28 ms·

> Are they? You make bombastic statements but the reality is, that there are not that many people anywhere. The US Bureau of Labor Statistics tracks this. I di

> Are they? You make bombastic statements but the reality is, that there are not that many people anywhere. The US Bureau of Labor Statistics tracks this. I did not make those numbers up. We can certainly agree that not everyone with a degree or experience is a good hire, but good talent is not one in a hundred. "Everyone" is not having a hard time hiring. There is a relatively decent labor market, but employers are having trouble for a variety of reasons that mostly has nothing to do with a lack of sufficient talent. Some factors: they have unreasonable requirements , such as years of experience in specific technologies; they can't or won't pay a competitive salary; they locate in high CoL cities where paying a competitive salary is even harder; they filter out people with bad hiring and interviewing processes; they demand excessive dedication (50+ hour weeks, for example); they don't post jobs where enough people are looking. [EDIT: Another good one that came up mind: the hiring process takes too long. One of the companies I work with now takes an average of six weeks simply to respond to a candidate!! I have had several people I know through my network put in and they all got jobs elsewhere before they were even contacted! This same company cannot understand why they have such trouble finding people - they also pay well below market rates.] I have been on the interviewer side of the table and made hiring decisions. I can tell you where I live, we don't really have problems. At my previous company, we didn't even pay well and got tons of great talent! (This did mean we had some high turnover, but there were always more.) > Programing != building for scale. This is only something experience can teach, but was majority of developers never touched any meaningful scale. I don't mean to demean what you do and again, you obviously know more about your product than I do, but no, scale is really not that hard, can be taught pretty readily, and the hardest problems of scaling are already solved by your cloud platform, which is half if not more of their value proposition in the first place. There are a lot of people operating at massive scales even without cloud services. It isn't that unusual. I sincerely think you should take a very close look about what your jobs really require and what most software engineers are actually capable of doing. I've no doubt it requires skill and experience. I just don't think it requires as much as you think it does.

doh 7y ago
> We can certainly agree that not everyone with a degree or experience is a good hire, but good talent is not one in a hundred. It also not 1 of 2. And that's quite the gist of it. We get plenty of interest, but people fall off quickly through the hiring process. We don't care what their prior experiences are. We don't demand 50+ hour work week (in fact, we actively discourage people to work past that including commute). Still takes time to find people that can learn to build for scale. > I don't mean to demean what you do and again, you obviously know more about your product than I do, but no, scale is really not that hard, can be taught pretty readily, and the hardest problems of scaling are already solved by your cloud platform, which is half if not more of their value proposition in the first place. There are a lot of people operating at massive scales even without cloud services. It isn't that unusual. They do take a huge load off the plate. That doesn't mean you can easily build for scale. There are plenty of companies that can put a few millions of rows in table and serve hundreds of concurrent users. Very few can do that in orders of magnitude higher. That's where we operate. > I sincerely think you should take a very close look about what your jobs really require and what most software engineers are actually capable of doing. I've no doubt it requires skill and experience. I just don't think it requires as much as you think it does This is hard to argue. I'm doing this for 5.5 years now, so would guess that I know what we need at this point.
- Amezarak 7y ago
  > There are plenty of companies that can put a few millions of rows in table and serve hundreds of concurrent users. Very few can do that in orders of magnitude higher. That's where we operate. This is actually not at all unusual, and even when it's not already the case, most well-architected applications serving hundreds of concurrent users can already scale to 1ks or 10ks without changing anything. I think this is where your disconnect is. In 2019, 10ks of concurrent users is not special. It's ordinary. Even 100ks or 1Ms are not that unusual. I am not a fan of cloud services or most of the new hotnesses in vogue today, but this is something that they've enabled even most new grads with an iota of training to do. Looking from the outside in, the problem (and arms race) involved in determining whether one video is actually a 'like'/edited version of another infringing video seems much more interesting.
  - doh 7y ago
    > I think this is where your disconnect is. In 2019, 10ks of concurrent users is not special. It's ordinary. Even 100ks or 1Ms are not that unusual. Really? So which services have 1M concurrent users, outside of the top 10? At 100M daily users a service would get 1.1k users per sec (on average). And those have quite large deployment and plenty of engineers.
    - Amezarak 7y ago
      A lot of services peak in the millions. I don't know what you consider the top ten, but a lot of chat apps - Discord, Skype, Whatsapp (even prior to Facebook acquisition, and which famously had what, 50 employees?), a lot of games, mobile and desktop, top apps in the app store, top couple hundred websites or so. If we just think about TPS, bajillions of applications you've never heard of are dealing with large loads (10-100k TPS) in every economic sector you can think of.
      - doh 7y ago
        WhatsApp stored no data and ran everything as through service, that's why they did it with so few engineers. Almost no games ever reach scale close to what you just wrote. The few that do are known with large technical teams. Same goes to everything else. The reason why you can name them almost all is because it's so rare. You are grossly overblowing the numbers. Only a very few services ever get to these numbers and once they do, they talk about the challenges of scale constantly [0] [1] [2] [0] https://blog.discordapp.com/using-rust-to-scale-elixir-for-11-million-concurrent-users-c6f19fc029d3 https://blog.discordapp.com/using-rust-to-scale-elixir-for-1... [1] https://blog.roblox.com/2012/04/supporting-millions-of-players-at-roblox/ https://blog.roblox.com/2012/04/supporting-millions-of-playe... [2] http://highscalability.com/blog/2014/2/26/the-whatsapp-architecture-facebook-bought-for-19-billion.html http://highscalability.com/blog/2014/2/26/the-whatsapp-archi...

doh 7y ago

> We can certainly agree that not everyone with a degree or experience is a good hire, but good talent is not one in a hundred. It also not 1 of 2. And that's quite the gist of it. We get plenty of interest, but people fall off quickly through the hiring process. We don't care what their prior experiences are. We don't demand 50+ hour work week (in fact, we actively discourage people to work past that including commute). Still takes time to find people that can learn to build for scale. > I don't mean to demean what you do and again, you obviously know more about your product than I do, but no, scale is really not that hard, can be taught pretty readily, and the hardest problems of scaling are already solved by your cloud platform, which is half if not more of their value proposition in the first place. There are a lot of people operating at massive scales even without cloud services. It isn't that unusual. They do take a huge load off the plate. That doesn't mean you can easily build for scale. There are plenty of companies that can put a few millions of rows in table and serve hundreds of concurrent users. Very few can do that in orders of magnitude higher. That's where we operate. > I sincerely think you should take a very close look about what your jobs really require and what most software engineers are actually capable of doing. I've no doubt it requires skill and experience. I just don't think it requires as much as you think it does This is hard to argue. I'm doing this for 5.5 years now, so would guess that I know what we need at this point.

Amezarak 7y ago
> There are plenty of companies that can put a few millions of rows in table and serve hundreds of concurrent users. Very few can do that in orders of magnitude higher. That's where we operate. This is actually not at all unusual, and even when it's not already the case, most well-architected applications serving hundreds of concurrent users can already scale to 1ks or 10ks without changing anything. I think this is where your disconnect is. In 2019, 10ks of concurrent users is not special. It's ordinary. Even 100ks or 1Ms are not that unusual. I am not a fan of cloud services or most of the new hotnesses in vogue today, but this is something that they've enabled even most new grads with an iota of training to do. Looking from the outside in, the problem (and arms race) involved in determining whether one video is actually a 'like'/edited version of another infringing video seems much more interesting.
- doh 7y ago
  > I think this is where your disconnect is. In 2019, 10ks of concurrent users is not special. It's ordinary. Even 100ks or 1Ms are not that unusual. Really? So which services have 1M concurrent users, outside of the top 10? At 100M daily users a service would get 1.1k users per sec (on average). And those have quite large deployment and plenty of engineers.
  - Amezarak 7y ago
    A lot of services peak in the millions. I don't know what you consider the top ten, but a lot of chat apps - Discord, Skype, Whatsapp (even prior to Facebook acquisition, and which famously had what, 50 employees?), a lot of games, mobile and desktop, top apps in the app store, top couple hundred websites or so. If we just think about TPS, bajillions of applications you've never heard of are dealing with large loads (10-100k TPS) in every economic sector you can think of.
    - doh 7y ago
      WhatsApp stored no data and ran everything as through service, that's why they did it with so few engineers. Almost no games ever reach scale close to what you just wrote. The few that do are known with large technical teams. Same goes to everything else. The reason why you can name them almost all is because it's so rare. You are grossly overblowing the numbers. Only a very few services ever get to these numbers and once they do, they talk about the challenges of scale constantly [0] [1] [2] [0] https://blog.discordapp.com/using-rust-to-scale-elixir-for-11-million-concurrent-users-c6f19fc029d3 https://blog.discordapp.com/using-rust-to-scale-elixir-for-1... [1] https://blog.roblox.com/2012/04/supporting-millions-of-players-at-roblox/ https://blog.roblox.com/2012/04/supporting-millions-of-playe... [2] http://highscalability.com/blog/2014/2/26/the-whatsapp-architecture-facebook-bought-for-19-billion.html http://highscalability.com/blog/2014/2/26/the-whatsapp-archi...