Content Copyright © 2023 Bloor. All Rights Reserved.
Also posted on: Bloor blogs
I don’t really associate OpenText with DevOps particularly but, of course, the acquisition of Micro Focus now makes it a major player. At OpenText World 2023 in Las Vegas, I was able to discuss the issues around enterprise-scale DevOps with Julian Fish, Director of Product Management Development, ADM/DevOps, and George Harot, Director Product Marketing, ADM/DevOps ,at Open Text. We all agreed that the key current issue with DevOps is delivering business outcomes to the business. DevOps can be a success, it can deliver quality software to the business through an automated pipeline, and the business can still ask, “why aren’t we moving faster and competing better?”.
This is because DevOps is an attitude and culture as much as technology, and it isn’t enough for just the developers to be good at DevOps. Delivering software fast is a good thing only if it is the right software – the software the business wants (or, rather, the software the business needs). Which implies that the feedback loop from deployed software to agile build has to extend out from business outcomes to design thinking. If any part of that bigger loop is a bottleneck, the business won’t see the business outcomes from DevOps it is expecting.
I do see an opportunity for AI here, and not only for speeding up coding (suggesting the next piece of code) and designing the rich set of test cases needed for the delivery pipeline, but also for finding emerging patterns in the business, that lead to interesting opportunities to develop at DevOps velocity. It will also be important to integrate mainframe applications with the DevOps culture – AI search should help – the mainframe is a source of high quality data for new Web apps, and this quality should be exploited, not lost through mixing it with lower quality data from the “data swamp”.
As an aside, it is worth remembering that in a regulated business (and, with privacy and personal data regulations, all businesses are, or will be soon), some things can’t be allowed to break. Some financial business regulations will close you down if you break them (and, I increasingly expect regulators – or hostile actors – to use AI to find such issues, if they escape into production). So, it is good to remember that Fish coined the phrase “move fast without breaking things” when he was at Serena, before its acquisition by MicroFocus. Serena, to my mind, had a good handle on Agile development at scale, in highly regulated organisations.
The major DevOps announcement at OpenText was the DevOps Aviator, which it hopes will revolutionise software delivery, taking advantage of generative AI to increase the velocity of software delivery. This will come as a result of feature prediction, automatic test creation and authoring and the identification of points of risk that impact quality. Despite all the hype, I really do think it will – but in conjunction with the tried and tested DevOps techniques that Serena and Micro Focus, now in OpenText, helped to develop.