4 ms·

I have 2 main concerns going forward with any Rails alternative (ie. Trailblazer and the like): community support and jobs. I have no idea how big the communit

by adamors 10y ago

I have 2 main concerns going forward with any Rails alternative (ie. Trailblazer and the like): community support and jobs. I have no idea how big the community behind each library/framework is and honestly I don't have the time to properly test-drive multiple and research the health of their respective ecosystems. With Rails I can rely on 10-15 gems that can help me with 80% of my projects. Perhaps I overestimate my reliance on 3rd party code, but that leaves me with my second issue. Ruby jobs are de facto Rails jobs. I have weekly email newsletters set up for Ruby jobs and in the last 3+ years only a handful were Sinatra/Rack based. It would be interesting to see a reliable source of non-Rails Ruby jobs out there.

chucksmash 10y ago
At some point you'll be in a position to decide which tech stack a company should use. I suspect you'll find there's a bit of a chicken and egg problem here - you'll want access to the largest pool of good candidates possible so you will also be disincentived from straying from the well worn path. Even if you grant that good programmers can pick up new languages or frameworks easily (which I'm not entirely convinced of - there is a wide gulf between "can follow the tutorial" and "successfully used in anger"), you'll incur a penalty in velocity while all your new hires learn Framework X. Anecdotal example of the perils of going off the beaten path: I often see a job advert on StackO for a company looking for Python/Django guys. When you click through to the job description, they are actually using Perl/Dancer. Not that there's anything wrong with that, but for me personally that's a deal breaker. I have no interest in investing the time to become proficient on that particular stack.
- pessimizer 10y ago
  > Even if you grant that good programmers can pick up new languages or frameworks easily (which I'm not entirely convinced of - there is a wide gulf between "can follow the tutorial" and "successfully used in anger"), you'll incur a penalty in velocity while all your new hires learn Framework X. I think the question is more "Does what we're doing require good programmers, or can we get by with shitty ones?" More often than not, you're not doing anything special, and you want to pick a platform that the cheapest, most common programmers out there can add features in a non-disastrous amount of time (different depending on industry + your competition), and not generate enough technical debt to collapse everything before the next rewrite/upgrade cycle.

by adamors 10y ago

chucksmash 10y ago

At some point you'll be in a position to decide which tech stack a company should use. I suspect you'll find there's a bit of a chicken and egg problem here - you'll want access to the largest pool of good candidates possible so you will also be disincentived from straying from the well worn path. Even if you grant that good programmers can pick up new languages or frameworks easily (which I'm not entirely convinced of - there is a wide gulf between "can follow the tutorial" and "successfully used in anger"), you'll incur a penalty in velocity while all your new hires learn Framework X. Anecdotal example of the perils of going off the beaten path: I often see a job advert on StackO for a company looking for Python/Django guys. When you click through to the job description, they are actually using Perl/Dancer. Not that there's anything wrong with that, but for me personally that's a deal breaker. I have no interest in investing the time to become proficient on that particular stack.

pessimizer 10y ago
> Even if you grant that good programmers can pick up new languages or frameworks easily (which I'm not entirely convinced of - there is a wide gulf between "can follow the tutorial" and "successfully used in anger"), you'll incur a penalty in velocity while all your new hires learn Framework X. I think the question is more "Does what we're doing require good programmers, or can we get by with shitty ones?" More often than not, you're not doing anything special, and you want to pick a platform that the cheapest, most common programmers out there can add features in a non-disastrous amount of time (different depending on industry + your competition), and not generate enough technical debt to collapse everything before the next rewrite/upgrade cycle.