Agile, DevOps, platform engineering confusion holds developers back

The open source China community team made its first live broadcast, telling the story behind the open source China community in the name of sharing."

Process is the goal: improve delivery in complex enterprise environments to influence change and achieve true continuous improvement.

Translated from Agile, DevOps, Platform Engineering Confusion Stalls Devs , author Charles Humble.

As a large global consulting firm, UST focuses on helping companies innovate and improve technology delivery, with much of its work stemming from DevOps practices . When working with customers, the company often encounters questions such as how to reduce time to market while improving code quality and engineer engagement.

Eugene Sazhin , Head of Engineering, Digital Agile Platforms and Solutions at UST Global, often finds himself thinking, “There should be an overarching set of principles that we can develop, adhere to, and advise our clients to adhere to as well, in order for them to achieve success on their transformation journey Better results.”

This is exactly the goal of UST's recently released Delivery Improvement Framework .

7 principles to improve software delivery

The framework combines Lean, Agile and DevOps with ideas derived from Conway's Law , Team Topology , DORA and UST's own experience. It contains seven core principles:

Flow comes first.
Positivity drives flow.
Team topology orchestrates flow.
Key Performance Indicators (KPIs) coach, not manage.
Lean principles drive optimization.
Purposefully following Conway’s Law]( https://thenewstack.io/context-reversing-conways-law-for-superior-service-management/). %E3%80%82)
Value trust over performance and learning over expertise.

The concept of “flow” is an important part of the framework, and UST uses flow in a lean/organizational sense rather than referring to the flow state of individual developers (although the two concepts are closely related). In defining "flow," UST writes:

"Drawing inspiration from lean principles, we prioritize optimizing the entire value stream, not just isolated parts. We eliminate waste, minimize context switching, and build a system that allows development work to flow seamlessly from ideation to delivery."

The entire framework is employee-centric. The second principle, “Positivity drives flow,” talks about work-life balance, autonomy, recognition, personal growth, and learning. Engaged and happy employees are more productive, but another factor is that successful organizational transformation requires the involvement of everyone involved, from senior leadership to lower-level employees. I have described elsewhere that three of the five organizational transformations I was involved in did not go as expected. In each case, a lack of leadership support or attention was a major factor in the failure.

Express your core principles

UST's logic in publishing its Delivery Improvement Framework draws on Simon Sinek 's idea that if you articulate your core principles publicly, others will be encouraged to join you. By making the framework public, the advisory group is also inviting dialogue, which should allow them to iterate and improve the framework. Rapid feedback loops are essential in any learning organization.

Sazhin also said releasing the framework would help the company's consulting business. If UST clients "understand where we're coming from when solving transformation problems, then we can connect and collaborate, and we have a greater chance of success," he said. Conversely, if clients don't agree with the principles, then "we have core differences. , this simply won’t work.”

If a company declares a set of core values but no one pays much attention to them, they mean nothing. "If you commit to the core values in the framework," Sazhin told The New Stack, "it means you are evaluating all your actions and ideas against those core values and should filter out those that are inconsistent. For example, if You are committed to the core value of positivity, so your communications with employees should follow this principle. You should not allow HR to issue threatening communications; doing so would violate one of the core principles you claim to uphold, and you will. Lost trust. It’s extremely difficult to restore lost trust.”

Developer Productivity Conundrum

In organizations, trust comes from the top. It's also disturbingly easy to lose trust; one way is by not demonstrating the values you espouse, but other ways include making mass layoffs when the company is profitable or tying performance reviews to easily manipulated statistics.

I suspect almost everyone has been asked to implement something they firmly believe won't work. As a responsible partner, you sometimes have to fight back—even if you're a consulting firm that relies on keeping clients happy. For example, UST customers often request a way to measure developer productivity and incorporate it into their processes. However, Sazhin said, "Developer productivity is widely misunderstood and abused in most companies."

McKinsey was heavily criticized for its article on measuring developer productivity . But the underlying attitude still prevails: programming can be reduced to typing, so productivity is equated with Git commits, lines of code, bugs fixed, or other meaningless statistics. The reality is that typing is not the hard part of programming, and, as Sazhin points out, "these things have nothing to do with the team's business goals."

For example, a senior engineer spending time helping more junior members of the team solve their problems is probably the best thing he can do to improve team performance. But if her performance evaluation is based on her Git commit statistics, none of this will be visible.

Sazhin said the desire to measure developers in this way stems from a desire to micromanage both team members and individuals. It stems from a lack of trust in the team's ability to organize itself. But this idea is inconsistent with the core principles of UST. Instead, leaders should allow the team to identify if anyone is not functioning, allow the team to safely voice their concerns, and help the person improve (or make more radical changes).

Experiment and avoid magical thinking

Sazhin believes that rather than micromanaging, managers need to “make sure the team understands what the goals are, remove roadblocks, and then measure whether what the team delivers meets those business goals.” This is reflected in the fourth principle in the framework, “KPIs Coach, not manage”.

This is important because the tech industry is prone to magical thinking . If nothing goes into production without the approval of a change control board (which only meets twice a year), IT can't move faster no matter how many microservices there are. But many organizations seem to think it will.

"Similarly, 'companies say they want agile,' but then they adopt SAFe [Scaled Agile Framework], which confuses me," Sazhin said. "It means they completely misunderstood what agile is. You can implement SAFe Years without any results - you could claim it's because you didn't implement SAFe correctly, but that's not the case."

As Agile consultant Dan North says , "SAFe's stated business model is 'a provider of frameworks, platforms, professional training content, and certifications.' It says nothing about customer success, nothing about accountability, and nothing about about whether the framework is effective." This is all a far cry from the Agile Manifesto's mission of " discovering better ways to develop software ."

Still, the industry has made progress, particularly in improving time to value or how to quickly advance ideas to test with customers.

Companies including Amazon , Intuit , and Netflix conduct thousands of experiments every year. “I encourage people [at Amazon] to hit dead ends and experiment,” Jeff Bezos said . "We've tried to create tools to reduce the cost of experimentation so that we can do more experiments. If you can increase the number of experiments from a hundred to a thousand, you will dramatically increase the number of innovations you generate."

You also need to make it safe for experimenters to fail, since a large number of experiments won't do what you expect. In turn, this means employees need to have strong trust and psychological safety at their core in order to have autonomy. To avoid monocultural thinking, companies need to recruit diversely, and flexible remote working plays an important role by increasing the diversity of the available talent pool.

What this means for accelerating software delivery

Software delivery, in particular, requires small, autonomous, cross-functional teams to build independently deployable units of software. Typically, this means microservices or functions as a service (FaaS), such as AWS Lambda. There is a need to release to production as frequently as necessary with low failure rates - and to recover quickly from any deployment failures. We've known this for twenty years or more: the DORA Acceleration Study, led by expert Nicole Forsgren , even provides a benchmark.

Sazhin emphasizes the importance of validating your ideas and the progress you've made so that any decisions are made based on data rather than intuition. "You need to specify a hypothesis, collect as much data as possible, and then use the scientific method to refine it," he said. "There is a common mindset that some things are too difficult or impossible to measure, and therefore collecting data for scientific analysis is too difficult or simply not feasible. This couldn't be further from the truth." He recommends the book _How to Measure Anything: Finding the Value of Intangible Assets in Business_, by Douglas W. Hubbard, shows how to do this with the help of calibrated estimates, statistical analysis, and simulation.

Finally, Sazhin recommends having a network of influential people who support your work. “If you don’t have them, you’re not going to get the cultural change you need to continuously improve,” he said. "You need to increase the number of change agents in your organization who are actively helping teams on this journey. They should be fully aligned and committed, but most importantly, they should be passionate about these changes."

For more insights, download UST's white paper Unleashing the Developer Process: A Framework for Improving Enterprise Software Delivery .

This article was first published on Yunyunzhongsheng ( https://yylives.cc/ ), everyone is welcome to visit.