contracts. Making such fixed price relationships Agile is possible, but not easy, and the culture in the external party is somewhat out of one’s control.
My take-aways on the introduction of OutSystems and DevOps at Rijk Zwaan, from the discussion, were:
- Buy-in by management is important, but so is bottom-up enthusiasm for the OutSystems approach.
- DevOps fits a “flatter” organisation better than it does a silo’d command and control structure.
- An organisation needs to make resources available for managing change.
- Introducing a low-code approach is a journey and for at least part of this journey, you may have to tolerate other approaches if they are delivering business value.
OutSystems seems to have put a lot of work into ensuring that the latest development trends are fully supported with this release – in particular DevOps or BizDevOps, as I’ve said. Of course, removing silos between developers, ops and the business shouldn’t be all that hard for an effective low-code solution – it is something it has always facilitated, by abstracting the underlying technology and by working at the business outcome level. Nevertheless, underlining this capability is good.
I also noted support for AI (I’ll blog about this in more detail in a later blog) and Blockchain. OutSystems’ approach to AI seems measured, but likely to produce good results – AI is seen as essential, in time, to achieving extreme agility. For now, OutSystems is concentrating on assisted data model design and assisted process flow – behaviour – development, where large benefits in time saved are easily available.
I am sure that its support for Blockchain is also pretty effective – except that no-one, even at NextStep, has really convinced me that Blockhain isn’t, mostly, a solution looking for a problem, so far. There are indeed real Blockchain applications, around maintaining trusted and reliable ledgers in domains where no one player is able to impose its control (and it is used to track the provenance of diamonds), but it is currently slow, with high latency for syncing the Blockchain globally. And, most if not all, of what you need Blockchain for can probably be implemented using conventional security-oriented databases.
NextStep was impressive (at least in the Developer Conference space) for the number of attendees and the real “buzz” it got going (although it is nowhere near the size of, say, an IBM developer conference, of course). Overall, I think OutSystems is now probably in the right frame of mind for serious growth.