4 ms·

I am finding that everything is just becoming more and more fragmented. Programming languages, ecosystems, frameworks, blah. I have had ideas for some applicat

I am finding that everything is just becoming more and more fragmented. Programming languages, ecosystems, frameworks, blah. I have had ideas for some applications, but can’t do them because the libraries I need are written in different languages, which don’t interoperate well (and one I am not familiar in). Every post here looking for recommendations has many responses with different packages/ecosystems doing the same thing. Sometimes I feel like there are too many developers and not enough of them really interested in the actual hard problems. So they just make another python package manager or webapp framework.

PaulDavisThe1st 5y ago
So a perfect use-case for the obligatory XKCD post about "too many standards, invent one more to try to solve the issue => too many standards plus one".
- onlyrealcuzzo 5y ago
  I don't think standards are the problem. I think a wave of new standards causes issues like the above, and eventually this gets worked out. Not all "standards" in practice become standard. We learn a lot of "standards" were bad ideas, and they eventually get tossed out and most people forget anyone was ever trying to make that a thing.
  - PaulDavisThe1st 5y ago
    "Standards" in this case just means "things that I/we developed that we think should be adopted by lots of other people because it's awesome-sauce"
pjmlp 5y ago
It isn't nice? Back to the 8 and 16 bit home computers, when we had plenty of choice and real fights on the school playground about which were better.
eternityforest 5y ago
I think they are too invested im hard problems, but not invested enough in tedious problems. I don't want a new language, or a new preprocesser, or a new way of thinking about programs... I just want a one click way to take a folder of HTML that looks like a static site, and package it up into cross platform apps with all the proper API access. I don't care about cryptocurrency and global decentralized databases, I just want to be able to run a site by buying a NAS appliances, putting files on it, and sharing the QR code on it without any signups and cloud accounts. The hard problems are mostly solved. There's probably some long dead random github repo that does anything you want. They're just not packaged for all platforms and maintained professionally. I don't need a phone without any binary blobs at all... I just want a mesh network feature and a fair trade stamp. Because... it sucks to maintain something novel even if it's all just pieced together from npm. It sucks to be the one responding to issues. It sucks to actually run a project and keep track of devops stuff. All the tech has been invented dozens of times over, but never polished, and always ruined with some unnecessary immutable log feature or performance destroying random access low latency routing thing like IPFS has.
- tacitusarc 5y ago
  This is why I like Golang. I think it’s the first time in my professional experience where if I see a package that hasn’t been updated in 3 years, that doesn’t mean it’s abandoned, it means it’s stable.
  - christophilus 5y ago
    Clojure is like that. I love it.
  - pnathan 5y ago
    common lisp is like that.
- TheOtherHobbes 5y ago
  There's a gap between developer culture and the giant corporate bureaucracies. The former is focussed on tinkering, nostalgia, tool-building, wheel reinventions, and half-finished build-and-forget spare time projects. The latter has successfully applied high-friction chokeholds around startup culture which makes the cost of entry for new business ideas far higher than it really needs to be. There is almost no blue sky development equivalent to the original Internet, the Mother of All Demos, and the desktop computing model - and associated projects - created at PARC in the 70s and early 80s. It's all very "mature", incremental, slow, and me-too. And so expensive, both financially and in terms of developer time. I'd love to see a renewed interest in attacking this from both ends - a movement to really push the imaginative limits of what's possible, while also being implacably humane and user-oriented and indifferent to ad tech, tinker tech, and pre-commercialised cloud tech.
tacitusarc 5y ago
I’m curious what you would classify as a hard problem. Personally, I think writing a good, easy to use Python package manager that both gains adoption and addresses the myriad of corner cases is a hard problem.
- sseagull 5y ago
  I work in scientific software. I tend to view really hard problems as problems where you don't know the solution, and don't even have a blueprint for how could be solved. Writing a package manager is hard in some sense, but you have an idea of what it needs to do, what features it should have, and what success ultimately looks like. But there are problems, large and small, which require new thinking and leaps of faith. There are people working on these, of course, but sometimes I feel they are being neglected. (and people working on those problems probably make 10-25% of what many FANG developers make...)
  - mportela 5y ago
    > I work in scientific software I envy you!

PaulDavisThe1st 5y ago

So a perfect use-case for the obligatory XKCD post about "too many standards, invent one more to try to solve the issue => too many standards plus one".

onlyrealcuzzo 5y ago
I don't think standards are the problem. I think a wave of new standards causes issues like the above, and eventually this gets worked out. Not all "standards" in practice become standard. We learn a lot of "standards" were bad ideas, and they eventually get tossed out and most people forget anyone was ever trying to make that a thing.
- PaulDavisThe1st 5y ago
  "Standards" in this case just means "things that I/we developed that we think should be adopted by lots of other people because it's awesome-sauce"

pjmlp 5y ago

It isn't nice? Back to the 8 and 16 bit home computers, when we had plenty of choice and real fights on the school playground about which were better.

eternityforest 5y ago

I think they are too invested im hard problems, but not invested enough in tedious problems. I don't want a new language, or a new preprocesser, or a new way of thinking about programs... I just want a one click way to take a folder of HTML that looks like a static site, and package it up into cross platform apps with all the proper API access. I don't care about cryptocurrency and global decentralized databases, I just want to be able to run a site by buying a NAS appliances, putting files on it, and sharing the QR code on it without any signups and cloud accounts. The hard problems are mostly solved. There's probably some long dead random github repo that does anything you want. They're just not packaged for all platforms and maintained professionally. I don't need a phone without any binary blobs at all... I just want a mesh network feature and a fair trade stamp. Because... it sucks to maintain something novel even if it's all just pieced together from npm. It sucks to be the one responding to issues. It sucks to actually run a project and keep track of devops stuff. All the tech has been invented dozens of times over, but never polished, and always ruined with some unnecessary immutable log feature or performance destroying random access low latency routing thing like IPFS has.

tacitusarc 5y ago
This is why I like Golang. I think it’s the first time in my professional experience where if I see a package that hasn’t been updated in 3 years, that doesn’t mean it’s abandoned, it means it’s stable.
- christophilus 5y ago
  Clojure is like that. I love it.
- pnathan 5y ago
  common lisp is like that.
TheOtherHobbes 5y ago
There's a gap between developer culture and the giant corporate bureaucracies. The former is focussed on tinkering, nostalgia, tool-building, wheel reinventions, and half-finished build-and-forget spare time projects. The latter has successfully applied high-friction chokeholds around startup culture which makes the cost of entry for new business ideas far higher than it really needs to be. There is almost no blue sky development equivalent to the original Internet, the Mother of All Demos, and the desktop computing model - and associated projects - created at PARC in the 70s and early 80s. It's all very "mature", incremental, slow, and me-too. And so expensive, both financially and in terms of developer time. I'd love to see a renewed interest in attacking this from both ends - a movement to really push the imaginative limits of what's possible, while also being implacably humane and user-oriented and indifferent to ad tech, tinker tech, and pre-commercialised cloud tech.

tacitusarc 5y ago

I’m curious what you would classify as a hard problem. Personally, I think writing a good, easy to use Python package manager that both gains adoption and addresses the myriad of corner cases is a hard problem.

sseagull 5y ago
I work in scientific software. I tend to view really hard problems as problems where you don't know the solution, and don't even have a blueprint for how could be solved. Writing a package manager is hard in some sense, but you have an idea of what it needs to do, what features it should have, and what success ultimately looks like. But there are problems, large and small, which require new thinking and leaps of faith. There are people working on these, of course, but sometimes I feel they are being neglected. (and people working on those problems probably make 10-25% of what many FANG developers make...)
- mportela 5y ago
  > I work in scientific software I envy you!