3 ms·

Why should we do anything new? I've been seeing this attitude more and more where developers think they're entitled to be convinced to use some new technology a

by zcw100 6y ago

Why should we do anything new? I've been seeing this attitude more and more where developers think they're entitled to be convinced to use some new technology as though their time and attention are so valuable it's someone else's responsibility to convince them to spend it. What I think differentiates this from normal outreach and education that every project needs to do to build a community is the sense of entitlement and the declaration that they're outright hostile and argumentative and then expect to be engaged.

amelius 6y ago
> Why should we do anything new? I think a problem exists when that something new is being pushed and hyped in order to create a self-fulfilling prophecy where everybody starts jumping on the bandwagon and an ecosystem is built.
hackerfromthefu 6y ago
Seems like a natural defensive mechanism against the hype train, which is real and various industries make their living from churn. Because of the bias of vested interests many new things that are pushed are not really in the developers interest. Some are, and some are worth the effort, but many aren't and how do you know which without expending your limited resource of attention?
moksly 6y ago
> Why should we do anything new? From a management positions this is probably the most important question you can ask. Your most expensive resource is your programmers, and a few talented operations guys, and you need a good reason to expend those resources. We work rather closely with other municipalities, and one in particular is very adoptive of new things. One of the things they adopted was kubernetes. But the thing is, we’re not Netflix, and kubernetes isn’t easy, and the result was they spend two years worth of man hours on something that didn’t make them anymore competitive than us because the advantages never really comes into play in our small enterprise organisations. We spent those two years worth of manhours improving our Azure strategy, getting our operations guys up to speed on powershell and our developers on python, and actually moved our organisation forward in a manner that was useful. Maybe Rust is useful, maybe it’s going to have a brighter web-future than .Net Core, but if you can’t show me how that is, or why I should adopt it, then it’s a really big risk that I’m happy to let other people take. We can always adopt it later. Of course you probably meant to ask the question in a different light, and I think that point is very valid. I just can’t stress enough, how little something being all the hype actually means. What if Rust turns out to be another Ruby on Rails or “Node.JS for everything” craze? Then I won’t lose any sleep over never having thouched it, and if you look at our local job market, then there hasn’t been a single Rust focused job in my entire country for two years now. Not even as a “nice if also you know” listing. That doesn’t exactly install me with confidence that Rust is ever going to be important.
- Dowwie 6y ago
  You're in a position right now where you probably ought to continue using Python. It's a great tool for many jobs, the community is super helpful and very capable, and the ecosystem is tremendous. Tech decisions are hard to make and managers try to choose carefully. It sounds like your team already bought into some of the costly hype by adopting Kubernetes (and probably microservices). It takes restraint not to go further and rewrite everything in Rust. It's been a while since microservices became a big deal, so the tech industry may be due for a cost-cutting correction where projects consolidate a constellation of microservices to efficient monolithic systems written in systems languages such as Rust or Go and use leaner hosting solutions. You can adopt Rust then, maybe.

by zcw100 6y ago

amelius 6y ago

> Why should we do anything new? I think a problem exists when that something new is being pushed and hyped in order to create a self-fulfilling prophecy where everybody starts jumping on the bandwagon and an ecosystem is built.

hackerfromthefu 6y ago

Seems like a natural defensive mechanism against the hype train, which is real and various industries make their living from churn. Because of the bias of vested interests many new things that are pushed are not really in the developers interest. Some are, and some are worth the effort, but many aren't and how do you know which without expending your limited resource of attention?

moksly 6y ago

> Why should we do anything new? From a management positions this is probably the most important question you can ask. Your most expensive resource is your programmers, and a few talented operations guys, and you need a good reason to expend those resources. We work rather closely with other municipalities, and one in particular is very adoptive of new things. One of the things they adopted was kubernetes. But the thing is, we’re not Netflix, and kubernetes isn’t easy, and the result was they spend two years worth of man hours on something that didn’t make them anymore competitive than us because the advantages never really comes into play in our small enterprise organisations. We spent those two years worth of manhours improving our Azure strategy, getting our operations guys up to speed on powershell and our developers on python, and actually moved our organisation forward in a manner that was useful. Maybe Rust is useful, maybe it’s going to have a brighter web-future than .Net Core, but if you can’t show me how that is, or why I should adopt it, then it’s a really big risk that I’m happy to let other people take. We can always adopt it later. Of course you probably meant to ask the question in a different light, and I think that point is very valid. I just can’t stress enough, how little something being all the hype actually means. What if Rust turns out to be another Ruby on Rails or “Node.JS for everything” craze? Then I won’t lose any sleep over never having thouched it, and if you look at our local job market, then there hasn’t been a single Rust focused job in my entire country for two years now. Not even as a “nice if also you know” listing. That doesn’t exactly install me with confidence that Rust is ever going to be important.

Dowwie 6y ago
You're in a position right now where you probably ought to continue using Python. It's a great tool for many jobs, the community is super helpful and very capable, and the ecosystem is tremendous. Tech decisions are hard to make and managers try to choose carefully. It sounds like your team already bought into some of the costly hype by adopting Kubernetes (and probably microservices). It takes restraint not to go further and rewrite everything in Rust. It's been a while since microservices became a big deal, so the tech industry may be due for a cost-cutting correction where projects consolidate a constellation of microservices to efficient monolithic systems written in systems languages such as Rust or Go and use leaner hosting solutions. You can adopt Rust then, maybe.