On one hand, being paid a dump truck of cash and not having to actually work might be nice for a little bit of time, but I can see how aggravating it was to the author to want to engineer and...
On one hand, being paid a dump truck of cash and not having to actually work might be nice for a little bit of time, but I can see how aggravating it was to the author to want to engineer and actually work and have no projects to work on.
It's easy for me to say I'd love a paid year to do nothing, but I bet I would similarly go crazy when I have nothing to do. That's pretty fucked that the manager metrics were about head count and not actual productivity. It seems that parts of the Silicon Valley TV show are more true than you'd think, ha.
Author here. I find it hard to complain too much because I did get an absolute fuck-ton of cash to do very little work. And the free therapy through the health plan was useful for the existential...
Author here. I find it hard to complain too much because I did get an absolute fuck-ton of cash to do very little work. And the free therapy through the health plan was useful for the existential dread of realizing my 20 year goal was a dead end. After a few months there I was having nightmares and panic attacks. I needed to tell myself I was going to quit to calm the anxiety.
I'm sorry to hear you were struggling with your mental health because of that experience, that sounds awful. If I'm affected by layoffs at my company, one thought of mine was to try to get a low...
I'm sorry to hear you were struggling with your mental health because of that experience, that sounds awful.
If I'm affected by layoffs at my company, one thought of mine was to try to get a low effort job at Google or Netflix or something for a year to pad my savings, but it sounds like that's a real decision to make and something I shouldn't do lightly, since I probably won't be happy in that environment, just like you weren't (I'm currently on a small team in a smallish company, I get to WFH full time, and I make good money and have little oversight (but do my job well)).
There's a point at which you make enough money to be comfortable and anything else past that is just nice to have. In tech, especially certain jobs, it can be fairly easy to reach that point. My...
There's a point at which you make enough money to be comfortable and anything else past that is just nice to have. In tech, especially certain jobs, it can be fairly easy to reach that point. My suggestion if you reach that point and find yourself dissatisfied with your existing work is to pivot somewhere which pays less but is more fulfilling. It sounds like this article is your journey to this realization.
Your 20 year goal is absolutely possible, just maybe not at the company you thought it would be. Build up a set of skills to learn when a team is a good fit, and apply broadly to find the right position for you. I've been lucky enough to never feel a need to chase big tech money and instead choose a career which is more fulfilling to me in other ways. Unsurprisingly there are times I think about taking a year or few at a big tech company to bank some money and just retire early (the grass is always greener) but ultimately I think I'm happier and healthier and more fulfilled continuing the same course.
It's anxiety-inducing to realize that you're in a much better situation than many people can dream of, and still feel incomplete. My advice is to do what you can to make the world a better place....
It's anxiety-inducing to realize that you're in a much better situation than many people can dream of, and still feel incomplete.
My advice is to do what you can to make the world a better place. See if you can find a job which applies your skills and helps others.
A year of being paid to do nothing just sounds like a year for my anxiety to really feel that axe at the back of my neck waiting for the 'We are letting you go' email to eventually drop. Even...
A year of being paid to do nothing just sounds like a year for my anxiety to really feel that axe at the back of my neck waiting for the 'We are letting you go' email to eventually drop. Even knowing that it is coming wouldn't help...
I hadn't heard of "dotted line manager" before, but it sounds like his manager on the loan team (implying a dotted line on the org chart, while his real manager has a solid line). Skip manager is...
Exemplary
I hadn't heard of "dotted line manager" before, but it sounds like his manager on the loan team (implying a dotted line on the org chart, while his real manager has a solid line).
Skip manager is your manager's manager (i.e., going upwards and skipping one manager).
'ticket to PR, LGTM, release, and get metrics in a hour' is saying that the lifecycle of a code change is very fast.
A ticket is a request for some specified work.
A PR is a pull request, which means you have finished implementing the code which fulfils the request in the ticket, and you would like someone to approve the PR so that your code can be merged into the main branch of code.
LGTM is "looks good to me", which is what reviewers customarily say when they agree to merging a PR into the main branch.
release is when your code has finished merging into the main branch and has been deployed into production systems which are actually running.
getting metrics means you are monitoring your production system in real time, to confirm that the changes you just merged in, are actually working correctly.
An hour is almost unbelievably fast for this process, I would expect days and sometimes weeks.
And a "production system" is the one that real users are actually using, the thing you're using when you go to google.com. I don't know if non-tech people know that term. Usually we run a separate...
And a "production system" is the one that real users are actually using, the thing you're using when you go to google.com. I don't know if non-tech people know that term.
Usually we run a separate copy of the site on our own computer, not accessible to the public, to test our changes while we're working on them, though I understand Google just throws those copies on servers too because when you're literally Google you get really good at setting up lots of servers really quickly.
I would certainly hope anyone with more than 5 people working for them runs seperate non-production servers. Especially with automated deployments, it's trivial to deploy to dev servers to test,...
though I understand Google just throws those copies on servers too
I would certainly hope anyone with more than 5 people working for them runs seperate non-production servers.
Especially with automated deployments, it's trivial to deploy to dev servers to test, then promote to prod in the same way.
Even in the age of containers, there's too many ways for "It works on my machine" to be a statement that gets uttered on a production failure.
That, and non-production is a great place for training other staff.
I'm referring to the local copy you're hopefully testing on before you push anywhere. It's absolutely painful to work on things when you can't run them at all without deploying. Iteration time for...
I'm referring to the local copy you're hopefully testing on before you push anywhere. It's absolutely painful to work on things when you can't run them at all without deploying. Iteration time for all the "I know this is unfinished but I want to see if this particular part works" intermediate steps now includes deployment time, and you can't easily run the thing in a debugger.
Google is special in that they outright don't support this, and put all the effort that would be used to support it into just doing development on a remote server from a chromebook. I'm very skeptical and would want to see an extensive demo before working there, but apparently it works for them.
Continuing the explaining-things-to-non-tech-people thing: Containers are a way to pack up all the software needed to run your thing and run it as a virtual computer inside another computer. They...
Continuing the explaining-things-to-non-tech-people thing:
Containers are a way to pack up all the software needed to run your thing and run it as a virtual computer inside another computer. They were a response to people repeatedly getting annoyed that things would work on one computer and break on another, so we kind of just went "what if we pack up the whole computer it works on and just deploy that to production? and then if we need more of them because we have lots of users, we can just make copies of it instead of setting up new servers from scratch". This works way better than you'd expect for how much of a bodge it sounds like, and now it's the standard way to run things on servers.
A debugger is a program that attaches itself to another program (the one you're working on) and gives you ways to pause it and look around at its insides while it's running. If you don't have access to one and you need to see internal state, the alternative is to add a bunch of "and then when you get to this part, throw this data in the logs so I can look at it afterward" to the code, which may be either totally fine or painfully slow depending on how long it takes between changing code and running the new code.
That's essentially what I do at my job, in my dev branch: Do some work, lint/commit/push, let pipeline get into dev server. Takes like 3-5 minutes and gives me a good excuse to stretch. On the...
That's essentially what I do at my job, in my dev branch:
Do some work, lint/commit/push, let pipeline get into dev server. Takes like 3-5 minutes and gives me a good excuse to stretch.
On the rare occasion that's too slow for troubleshooting iterations, I mount the local build folder to the container via SSHFS.
I guess it helps that the apps I'm building touch heavily locked down systems, the kind that they don't want dev's machines to have direct access to. The dev servers that run the apps have tight logging and monitoring systems in place.
Technically "computer science" is the theoretical side of the making-computers-do-stuff field, which lots of us have degrees in but don't interact with much after graduating. This is "software...
Technically "computer science" is the theoretical side of the making-computers-do-stuff field, which lots of us have degrees in but don't interact with much after graduating. This is "software development" or "software engineering", depending on how fancy you want to sound. It's kind of like "physics" vs the thing normal people mean when they say "engineering" (it's common to just call us "engineers" when there are no Real Engineers at the company).
Meta: I think it's actually a fascinating exercise to explain this stuff to people with no exposure to it. I would have so much fun explaining human culture(s) to English-speaking aliens.
Ditto. I've never served, but Military History has been a passion of mine for decades, and I've even been known to watch the occasional Military Academy lecture too. Love the username, BTW. Navy...
let me know if you ever want any obscure, arcane military concepts, doctrine or over-complicated slides
Ditto. I've never served, but Military History has been a passion of mine for decades, and I've even been known to watch the occasional Military Academy lecture too. Love the username, BTW. Navy or Coast Guard reference, or Iain M Banks fan? Either way 👍 ;)
Sorry it wasn’t clear. Series A is supposed to mean the company has gone through one round of fundraising. Really it’s more complicated than that as there can be one or two preliminary rounds...
Sorry it wasn’t clear. Series A is supposed to mean the company has gone through one round of fundraising. Really it’s more complicated than that as there can be one or two preliminary rounds before that. But the idea I was conveying was that we were working from the very start of the company with everyone being one of the original team members. This was the second startup. The first startup was series B and has gone on to series C since I left.
I also started with a team I didn't like, but instead of looking outside Google, I looked for internal transfers. Fortunately there was nobody saying they didn't like 20% projects, so I built an...
I also started with a team I didn't like, but instead of looking outside Google, I looked for internal transfers. Fortunately there was nobody saying they didn't like 20% projects, so I built an internal search engine for Java (before they had decent code search) and ended up demoing it to another team, which got me a transfer to developer tools.
But there was still a lot of bouncing around and thinking I really should be doing something that has more impact, somehow, if I found the right team and applied myself.
you're at Google, an insanely profitable, genius-filled enterprise. You're well-paid, well-fed, and live in the eternal spring paradise of Silicon Valley. What's the best thing you can do with this insane luck?
And yet, when you try to actually do something, it somehow slips away; the good projects always seem to be in another castle.
But I can't really complain too much about the company that funded my retirement. Now I can screw around doing whatever I want.
I think if I could have joined 10 years earlier I would have loved it while rapidly building a retirement portfolio. But I really want to be in the building phase. Building without “process...
I think if I could have joined 10 years earlier I would have loved it while rapidly building a retirement portfolio. But I really want to be in the building phase. Building without “process scars”. Without monopoly-sized code mistakes. There’s a master logging protobuf with a billion extensions and considerations because of all of the things it’s getting used for. Any reasonably sized company couldn’t have that kind of tech debt.
That part really got to me - the tech debt slowing me down when I wasn’t sure if the cause of that debt was even legal. It may actually be perfectly legal, but I think if you find an infinite money glitch like Google has the government should step in and slow it down on principle.
That blog post you linked actually spurred me to write my own.
There was plenty of process that smaller companies don't have even back then. The logging protobuf was there. But there was also lots of churn because some of the foundations hadn't been worked...
There was plenty of process that smaller companies don't have even back then. The logging protobuf was there. But there was also lots of churn because some of the foundations hadn't been worked out yet. I thought the launch of App Engine was pretty great (despite some flaws) because I could build simple internal websites without having to deal with the problems production apps have.
But the main difference is that the tech backlash hadn't started when I joined. That was only for the last few years.
This seems like a massive "got-mine" attitude. Do you really not believe that a major portion of the money you made came from imposing massive externalities onto society at large? Ignoring basic...
But I can't really complain too much about the company that funded my retirement. Now I can screw around doing whatever I want.
This seems like a massive "got-mine" attitude. Do you really not believe that a major portion of the money you made came from imposing massive externalities onto society at large?
Ignoring basic customer support, rugpulling products, WEI, widevine, Project dragonfly, dragnet surveillance, leveraging youtube as an anticompetitive loss leader, paying billions of dollars to suppress search alternatives, etc.?
E: And to clarify, there's nothing wrong with pointing out that Google does good things and has offered some nice services to the world, but it's uncritically ignoring the considerable harms that I find objectionable.
You’re misreading that. I wasn’t always happy when I worked at Google, there are consequences of that, and I have plenty of criticisms. But what I meant to say is that overall I think they treated...
You’re misreading that. I wasn’t always happy when I worked at Google, there are consequences of that, and I have plenty of criticisms. But what I meant to say is that overall I think they treated me very well.
The world is not fair and never was. Compensation in Silicon Valley, especially, is famous for not being fair, for certain janitors and receptionists becoming rich because they were early employees at the right company. A gold rush has nothing to do with fairness. People are attracted by unfairness, hoping to get some unfair results for themselves.
The time it really sunk in was during the dot-com era when a co-worker sold a domain name for millions and retired. (I didn’t speculate in domain names myself because it seemed somehow uncouth.) I remember reading articles about suddenly rich young people dealing with the guilt. That’s a nice problem to have!
Because I actually do care about fairness, despite everything, I always felt a little guilty about how much I was being paid while working at Google (given how much I accomplished, which was middling), and that was even before I realized how much the stock would be worth. However, looking back on it, it was actually quite hard to be productive at Google; there were lots of obstacles and traps for the unwary. It was rather weird for an institution to pay so much and seemingly care so little about productivity. (Which isn’t to say nobody cared; there were plenty of company and grass-roots initiatives. It’s hard to fix.)
Also, the money mostly comes from advertisers. They seem like an unlikely candidate for an exploited class. Better to “pay it forward” in some way. It’s a reason I got interested in effective altruism. Guilt money, basically.
This is all from an individualistic perspective. Systemic problems need systemic solutions, but from an individual point of view, fixing any single systemic problem is an ambitious project.
I don't think that there isn't really a good way or reason to hold the individual programmers and software engineers who code Google's abominations into being. Google has massively changed...
I don't think that there isn't really a good way or reason to hold the individual programmers and software engineers who code Google's abominations into being.
Google has massively changed society, and certainly for the worst. Unfortunately, holding low level actors responsible for this wouldn't change anything here. If it wasn't OP, it would be some other programmer or engineer dealing with the same issues.
The problem is systematic, and more evident of human nature as a whole.
I do actually think there's a step we can take as individuals in the industry, which is the OP's thesis: spreading the message that the money isn't worth it. Beyond that, you're right. There was a...
I do actually think there's a step we can take as individuals in the industry, which is the OP's thesis: spreading the message that the money isn't worth it. Beyond that, you're right.
There was a time was a time where I personally would've worked for a FAANG, but I was lucky enough to have friends and mentors who had been down that route, told me their stories, and steered me away. The less talent Google has, the sooner it crumbles.
I work for the non-toxic breed of startup mentioned in the piece, and I'm solemnly convinced that decentralizing tech into smaller companies in the <200 headcount range is both a viable and superior form of social organization. Less busy work, less stress, more innovation, more consumer choice, and more worker choice.
Basically, there's a better local optimum right over the next hill, and once society as a whole knows about it, we can start realigning those incentives that individuals are currently following to point us to that greener pasture.
I know somebody who worked for google. He started on ads, but then switched to chromebook graphics drivers. Then, when he found out what google was doing with chromebooks (I assume he started...
I know somebody who worked for google. He started on ads, but then switched to chromebook graphics drivers. Then, when he found out what google was doing with chromebooks (I assume he started before the strategy was obvious), he quit.
Thanks for sharing! I'm in a somewhat similar life situation to yours pre-Google, and I found this very insightful. I've often wondered if a jump to such a mature company would help me grow or...
Thanks for sharing! I'm in a somewhat similar life situation to yours pre-Google, and I found this very insightful. I've often wondered if a jump to such a mature company would help me grow or kill my enthusiasm.
It helps illustrate the case-by-case nature of large organizations: some teams are fantastic and some are downright miserable, and it's not always easy to know which is which beforehand. Sometimes you take the plunge and find out the hard way.
It's also helpful to realize the importance of expectations when thinking about famous industry players. No matter the legacy, they're all normal businesses at the end of the day.
The blog post titled "I Skipped to the Ending" on danangell.com reflects on the author's journey from early web development with a friend, through various jobs in Silicon Valley startups, to finally landing a job at Google. Despite the prestige and pay at Google, the author faced bureaucracy, mismanagement, and a lack of direction that led to a realization: the scale and compensation of FAANG companies did not equate to meaningful work or learning experiences for him. Ultimately, the author left Google, valuing hard work and learning over compensation and status ➊.
I have a friend who went through this exact process. The biggest driver of his leaving was the culmination of this and the last round of layoffs where he was spared (to be promoted) and his...
I have a friend who went through this exact process. The biggest driver of his leaving was the culmination of this and the last round of layoffs where he was spared (to be promoted) and his mentor/supervisor who was there for 10 years was laid off.
I also have several other friends who work at Google, and they are all just complacent at this point. The joke is that you "retire to Google" for a reason, I suppose, you just work and don't care whether what you do makes a real difference. All of them are basically just doing the "quiet quitting" (gods I hate that term, but I'm too lazy to explain) thing and have been for the last 5 years.
I remember having such a high opinion of Google as a student. (The jokes were about IBM.) Though I've not worked there, it's disappointing to watch a place you admire become...
I remember having such a high opinion of Google as a student. (The jokes were about IBM.) Though I've not worked there, it's disappointing to watch a place you admire become entrenched/unproductive.
The company I work for sounds a lot like the author's first Silicon Valley startup. There's a healthy and transparent culture, people are engaged, and my colleagues are great. Coincidentally I also happened to work alongside the project founder for a year or two. The longer I stay, the less I want to leave! Articles like this make me appreciate what I have and temper the (financial) allure of a mature company for sure.
It's pretty hard to turn down oodles of cash though. o_O I work to pay the bills mostly, and can find engaging tasks outside of software... civic or social volunteering mainly, that sort of thing.... I would have few qualms being paid to do nothing. I feel like most jobs I've had involve a lot of doing nothing, even the fun ones.
It's funny how so many people I know who haven't worked in software have backwards ideas about Silicon Valley work culture. Both startups I worked at had very healthy work cultures. Unlimited PTO...
It's funny how so many people I know who haven't worked in software have backwards ideas about Silicon Valley work culture. Both startups I worked at had very healthy work cultures. Unlimited PTO where there was no pressure to not take it. I think I generally took 4-5 weeks off each year and no one even said a word. Maybe a couple weeks each year I'd need to do 50 hours, but it always felt like an accomplishment afterwards. Most other weeks I'd clock 30 hours across 5 days. When I told friends' parents and such about my working for a startup they'd ask me how I'm handling the grind culture.
Google still maintains an illusion that it has retained its "Googley" culture from the earlier years. But really their culture is the worst I've dealt with since graduating college.
Those companies absolutely exist, they just also pay through the nose. It's the trade off; make an absolute assload of cash for a short period of time until you burn out and quit abruptly, or find...
Those companies absolutely exist, they just also pay through the nose. It's the trade off; make an absolute assload of cash for a short period of time until you burn out and quit abruptly, or find a more sustainable job that you can work at for a while.
I don’t know if it’s a good general rule to say grind culture is well compensated for. I’ve interviewed at places that clearly were grindy. So I have some points of reference. One such place was...
I don’t know if it’s a good general rule to say grind culture is well compensated for. I’ve interviewed at places that clearly were grindy. So I have some points of reference. One such place was the Alliance of American Football, a now defunct alternative football league. I went to their office for my interview. When they brought me into the main room where the employees worked, every single one looked like they’d hardly slept and no one moved their eyes in my direction. I got an almost identical offer from that company as I did from the first startup I worked at which had a great culture.
The AAF shut down less than 6 months after I got their offer.
That's fair, I suppose I will say the companies that exist long term that are known for that kind of culture (Amazon comes to mind, or fintech) also pay through the nose or hire exclusively fresh...
That's fair, I suppose I will say the companies that exist long term that are known for that kind of culture (Amazon comes to mind, or fintech) also pay through the nose or hire exclusively fresh grads or have some other 'thing' that makes them desirable.
Make no mistake, high attrition rates are horrible for the health of any engineering org in the long run, but a place like Amazon or, say, the games industry get enough fresh blood that wants to work there that they can absorb the impact of that poor health by just throwing more bodies at the problem.
On one hand, being paid a dump truck of cash and not having to actually work might be nice for a little bit of time, but I can see how aggravating it was to the author to want to engineer and actually work and have no projects to work on.
It's easy for me to say I'd love a paid year to do nothing, but I bet I would similarly go crazy when I have nothing to do. That's pretty fucked that the manager metrics were about head count and not actual productivity. It seems that parts of the Silicon Valley TV show are more true than you'd think, ha.
Author here. I find it hard to complain too much because I did get an absolute fuck-ton of cash to do very little work. And the free therapy through the health plan was useful for the existential dread of realizing my 20 year goal was a dead end. After a few months there I was having nightmares and panic attacks. I needed to tell myself I was going to quit to calm the anxiety.
I've been learning coping mechanisms.
I'm sorry to hear you were struggling with your mental health because of that experience, that sounds awful.
If I'm affected by layoffs at my company, one thought of mine was to try to get a low effort job at Google or Netflix or something for a year to pad my savings, but it sounds like that's a real decision to make and something I shouldn't do lightly, since I probably won't be happy in that environment, just like you weren't (I'm currently on a small team in a smallish company, I get to WFH full time, and I make good money and have little oversight (but do my job well)).
I do know people there that like their jobs. It’s a big company - it can’t all be bad.
There's a point at which you make enough money to be comfortable and anything else past that is just nice to have. In tech, especially certain jobs, it can be fairly easy to reach that point. My suggestion if you reach that point and find yourself dissatisfied with your existing work is to pivot somewhere which pays less but is more fulfilling. It sounds like this article is your journey to this realization.
Your 20 year goal is absolutely possible, just maybe not at the company you thought it would be. Build up a set of skills to learn when a team is a good fit, and apply broadly to find the right position for you. I've been lucky enough to never feel a need to chase big tech money and instead choose a career which is more fulfilling to me in other ways. Unsurprisingly there are times I think about taking a year or few at a big tech company to bank some money and just retire early (the grass is always greener) but ultimately I think I'm happier and healthier and more fulfilled continuing the same course.
It's anxiety-inducing to realize that you're in a much better situation than many people can dream of, and still feel incomplete.
My advice is to do what you can to make the world a better place. See if you can find a job which applies your skills and helps others.
A year of being paid to do nothing just sounds like a year for my anxiety to really feel that axe at the back of my neck waiting for the 'We are letting you go' email to eventually drop. Even knowing that it is coming wouldn't help...
I hadn't heard of "dotted line manager" before, but it sounds like his manager on the loan team (implying a dotted line on the org chart, while his real manager has a solid line).
Skip manager is your manager's manager (i.e., going upwards and skipping one manager).
'ticket to PR, LGTM, release, and get metrics in a hour' is saying that the lifecycle of a code change is very fast.
An hour is almost unbelievably fast for this process, I would expect days and sometimes weeks.
And a "production system" is the one that real users are actually using, the thing you're using when you go to google.com. I don't know if non-tech people know that term.
Usually we run a separate copy of the site on our own computer, not accessible to the public, to test our changes while we're working on them, though I understand Google just throws those copies on servers too because when you're literally Google you get really good at setting up lots of servers really quickly.
I would certainly hope anyone with more than 5 people working for them runs seperate non-production servers.
Especially with automated deployments, it's trivial to deploy to dev servers to test, then promote to prod in the same way.
Even in the age of containers, there's too many ways for "It works on my machine" to be a statement that gets uttered on a production failure.
That, and non-production is a great place for training other staff.
I'm referring to the local copy you're hopefully testing on before you push anywhere. It's absolutely painful to work on things when you can't run them at all without deploying. Iteration time for all the "I know this is unfinished but I want to see if this particular part works" intermediate steps now includes deployment time, and you can't easily run the thing in a debugger.
Google is special in that they outright don't support this, and put all the effort that would be used to support it into just doing development on a remote server from a chromebook. I'm very skeptical and would want to see an extensive demo before working there, but apparently it works for them.
Continuing the explaining-things-to-non-tech-people thing:
Containers are a way to pack up all the software needed to run your thing and run it as a virtual computer inside another computer. They were a response to people repeatedly getting annoyed that things would work on one computer and break on another, so we kind of just went "what if we pack up the whole computer it works on and just deploy that to production? and then if we need more of them because we have lots of users, we can just make copies of it instead of setting up new servers from scratch". This works way better than you'd expect for how much of a bodge it sounds like, and now it's the standard way to run things on servers.
A debugger is a program that attaches itself to another program (the one you're working on) and gives you ways to pause it and look around at its insides while it's running. If you don't have access to one and you need to see internal state, the alternative is to add a bunch of "and then when you get to this part, throw this data in the logs so I can look at it afterward" to the code, which may be either totally fine or painfully slow depending on how long it takes between changing code and running the new code.
That's essentially what I do at my job, in my dev branch:
Do some work, lint/commit/push, let pipeline get into dev server. Takes like 3-5 minutes and gives me a good excuse to stretch.
On the rare occasion that's too slow for troubleshooting iterations, I mount the local build folder to the container via SSHFS.
I guess it helps that the apps I'm building touch heavily locked down systems, the kind that they don't want dev's machines to have direct access to. The dev servers that run the apps have tight logging and monitoring systems in place.
I like to replace this in my head with "Let's gamble, try merging."
“Fuck it, ship it”
My favorite is to ask in #general if we feel confident with a “YOLO deploy”
Technically "computer science" is the theoretical side of the making-computers-do-stuff field, which lots of us have degrees in but don't interact with much after graduating. This is "software development" or "software engineering", depending on how fancy you want to sound. It's kind of like "physics" vs the thing normal people mean when they say "engineering" (it's common to just call us "engineers" when there are no Real Engineers at the company).
Meta: I think it's actually a fascinating exercise to explain this stuff to people with no exposure to it. I would have so much fun explaining human culture(s) to English-speaking aliens.
Ditto. I've never served, but Military History has been a passion of mine for decades, and I've even been known to watch the occasional Military Academy lecture too. Love the username, BTW. Navy or Coast Guard reference, or Iain M Banks fan? Either way 👍 ;)
I didn't know there was a real term for this - I've been using "grandboss" for years.
Sorry it wasn’t clear. Series A is supposed to mean the company has gone through one round of fundraising. Really it’s more complicated than that as there can be one or two preliminary rounds before that. But the idea I was conveying was that we were working from the very start of the company with everyone being one of the original team members. This was the second startup. The first startup was series B and has gone on to series C since I left.
I also started with a team I didn't like, but instead of looking outside Google, I looked for internal transfers. Fortunately there was nobody saying they didn't like 20% projects, so I built an internal search engine for Java (before they had decent code search) and ended up demoing it to another team, which got me a transfer to developer tools.
But there was still a lot of bouncing around and thinking I really should be doing something that has more impact, somehow, if I found the right team and applied myself.
I read another blog post captured this well:
And yet, when you try to actually do something, it somehow slips away; the good projects always seem to be in another castle.
But I can't really complain too much about the company that funded my retirement. Now I can screw around doing whatever I want.
I think if I could have joined 10 years earlier I would have loved it while rapidly building a retirement portfolio. But I really want to be in the building phase. Building without “process scars”. Without monopoly-sized code mistakes. There’s a master logging protobuf with a billion extensions and considerations because of all of the things it’s getting used for. Any reasonably sized company couldn’t have that kind of tech debt.
That part really got to me - the tech debt slowing me down when I wasn’t sure if the cause of that debt was even legal. It may actually be perfectly legal, but I think if you find an infinite money glitch like Google has the government should step in and slow it down on principle.
That blog post you linked actually spurred me to write my own.
There was plenty of process that smaller companies don't have even back then. The logging protobuf was there. But there was also lots of churn because some of the foundations hadn't been worked out yet. I thought the launch of App Engine was pretty great (despite some flaws) because I could build simple internal websites without having to deal with the problems production apps have.
But the main difference is that the tech backlash hadn't started when I joined. That was only for the last few years.
This seems like a massive "got-mine" attitude. Do you really not believe that a major portion of the money you made came from imposing massive externalities onto society at large?
Ignoring basic customer support, rugpulling products, WEI, widevine, Project dragonfly, dragnet surveillance, leveraging youtube as an anticompetitive loss leader, paying billions of dollars to suppress search alternatives, etc.?
E: And to clarify, there's nothing wrong with pointing out that Google does good things and has offered some nice services to the world, but it's uncritically ignoring the considerable harms that I find objectionable.
You’re misreading that. I wasn’t always happy when I worked at Google, there are consequences of that, and I have plenty of criticisms. But what I meant to say is that overall I think they treated me very well.
The world is not fair and never was. Compensation in Silicon Valley, especially, is famous for not being fair, for certain janitors and receptionists becoming rich because they were early employees at the right company. A gold rush has nothing to do with fairness. People are attracted by unfairness, hoping to get some unfair results for themselves.
The time it really sunk in was during the dot-com era when a co-worker sold a domain name for millions and retired. (I didn’t speculate in domain names myself because it seemed somehow uncouth.) I remember reading articles about suddenly rich young people dealing with the guilt. That’s a nice problem to have!
Because I actually do care about fairness, despite everything, I always felt a little guilty about how much I was being paid while working at Google (given how much I accomplished, which was middling), and that was even before I realized how much the stock would be worth. However, looking back on it, it was actually quite hard to be productive at Google; there were lots of obstacles and traps for the unwary. It was rather weird for an institution to pay so much and seemingly care so little about productivity. (Which isn’t to say nobody cared; there were plenty of company and grass-roots initiatives. It’s hard to fix.)
Also, the money mostly comes from advertisers. They seem like an unlikely candidate for an exploited class. Better to “pay it forward” in some way. It’s a reason I got interested in effective altruism. Guilt money, basically.
This is all from an individualistic perspective. Systemic problems need systemic solutions, but from an individual point of view, fixing any single systemic problem is an ambitious project.
Well-considered response - thanks for typing out your take.
I don't think that there isn't really a good way or reason to hold the individual programmers and software engineers who code Google's abominations into being.
Google has massively changed society, and certainly for the worst. Unfortunately, holding low level actors responsible for this wouldn't change anything here. If it wasn't OP, it would be some other programmer or engineer dealing with the same issues.
The problem is systematic, and more evident of human nature as a whole.
I do actually think there's a step we can take as individuals in the industry, which is the OP's thesis: spreading the message that the money isn't worth it. Beyond that, you're right.
There was a time was a time where I personally would've worked for a FAANG, but I was lucky enough to have friends and mentors who had been down that route, told me their stories, and steered me away. The less talent Google has, the sooner it crumbles.
I work for the non-toxic breed of startup mentioned in the piece, and I'm solemnly convinced that decentralizing tech into smaller companies in the <200 headcount range is both a viable and superior form of social organization. Less busy work, less stress, more innovation, more consumer choice, and more worker choice.
Basically, there's a better local optimum right over the next hill, and once society as a whole knows about it, we can start realigning those incentives that individuals are currently following to point us to that greener pasture.
I know somebody who worked for google. He started on ads, but then switched to chromebook graphics drivers. Then, when he found out what google was doing with chromebooks (I assume he started before the strategy was obvious), he quit.
Thanks for sharing! I'm in a somewhat similar life situation to yours pre-Google, and I found this very insightful. I've often wondered if a jump to such a mature company would help me grow or kill my enthusiasm.
It helps illustrate the case-by-case nature of large organizations: some teams are fantastic and some are downright miserable, and it's not always easy to know which is which beforehand. Sometimes you take the plunge and find out the hard way.
It's also helpful to realize the importance of expectations when thinking about famous industry players. No matter the legacy, they're all normal businesses at the end of the day.
ChatGPT summary of the link:
I have a friend who went through this exact process. The biggest driver of his leaving was the culmination of this and the last round of layoffs where he was spared (to be promoted) and his mentor/supervisor who was there for 10 years was laid off.
I also have several other friends who work at Google, and they are all just complacent at this point. The joke is that you "retire to Google" for a reason, I suppose, you just work and don't care whether what you do makes a real difference. All of them are basically just doing the "quiet quitting" (gods I hate that term, but I'm too lazy to explain) thing and have been for the last 5 years.
I remember having such a high opinion of Google as a student. (The jokes were about IBM.) Though I've not worked there, it's disappointing to watch a place you admire become entrenched/unproductive.
The company I work for sounds a lot like the author's first Silicon Valley startup. There's a healthy and transparent culture, people are engaged, and my colleagues are great. Coincidentally I also happened to work alongside the project founder for a year or two. The longer I stay, the less I want to leave! Articles like this make me appreciate what I have and temper the (financial) allure of a mature company for sure.
It's pretty hard to turn down oodles of cash though. o_O I work to pay the bills mostly, and can find engaging tasks outside of software... civic or social volunteering mainly, that sort of thing.... I would have few qualms being paid to do nothing. I feel like most jobs I've had involve a lot of doing nothing, even the fun ones.
It's funny how so many people I know who haven't worked in software have backwards ideas about Silicon Valley work culture. Both startups I worked at had very healthy work cultures. Unlimited PTO where there was no pressure to not take it. I think I generally took 4-5 weeks off each year and no one even said a word. Maybe a couple weeks each year I'd need to do 50 hours, but it always felt like an accomplishment afterwards. Most other weeks I'd clock 30 hours across 5 days. When I told friends' parents and such about my working for a startup they'd ask me how I'm handling the grind culture.
Google still maintains an illusion that it has retained its "Googley" culture from the earlier years. But really their culture is the worst I've dealt with since graduating college.
Yeah, I'm sure people aren't lying about startups that have grind cultures, but somehow those were never the ones I ended up working for.
Those companies absolutely exist, they just also pay through the nose. It's the trade off; make an absolute assload of cash for a short period of time until you burn out and quit abruptly, or find a more sustainable job that you can work at for a while.
I don’t know if it’s a good general rule to say grind culture is well compensated for. I’ve interviewed at places that clearly were grindy. So I have some points of reference. One such place was the Alliance of American Football, a now defunct alternative football league. I went to their office for my interview. When they brought me into the main room where the employees worked, every single one looked like they’d hardly slept and no one moved their eyes in my direction. I got an almost identical offer from that company as I did from the first startup I worked at which had a great culture.
The AAF shut down less than 6 months after I got their offer.
That's fair, I suppose I will say the companies that exist long term that are known for that kind of culture (Amazon comes to mind, or fintech) also pay through the nose or hire exclusively fresh grads or have some other 'thing' that makes them desirable.
Make no mistake, high attrition rates are horrible for the health of any engineering org in the long run, but a place like Amazon or, say, the games industry get enough fresh blood that wants to work there that they can absorb the impact of that poor health by just throwing more bodies at the problem.