After going back and forth between in-person and remote jobs, my conclusion:
* Remote work is much better if you know what you're building and "just" need to execute.
* In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
Obvious, right? But I think what I've realize is the dimension of "we know what we need to do" is actually pretty much orthogonal to the product, size of the company, tenure of the employees, etc. I've been in early stage startups with no PMF that nonetheless have strong product-centric leadership who can set down an unambiguous vision for what they believe needs to be done. And I've been in startups that have PMF yet have no vision, and everyone is standing around in their remote offices twiddling their thumbs on "what to build next".
For large companies like FAANG, I actually think they mostly fall in bucket 2. I worked at Meta remotely for a year. My instructions when I joined were something like "Yeah so just talk to a bunch of people and look for opportunities to contribute". WTF? No wonder remote is not working well for them. You had to go 5-6 layers up the reporting chain to find anyone with any sort of holistic sense of what needed to be done, and they were not exactly empowered to share that vision lest it conflict with someone elses.
TL;DR I think remote work is well set up for companies with leadership that resembles a benevolent dictatorship. I think if you are all-remote and your leadership is effectively "managerial" in nature, NGMI.
> In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
true.
> For large companies like FAANG, I actually think they mostly fall in bucket 2
Also true most fall in the "figure it out" bucket... But here's where it gets funny: just meet in-person you say?
No, this is FAANG. You're in MTV, but the people you need to collaborate are in NYC, LAX, SEA, AUS, SFO...
So your in-person experience is you booking a room just for yourself so you can video conference all those people who also booked a room for themselves.
Because every other team is in same situation, finding a meeting room is a challenge in itself.
Take the call from your desk then?
No, this is FAANG, every building is open office layout, it'd be rude & disrespectful to your colleagues around who are busy coding.
A lot of companies want the benefits of distributed teams, but also want to force everyone to come to offices and do conference calls in meeting rooms instead of work remote.
It's mind boggling to me. All I can think is that there is a lot of ego tied up in offices for some reason.
Working for a FAANG right now and since RTO was mandated, I noticed that our team have been spending ~5 mins before the beginning of the meetings finding a room and setting up the audio/visual system for conference calls with other teammates across the country. This is for a company that has a lot of office space in its birth town and has HQs in a few other cities. A couple of teammates (out of 8 total) privately told me that they hate the RTO.
I personally prefer fully remote because I hate the distraction in the open office plan and associated pains like having to find meeting rooms, having to share restrooms with others (I admit this is just my personal pet peeve), and not being able to avoid distraction when I am in the zone for working. Most of all, I don't like the commute (I have to commute 1 hour each way, so everyday, I am spending 2 hours; when I worked from home, I started early and end up working an extra hour or two because I don't have to commute; these days, I try to arrive at 8-9am and leave by 5-6pm to beat the traffic and once I get home, I don't do any work). I consider quitting someday soon and am waiting for my wife to complete her training. Once she has done that, we will move to a place where she can find work, and I will quit my FAANG job for a fully-remote one (sure, my income will take a small hit, but quality of work life is more important for me).
If I went into "my" office I probably wouldn't recognize a single person at this point and I'm not sure there's anyone who I regularly work with who is assigned to that office, much less goes in.
>No, this is FAANG, every building is open office layout, it'd be rude & disrespectful to your colleagues around who are busy coding.
Individual offices have not been the norm for everyone at large companies maybe ever? Cubicles are "better" but not that much better. (Source: I had one for years--both full-height and then shorter.) And the fact is that lots used to spend a lot more time on the phone than they generally need to today given other communication channels. When I've been in an office I absolutely make calls from my desk and everyone else does too.
Sure this happens but it’s not like it’s everyone. For example right now 80% of the people I collaborate with are in the same office as me. So it’s still noticeably better to come in the office.
The downside is it’s hard to get actual coding down because of all the “collaboration”. So my ideal would be 1-2 days in office to meet and then 3-4 days to code ins peace at home.
> No, this is FAANG. You're in MTV, but the people you need to collaborate are in NYC, LAX, SEA, AUS, SFO...
I don't think this is really that common. I spent a decade in MTV on a wide variety of teams. I had VCs with people in different locations a handful of times and I never had one by myself. And getting deep focus with those people was so valuable that it was easier to just go to NYC, MUC, or SYD once a quarter than it was to jump on a call.
Even FAANGs do not apply perfect hashing to distribute teams across their office locations (well, for all I know, Google does, but the one I work at doesn‘t). And often it already helps if some of the team is co-located.
My best experiences working remotely—both during the pandemic and with folks in other locations before that—has been with research-oriented work. The reason? We had the time and space to work and communicate the way that made sense to us. We had plenty of shared pair/group sessions to work out design ideas and experiments, iterate on the system, debug trickier issues... etc. Everything else we could do well asynchronously. Since this was longer-term, more speculative work, we didn't have to worry about tickets or product/management breathing down our necks every day.
On the other hand, the projects that suffered the most were the ones where folks had less scope and autonomy. If you can't unilaterally make decisions for anything that spans more than a day, and you're forced into some ticket-oriented process where everyone has "their" task to do, communication naturally suffers. If you're all in the office, there's a bit more slack to talk quickly and informally outside your immediate silo; remotely, this takes more intentionality, so the friction imposed by your culture and process has an outsize effect.
So my conclusion has been, pretty much, the opposite: remote works better when people have real scope, independence and autonomy. If engineers and small teams can make decisions over longer time-frames, they'll have the space to adapt and work effectively. They might need somebody to set the example or foster the right habits, but it's fundamentally healthy and effective. The frictions with remote work come up when it's thrust against a less flexible culture that makes it harder for folks to adapt.
I think you're using different phrasing to describe the same effect. Autonomy and the org knowing what it wants to do/needs to do go hand in hand. In your long-term speculative research oriented work, you had autonomy because you knew what you were working on - not what it was in terms of tickets per se, but it sounds like you knew your scope and your objectives.
At 10+ person org sizes, most people can't have long-term scope and autonomy if the management don't know what it wants/needs, since they're just going to change their mind in a week or two. Once those decisions are made you can start delegating responsibility for execution.
The easy thing to do is to blame this all on poor management, but I think that's too pat. I think figuring out the best way to use a team of 20, 100, 1000 engineers to accomplish business goals is just an incredibly hard problem... and one that's often an ongoing, Sisyphean task, where high-bandwidth, low-latency communications are incredibly useful.
I think you articulated a sentiment a lot of people have very well. For the non-decision makers, there is a strange dichotomy I see with coding.
Remotely, I’m vastly more productive in getting coding work done. I’m sure most people would agree. But that productivity is almost erased by onboarding new hires. There are so many aspects of software engineering where it’s so much easier to be sitting next to someone, pointing at their monitor, and discussing what and why you are doing something. Instead, you need to hop on a call and awkwardly screen share on a call or write out paragraphs of text to impart the knowledge. In my experience, having such barriers to doing these things makes new hires much less likely to reach out and put themselves into the situations where they can learn, no matter how much they are told it’s ok to bother people.
To me, this is why remote works so much better at smaller companies with less bespoke tooling and legacy code. There’s no better way to learn than in person and there is a huge productivity boost when doing so. That said, personally, I’m willing to deal with the lower productivity in my work life to get the higher productivity in my real life by being able to do laundry during the day and not sit in traffic.
> ’s so much easier to be sitting next to someone, pointing at their monitor, and discussing what and why you are doing something. Instead, you need to hop on a call and awkwardly screen share on a call or write out paragraphs of text to impart the knowledge
For me, it's the exact opposite.
Before the pandemic, I always found the practice (and even the idea!) of "pairing" quite off-putting. You're standing behind someone, breathing in their body odor, awkwardly sweating next to each other, bending over if you want to type something... a really weird experience.
Remotely, none of those apply! You each get your own keyboard, your own screen, your own mouse. Both of you are 100% comfortable in your own environment. Using decent software helps as well, I recommend Tuple which makes sharing the mouse & keyboard seamless.
I am at the point where remote works better for me for a lot of what are ostensibly team / collaborative activities.
For example, incident handling has been lot more efficient for us with the team looking at different aspects and sharing on a call/huddle than what we often used to naturally do (involving a gaggle forming around 1 person's computer with different people offering suggestions of varying usefulness).
Demos and knowledge sharing I find also work better remote now than they used to in person now that people are used to the tooling.
I slightly prefer onboarding remote also, but that is because I'm the kind of person who gets distracted by wondering if I'm sitting too close or maybe I have not enough or too much deodorant etc.
The only thing that suffers remotely for me at this point is engagement. There's something about physically being with other people that makes their problems and desires seem more important to me, and I don't have any good hacks for giving more of a &"@$.
> Remotely, I’m vastly more productive in getting coding work done. I’m sure most people would agree.
I certainly don't agree at all. Home is where all the distractions are. Home is where i can drift off to make a cup of tea and end up spending an hour reading the newspaper without anyone noticing.
I think it would be useful to understand the distribution of peoples' feelings and experience about this. I have seen so many remote work enthusiasts just taking it as read that everyone prefers remote work because they do.
Yes, I am going through this right now as a new hire onboarding to a boutique codebase. As soon as you are 100% reliant on internal docs, you realize just how large the gap is between them and mainstream OSS libraries.
I think a decent middle-ground between "Document everything" and "Organically relay all information" is to just write "doc-stubs". Just a list of everything out there with no real content. It gives people a jumping off point to ask questions and lets them know what they don't know.
> t’s so much easier to be sitting next to someone, pointing at their monitor, and discussing what and why you are doing something
This feels like a tech choice issue to me. I'm constantly on calls with either slack/pop sharing where anyone can draw on the shared screen, or shared coding session like in vscode where you can see each other's cursors. It... works the same for me as pointing at the screen. Just without it being awkward when it's more than 2 people getting to cram into a a small space.
I don’t think in-person is better if you don’t know what you need to build. I’ve experienced teams where people were totally without direction despite being in-person. I think in-person just makes it easier to look like you are busy when you have no direction, and remote lays the fact bare.
Agree that though it's better in person during brainstorm / discussion, it isn't decisive. When people are having zero ground during the meeting, it won't provide any meaningful result no matter what you do. And if it resulting in something, usually it'll be bad.
What I found to be the best approach is one or two people already established a concept at what to make or at least the image of it, and the discussion can move to the details and implementation plan. And that doesn't need to be in person
In-person is much better if you don't know what you need to build..
Can you explain how some of the greatest technical research projects of our time like the internet, Linux, etc happened without a big office then?
It's clearly nonsense to suggest you have to be collaborating in-person to build big unknown things. There are thousands of examples throughout open source and more where that hasn't been the case, and millions of examples of projects from company offices that are awful.
The physical geography of the team is not an indicator of success.
Is your average company made of some of the most brilliant and driven and experienced people in the world in their fields? Do they have no actual business goals but a wide open field to just do something big and different and impactful? If so then literally nothing else matters except not getting in their way. But there's very few cases like this.
while linux or git was created relatively sololy by a single person (at least at the beginning), i dont think you can call the internet a solo production.
> * In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
My anecdotal evidence is different.
I've spent the last 2 years building a large, complex, and ambiguous system working almost exclusively remote, and half of it from another continent. For the first year I never met a colleague in person.
However, all this time I have been working closely with another engineer. Both of us are senior and very tenured, got good rapport and were highly aligned and motivated. We were able to execute efficiently, sometimes spending hours huddling in Slack per day.
So my feeling is that remote works in some scenarios and not others, particularly involving new or junior engineers.
1. You do not have home problems that you will end up focusing on when you are there - so if you are single and do not own your home, or you have a great relationship and you do own your home but there are no problems the remote work is probably better. But just like being in an office and going into a private room can help productivity if you have problems with home life going to office to work can help with productivity.
2. Remote work is much better if you are a person who is motivated by love of what you are doing and you actually love what you are doing, if you start to not love what you are doing but you have not left the company yet etc. there can be some slight loss of productivity, not because intentionally goofing off so much but because like #1 above working remotely makes it harder to get focused on something one does for money and not love if you are a love focused worker.
Why does it have to be because some want to go in an office so we all do?
If you want to work in an office go work in one. Join a co-working space or if there is sufficient momentum have everyone who wants an office join the same one in your area.
But don’t force it on everyone. Let people choose. That’s the thing I hate about these conversations is it should just be personal choice for the individual. Make async practices and communication better and re-tool the culture if need be, but remote first thinking benefits office workers too.
During the Covid lockdowns and everybody was forced to work from home, the single people in the team seemed to be the most depressed about the situation. Some of them were really eager to get to the office as soon as possible. The people with families were stressed if they suddenly had to homeschool their kids, but otherwise they were fine.
Of course if your spouse also works from home, it's important to have a home with more than a single room. Not everybody has that luxury.
This always gets to be such a dogmatic discussion that will to someone claiming anyone who likes working in an office has no social life.
It’s just a preference. Remote work is not overall objectively better than office work (but you can create a set of goals for which this is true). Office work is not the right choice (but you can create a set of goals for which this is true).
#1 can be solved by renting a co-working space if troubles at home are really that severe. I doubt that #2 is really any different than if you are unmotivated while in the office. If anything, one unmotivated employee in the office can more easily spread that lack of motivation to their co-workers.
> You do not have home problems that you will end up focusing on when you are there
Not sure why this needs to be repeatedly brought up.. I mean it is called "home office" for a reason. If you don't have a real one, remote can suffer.. no surprise.
> * Remote work is much better if you know what you're building and "just" need to execute.
> * In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
This is why the other day I ranted about how my actual preference is in office but having a door that I can close. Realistically we shift between these two modes. In the first mode I need to reduce distractions and isolate myself. This is hard with slack and hard with an open office or cubicles. In person let's me even ignore slack as someone can knock on my door if it is urgent! But that's not all the time, so I open the door and communicate that it is collaboration and discussion phase.
I don't know why companies have become so allergic to doors. They are amazing and can let employees be in control of the chaos of work schedules. They'll know far better than anyone else if they're more productive that day/hour/minute in isolation or in collaboration. Micromanaging doesn't help, a door does.
> In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
In-person has benefits in those situations, but whether or not it's better varies a lot based on the person and the office. I find it VERY distracting to have people talking around me. As such, I get a lot less done when in an office, because I can't concentrate as well. When you add in the other negative (commute time, in person interruptions, etc), going into an office for work can far outweigh any benefits of being around the people you work with.
I suspect what your describing is less the cause but a symptom of the real case.
In my experience(i have worked from branch offices most my career) the differentiating factor is literacy, ie how comfortable/confident are the organization with recorded text vs private spoken word and the more the organization writes the more effective the remote staff will be.
And one of the side effect of a written culture is that a lot of the informal conventions/conversations become formal structure so the same level of top down planning can look very differently based on weather the culture for disseminating it is literate or oral.
> implicit fast feedback loop of standing next to each other.
I have a hard time grasping this one. Is it brainstorming ?
For things like pair programming or specification of a product, having separate screens and independent access to the same documents makes things a lot faster and efficient.
Same way meetings tend to go smoother all participants can access whatever data they need independently from the presenter, and move through the docs at their own pace.
Basically a mix of synchronous communication and access to all needed resources seems to me to perform better than IRL binding of two or more people looking at each other.
I wouldn't call it brainstorming. More like "alignment". Fleshing out the vision you have for what you need to do, and putting that up against your coworkers vision, and figuring out where the disagreements/gaps are. You can do this quickly if you have an experienced and disciplined PM guiding the discussion, but most people don't, so they just have long conversations. Which works fine IRL, but it's very uncomfortable to be on a zoom call for more than about 45 minutes at a time.
My experience leading an older startup is that category two is hard to escape whether or not you’re working in person. It has more to do with the “Innovator’s Dilemma” - once a company has found product market fit and is generating profits, everything is fine-tuned for that purpose. The company’s reason for existing is to generate profits from that one thing that really worked out.
Some companies are able to add great new products over time, but I sense that it’s more common for companies to have one great play, after which they either die or merge with another to enjoy better economies of scale.
If you’re big enough and rich enough, you tend to burn a ton of money on internal research groups, but very little of what gets researched is ever a successful product. Doing this as a smaller organization is nearly impossible, which is why I think so many startups just exit rather than finding entirely new things to build.
How often are entire teams at the same building ? Like the other person mentioned - you're going to meet in conference rooms with video - this was norm pre COVID.
I would say biggest benefits of in office is spontanious meet/discussion, assisting less senior coworkers etc. If I have a heavy day of work to plow through alone I avoid the office. If I need to sync or work on something with others in-person beats all the tools.
Your experience matches mine, where yeah, it could work even in situations where it commonly fails if there was "just" better management... but... well, yeah.
EDIT: and so many of the challenges people are running into now are not new, either - companies were incentivized to make remote work work for two decades pre-Covid, and for a while pundits were all convinced that software jobs were all going away overseas as a result, but despite the salary savings the ROI just wasn't there for all but the most basic of work.
I don't think it's true. This whole office/in office/rto has to do with people being dishonest about why the rto needs to happen and being stuck romanticising the good old days.
There is huge overhead to in person work and the benefits are wildly oversold.
IMHO a professional software developer in a company with good management/clear direction will thrive in a remote work scenario.
Not to mention that even in-person workplaces (especially at large companies) still involve plenty of remote interaction due to the global nature of the workforce. Plenty of Googlers being required to be in-office will actually be working alone in-office, with their teammates far away...
If this is true then there is no need to worry: companies with thriving WFH developers will surely overcome the stuck in the old ways ones. Not only their developers have same or greater performance due to constant thriving but their expenses will be much less so they will be able to offer better compensation and steal the talent from the in-office dinosaurs.
The only other place where I encountered such a "system" was in one (small) company, where the CEO's dress code did not include pants and I wish I never went there in the first place.
> "Yeah so just talk to a bunch of people and look for opportunities to contribute."
Hard relate. Had my first internship WFH at Big Tech during the pandemic. I had expected work to be more on the lines of: there's always a lot of work to be done, and people will let you know about that. Instead it was the opposite, exactly as quoted above.
While it meant I had more freedom, I would have preferred if someone just doled out concrete tasks to me. I felt a bit lost. But then again, grass is greener on the other side.
> You had to go 5-6 layers up the reporting chain to find anyone with any sort of holistic sense of what needed to be done, and they were not exactly empowered to share that vision lest it conflict with someone elses.
OTOH, from personal experience working on-site hasn't really ever helped particularly well in that scenario (which I'mm all too familiar with.)
I used to work for Amzn and the management was totally idiotic about remote work. We had to go to the office to be on call for datacenters 1000kms from the office. Makes perfect sense right?
In person makes no sense for any FAANG, because they are geographically distributed. All you do is come into the office to sit on video calls all day with people in other locations, unless you're at a very low level and your influence is very limited to just your team.
I see where you’re going with this but pretty much all L3s and L4s mostly deal with just their team. That’s more than 50% of the company if I’m not mistaken.
I'm curious if you've considered that there is a pretty major cost difference between the two options? (at least for the company) Would you conclude that In Person is much better , in excess of the additional costs ?
The cost difference is probably rather small. CoWorking spaces in large cities charge something like $500/desk/month which we can assume is a reasonable cost of office space. That comes out to $6k/year/employee. Google's median employee pay is around $300k so we're talking a 2% difference in cost.
I think that's only one of many models where remote work works out. The other that I've seen function well across many companies is remote work where team members are highly skilled and have a lot of freedom in choosing what to work on. This essentially lets every domain expert identify what the biggest pain points are in their field and then figure those out. You don't need top down dictatorship for this to work, but you do need a clear vision of where you want to be.
> In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
Or sharing a zoom session. Bonus is that I don’t have to commute to an office on another continent to engage in this second modality - more than half of my teammates are in other cities and I don’t see a benefit in the standing next to each other.
Tech pays Industrial Psychologists, I don't know why we don't put them to work on this.
I restarted my software career doing remote work. I didn't have a problem finding things to do because my responsibilities were clear (read: I had a good manager) and I was self-taught so I already had a muscle for seeking knowledge on my own.
I suspect that remote work success is highly influenced by an equation involving how good your manager is and how much you innately know about what things your role is responsible for which can be propped up or replaced by seniority. I also suspect that new grads don't have muscles for finding/seeking internal and external information on their own, which feels powerless. There's ways to solve all these things without treating all of these folks the same.
Frankly, I think these are all the problems you and I relate to, but what's really solved for the business by returning to an office isn't productivity or happiness. It's a real estate problem.
>I suspect that remote work success is highly influenced by an equation involving how good your manager is
I suspect the people on here who like to hate on managers either have a bad one or just don't realize how much crap they're being shielded from--a lot of which is inevitable in a large organization (where you can't tell everyone "hey figure it out on your own").
> My instructions when I joined were something like "Yeah so just talk to a bunch of people and look for opportunities to contribute". WTF? No wonder remote is not working well for them. You had to go 5-6 layers up the reporting chain to find anyone with any sort of holistic sense of what needed to be done, and they were not exactly empowered to share that vision lest it conflict with someone elses.
That just sounds like poor leadership to me. If you've got a holistic sense of what needs to be done, please share it with the people who need to be doing it.
That, or cut it up into concrete chunks and tell people to build that. But telling people to "figure it out" and then not sharing the overall vision, sounds incredibly dysfunctional to me.
having worked for a fully remote company that was successfully acquired for a decent sum by a much larger respectable company, remote depends on the top levels of management communicating the vision regularly and being transparent about direction, changes in direction, and reasons for changes. add in someone skilled whose responsibility is for facilitating remote culture and you have a recipe for success.
the problem is there is a ton of terrible leadership out there who want to rule by fiat, ignore their employees, and don't trust their employees (who they either directly or indirectly hired) to so their jobs.
People who have been in the industry and with a company for a long time largely have a network and don't really have problems connecting with people. (And at larger companies that network is often pretty distributed anyway.)
But the--admittedly not universal--consensus is that new folks struggle with onboarding and generally making progress in their careers. And even coming into an office doesn't really help if their managers and more senior co-workers are mostly not there. It's certainly hard for me to imagine graduating school and working remotely.
I graduated from uni and all my jobs were 100% remote, and I don't really feel I had any serious problems in onboarding or advancing in my career at least yet :)
I'm not gonna pretend that every project had 100% coverage and up-to-date documentation, but you should overcome fear of appearing weak/dumb and just ask a question, not even waiting for the next daily meetup. If culture is good enough, people are gonna realize it isn't you dumb, but docs and processes suck ass and will do something about it, maybe even asking you to participate. If not, just leave when the next offer lines up, I did that once when people straight up refused answering to my questions directly, and instead nagging that I didn't ask them properly.
Building up network is obviously more reliable when meeting in person, but it doesn't defeat the idea of getting to know each other through daily interactions, retros, 1-1s, whatever. And moreover if you appear professional and capable in eyes of colleagues by getting shit done, they're would want at least keep in touch
A lot of that struggle has to do with what others point out already: the onboarding processes and documentation being of shoddy quality and relying on word-of-mouth to transfer knowledge and repeatedly check in.
If you start raising future generations to be, you know, adults, instead of mentally stunting them into being eternal teens well into their 30s, and companies make an effort not to have a horrible onboarding and rely on 3 year tenures (and paradoxically do nothing to keep said employees that long), the problem is far less prevalent. Heck, most CS students are expected to do far more complex things at school without any assistance these days.
I was a Data Scientist. Personally I found the autonomy immensely frustrating. Combined with the stack ranking performance review system, it was effectively a way of pushing the manager's job on to the employee and forcing them to hold themselves accountable for their own priorities. Makes sense in theory, but once you add huge organization churn + being remote, it was difficult to get a sense of what priorities were exactly, or would be next month. It made me feel insecure at all times that perhaps I was working on the wrong thing and would be punished for it.
I transferred within my company to the much bigger company that bought us out and this is basically what my boss told me. Actually his exact words were "you need to create your own job." I hated it. Give me a freaking task, don't make me network and sell myself (remotely of course, no one I worked with was anywhere near me) just for something to do.
I created my own job, just with a startup. For a year I had two ostensibly full time jobs. By the end the startup wasn't doing much either and I considered going for three full time jobs.
Currently I’m a GTPM at Meta and this is the expectation we have for onboarding to a new team unless you’re early in career. I’ve seen this be the case for SWE and other roles in the technical domain. Typically you have an expertise area and at least on my team a starter project or 2 with a tangible deliverable to get you going on something while you ramp and meet others and figure things out. Meta is a big place though so could be other expectations in other organizations.
It’s less a job title (I’m a “software engineer” like everyone else) but the scope of the problem your role is aiming to solve. Look for places that 1. have an unsolved hard problem with 2. no real direction on where to go with it and 3. buy-in from someone to get it solved.
> In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
I'm curious how often anyone actually finds this situation to happen and improve the feedback loop. I feel like it's more of an illusion of productivity then actual productivity because it's a pile of constant interruptions.
Like the given example with Meta doesn't seem like it would be better in-person, it sounds broken regardless.
I have the same conclusions as well however, in-person collaboration can be done remotely but it requires less professionalism and more trust. You have to allow for unforeseen moments and allow each participant to have their time to weigh in. Take breaks. It’s a 2.5x time slog vs in-person but it can be done with whiteboard tools and open honest communication. It’s the ones that expect excellence that struggle with remote’s imperfections.
With your Meta experience I am curious - would the people you had to talk to been in the office near you anyways, even if everyone was in office? It seems like with a huge global company the people who would be hard to get context from (so not the people you might have a video standup with daily or weekly) might be in a different office or building requiring a video call anyways
Not everyone, but there would have been enough people around to have dramatically improved my experience. The problem there was Meta does not have a strong culture of documentation, and I didn't know what I didn't know. It was common to spend 2-3 days working on something only to find out on a random zoom call there was an entire team working on the same thing. Discoverability didn't exist. I found the environment untenable for doing anything of value.
> * In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
Beyond the practical scope of this point, we also find the need for mentoring (building a sustainable team).
This is a good insight, but the interesting counter-weight to the reality you describe is that many people who want "remote" and the independence of work, don't really want the same "benevolent dictatorship" either.
The first part of this is patently untrue. The most successful startup I worked at (defined as having IPO’d, and having their products effectively everywhere) was fully remote. This was very much 0-1, 1-10 type work. The only requirement to ever physically be in a given place was when visiting a customer.
The FAANG I work for now similarly has lots of 0-1, 1-10 type work (at least for my team), and a fully remote team. It works fine too.
The part about more junior employees is _more_ valid, though we had no problem on boarding juniors in either of those two situations.
Ultimately whether remote work is successful depends on whether the people doing the work have sufficient self-discipline to get the job done, and whether the hypothetical office alternative is set up for success. Most offices are not remotely set up for success, but optimize instead for middle management (toxin or cancer? [1]) being able to view their fifedoms.
My wife worked for several years for a company with only about a half-dozen employees. It was full remote.
She left there in 2011, after rising through the ranks to the level of division director.
They did almost all their business over email and Skype, because at the time, Skype was still the best voice/video conferencing software available without dedicated hardware.
If it was possible then for a small group of non-tech-focused people, it's certainly possible now, over a decade later and with an entire pandemic forcing the rapid improvement of all sorts of remote-work technology. That doesn't mean it's for everyone, but boldly stating that it's a "non-starter" is just unsupportable.
Big companies are also ridiculously overstaffed, as Elon Musk proved empirically. And many of them have only one thing that works and ambitions of doing more than will never work out. So there's a lot of wasted talent, in other words, but this ability to waste talent necessitates in person communication, just to make sure everyone is stack ranked according to the current arbitrary and probably mindless goals.
Quick search on this thread didn't reveal the word "flame" (as in "flame war"), so I can´t resist: That's what the remote vs on-site debate is. About as interesting and ultimately inconclusive as Emacs vs Vi. Neither is universally better. It depends, _primarily_ on the people involved and their preferences. My advice would be to try and work with people that are on the same page about this with you - if possible. If everybody can muster the level of acceptance and respect for their colleagues required to make a hybrid environment work, that'd also do the trick.
What I find fascinating about the remote vs on-site flame war are the _stakes_ though, I don't think we see anything like it in dynamic vs static typing and all the others.
If remote loses, remote workers have to figure out how to relocate their entire families to a place they don't really want to live in, or accept severely limited job opportunities - on top of having to work in an environment they might not be productive or happy in.
If on-site loses, people who are fetching large salaries and bought a house in an area with insane real estate prices face a future where their house value shoots lower than the credit they took to buy it (more towards the national average), and where it's increasingly difficult for them to prove that they earn that large salary they're getting. That's on top of working in an environment they might not be productive or happy in.
Both sides have a lot to lose if it all goes either one way or another, that's unlike any other flame war I can think of. I wonder how much of this is at play at Google. Certainly seems to be escalating if someone feels the need to pull a stunt like that (with the performance reviews).
> Quick search on this thread didn't reveal the word "flame" (as in "flame war"), so I can´t resist: That's what the remote vs on-site debate is. About as interesting and ultimately inconclusive as Emacs vs Vi. Neither is universally better. It depends, _primarily_ on the people involved and their preferences
You nailed it. Ideally, both employees and employers should have the freedom to choose whether to work onsite or remotely, based on their preferences (again, this is subjective and not scientific). I believe the market could self-regulate in such a way. For instance, if the demand for remote work exceeds the supply, a savvy employer with a fit for remote work could take advantage of it by attracting better/more/cheaper talent through offering remote positions.
However, what needs to be debated and critizised is when employers change the rules midway through a game. It's unfair to hire people for a "remote role" and then force them to work in the office. This creates real-world hardships for individuals, such as the need to relocate.
Using your analogy, it would be like initially giving your future employees the choice between Emacs and Vim, but later giving Vim users extra points in performance reviews.
> both employees and employers should have the freedom to choose whether to work onsite or remotely, based on their preferences
Not that I'm antsy to get back to the office, but that scenario is not an "everyone wins" scenario, it's just another version of remote work. Onsite is only onsite if everyone is onsite. If it's some people in the office and some people at home, the people in the office are just working remotely from the office. All your meetings will be virtual, you don't get the benefits of onsite like being able to gauge someone's availability by just looking at them, you're gonna be screen sharing instead of just pulling up a chair next to someone, etc etc etc.
> Using your analogy, it would be like initially giving your future employees the choice between Emacs and Vim, but later giving Vim users extra points in performance reviews.
That analogy doesn't quite apply to this Google situation.
Google has a distinction between "full remote" and "hybrid" employees. Attendance will NOT be considered in the performance reviews of full remote employees.
For the politicians and influencers, however, on-site work is vital (especially if they are good looking or charismatic). They need meetings to shine optically and verbally and dominate the real workers.
This parasitical category of people usually has the most influence, so the move to return to on-site work is not surprising.
I know what you mean, but this kind of negative spin probably doesn't help the situation :D
Some people are better at creating (real or fake) value face to face. I've seen what you describe in remote/distributed setting as well, in internal message boards for example. Some folks are shy and/or incoherent in person and master manipulators in writing.
I’m a good looking and charismatic programmer, one who understands we evolved as a species solving problems with other humanoids on the planes, and it makes perfect sense to me that a hybrid work situation is the optimal way for most teams to solve problems most of the time. I’m sorry you feel dominated, consider developing the non programmer side of your spirit.
It might be useful to "all be in the room" when brainstorming. Once the coding starts, this is just a set of distractions I don't need. I suspect many/most of the developers reading HN want peace and quiet far more than the chance of magical hallway conversations.
And it's only an anecdote, but we debugged a problem today where people were in two provinces of Canada and the UK. Were perfectly able to share screens, run test calculations and "interact" using existing technology. Companies either trust their staff or they don't and the staff should act accordingly.
Fully remote with 2-3 day quarterly in-person meet ups would provide tech companies with the in person collaboration they seek and be cheaper. Remote first companies do this all the time. Employees feel way better about it if the company provides hotel and accommodations when doing destination meet ups.
I haven’t found in-person meetups to be that useful. In my experience they only really started happening in late 2021/early 2022 due to the pandemic, and since then they’ve been severely curtailed due to the market downturn. So when you do meet up it’s more about finally seeing people in person, team building, and restating your team’s goals and mission. Any brainstorming I’ve seen happens in over-long sessions with 30+ people in attendance, where a ton of ideas are generated but there’s no actual momentum once everyone gets on the plane back home. The real collaboration happens in the interstitial times- lunch, breaks, and so on. Which again points to the importance of having at least part of the team co-located.
Full disclosure, this is at a company that is hybrid and not remote-first. But I’ve seen the pattern repeated enough times across various teams that I don’t think my experience is unique. And I say this as someone who personally prefers to work remotely- I’m just not unaware of what’s lost in that environment.
We used to do this in a startup I worked at. Was a lot cheaper to book a big AirBnB for the meet-ups than to have a fixed office space. And we could pick interesting meet-up locations.
> Fully remote with 2-3 day quarterly in-person meet ups would provide tech companies with the in person collaboration they seek and be cheaper
I am a fan of hybrid (people can choose whether or not they're in the office).
However, you want to be able to collaborate effectively at any time and not have to wait for quarterly gatherings. I've got complex problems to solve together with my team mates!
> It might be useful to "all be in the room" when brainstorming. Once the coding starts, this is just a set of distractions I don't need.
This is not the dichotomy you make it look. There are lots of other tasks to do, and for those I personally would often go in a cafeteria or public space, since I didn't mind the distractions -- reading code, doing code reviews, reading docs, writing docs, responding to emails. "Coding" was a relatively small part of my day as a SWE at [FAANG].
The problem is just that there are some people who are able to concentrate while lots of other people are having lots of non-related conversations around them in a cafeteria (sounds like this might be you), while others can't. And the latter will prefer home office to a cafeteria or even to the usual open-plan office...
I find debugging to be easier when everyone has their own machine. You can screen share the primary person's screen but then have others try out little things and do exploration on the side. And I'm far better with my desktop setup than I would be on a laptop that might not even have a mouse.
> It might be useful to "all be in the room" when brainstorming.
Really? what does it bring that you can't do remotely? I can hardly think of a problem where "being in the same room" is a requirement to have a productive discussion
I'm pro work from anywhere so long as shit gets done but...
Picking up on social cues. Bouncing ideas around using a whiteboard. Speaking quickly without that ever so slight lag on Zoom which means you struggle to get a word in without potentially talking over somebody. Having food or drinks before or afterwards to unwind and build relationships and kick things around informally.
+1 every time we have these in-person sprints they are dominated by the managerial types trying to make the engineers and designers engage in trite 'activities' they read about online or in some book. And even when the result of the sprint is one poorly thought out doomed to fail concept, the managerial types parade around their ability to run a sprint, the outcome of the thing not really relevant at that point.
It's not a requirement, but it helps a lot. The biggest factor might be latency, which makes it much harder to informally coordinate multiple people talking without speaking over each other.
Even for that I feel like asynchronous works better. You often get stuck on someone’s pet peeve and waste an hour when on something like a Google doc you can have 30 different threads for comments concurrently
>It might be useful to "all be in the room" when brainstorming.
Hasn't every study on brainstorming found that it's not a good way to generate good ideas? With having people breathing the same air in a small space believed to be a contributing factor to that.
Yeah high CO2 concentration is probably making people less sharp, but mostly I've heard it's about people being less creative due to consensus-driven thinking
> People in groups self-censor and anchor on each others ideas, reducing creativity.
Also can be anti-inclusive when "anchor on each other idea" is a euphemism for that guy repeating what a more shy person said, but louder, stealing credit.
With remote, usually comes a more doc-driven culture, where someone's original idea can be easier to identify and make sure they get credit.
1. Brainstorming questions, not ideas (then go answer them)
2. Soliciting responses anonymously and quietly. This avoids both hierarchical biases and "Loudest person wins" bias. And not to excessively disparage verbal processors (I am one myself). But it's important for them to realize that the constant talking can drown out others' thinking, or their sharing, or their will to live.
I’ve experienced recently people thinking they were doing some creative brainstorming and being productive by being in person collaborating in a big room, but the output ends up being basically the same solution or idea that was created using remote collaboration. And of course everyone’s exhausted at the end after a week of “collaborating” for 8-10 hours a day. It seems like it was mostly theatre.
Worked from home ~5 of my ~10 years in development.
Obviously, the "pros" of WFH are massive. No commute. Comfortable office. Food in the fridge. Daytime exercise + showers. Etc. etc.
I found however that too much of a good thing, is bad. For me, the lack of in person communication started to atrophy my enjoyment of life. I wouldn't have been able to really tell you it for the first few years, but I was slowly moving toward being mildly depressed.
Returning to office restored my excitement for my work. I credit my colleagues and the process of working toward a big goal with other people. We really underestimate how important this is for mental health.
If you're in office and you now what I mean, show some love to a fellow co-worker. You gotta admit it's nice to see each other.
The obvious counterpoint is that many people were and are miserable commuting into a workplace and then putting in the grind. It's actually humorous that most tech "social media" like HN, Reddit, Digg, and so on were much busier back when most of us were in offices, because we all would bide our days doing anything but work. You could see the bursts as the primary demographic timezones started their "work" days.
Not everyone, of course: I had a peer once who absolutely couldn't stand his family, and he would be at the office before everyone, staying until late at night. Volunteered on weekends too. Almost all of his contributions were the sort of thing that could have been easily automated, but he was happy just wiling hours away. Mileage varies and everyone's situation is unique.
Then of course there's the change element itself being a temporary boost. Turn the lights down and productivity increases. Turn them up and productivity increases. Return them to the normal and productivity increases. Having a varied life is essential. HN's history is littered by posts of people who changed some element of their life and This Changes Everything...at least during the initial period. Then it fades and they abandon it.
None of this is to diminish your personal experience that you obviously enjoy.
> It's actually humorous that most tech "social media" like HN, Reddit, Digg, and so on were much busier back when most of us were in offices, because we all would bide our days doing anything but work.
Do you have a source on this? I’m both curious and skeptical.
(And ironically, for many in Europe and Eastern NA/SA it’s the work day right now.)
>For me, the lack of in person communication started to atrophy my enjoyment of life.
I would suggest that it's possible to interact with people outside of work. It's genuinely sad to me how many of these "if I don't go into the office I have no one to interact with" posts I see.
This is so common on HN. I find it rather dystopian, as if we’ve all finally given our full selves to our corporate jobs and can’t piece our own lives together. I can feel the “but you’re at work 8 hours a day” commenter typing away as I type this, but the point is that all your social interaction doesn’t and shouldn’t come from your job regardless. That isn’t healthy.
People are often in a weird headspace at work too and a bit guarded. I don’t know why that as a social setting is inherently healthier than surrounding yourself with people outside work.
I have a circle of friends and a habit that we keep since 20+ years: every Friday morning we meet for a coffee in a physical coffee shop. Location varied along time, I lived as close as 2 minutes walk from my apartment to (now) 1 hour drive (because of the traffic).
It's not a religious thing, people miss meetings. Got other stuff, or just not being in the mood (like I drank till 2 AM, no way I'm waking up at 6 so I can be out the door at 7 so I'll be in the coffee shop at 8, stay an hour then back home). But the habit has stuck and we kept our friendship. Some people have left the group (one died), others joined it. It's easy to expand a group once you have a stable kernel: just go out for some coffees at day or drinks at night and talk to people.
I get along well with work colleagues as well, mostly I'm asking the question: if I leave this company, will I be ever talking to these guys ever again? If the answer is yes (stay in touch through WhatsApp and the occasional beer in a bar) then they're more than acquaintances and useful connections in case I need a job. Like recently got laid out (whole office closed, some 200 people fired), found a new job in less than a week. Connections, people.
So office has some value but it's far from being "the whole world".
While true, I think you are missing the time commitment and the relationships. Once the kids get home and you have dinner and soccer and karate and chores, a lot of people don't have time to squeeze in "find friends" time. Additionally, they may want to spend that limited time with their kids. Work interactions are free from a time commitment and that may be the best chance for some people to have adult interaction.
When you go into the office, you can build quite a history with people over time. I have good friends that I've worked with (off and on) for over 25 years. And we are not just "work friends".
I've also felt much a part of a "team" working in person than wfh and those connections and trust can manifest in the work too.
I think that we need to continue to acknowledge that this is a human preference. And we all have quite different history, goals and situations.
That’s not how I read it. I have a fine social life outside work, but when my wife heads out every morning and I’m alone from roughly 8-6 every weekday, I feel exactly how that comment described.
It’s not that I don’t see friends and family outside those hours, it’s just that those hours constitute like half of the awake hours of my life, and the grind sucks when I do it alone.
Also the fact that the change of pace helps in general, you can turn his anecdote around and it still sounds plausible (worked in the office for 5y and started to get slightly depressed, really appreciated starting WFH once it came, etc).
That's great that you get something out of being in the office.
I really wish the powers that be would put some data to this. You and I are complete opposites; paying exhorbitant rent, transportation, and living costs usurped much of the enjoyment I could get out of going to an office. I build infra products and in an office setting I usually get a lot of interrupts because product folks generally lack some amount of knowledge/background to fill in gaps on how to use products like mine. When I'm remote they have to follow a support queue, read docs, learn more about infra basics, and attend talks on infra. When I'm in person they can just stop by my desk.
My hypothesis is that folks who like going into an office generally fall into a couple categories:
- people who get/enjoy a substantial amount of social interaction out of work and work adjacent activities
- people who work on public facing products who don't have many internal customers
- people with disruptive home environments
The thing that upsets me is that we're being treated the same and that most of this is to save face for their real estate investment. Not to mention all the half-truths/gaslighting about the universalness of productivity.
After trying both for years I don't really know what we're talking about here.
I can go to an office where we can schedule room to have a video call to sit around and discuss what tickets we're going to create and who will work on them. Those tickets will get worked on and tracked then we end up creating a product or feature.
I can stay at home to join the video call to sit around and discuss what tickets we're going to create and who will work on them. Those tickets will get worked on and tracked then we end up creating a product or feature.
I don't know what this mysterious in-office inspiration cycle is. It was never polite to just wander up to people working on something and bother them with random junk.
Offices exist for companies where management has sufficiently removed the ability of the people they pay to make decisions. Now you need to come to them and ask to be allowed to do things, so your manager can ponder whether that thing makes sense in whatever weird inter-department power struggle they are working in.
> It was never polite to just wander up to people working on something and bother them with random junk.
A lot of 'management' doesn't build communication or workflows more advanced than that. And when people are never more than a short walk away, they don't have to.
Slack is just the online version of that. A noisy conversation always within earshot. One that you can usually ignore, but should still be a little aware of what's going on overall, and occasionally you get pulled in to the larger conversation.
While I enjoy this whole debate in HN about remote Vs in-person, the point here majority seems to be missing is that the sole attendance is used as a threat for performance review. Effectively a joke. If Google had a problem with people and teams hiding behind a monitor not doing much, I'm sure you can recall a pre-pandemic world where this was also the norm when in the office.
Now that my company has started enforcing RTO, I noticed patterns in some of my colleagues, such as arriving late and leaving early, leaving the laptop in their desk disappearing for a long while, not answering messages for hours with nowhere to be seen, and very vague status updates when asked.
It turns out slackers will slack, be it at home or at the office.
I love the fantasy that some management hold that back in my day you couldn't slack in the office and it was a thing that was invented on the first day someone "worked from home".
Over a few drinks they will also tell you stories of the old days where you would leave your jacket on the back of your chair and go to the pub at 11am come back to the office at some point in the afternoon.
You see someone doing this and think: "slacker"...?
I come into the office around 9am and leave at 3:30pm to beat traffic and get a quick gym session in without the crowds. And then I hope back online to do some work. Am I slacking?
I don't always slack off, but I do often need time to rest and gather my thoughts. Sometimes my brain is just not ready or already fatigued and needs time. At home I'll chill out watching some mindless Youtube video or go for a walk. At the office I used to pretend to work by clicking around in some emails or something that was equally unproductive. I prefer not having to pretend to work.
The idea that Urs was hard core about colocation is absolutely false. I was a direct report to Urs when he was VP engineering and I telecommuted one day every week. Urs successor, Wayne Rosing, was much stricter than Urs ever was.
No, remote work is still allowed. It might be a bit more difficult to switch to a remote work contract now (but still possible), but people who are already remote will continue like this.
> people who are already remote will continue like this
My understanding is that everyone has to get approval even people who are currently remote. For those who are near an office, they are being told to expect to be asked to come in frequently.
This is just another form of layoffs, I think. The problem is that many other execs will follow. It's bizarre how this is purely about making life worse for the engineers without any rational reason.
>It's bizarre how this is purely about making life worse for the engineers without any rational reason
There's absolutely a rational reason: many people in management went into management because they enjoy having power over people, and they can have more power over employees who are physically in office.
Readit News
* Remote work is much better if you know what you're building and "just" need to execute.
* In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
Obvious, right? But I think what I've realize is the dimension of "we know what we need to do" is actually pretty much orthogonal to the product, size of the company, tenure of the employees, etc. I've been in early stage startups with no PMF that nonetheless have strong product-centric leadership who can set down an unambiguous vision for what they believe needs to be done. And I've been in startups that have PMF yet have no vision, and everyone is standing around in their remote offices twiddling their thumbs on "what to build next".
For large companies like FAANG, I actually think they mostly fall in bucket 2. I worked at Meta remotely for a year. My instructions when I joined were something like "Yeah so just talk to a bunch of people and look for opportunities to contribute". WTF? No wonder remote is not working well for them. You had to go 5-6 layers up the reporting chain to find anyone with any sort of holistic sense of what needed to be done, and they were not exactly empowered to share that vision lest it conflict with someone elses.
TL;DR I think remote work is well set up for companies with leadership that resembles a benevolent dictatorship. I think if you are all-remote and your leadership is effectively "managerial" in nature, NGMI.
true.
> For large companies like FAANG, I actually think they mostly fall in bucket 2
Also true most fall in the "figure it out" bucket... But here's where it gets funny: just meet in-person you say?
No, this is FAANG. You're in MTV, but the people you need to collaborate are in NYC, LAX, SEA, AUS, SFO...
So your in-person experience is you booking a room just for yourself so you can video conference all those people who also booked a room for themselves.
Because every other team is in same situation, finding a meeting room is a challenge in itself.
Take the call from your desk then?
No, this is FAANG, every building is open office layout, it'd be rude & disrespectful to your colleagues around who are busy coding.
It's mind boggling to me. All I can think is that there is a lot of ego tied up in offices for some reason.
I personally prefer fully remote because I hate the distraction in the open office plan and associated pains like having to find meeting rooms, having to share restrooms with others (I admit this is just my personal pet peeve), and not being able to avoid distraction when I am in the zone for working. Most of all, I don't like the commute (I have to commute 1 hour each way, so everyday, I am spending 2 hours; when I worked from home, I started early and end up working an extra hour or two because I don't have to commute; these days, I try to arrive at 8-9am and leave by 5-6pm to beat the traffic and once I get home, I don't do any work). I consider quitting someday soon and am waiting for my wife to complete her training. Once she has done that, we will move to a place where she can find work, and I will quit my FAANG job for a fully-remote one (sure, my income will take a small hit, but quality of work life is more important for me).
>No, this is FAANG, every building is open office layout, it'd be rude & disrespectful to your colleagues around who are busy coding.
Individual offices have not been the norm for everyone at large companies maybe ever? Cubicles are "better" but not that much better. (Source: I had one for years--both full-height and then shorter.) And the fact is that lots used to spend a lot more time on the phone than they generally need to today given other communication channels. When I've been in an office I absolutely make calls from my desk and everyone else does too.
The downside is it’s hard to get actual coding down because of all the “collaboration”. So my ideal would be 1-2 days in office to meet and then 3-4 days to code ins peace at home.
But at the same time it's perfectly okay to just come up to your busy colleague and pester them about something, because collaboration.
I don't think this is really that common. I spent a decade in MTV on a wide variety of teams. I had VCs with people in different locations a handful of times and I never had one by myself. And getting deep focus with those people was so valuable that it was easier to just go to NYC, MUC, or SYD once a quarter than it was to jump on a call.
On the other hand, the projects that suffered the most were the ones where folks had less scope and autonomy. If you can't unilaterally make decisions for anything that spans more than a day, and you're forced into some ticket-oriented process where everyone has "their" task to do, communication naturally suffers. If you're all in the office, there's a bit more slack to talk quickly and informally outside your immediate silo; remotely, this takes more intentionality, so the friction imposed by your culture and process has an outsize effect.
So my conclusion has been, pretty much, the opposite: remote works better when people have real scope, independence and autonomy. If engineers and small teams can make decisions over longer time-frames, they'll have the space to adapt and work effectively. They might need somebody to set the example or foster the right habits, but it's fundamentally healthy and effective. The frictions with remote work come up when it's thrust against a less flexible culture that makes it harder for folks to adapt.
At 10+ person org sizes, most people can't have long-term scope and autonomy if the management don't know what it wants/needs, since they're just going to change their mind in a week or two. Once those decisions are made you can start delegating responsibility for execution.
The easy thing to do is to blame this all on poor management, but I think that's too pat. I think figuring out the best way to use a team of 20, 100, 1000 engineers to accomplish business goals is just an incredibly hard problem... and one that's often an ongoing, Sisyphean task, where high-bandwidth, low-latency communications are incredibly useful.
Remotely, I’m vastly more productive in getting coding work done. I’m sure most people would agree. But that productivity is almost erased by onboarding new hires. There are so many aspects of software engineering where it’s so much easier to be sitting next to someone, pointing at their monitor, and discussing what and why you are doing something. Instead, you need to hop on a call and awkwardly screen share on a call or write out paragraphs of text to impart the knowledge. In my experience, having such barriers to doing these things makes new hires much less likely to reach out and put themselves into the situations where they can learn, no matter how much they are told it’s ok to bother people.
To me, this is why remote works so much better at smaller companies with less bespoke tooling and legacy code. There’s no better way to learn than in person and there is a huge productivity boost when doing so. That said, personally, I’m willing to deal with the lower productivity in my work life to get the higher productivity in my real life by being able to do laundry during the day and not sit in traffic.
For me, it's the exact opposite.
Before the pandemic, I always found the practice (and even the idea!) of "pairing" quite off-putting. You're standing behind someone, breathing in their body odor, awkwardly sweating next to each other, bending over if you want to type something... a really weird experience.
Remotely, none of those apply! You each get your own keyboard, your own screen, your own mouse. Both of you are 100% comfortable in your own environment. Using decent software helps as well, I recommend Tuple which makes sharing the mouse & keyboard seamless.
For example, incident handling has been lot more efficient for us with the team looking at different aspects and sharing on a call/huddle than what we often used to naturally do (involving a gaggle forming around 1 person's computer with different people offering suggestions of varying usefulness).
Demos and knowledge sharing I find also work better remote now than they used to in person now that people are used to the tooling.
I slightly prefer onboarding remote also, but that is because I'm the kind of person who gets distracted by wondering if I'm sitting too close or maybe I have not enough or too much deodorant etc.
The only thing that suffers remotely for me at this point is engagement. There's something about physically being with other people that makes their problems and desires seem more important to me, and I don't have any good hacks for giving more of a &"@$.
I certainly don't agree at all. Home is where all the distractions are. Home is where i can drift off to make a cup of tea and end up spending an hour reading the newspaper without anyone noticing.
I think it would be useful to understand the distribution of peoples' feelings and experience about this. I have seen so many remote work enthusiasts just taking it as read that everyone prefers remote work because they do.
I think a decent middle-ground between "Document everything" and "Organically relay all information" is to just write "doc-stubs". Just a list of everything out there with no real content. It gives people a jumping off point to ask questions and lets them know what they don't know.
This feels like a tech choice issue to me. I'm constantly on calls with either slack/pop sharing where anyone can draw on the shared screen, or shared coding session like in vscode where you can see each other's cursors. It... works the same for me as pointing at the screen. Just without it being awkward when it's more than 2 people getting to cram into a a small space.
What I found to be the best approach is one or two people already established a concept at what to make or at least the image of it, and the discussion can move to the details and implementation plan. And that doesn't need to be in person
Can you explain how some of the greatest technical research projects of our time like the internet, Linux, etc happened without a big office then?
It's clearly nonsense to suggest you have to be collaborating in-person to build big unknown things. There are thousands of examples throughout open source and more where that hasn't been the case, and millions of examples of projects from company offices that are awful.
The physical geography of the team is not an indicator of success.
Stating that it’s possible is very different to stating which model might be more effective or faster.
Unfortunately we don’t have a parallel universe where torvalds lived on the same street as a bunch of other collaborators to compare
while linux or git was created relatively sololy by a single person (at least at the beginning), i dont think you can call the internet a solo production.
My anecdotal evidence is different.
I've spent the last 2 years building a large, complex, and ambiguous system working almost exclusively remote, and half of it from another continent. For the first year I never met a colleague in person.
However, all this time I have been working closely with another engineer. Both of us are senior and very tenured, got good rapport and were highly aligned and motivated. We were able to execute efficiently, sometimes spending hours huddling in Slack per day.
So my feeling is that remote works in some scenarios and not others, particularly involving new or junior engineers.
Remote Work is better if
1. You do not have home problems that you will end up focusing on when you are there - so if you are single and do not own your home, or you have a great relationship and you do own your home but there are no problems the remote work is probably better. But just like being in an office and going into a private room can help productivity if you have problems with home life going to office to work can help with productivity.
2. Remote work is much better if you are a person who is motivated by love of what you are doing and you actually love what you are doing, if you start to not love what you are doing but you have not left the company yet etc. there can be some slight loss of productivity, not because intentionally goofing off so much but because like #1 above working remotely makes it harder to get focused on something one does for money and not love if you are a love focused worker.
If you want to work in an office go work in one. Join a co-working space or if there is sufficient momentum have everyone who wants an office join the same one in your area.
But don’t force it on everyone. Let people choose. That’s the thing I hate about these conversations is it should just be personal choice for the individual. Make async practices and communication better and re-tool the culture if need be, but remote first thinking benefits office workers too.
Of course if your spouse also works from home, it's important to have a home with more than a single room. Not everybody has that luxury.
Remote Work is better if
1. You prefer remote work
This always gets to be such a dogmatic discussion that will to someone claiming anyone who likes working in an office has no social life.
It’s just a preference. Remote work is not overall objectively better than office work (but you can create a set of goals for which this is true). Office work is not the right choice (but you can create a set of goals for which this is true).
Not sure why this needs to be repeatedly brought up.. I mean it is called "home office" for a reason. If you don't have a real one, remote can suffer.. no surprise.
> * In-person is much better if you don't know what you need to build, are trying to "figure it out as you go along", and are relying on the implicit fast feedback loop of standing next to each other.
This is why the other day I ranted about how my actual preference is in office but having a door that I can close. Realistically we shift between these two modes. In the first mode I need to reduce distractions and isolate myself. This is hard with slack and hard with an open office or cubicles. In person let's me even ignore slack as someone can knock on my door if it is urgent! But that's not all the time, so I open the door and communicate that it is collaboration and discussion phase.
I don't know why companies have become so allergic to doors. They are amazing and can let employees be in control of the chaos of work schedules. They'll know far better than anyone else if they're more productive that day/hour/minute in isolation or in collaboration. Micromanaging doesn't help, a door does.
In-person has benefits in those situations, but whether or not it's better varies a lot based on the person and the office. I find it VERY distracting to have people talking around me. As such, I get a lot less done when in an office, because I can't concentrate as well. When you add in the other negative (commute time, in person interruptions, etc), going into an office for work can far outweigh any benefits of being around the people you work with.
In my experience(i have worked from branch offices most my career) the differentiating factor is literacy, ie how comfortable/confident are the organization with recorded text vs private spoken word and the more the organization writes the more effective the remote staff will be.
And one of the side effect of a written culture is that a lot of the informal conventions/conversations become formal structure so the same level of top down planning can look very differently based on weather the culture for disseminating it is literate or oral.
I have a hard time grasping this one. Is it brainstorming ?
For things like pair programming or specification of a product, having separate screens and independent access to the same documents makes things a lot faster and efficient.
Same way meetings tend to go smoother all participants can access whatever data they need independently from the presenter, and move through the docs at their own pace.
Basically a mix of synchronous communication and access to all needed resources seems to me to perform better than IRL binding of two or more people looking at each other.
Some companies are able to add great new products over time, but I sense that it’s more common for companies to have one great play, after which they either die or merge with another to enjoy better economies of scale.
If you’re big enough and rich enough, you tend to burn a ton of money on internal research groups, but very little of what gets researched is ever a successful product. Doing this as a smaller organization is nearly impossible, which is why I think so many startups just exit rather than finding entirely new things to build.
I would say biggest benefits of in office is spontanious meet/discussion, assisting less senior coworkers etc. If I have a heavy day of work to plow through alone I avoid the office. If I need to sync or work on something with others in-person beats all the tools.
EDIT: and so many of the challenges people are running into now are not new, either - companies were incentivized to make remote work work for two decades pre-Covid, and for a while pundits were all convinced that software jobs were all going away overseas as a result, but despite the salary savings the ROI just wasn't there for all but the most basic of work.
There is huge overhead to in person work and the benefits are wildly oversold.
IMHO a professional software developer in a company with good management/clear direction will thrive in a remote work scenario.
Hard relate. Had my first internship WFH at Big Tech during the pandemic. I had expected work to be more on the lines of: there's always a lot of work to be done, and people will let you know about that. Instead it was the opposite, exactly as quoted above.
While it meant I had more freedom, I would have preferred if someone just doled out concrete tasks to me. I felt a bit lost. But then again, grass is greener on the other side.
OTOH, from personal experience working on-site hasn't really ever helped particularly well in that scenario (which I'mm all too familiar with.)
I used to work for Amzn and the management was totally idiotic about remote work. We had to go to the office to be on call for datacenters 1000kms from the office. Makes perfect sense right?
If you are talking about cost, then the only factor here is the market.
More likely is the "prestige HQ" which is deliberately made more expensive than it needs to be for status.
Or sharing a zoom session. Bonus is that I don’t have to commute to an office on another continent to engage in this second modality - more than half of my teammates are in other cities and I don’t see a benefit in the standing next to each other.
I restarted my software career doing remote work. I didn't have a problem finding things to do because my responsibilities were clear (read: I had a good manager) and I was self-taught so I already had a muscle for seeking knowledge on my own.
I suspect that remote work success is highly influenced by an equation involving how good your manager is and how much you innately know about what things your role is responsible for which can be propped up or replaced by seniority. I also suspect that new grads don't have muscles for finding/seeking internal and external information on their own, which feels powerless. There's ways to solve all these things without treating all of these folks the same.
Frankly, I think these are all the problems you and I relate to, but what's really solved for the business by returning to an office isn't productivity or happiness. It's a real estate problem.
I suspect the people on here who like to hate on managers either have a bad one or just don't realize how much crap they're being shielded from--a lot of which is inevitable in a large organization (where you can't tell everyone "hey figure it out on your own").
That just sounds like poor leadership to me. If you've got a holistic sense of what needs to be done, please share it with the people who need to be doing it.
That, or cut it up into concrete chunks and tell people to build that. But telling people to "figure it out" and then not sharing the overall vision, sounds incredibly dysfunctional to me.
Not everyone is the same office - it’s a joke
For me, the extroverts that insist in socializing, gossiping and politicking in person make office work an unproductive nightmare.
the problem is there is a ton of terrible leadership out there who want to rule by fiat, ignore their employees, and don't trust their employees (who they either directly or indirectly hired) to so their jobs.
People who have been in the industry and with a company for a long time largely have a network and don't really have problems connecting with people. (And at larger companies that network is often pretty distributed anyway.)
But the--admittedly not universal--consensus is that new folks struggle with onboarding and generally making progress in their careers. And even coming into an office doesn't really help if their managers and more senior co-workers are mostly not there. It's certainly hard for me to imagine graduating school and working remotely.
I'm not gonna pretend that every project had 100% coverage and up-to-date documentation, but you should overcome fear of appearing weak/dumb and just ask a question, not even waiting for the next daily meetup. If culture is good enough, people are gonna realize it isn't you dumb, but docs and processes suck ass and will do something about it, maybe even asking you to participate. If not, just leave when the next offer lines up, I did that once when people straight up refused answering to my questions directly, and instead nagging that I didn't ask them properly.
Building up network is obviously more reliable when meeting in person, but it doesn't defeat the idea of getting to know each other through daily interactions, retros, 1-1s, whatever. And moreover if you appear professional and capable in eyes of colleagues by getting shit done, they're would want at least keep in touch
If you start raising future generations to be, you know, adults, instead of mentally stunting them into being eternal teens well into their 30s, and companies make an effort not to have a horrible onboarding and rely on 3 year tenures (and paradoxically do nothing to keep said employees that long), the problem is far less prevalent. Heck, most CS students are expected to do far more complex things at school without any assistance these days.
Do you mind sharing the job title for something like this? Sounds like a dream job.
I created my own job, just with a startup. For a year I had two ostensibly full time jobs. By the end the startup wasn't doing much either and I considered going for three full time jobs.
I'm curious how often anyone actually finds this situation to happen and improve the feedback loop. I feel like it's more of an illusion of productivity then actual productivity because it's a pile of constant interruptions.
Like the given example with Meta doesn't seem like it would be better in-person, it sounds broken regardless.
Beyond the practical scope of this point, we also find the need for mentoring (building a sustainable team).
Similarly, what career stage one is in? Young workers who are in 0-1, and to certain extent 1-10, remote really does not work.
The FAANG I work for now similarly has lots of 0-1, 1-10 type work (at least for my team), and a fully remote team. It works fine too.
The part about more junior employees is _more_ valid, though we had no problem on boarding juniors in either of those two situations.
Ultimately whether remote work is successful depends on whether the people doing the work have sufficient self-discipline to get the job done, and whether the hypothetical office alternative is set up for success. Most offices are not remotely set up for success, but optimize instead for middle management (toxin or cancer? [1]) being able to view their fifedoms.
[1]: https://www.talkfrom.com/video?name=leadership-without-manag...
She left there in 2011, after rising through the ranks to the level of division director.
They did almost all their business over email and Skype, because at the time, Skype was still the best voice/video conferencing software available without dedicated hardware.
If it was possible then for a small group of non-tech-focused people, it's certainly possible now, over a decade later and with an entire pandemic forcing the rapid improvement of all sorts of remote-work technology. That doesn't mean it's for everyone, but boldly stating that it's a "non-starter" is just unsupportable.
Deleted Comment
More team work versus individual work?
Deleted Comment
* Remote work is better if the management has a common goal and everyone can swim towards it.
* In-person work is better if the management is adversarial or competitive with each other, or doesn't have a common goal to work towards.
In either case, it seems like the problem is management, not the work location. Am I wrong here?
What I find fascinating about the remote vs on-site flame war are the _stakes_ though, I don't think we see anything like it in dynamic vs static typing and all the others.
If remote loses, remote workers have to figure out how to relocate their entire families to a place they don't really want to live in, or accept severely limited job opportunities - on top of having to work in an environment they might not be productive or happy in.
If on-site loses, people who are fetching large salaries and bought a house in an area with insane real estate prices face a future where their house value shoots lower than the credit they took to buy it (more towards the national average), and where it's increasingly difficult for them to prove that they earn that large salary they're getting. That's on top of working in an environment they might not be productive or happy in.
Both sides have a lot to lose if it all goes either one way or another, that's unlike any other flame war I can think of. I wonder how much of this is at play at Google. Certainly seems to be escalating if someone feels the need to pull a stunt like that (with the performance reviews).
You nailed it. Ideally, both employees and employers should have the freedom to choose whether to work onsite or remotely, based on their preferences (again, this is subjective and not scientific). I believe the market could self-regulate in such a way. For instance, if the demand for remote work exceeds the supply, a savvy employer with a fit for remote work could take advantage of it by attracting better/more/cheaper talent through offering remote positions.
However, what needs to be debated and critizised is when employers change the rules midway through a game. It's unfair to hire people for a "remote role" and then force them to work in the office. This creates real-world hardships for individuals, such as the need to relocate.
Using your analogy, it would be like initially giving your future employees the choice between Emacs and Vim, but later giving Vim users extra points in performance reviews.
Not that I'm antsy to get back to the office, but that scenario is not an "everyone wins" scenario, it's just another version of remote work. Onsite is only onsite if everyone is onsite. If it's some people in the office and some people at home, the people in the office are just working remotely from the office. All your meetings will be virtual, you don't get the benefits of onsite like being able to gauge someone's availability by just looking at them, you're gonna be screen sharing instead of just pulling up a chair next to someone, etc etc etc.
That analogy doesn't quite apply to this Google situation.
Google has a distinction between "full remote" and "hybrid" employees. Attendance will NOT be considered in the performance reviews of full remote employees.
For the politicians and influencers, however, on-site work is vital (especially if they are good looking or charismatic). They need meetings to shine optically and verbally and dominate the real workers.
This parasitical category of people usually has the most influence, so the move to return to on-site work is not surprising.
Some people are better at creating (real or fake) value face to face. I've seen what you describe in remote/distributed setting as well, in internal message boards for example. Some folks are shy and/or incoherent in person and master manipulators in writing.
And it's only an anecdote, but we debugged a problem today where people were in two provinces of Canada and the UK. Were perfectly able to share screens, run test calculations and "interact" using existing technology. Companies either trust their staff or they don't and the staff should act accordingly.
Full disclosure, this is at a company that is hybrid and not remote-first. But I’ve seen the pattern repeated enough times across various teams that I don’t think my experience is unique. And I say this as someone who personally prefers to work remotely- I’m just not unaware of what’s lost in that environment.
I am a fan of hybrid (people can choose whether or not they're in the office).
However, you want to be able to collaborate effectively at any time and not have to wait for quarterly gatherings. I've got complex problems to solve together with my team mates!
Deleted Comment
This is not the dichotomy you make it look. There are lots of other tasks to do, and for those I personally would often go in a cafeteria or public space, since I didn't mind the distractions -- reading code, doing code reviews, reading docs, writing docs, responding to emails. "Coding" was a relatively small part of my day as a SWE at [FAANG].
Really? what does it bring that you can't do remotely? I can hardly think of a problem where "being in the same room" is a requirement to have a productive discussion
Picking up on social cues. Bouncing ideas around using a whiteboard. Speaking quickly without that ever so slight lag on Zoom which means you struggle to get a word in without potentially talking over somebody. Having food or drinks before or afterwards to unwind and build relationships and kick things around informally.
There's lots more.
Hasn't every study on brainstorming found that it's not a good way to generate good ideas? With having people breathing the same air in a small space believed to be a contributing factor to that.
Per https://twitter.com/emollick/status/1194301106584616960:
> People in groups self-censor and anchor on each others ideas, reducing creativity.
Also can be anti-inclusive when "anchor on each other idea" is a euphemism for that guy repeating what a more shy person said, but louder, stealing credit.
With remote, usually comes a more doc-driven culture, where someone's original idea can be easier to identify and make sure they get credit.
1. Brainstorming questions, not ideas (then go answer them)
2. Soliciting responses anonymously and quietly. This avoids both hierarchical biases and "Loudest person wins" bias. And not to excessively disparage verbal processors (I am one myself). But it's important for them to realize that the constant talking can drown out others' thinking, or their sharing, or their will to live.
Obviously, the "pros" of WFH are massive. No commute. Comfortable office. Food in the fridge. Daytime exercise + showers. Etc. etc.
I found however that too much of a good thing, is bad. For me, the lack of in person communication started to atrophy my enjoyment of life. I wouldn't have been able to really tell you it for the first few years, but I was slowly moving toward being mildly depressed.
Returning to office restored my excitement for my work. I credit my colleagues and the process of working toward a big goal with other people. We really underestimate how important this is for mental health.
If you're in office and you now what I mean, show some love to a fellow co-worker. You gotta admit it's nice to see each other.
Not everyone, of course: I had a peer once who absolutely couldn't stand his family, and he would be at the office before everyone, staying until late at night. Volunteered on weekends too. Almost all of his contributions were the sort of thing that could have been easily automated, but he was happy just wiling hours away. Mileage varies and everyone's situation is unique.
Then of course there's the change element itself being a temporary boost. Turn the lights down and productivity increases. Turn them up and productivity increases. Return them to the normal and productivity increases. Having a varied life is essential. HN's history is littered by posts of people who changed some element of their life and This Changes Everything...at least during the initial period. Then it fades and they abandon it.
None of this is to diminish your personal experience that you obviously enjoy.
Do you have a source on this? I’m both curious and skeptical.
(And ironically, for many in Europe and Eastern NA/SA it’s the work day right now.)
This is basically the essence of life
It’s sometimes called Hedonistic Adaptation
And it’s also almost the same concept as the endless arising and passing, the source of “suffering”, in many eastern spiritual traditions
I would suggest that it's possible to interact with people outside of work. It's genuinely sad to me how many of these "if I don't go into the office I have no one to interact with" posts I see.
People are often in a weird headspace at work too and a bit guarded. I don’t know why that as a social setting is inherently healthier than surrounding yourself with people outside work.
I have a circle of friends and a habit that we keep since 20+ years: every Friday morning we meet for a coffee in a physical coffee shop. Location varied along time, I lived as close as 2 minutes walk from my apartment to (now) 1 hour drive (because of the traffic).
It's not a religious thing, people miss meetings. Got other stuff, or just not being in the mood (like I drank till 2 AM, no way I'm waking up at 6 so I can be out the door at 7 so I'll be in the coffee shop at 8, stay an hour then back home). But the habit has stuck and we kept our friendship. Some people have left the group (one died), others joined it. It's easy to expand a group once you have a stable kernel: just go out for some coffees at day or drinks at night and talk to people.
I get along well with work colleagues as well, mostly I'm asking the question: if I leave this company, will I be ever talking to these guys ever again? If the answer is yes (stay in touch through WhatsApp and the occasional beer in a bar) then they're more than acquaintances and useful connections in case I need a job. Like recently got laid out (whole office closed, some 200 people fired), found a new job in less than a week. Connections, people.
So office has some value but it's far from being "the whole world".
When you go into the office, you can build quite a history with people over time. I have good friends that I've worked with (off and on) for over 25 years. And we are not just "work friends". I've also felt much a part of a "team" working in person than wfh and those connections and trust can manifest in the work too.
I think that we need to continue to acknowledge that this is a human preference. And we all have quite different history, goals and situations.
It’s not that I don’t see friends and family outside those hours, it’s just that those hours constitute like half of the awake hours of my life, and the grind sucks when I do it alone.
I really wish the powers that be would put some data to this. You and I are complete opposites; paying exhorbitant rent, transportation, and living costs usurped much of the enjoyment I could get out of going to an office. I build infra products and in an office setting I usually get a lot of interrupts because product folks generally lack some amount of knowledge/background to fill in gaps on how to use products like mine. When I'm remote they have to follow a support queue, read docs, learn more about infra basics, and attend talks on infra. When I'm in person they can just stop by my desk.
My hypothesis is that folks who like going into an office generally fall into a couple categories:
- people who get/enjoy a substantial amount of social interaction out of work and work adjacent activities
- people who work on public facing products who don't have many internal customers
- people with disruptive home environments
The thing that upsets me is that we're being treated the same and that most of this is to save face for their real estate investment. Not to mention all the half-truths/gaslighting about the universalness of productivity.
I can go to an office where we can schedule room to have a video call to sit around and discuss what tickets we're going to create and who will work on them. Those tickets will get worked on and tracked then we end up creating a product or feature.
I can stay at home to join the video call to sit around and discuss what tickets we're going to create and who will work on them. Those tickets will get worked on and tracked then we end up creating a product or feature.
I don't know what this mysterious in-office inspiration cycle is. It was never polite to just wander up to people working on something and bother them with random junk.
Offices exist for companies where management has sufficiently removed the ability of the people they pay to make decisions. Now you need to come to them and ask to be allowed to do things, so your manager can ponder whether that thing makes sense in whatever weird inter-department power struggle they are working in.
A lot of 'management' doesn't build communication or workflows more advanced than that. And when people are never more than a short walk away, they don't have to.
Slack is just the online version of that. A noisy conversation always within earshot. One that you can usually ignore, but should still be a little aware of what's going on overall, and occasionally you get pulled in to the larger conversation.
It turns out slackers will slack, be it at home or at the office.
Over a few drinks they will also tell you stories of the old days where you would leave your jacket on the back of your chair and go to the pub at 11am come back to the office at some point in the afternoon.
But kidz these days...
You see someone doing this and think: "slacker"...?
I come into the office around 9am and leave at 3:30pm to beat traffic and get a quick gym session in without the crowds. And then I hope back online to do some work. Am I slacking?
Deleted Comment
My understanding is that everyone has to get approval even people who are currently remote. For those who are near an office, they are being told to expect to be asked to come in frequently.
There's absolutely a rational reason: many people in management went into management because they enjoy having power over people, and they can have more power over employees who are physically in office.
that's superficial power
This is absolutely another form of layoffs. Google get the double benefit of 1) loose head count 2) only keeping dedicated* people.
They do loose people with the highest agency as they will be the first to leave.
* Those dedicated people could also just be most in need.