In-Office vs. Remote: Tackling the Productivity Puzzle in a World of Waste
Big Time Office Waste Mush Blob'in'. Also: John Willis on Deming and DevSecOps, and kubernetes in big organizations.
Re: Iron Law of Remote Work
Last time, I suggested that bosses generally want in-office people, and workers want to work from home. This is regardless of any data, studies, or “reality.” It’s pure human preference. One of you asked me if this was because of an increase in productivity or a decrease. I took that to mean, do bosses like in-office because it increases product, and do employees like remote because it allows them to work less. You could go the other direct as well, or whatever.
I don’t think either model is better or worse for productivity. With exceptions that highlight the normal (as always!), I think most of the work knowledge workers do is full of waste. Much of this is unavoidable because, like, we can't all be robots that only focus on work. We need to have small talk and do all the social engineering (consensus building, etc.) that comes with corporate work.
If you're not a "producer" of your organization's product (in our world, a developer, operations person, maybe a few others), most of your time is spent understanding what's going on, coming up with what to do next, communicating all of that context to different groups and people, trying to persuade people to do what you think should happen, and also all the meta-corporate stuff like ergonomics assessments and understanding annual insurance re-enrollment. Plus, you have to transition between all of your activities - "context switching" as us programers learned back in the 90s/00's.
I should come up with a good example, but, here are some crude ones: imagine the process to decide whether or not you should open a new office in another city. If you’re more into this kind of thing, imagine the process of deciding to acquire a company, or, even more complicated, sell of part of your business. More relevant: imagine deciding how many and who to layoff. Even if you have some kind of power structure where one to three people swiftly make al the decisions, you still have to do all that work and make the hundreds of little choices that make it all happen. More than likely, you’re not in an organization where an executive or two can make a quick decision, so you have to do a lot of work to decide that before you get to the hundreds of other decisions.
In other words you’re going to spending a lot of time in "meetings." For all of the productivity books and ideas over my entire career, I've never seen meetings that work well. They all are just a big pile of mush. The six page memo may make a lot of sense: but are there more than one or five (large) organizations that actually do it as part of their daily work? Maybe pre-existing briefing based organizations like militaries and spies? There's so much optimization that needs to happen for in-office knowledge work, that it's not really worth comparing to anything else.
This meetings problem won’t really be fixed by being in office or remote. Rather than caring about the effect of in-office versus remote work, I’d focus on the bigger problem of all this decision-making waste. Until that problem is fixed, who cares? You’re likely arguing over marginal costs and benefits while that giant blob of waste I just gestured towards is still pulling you down.
My theory is that it’s a little better when you’re remote because you eliminate all the hallway distractions. You know, you’re done with the meeting and you’re walking to get coffee, and someone from the meeting catches up with you and you start discussing something. That doesn’t happen so much in remote work.
For a worker (and the boss!) you also eliminate commute time. Officially, this isn’t supposed to be “on the clock,” but, come on, we all know that you need to leave the office by 4:30 to beat traffic, right? Also, I don’t know about you, but I find that I start work a lot sooner when I’m at home. I just have to walk to my desk. There’s also no time spent going out to lunch. Eliminating the commute is a huge bonus for the individual, and I suspect some of that spills out for the corporation.
But, working at home isn't perfect either. It does subtract out some of the mush, but you'll likely still have the same amount of waste in the system. Instead of taking 30 minutes to talk with someone after the meetin, you get distracted by picking up the kid's room for 30 minutes. Sure, you can listen to the two day, 10 hour annual kick-off meeting while you wash dishes or even mow the lawn...but then you can't take notes, start thinking of how you'll fit into the plan, you’ll zone out, etc. And, yes, it's super-easy to goof off and just straight up not do work.
But, those "problems" exist in the wasteful office as well. You could end up having a two hour lunch where discuss/complain about the new architectural policy. This seems like work, sort of, but it's in that trash-it quadrant of the Eisenhower matrix. And if you're sitting in person for that ten hour kick-off, are you really going to be doing all of those things, or just "checking your email" and day-dreaming.
I think both systems are flawed in different ways, but the effect of those flaws is pretty much the same, uh, magnitude. All that matters is addressing that big pile of waste in how most knowledge work is done. Once that waste is removed, we can argue about, you know - whatever the numbers are - an extra 2% to 5% productivity increase.
My Content
Here’s a few things from me this week, I’ve been busy!
DevSecOps in Practice with VMware Tanzu: A Discussion with the Authors - Tanzu Talk podcast - What is VMware Tanzu? I get asked this question a lot and, you know, I try to explain it. If you want a really good explanation, you should check out a new book on the topic, DevSecOps in Practice with VMware Tanzu. It’s expansive and in-depth, not only on the parts of Tanzu, but also the theory, ideas, and ways of working that Tanzu embodies. Cora Iberkleid joins me to interview the authors of the book, Parth Pandit and Rob Hardt. Video above, if you prefer that.
John Willis on Deming, DevOps, Platform Engineering, and DevSecOps - Software Defined Talk - John Willis joins Matt Ray and me for a discussion in this episode. We discuss John's upcoming book on Deming; the progress of automating audit, security; and compliance with DevOps-think, and then the general state of DevOps and platform engineering.
📚🆓🎉 Free Book: Securing Cloud Applications.
I perused an early version of Adib's new book, Securing Cloud Applications. It provides an excellent synopsis of crafting secure apps in kubernetes, with Java, Spring, etc. If memory serves, it addresses my longtime favorite topics - authentication and authorization - featuring modern techniques such as Face ID, among others.
Here’s more topics covered: configuring standard security protocols, troubleshooting errors from security libraries, utilizing the Google Tink cryptography library, implementing X.509 digital certificates, establishing passwordless logins with WebAuthentication protocol, implementing single sign-on using OpenID Connect protocol, setting up authentication and authorization services with Spring Authorization Server, leveraging popular secret storage solutions, and utilizing Kubernetes' security features to safeguard deployed applications.
Despite the emphasis on security in the cloud, figuring it out in development is hard. Take a look at Adib's book to find out. You can get it free thanks VMware, by clicking on this link.
🧩 ALSO!🚀 Adib is a busy dude, he’s co-hosting two seminars on Azure Spring Apps. As the free sign-up page says: “If you're an application developer, architect, or platform engineer looking to build and scale enterprise-ready apps in the cloud, Azure Spring Apps Enterprise is a fully managed service that can help you reach your goals sooner − all while minimizing busywork and maximizing security with Azure. Want to learn more?” If you’re doing Java development it’ll be worth your time to check out. Sign up for free.
Wastebook
And then he then proceeded to sit in an assortment of crustaceans.
I was talking about the “secure software supply chain” concept with someone yesterday, and I slipped in talking about a “factory” for doing software. Historically, people don’t seem to like the word factory in reference to all the stuff that goes into software. But, you know - it kind of works really well. Is the stink of it yet? // Re: from Luke.
What you always have to ask is: “as opposed to what?”
“The ratio of people who agree with this but don’t act upon it looks like a walrus on a pogo stick.” Here.
Relevant to you’re interests
GPT-4 Has Arrived: Here’s What to Know - I’m trying to think of some uses for a long-lived, multi day chat session. A chat that summarizes articles is interesting. Maybe one that writes journal entries for you; it’s interesting to ask it comment on common themes. Maybe experiments in, like, social media posting. What else? // “GPT-4 can now handle 25,000 words of text, allowing for use cases like long-form content creation and long document search and analysis. It also has a longer working memory of about 64,000 words, or about 50 pages, so it can remember and refer back to things from earlier in a conversation. The old version of ChatGPT could only remember content as far back as 8,000 words, or four to five pages.”
Content was king - “Content trying to reach a busy audience requires packaging that tailors to their needs. Word of mouth doesn’t have the same reach anymore because there are too many words and mouths. Trust carries a lot of weight…. If you’re making content today and not putting effort into packaging then you are doing your audience a disservice. In the world of SEO optimizations, hashtags, and social media—your content will fail to convince anyone it’s worth the time to click. You can’t reach an audience who isn’t reaching back.”
The Impact of AI on Productivity - Developers using CoPilot are a lot faster: “This represents a 55.8% reduction in completion time.”
Kickstarting TikTok: 55,500 Followers & 7m views in 6 Weeks - This has a good video workflow overview.
Logoff
I’ve been a little too precious about the whole concept here for the wastebook. That is: perfect is the enemy of clicking publish. That piece above is more cynical than I’m tryin’ be, but, you know: publish!
I’ve been scheduling several new interviews and planning some new content: a overview of modernizing mainframe apps with my Legacy Trap co-author, a write-up of my platfom talk, an interview with one our consultants about getting kubernetes usage to stick once you’ve set it up in large organizations, and then an analysis of our upcoming State of Kubernetes survey (here’s last year’s). Oh, also, next week I owe a reporter some answers to questions about our State of Spring Survey, so I’ll probably put the unedited ones here like I used to do long ago.
Tomorrow is the picture clear-out wastebook episode. There’s a good one of an egg sandwich with some real sandwich innovation in it. Keep your eyes - like your boiled eggs when you mix them up into a sandwich - peeled.