Hacker News Reader

Yes! You're right that companies say they want ideas and then dismiss them, and that this is something co-workers often do to each other. I doubt this is a scrum/not-scrum issue. I think a manager (or master) can be dismissive within almost any work organization structure. The example you give looks like someone a) feeling entitled to blast you (What gave them that idea!?! I'm not an agile expert, but I'm sure "Chew people out" is not a part agile.) b) They blasted before understanding your thinking and decision making, thereby appearing to be stupid to you and to all us Hacker News readers years later.

That sounds like poorly run scrum more than anything else. I can't count the number of times I've thrown an "investigation" task on the backlog that may or may not have panned out.

Scrum is more about keeping your fellow engineers informed on what they are working on so incompatible things rise quickly. It shouldn't be used as a tool for micro management but instead reviewed at the end of each sprint.

And this is exactly why "agile" has become a religion more than anything — if it's not working, it's because you're doing it wrong. You just need to agile harder.

Well, there are no absolutes in this industry, if there were then we all wouldn't have a job since it would just be paint by numbers.

If a process isn't getting you what you want then change it. Is it any wonder that repeating a broken process gets you broken results each time?

To be fair, it really does sound like bad management - management shouldn’t be blasting an individual during scrum. Management also shouldn’t care if other work gets done as long as the core commitments get finished. Good management shouldn’t hamstring you like that - engineering doesn’t work like that.

That is regardless of agile or whatever development methodology you use.

Management really shouldn't even be participating in any of the scrum meetings other than the sprint review. All of the other meetings are supposed to be just for the team itself.

I've never seen agile work with sprints. Things change too fast and usually there are unknowns. I have seen Kanban work because it takes into account the reality that priorities change, and estimates are never accurate.

I am also a big believer in hard short deadlines when it comes to doing releases. When the deadline comes, release what you have ready and put it behind a feature toggle if needed to do a phased rollout.

If it's not working, why isn't it addressed in the retrospective? If you have a proper retrospective, you can change anything, even going from scrum to kanban.

This is 100% true. Who is holding you back to address things if they aren't working? I've literally never in my professional life have encountered a situation where honest feedback wasn't appreciated.

If you encounter situations where you are being micromanaged, try to place yourself in the role of the one who is micromanaging and try to see his challenges. That person is just somebody who is trying do his best. No one wants to suck the joy and energy out of his or her colleagues.

There is very little you have to add to "keeping everyone informed of what you are doing" to end up at micromanagement.

Notice I said engineers and not management or everyone, very small but important difference.

The thing is though, the number of times I've seen engineers work on conflicting/redundant things without realizing it (outside of the agile/scrum environment) has been vanishingly small. And the places I've seen it it's always been on separate teams that would never have been part of the same scrum group in the first place (ie, it was an inter-team problem rather than an intra-team problem)

This rings true for me. At one of my former employers our scrum masters came up with this idea of a ‘scrum-of-scrums’ where team leads would get together once every month and have a standup to talk about things they’re working on that could affect other teams. But it became a painful waste of time because some of the participants didn’t understand that we didn’t want to hear them talk about every single task their team worked on. Then we tried ‘release-train’ planning where the whole company would do 2 day company-stand ups. That was even worse. The real problem was that the company had a weird cultural and engineering mismatch between hardware and software engineers and the upper management was too clueless and careless to coordinate between them.

> You're doing it wrong.

This is the nonfalsifiability of a religion surfacing.

THIS is the problem. This very thing.

The implement is ALWAYS blamed.

The problem with "investigation" tasks or "spikes" is that it still boils down to essentially having to ask permission to do something. I really think extreme one-sided work transparency and a culture where everything is measured and permission has to be sought to work on something is entirely antithetical to encouraging creativity. Innovation tends to come from playfulness, and having to schedule a Jira task with a project manager for every idea is going to have a chilling effect.

You’ve hit my first main point of hatred against agile. It’s really a form of micromanaging in disguise. They sell it as developer empowerment but that’s just cognitive dissonance.

That depends on the company. Some companies (in my experience) do engineer-only scrums where programmers can talk about their blockers and ask for advice from seniors; others do mixed scrums where the engineers are basically reporting their progress to the product manager or someone, and where often technical questions are discouraged.

I'd tend to think the latter are "doing it wrong", but then again, who am I to say? I just know I much preferred working for the companies who did it the former way.

This idea that only managers micromanage and fellow engineers never do that us one of biggest agile fallacies.

Also, need to ask question should not require whole team meeting.

This is not a problem of agile, it's a problem with the implementation of agile. A company that tends to micromanage that implements agile will use agile to micromanage. It's unavoidable.

I've heard various forms of this response before but I'm always confused by it. If agile isn't the offending processes in question then what is agile.

And if the response is it's a verb not not a noun then it's really nothing but good hackers employing tried and true practices.

Anytime someone complains about agile and then a "true believer" defends it as "that's not agile", it reminds me of the "No True Scotsman"

Quite. I’m sure I’ve seen “no true scrummaster” used in the past to describe this form of debate.

I can't imagine taking up an investigation of any significant size (say, more than a day) that I don't mention to my team. And yes that means putting it on a board. This isn't asking for permission. It's making others aware. There's a good chance someone on my team has some valuable input. Or awareness of what time doing may affect what they choose to do somehow.

Granted, my team is small and close-knit. But the idea that keeping others informed of what you're doing is considered "asking permission" points to some serious pathology on your team.

And that pathology is not agile or scrum. We do scrum, and I'm sort of lukewarm about how much I like scrum. But whether we're doing scrum or something else, I am completely empowered to make my own decisions about my work. And I simultaneously value that the rest of my team keeps me informed of what they're doing. This is how we kick ass.

Nothing in agile stops you from creating experimentation tasks or simply spending time on whatever you want. It just asks you to honestly be aware of how you spend time.

Well, I'm pretty well aware of how I spend my time. Agile serves no purpose for me in that regard. The point of agile is so that others can track how I spend my time. Whether that's reasonable and if/when that's a good idea is entirely about context; but much like quantum mechanics, observing behavior can change it (positively or negatively).

When I've done R&D in an agile setting, usually the research was done outside of the scrum process, and once that was done then the development would be done in a standard scrum methodology.

The reason you can't really apply "Agile" to research is that having a useful product at the end of each sprint is a central to Agile development but runs contrary to the purpose of research.

Now, you still need to manage your research process. You need to identify the specific research questions you need to answer, and you need to bound the time investment. But let's not try too hard to shoehorn a methodology primarily oriented towards IT consulting into every other domain.

How many points did you give those tasks? We're the points agreed upon by the team? Did the team and the business agree to the tasks being included in the Sprint during Sprint planning?

It sounds like you created the tasks yourself and added them to the backlog mid-sprint. If that's what you did I'd be totally cool with it but I can tell you a lot of "agile shops" would blow a gasket over that.

Having been on a lot of teams with various processes, it seems to ultimately come down to trust. I’ve seen teams where there’s a ton of trust and mutual respect, where even a very structured agile approach has allowed for sprints focused entirely on innovation and exploration. Process is just another tool; how it’s used — and the environment — matters a lot.

I think the flip side of this is that Agile does not add much in the best case (people trust each other), but it can cause a lot of damage in the worst case (nobody trusts each other, but people can be productive thanks to a lack of transparency).

> There have been a lot of times in other jobs where we'd have a slack in our schedule and I used the down time to make a change that was really necessary and took a minimal amount of time, mentioned it in a standup meeting, and got blasted for not working on our "commitments" for the "sprint" (even when I had nothing to do).

That doesn't really have anything to do with agile or scrum though. Just add a refactor/investigation task to the backlog for the sprint. Those types of activities should be justifiable too. If your team doesn't allow those types of tasks for some reason, that's more of a team problem than a problem with agile.

"Justifiable" is the key word though. Why do I need to justify what I'm working on daily unless I'm not trusted? I mean, like most programmers, I have a four year computer science degree that was definitely not easy to obtain, I have like 10 years of experience, I've shown an understanding for prioritizing things that actually are important for the business, and they pay me a six figure salary for my expertise, so having to justify myself daily seems, frankly, insulting. I've worked in places where I didn't have to do that, and you know what? The sky didn't fall. Work still got done. (I would argue, more work, because people were more engaged since they felt a sense of ownership). We just had extra time in the day from fewer useless meetings, and a lot less politics from people trying to game the numbers they pull out of Jira and stuff like that.

This is a toxic attitude. Your 10 years of experience and 4 year education don't make you infallible. A high functioning team works closely, and that involves everyone's decisions being subject to scrutiny. If you can't justify your work, then it probably shouldn't get done. If you won't justify your work, then you don't deserve the six figures.

I never claimed to be infallible, and I'm happy to explain what I'm doing if someone asks, but ask yourself, how many other skilled highly educated professions are asked to provide such a humiliating measure of unnecessary transparency? (Hint: not most of them). Does your manager give you such transparency into what they're doing day to day? (Generally speaking: nope!) How do I know they're not just spending their afternoon's playing golf? (I don't: but I wouldn't work for them if I didn't think they were mature capable professionals).

Think of it this way, in relationships, business or otherwise, what's sub-communicated is just as important as what's explicitly communicated. Often it's more important. And what's communicated via most agile processes is that programmers are unimportant cogs with no business knowledge that can't be trusted to not stray from important tasks. I think that's ugly.

When a lawyer or consultant bills a client he has to enumerate how each hour is spent. As an engineer in another field I submitted all my calculations for peer review, and all my drawings had to be signed by a manager.

When you buy a grocery product you get a list of every ingredient and all the nutritional info.

It's nothing more than transparency and accountability. The one paying the money wants and deserves to know what it's for.

Your example is much closer to a code or design review than a daily scrum meeting. I don't think he's against design meetings or reviews.

Also, software development is quite a bit different than other engineering disciplines, at least at the application development level.

This made me think of the book "checklist manifesto" which argues that surgeons should be forced to follow a checklist to make sure they don't miss anything.

Part of the argument is that often the more junior members of them team know the surgeons are making mistakes, but there's no way for them to point this out without being bullied. Anectodes include the wrong leg being cut off while members of the medical team know it's the wrong one, but darent risk their career to mention it.

I'm not sure how this relates to your comment, except your line about other professions reminded me.

I would think the equivalent of that kind of thing in the software world would be things like automated testing and code reviews. (IE, processes that recognize people are fallible and provides mitigation). I think scrum is a different beast from those kinds of things (especially given that there's very little evidence for scrum's supposed effectiveness)

The problem is in your team or your corporate culture. Not agile or scrum as a whole.

On my team, agile is a process used among peers to coordinate. No one is a cog. No one is checking up on anyone. We keep each other aware of what's going on because you have to in order to coordinate and collaborate. My manager is just one voice among many on the team. He isn't tasked with monitoring my work because I don't need a babysitter.

I'm sorry that you work somewhere where you feel someone's monitoring or checking up on you. Where I work that kind of situation is called a Performance Improvement Plan. If you're not in that situation and are a highly productive individual, then your culture is broken and you should look for a different job.

But you're wrong that agile and scrum are at fault. If you used any other process, you'd have the same problem.

Well this is an aggregate opinion based on the last three places I've worked, mixed with a lot of conversations with peers at other places. It's certainly possible that I've just been unlucky and my colleagues have also been unlucky, but I would also point out that I've seen pretty much zero in the way of objective proof that agile actually improves anything, and a lot of subjective experience that agile doesn't actually benefit engineers. Frankly, I think it's up to the agile proponents to objectively demonstrate why this is better -- but 99% what I see is vague promises mixed with attempts at shaming people for "doing it wrong" when people say it doesn't work. (Not to mention the whole "waterfall" boogie man. As far as I can tell in the agilista world-view if you're not doing agile, you're 1970s IBM and you wear a blue suit to work every day)

Anecdotally, the last place I worked, I was on a small team that didn't really do agile. This was composed of some of the most competent, extremely smart people I've ever worked with that had decades of experience in the CAD field and had worked together for a very long time. We're talking people with like dual PhDs from very respected universities that have been working in the field for like 20 years; kept very sharp by going to C++ conferences, were very particular about testing and code reviews, and so on. I remember the corporate office tried to impose agile on these people, and after humoring it for about a month the whole thing was mercifully abandoned because everyone felt like it was ridiculous and it was slowing things down and causing a bunch of friction when some ridiculous consultant wanted us to do "planning poker". The only people that really wanted it in the first place were essentially remote project managers that were annoyed with these uppity engineers that thought they knew better (which frankly: they did; 20 years writing CAD software and you're going to learn a thing or two, whereas the project managers were pretty much just dabblers -- and when they couldn't convince people to do what they wanted by merits of their argument, they just wanted to install "process" to essentially strong-arm people into doing their will)

Yikes... this is a toxic attitude. I'd expect that kind of scrutiny in a hard-hat and lunch pail assembly line job, not a purportedly professional career.

Professional doctors, lawyers , and plumbers are all accountable for their time.

Being a professional means having the self confidence and emotional maturity to justify one's work whenever it is called into question. In fact, one should do the due diligence (including cost benefit analysis) to be able to answer the question with ease.

If you can't answer "why are you working on this?" easily then you're not doing your job. And if you resent being asked the question, then you either are self employed or you forgot who pays your salary.

You may be saying the same things. It seems like the person you’re replying to emphasized _daily_ justification of work, which does seem unnecessary to me.

Being open to having to justify anything that's called into question, and being capable of doing so, and being happy to... are different from forced to.

I sort of agree with you. A lot of "justification" ends up being such a frustrating waste if time. Usually necessary though.

Yeah, not once have I been involved with a project or team attempting some forced process where an objective benefit to cost analysis was done on the process itself.

This is spot on. It's the Enterprise's Holy Grail; the attempt to commoditize software development.

> The amount of micro-management that goes into most programmers day to day work makes it extremely hard to be innovative, because you essentially have to justify what you're working on daily -- and any creative person knows that not every idea pans out. You need managerial support to occasionally make mistakes.

There have been a lot of times in other jobs where we'd have a slack in our schedule and I used the down time to make a change that was really necessary and took a minimal amount of time, mentioned it in a standup meeting, and got blasted for not working on our "commitments" for the "sprint" (even when I had nothing to do).

> The worst part is that it's not just management creating this mentality, it's the other programmers. I've noticed that younger more inexperienced programmers tend to crave the structure of agile, whereas the older more experienced programmers chafe at it.

> I think if programming had more of a culture of apprenticeship and mentorship, and a deeper appreciation that this is hard and it takes a long time to get good at it, our industry would be a lot healthier.