4 ms·

> - On the other hand, even if some things in programming are fleeting (the currently popular frameworks etc), there is a core of fundamental knowledge that you

> - On the other hand, even if some things in programming are fleeting (the currently popular frameworks etc), there is a core of fundamental knowledge that you gain over the course of your career that will always apply. Spot on. Ignore the popular framework of the month. A lot of developers focus only on the surface (popular tools) and neglect fundamentals, algorithms, OS internals, how CPUs work, networks...

Kiro 6y ago
I find keeping up with the popular frameworks improve my knowledge of best practices and patterns.
singlow 6y ago
You can't ignore it. If you ignore the framework that is a requirement for 20-30% of advertised positions, then you are limiting your ability to get hired. That may be good advice if you are the one architecting the new project, but not if you are looking for a coding job. If you are looking for a new coding job, look at what people are hiring for and at least get familiar with a good number of those things. Of course if you aren't having trouble finding a job then you don't need to follow this advice.
- jerf 6y ago
  "If you ignore the framework that is a requirement for 20-30% of advertised positions," The "framework of the month" is not "a requirement for 20-30% of advertised positions". The "framework of the month" requires a minimum of a year to get to that point and 2-3 years is a lot more common, even in the JS frontend space. By the time something gets to 20-30% penetration it's past that phase.
  - wolco 6y ago
    What's the path around this? I've tried to keep one main language and work with something different each time. It creates diversivity in my resume but allows me to use my core anchor language to find a role. How do you pivot to something brand new?
    - jerf 6y ago
      Well, one option is just to ignore framework of the month entirely. It is perfectly feasible to just pay attention to things that prove themselves to the point that they're 10% of the relevant listings and have sustained that for a year or two. A mature engineer doesn't need to be bleeding edge. It's perfectly fine to merely be tracking technologies that have gotten to their high-growth phase. The other thing I try to do is have coverage, such that even if I don't know a particular tech I have a story that says I can learn it easily. I've used enough DBs that I don't need to chase the latest things; I'm confident that even if I've never used a columnar database that I can pick it up quickly if I need it. I've used enough programming languages that I don't need to go chasing the latest one, because it is still frankly mostly a different spelling of things I've already used. And so on. So I don't need to go chasing everything all the time. I generally don't pivot into anything brand new, because within the domains I work in, there isn't much fundamentally new stuff left for me to pivot into. (A lot of my recent growth involves learning how to do engineering while being more directly tied into the business and interacting with business people, and learning how to be an interface between business and tech, rather than more types of tech. I'm doing this from an engineering perspective rather than a "management" perspective, because it turns out there is quite a difference between the two once a company scales up enough.)
  - singlow 6y ago
    I was taking "framework of the month" a little less literally.
    - jerf 6y ago
      Yeah, but that's a pretty big gulf.
- eru 6y ago
  You can still ignore the framework. You won't apply to to 100% of jobs. So you already make a selection. Some jobs are require highly specialized skills and knowledge. Keeping up with the framework-of-the-month costs time as well. Time you could have spent getting better at Linux kernel programming or Cobol.
- ex_amazon_sde 6y ago
  I interviewed a whole lot of people when I was in Amazon and no candidate was rejected, in my memory, for not knowing a tool, a framework, one language or one category of technologies. The interview topics of on-sites are general enough and focused on principles, and, by far, the large majority of candidates are rejected for giving vague and superficial answers. My experience convinced me that, as a candidate, I don't want to divide my time preparing and interviewing for many "typical" companies. I instead focus on very few interesting ones and maximize my chances there. "shotgun versus sniper" if you will: you need one good offer, not 10 so-so ones. Among other reasons, the typical buzzword-driven company is less likely to be an interesting environment to work in.
nine_k 6y ago
Why ignore? Take a look, have an idea. Add another data point to the chart of where the industry is going. Once you have seen, say, 15 frameworks, it becomes easier to map a 16th in the coordinate space, and quickly learn it when needed by reusing the knowledge you already have.

Kiro 6y ago

I find keeping up with the popular frameworks improve my knowledge of best practices and patterns.

singlow 6y ago

You can't ignore it. If you ignore the framework that is a requirement for 20-30% of advertised positions, then you are limiting your ability to get hired. That may be good advice if you are the one architecting the new project, but not if you are looking for a coding job. If you are looking for a new coding job, look at what people are hiring for and at least get familiar with a good number of those things. Of course if you aren't having trouble finding a job then you don't need to follow this advice.

jerf 6y ago
"If you ignore the framework that is a requirement for 20-30% of advertised positions," The "framework of the month" is not "a requirement for 20-30% of advertised positions". The "framework of the month" requires a minimum of a year to get to that point and 2-3 years is a lot more common, even in the JS frontend space. By the time something gets to 20-30% penetration it's past that phase.
- wolco 6y ago
  What's the path around this? I've tried to keep one main language and work with something different each time. It creates diversivity in my resume but allows me to use my core anchor language to find a role. How do you pivot to something brand new?
  - jerf 6y ago
    Well, one option is just to ignore framework of the month entirely. It is perfectly feasible to just pay attention to things that prove themselves to the point that they're 10% of the relevant listings and have sustained that for a year or two. A mature engineer doesn't need to be bleeding edge. It's perfectly fine to merely be tracking technologies that have gotten to their high-growth phase. The other thing I try to do is have coverage, such that even if I don't know a particular tech I have a story that says I can learn it easily. I've used enough DBs that I don't need to chase the latest things; I'm confident that even if I've never used a columnar database that I can pick it up quickly if I need it. I've used enough programming languages that I don't need to go chasing the latest one, because it is still frankly mostly a different spelling of things I've already used. And so on. So I don't need to go chasing everything all the time. I generally don't pivot into anything brand new, because within the domains I work in, there isn't much fundamentally new stuff left for me to pivot into. (A lot of my recent growth involves learning how to do engineering while being more directly tied into the business and interacting with business people, and learning how to be an interface between business and tech, rather than more types of tech. I'm doing this from an engineering perspective rather than a "management" perspective, because it turns out there is quite a difference between the two once a company scales up enough.)
- singlow 6y ago
  I was taking "framework of the month" a little less literally.
  - jerf 6y ago
    Yeah, but that's a pretty big gulf.
eru 6y ago
You can still ignore the framework. You won't apply to to 100% of jobs. So you already make a selection. Some jobs are require highly specialized skills and knowledge. Keeping up with the framework-of-the-month costs time as well. Time you could have spent getting better at Linux kernel programming or Cobol.
ex_amazon_sde 6y ago
I interviewed a whole lot of people when I was in Amazon and no candidate was rejected, in my memory, for not knowing a tool, a framework, one language or one category of technologies. The interview topics of on-sites are general enough and focused on principles, and, by far, the large majority of candidates are rejected for giving vague and superficial answers. My experience convinced me that, as a candidate, I don't want to divide my time preparing and interviewing for many "typical" companies. I instead focus on very few interesting ones and maximize my chances there. "shotgun versus sniper" if you will: you need one good offer, not 10 so-so ones. Among other reasons, the typical buzzword-driven company is less likely to be an interesting environment to work in.

nine_k 6y ago

Why ignore? Take a look, have an idea. Add another data point to the chart of where the industry is going. Once you have seen, say, 15 frameworks, it becomes easier to map a 16th in the coordinate space, and quickly learn it when needed by reusing the knowledge you already have.