To be fair, the evidence provided for why not to return to office also contains no data or numbers and is based purely on anecdotal personal opinions and feelings.

Agreed.

There are studies around this topic. My issue is that, they are never cited when these decisions are announced.

There's no evidence to provide if that's the preferred way to work, which we know because we've had a ton of internal surveys about it. People prefer working remotely, that part, at least for AWS, is not up for debate.

WFH has no commute. That is a fact. We can argue if that is good or bad.

I think it is better to have no commute, since it is unpaid, and pollutes.

To me, an unpaid, let‘s say 1h, commute is worse on principle, and requires a strong case before accepting. WFH is better by default based on this principle.

i will return to office 100% of the time the day companies start to pay for commute time. they are making me spend my own time going to the office, so they should pay me for it.

Are you paid by the hour? What would it mean for your job to “pay for commute time”?

I'm salaried, but this could fit with another idea which is in the air at the moment - a reduced working week.

I have been WFH since before the pandemic, but technically there is an office I could commute to. It would take me about an hour each way. If I did that Monday to Thursday, that's 8 hours - so four 10-hour days if commuting was part of my "on the clock" time. If my employer told me I had the option to make Friday a weekend day in exchange for commuting Monday to Thursday, I would strongly consider it and think of it as a reasonable compromise.

I would have less convenient and more expensive days for the working week, but more time off. They get 8 fewer hours of me in front of a keyboard, but do get the supposed benefits of office-based employees.

Heh. That's not a solid argument, though.

A full time employee is paid for 40h per week. So using commonly available tools we can determine the average commute time and in many countries travel time is considered work time.

So on paper but generally rarely enforced, you should be able to subtract your commute time from the 40h and end up with actual work time.

Everything on top being overtime and should be paid according to rates mandated legally.

Except for a handful of companies and state employees, I've never seen overtime being paid, though.

> A full time employee is paid for 40h per week.

That’s not true, at least in the US. The amount of hours I work weekly has never been tracked at any company big or small.

You’re not paid for 40 hours a week, you’re paid to keep up with the general standard of your peers. 40 hours is a very vague estimate of the time required to do so.

In Europe, contracts generally specify a number of hours per week. For instance in Austria, my contract says 38.5h/week - this is what people usually think of here when they say "full time".

We have a time tracking system and overtime is either paid (not in my company, alas) or you can take it as free time.

The US is an outlier for the developed world.

We are talking about a US company.

With hundreds of thousands of employees outside the US.

I can get 5h of work time with a commute or 7h without due to family reasons. Mandating the office means I quit because the final 3h of wfh after 2h commute turn my life into hell.

You'd work a 2h shorter day, get there later/leave earlier than currently or than that you'd be working in your home office if it was a remote job.

I like driving; maybe I should then look for jobs about 150 miles away.

Not saying whether I believe it or not, and I personally work remote 100%, but one of the arguments for RTO is that remote work diminishes "social capital" in a company, where the social capital of a company is the combined value of all the social connections between the employees. In other words, if Bob has to fix the database and he knows that Zoe created it and is in the next cube over, he's going to be able to get through that problem a lot faster than if they've never met and he doesn't know what she works on. The idea being, of course, that at scale it could have a big impact on the effectiveness of the company.

(Edit: Just to be clear, last time I saw a presentation on this, they did have actual metrics. But numbers on a slide deck should always be taken with a big grain of salt.)

But that seems like a faulty reasoning, assuming I'm Bob, Slack and Email allow me to reach Zoe just as easily with the bonus that I'm not interrupting her flow.

> scale it could have a big impact on the effectiveness of the company

It does not though, if the culture is anything like Amazon's, Zoe will ask you to cut her a ticket instead of directly talking to her. There is nothing with in-person interactions that "scale".

That's not what social capital is.

A message or email works differently when you have an established relationship with the other party. It's one thing when I have to do something "because the company needs it" but totally different when I'm helping a flesh-and-blood coworker. You don't build relationships and trust through jira tickets and emails, you build them through unstructured messy human interactions. They pay off later.

> You don't build relationships and trust through jira tickets and emails, you build them through unstructured messy human interactions. They pay off later.

Assuming you're neurotypical.

Which most people are. You make structure how your business works on the typical case. That will provide the most transformation.

> Which most people are.

Definitely not in actual developers or sysadmins.

Most of them are in this manner.

Almost all the ones I've met.

We do an occasional get-together every few months (with a big break due to the plague). Seems to work just as well as when I worked at the office.

I absolutely have close work (and friend) relationships with people on far flung teams, and even companies I no longer work for in countries I'll never visit, whom I've never met in meatspace (or only a couple times at conferences or whatever). These relationships were built mainly by IRC/jabber/slack (depending on the vintage) and phone calls/videocon though, not tickets and email. Time zone probably matters more than anything -- I keep in touch with colleagues from India who I met when working third shift in web hosting support, but couldn't tell you the names of the daywalkers from the same office I was in.

Modern corporate environments are spread out all over the planet. If you can't build relationships without sharing air, you will be at a massive disadvantage. That's just how it is, and it isn't something that was caused by the pandemic, it's a consequence of intense globalization over the past ~40 years and an aggressive investor-driven acquisitions culture.

Companies who treat employees as meat-based dysfunctional robots diminish their social capital... If they want a "social capital, people who feel to be part of the company, something like a family" and so on, it's not WFH the point but how companies hire and behave.

> remote work diminishes "social capital"

Stack ranking forcing a number of your employees into "low performance" and putting them on PIP is awesome for social capital however. As are layoffs.

"it's all bullshit, and it's bad for you"

I think you're misunderstanding the point. "Social capital" isn't another way to say "happy workers". There's probably a ton of it in a gulag, for example. It's better for a prisoner to know who can patch a shoe versus having to walk their toes off in the snow, and it's better for the guards because they won't be short a log hauler. Social capital manifest, without a smile in sight.

A layoff is going to sever some of those connections in the short term, but the number and quality of connections isn't a function of how people feel about their job. Stack ranking and PIPs are tools that the company uses to mitigate the impact of letting people go. In other words, they're there to protect social capital (for the good of the company, don't get me wrong) versus doing something like picking a random 20% of people to eliminate.

I understand full well what "social capital" is supposed to mean.

My point is that fostering a culture of mistrust and competitiveness among the employees is antithetical to protecting social capital.

I won't effectively collaborate with you if it helps you being the low performer that will be laid off instead of me.

That's a fair point, but based on my experience I don't think most people are as deliberate as that. I've survived more layoffs than I care to count over the years, and it's always seemed like there were plenty of people on the far left of the bell curve without anyone needing to actively sabotage them.

That has been my observation as well.

And curiously enough, I normally don't care about being PIP'd or layoff, as I engage in healthy job hopping so that I was never actually laid off over my relatively long career. Typically I'm one to see the early signs of trouble ahead and line up a new job in the next month or so.

But it wouldn't surprise me if it was found out that the culture of PIP and Stack Ranking helped to select for the ones that are best at navigating the corporate culture in the way I described above.

> But it wouldn't surprise me if it was found out that the culture of PIP and Stack Ranking helped to select for the ones that are best at navigating the corporate culture in the way I described above.

Well, when you get down to it, if you find yourself back-stabbing a coworker to keep yourself out of the bottom 10% during a RIF, that's not a huge badge of honor either.

I don't expect people to be after badges of honor, generally speaking.

* * *

Yeah I think it is this. Teams glue easier when working physical together. Person in the office also buys managers bullshit and company peculiarities because there is social pressure to buy-in those . So in the end keeping remote workers is way harder and work is more transactional.

"Teams glue easier when working physical (sic) together."

You can't state this as a fact. My company has two main offices. Two of the other SREs I work with are in a different office/state. We're as tight as can be. A third SRE is in yet another city. Our teams performance since the start of remote work has been outstanding.

There are no "facts" to this RTO fetish. It's just a management fetish, just like Jack Welch's BS that spread like wildfire.

"and he knows that Zoe created it"

There's a lot of ways to share that piece of knowledge other than relying on physical proximity. Documentation, metadata, AOB in the team meeting.

Yeah, except one has seen people next to each other in cubicles working on the same project not really talking with each other unless dragged into a meeting room (that also could be virtual later during pandemic) ... and others who drop just few chat messages around to identify their person of interest and just reach out regardless of whether they are near by or in another coorp office in another city? I believe it is a people are different problem, and also this is just belief ;)

Knowing who is in charge of what service is something that should be solved by a service registry, not by a flaky informal gossip network. If people believe they can't solve this problem without overhearing things in the office, they are probably spending their social capital on the wrong things.

What makes it particularly irksome is how Amazon fashions itself as a data driven company, requiring reams of evidence from employees to get promoted - but that doesn't apply to the S-team and their decisions, because they are obviously just applying the "Are right a lot" LP!

> The announcement comes after [the company observed data from] nearly three years of experimenting with different models of hybrid and fully remote work

Data which they haven't shared. Don't you think that if it was convincing, they would gleefully publish it?

What makes you think they don't have data? It's just data they can't share because it's not morally palatable.

Does anyone actually buy data driven? Seems like a great methodology to adopt if you need to absolve yourself of responsibility for your failures. Most questions people try to apply “data driven” to are so ridiculously complex I’d need to see some incredible methodology and ground breaking understanding of human behavior to put any faith in them. There’s just too many unknowns and confounds.

Data driven is great when you’re monitoring computer performance, but that’s a domain humans have nearly built from the ground up. And even then it can still be very hard to utilize that data. Trying to apply the same to systems we barely understand seems fraught with error.

Is there a company that doesn’t fashion itself as “data driven” these days? The term lost all meaning a long time ago.

Yeah I'm expecting another round of layoffs from Amazon soon.

In my experience WfH is perfectly fine for the old hands, their productivity goes up due to lack of commute. However, it's absolutely awful for onboarding juniors and knowledge transfer.

Normal churn hits you MUCH harder if you are remote and even if you have a good documentation culture (hah, that's a tiny minority) there's a lot of intangible knowledge that doesn't get written down.

So you tell your juniors and seniors to communicate. I'm spending some time every week talking to either juniors or peers about things they're learning. If your seniors don't initiate teaching and your juniors don't ask enough questions... That's not an issue with a remote system.

"Tell them to communicate" is much less effective than having an environment that makes it natural and encourages it. E.g. the reason way more people are fat now is not because being slim isn't possible or because willpower has gone down, its because the environment is different. You can just tell people to eat less or you can make eating healthily and less easy.

Given the state of the fresh grads and the company culture in my previous company, wfh is much worse onboarding.

"An environment that makes it natural and encourages it."

In 24 years of IT, I've never seen this unicorn environment.

My previous company managed it with the expectation that people take regular breaks and pick someone to take that break with.

I think you're probably right, although I can't shake the fact that I picked up a lot of this stuff myself or through online interactions (e.g., IRC). Given how much more software holds your hands these days I'd expect a lot of juniors bent on becoming seniors to do quite well.

I can't believe people are really on here thinking that zero data was considered and it's just a bunch of execs going off random "feels".

It must have been sh1t data then if you can use that data to impose the same policy on all people.

And yes, I can believe that. With few exceptions, the execs with power are focused on sales. They couldn't give a sh!t about WFH. In fact, the reason this is probably an issue at all is because one too many of these same execs got p!ssed that they had no one to turn to to help change anemail setting or a printer cartridge and they had no idea how to do it. Not knowing sh!t gets old real quick (translation: "inefficient working practice which we need to do something about").

But I'm not bitter.

If they have data, share it, in the absence of said data we can only assume it's poppycock. Exec teams do that all the time, work on feel, why should we believe it's any different here?

Of course data was used, but you may have the order of operations wrong; execs often make the decision before the data.