Ask Lemmy

first, you're talking about software "engineers" which means you aren't talking about engineers in general.

and there's a good chance none of them have ever had an engineering course in their life. they're hackers who are good at making code.

the reason they probably seem reluctant to share is that what they've cobbled together with bubble gum and bailing wire is difficult to explain quickly and thoroughly AND they'd be taking time away from their assigned tasks to do so without having any change to their deadlines.

stop blaming them and start blaming their management for not giving them the time and permission they need to help you. go to the management and say you need so-and-so to be assigned 40 or 80 hours specifically to help you understand these widgets.

and in the future you need to push for clean up, documentation, lessons learned, and training to be part of every project estimate.

As an automation/software engineer that works with sysadmins I think there's a natural resistance to top-down initiatives or others meddling with their processes (i.e. they're the SMEs so just let them work). I could also see your line of questioning go into a decision to restructure and eliminate jobs.

In the first case (general change resistance) I think you'd need to come with numbers that show how expensive a dept is compared to industry standards or how inefficiency drives issues downstream.

In the second case the best way to show jobs as secure is to detail all the work that needs to be done, implying how foolish it would be to cut staff and miss even more deadlines.

It sounds like your job might be needlessly hard.

Coming up with a design for a new process that will fix all issues at once is very hard; you're very likely to miss something important. Making such a process change in one go is also hard, even if you somehow happened to end up with a improbably good spec. Doing it by interviewing people sounds kinda doomed.

An easier path might be to take whatever holistic understanding you have right now and start in some corner of the problem where there are clear issues. Bring engineers and people who use the system together. Have the people who use the system walk through their common workflow together with the engineers, noting what parts are usually hard or slow them down. Keep people focused on improving things rather than arguing about how you got here.

Together come up with small achievable process or software fixes you can implement and evaluate quickly (like in a week or two). If it works out, you have now made a real improvement. If it didn't work out, you understand the limitations a bit better and can try again, as it was pretty quick.

Helping to deliver real improvements in a way that's visible both to the involved engineers and the people using the system will buy you a lot of credibility for the next step.

I’ve tried to make it clear that I’m not interested in punishing anyone for past decisions or mistakes—on the contrary, I want to learn from them to create a better process moving forward. My goal is to collaborate and make their jobs easier, not harder, but I think building trust and comfort will take more time.

I'd wager that the engineers have experienced such promises in the past and got burned. Engineers, by nature, are very analytical. Re-gaining trust that was once burned will take a lot of work. And managers like you are exactly the kind of people that burn engineers.

Good luck to you. Sounds like you're working at the intersection of management meets reality, and nobody has extra love for a scrum master.

I can recommend honestly and incremental adoption. It will be difficult to eat this whole sandwich at once.

https://media.tenor.com/rGSj2L2GVecAAAAM/on-the.gif

I think people have already done a god job of covering the likely concerns. Here are the things I would emphasize.

Bear in mind that a lot of developers just hate doing documentation. :-}

Make sure that their management has made working with you a part of the engineers' work load and goals. No one is going to provide good information when every minute they spend is putting them behind on things that directly affect their careers.

Provide them with a context for what you are trying to accomplish. Tell them the why and how, not just the what. That information can be very general or it can be at the level of providing specific examples of how you intend to present the information you gather. Find out what they would like to know, particularly since it's likely to vary from person to person.

Keep in mind how different people can be. There are reasons for the stereotypes about developers, but their are pointy ends on every bell curve. You are likely to find a few people who communicate very well and can help you get the information you need from those who do not.

You sound like you have good intentions and the skill set for doing this kind of work. Don't let negative responses discourage you. Work with the people you have, treat them with respect, and make sure they get credit for the work they do with you. Let them see what you're doing and ask for feedback. There are going to be things you can't control in the process, but if you work openly and in good faith people will usually respond in kind.

Sounds like a good old leadership trust issue. Unfortunately the only thing that solves that is time, beer (or other social activity), making yourself useful to them in other ways, and being honest with them.

If they're afraid of punishment you can always try an amnesty box. They put what they would say in the box, anonymously, and you discuss it without trying to figure out who submitted it. Even if it's obvious. Then they don't have to trust you so much as the process.