5 ms·

Same old story: Developers aren't interested in it if it isn't friendly to the old school OOP + MVC approach. If you give most professional developers something

by drcode 12y ago

Same old story: Developers aren't interested in it if it isn't friendly to the old school OOP + MVC approach. If you give most professional developers something like Clojure, their first thought is "How do I architect a large application this way?" and they don't know the answer. (This is speaking as someone who thinks Clojure is the best language on the planet right now...)

bdavisx 12y ago
So it sounds like Clojure needs a really comprehensive guide/book on how to architect a large application.
- drcode 12y ago
  Agreed... but more than a book, more like a marketing campaign with some PR friendly buzzwords :-)
- dominotw 12y ago
  I think it needs a 'killer app' like rails/hibernate etc which are compelling enough for me to look at a new language.
dminor 12y ago
I think it probably has more to do with static typing than anything.
- asdf1234 12y ago
  Bingo. Clojure gets a lot of things right, but there's a lot of us that find dynamic typing absolutely unacceptable.
benjaminpv 12y ago
I find this to be true of a lot of promising new technologies. None of my day-to-day work is the kind of flashy/bleeding edge stuff that HN tends to cater to, but I'm always on the lookout for new tools that'll make things easier or more reliable. Oftentimes I'll go to evaluate something new only to find that their examples are the tried and true "get some tweets and do something with them" or "here's how you make yet another blog platform." Great. If my job consisted of making blogging platforms I might be interested. Show me how your product will free me from having to handle the drudgery of setting up yet another similar-yet-slightly-different form. Show me how your new framework makes it easier to do integration with several developers. Show me something that'll let me ease the technical debt we've accrued. That kind of thing.
hibikir 12y ago
While I have absolutely no problem with your opinions on programming languages (or anyone's, for that matter), I do not believe you are being very accurate when representing the views of the majority of developers. When you give most professional developers something like clojure, they ask: Where are my objects? where are my types? Going back to a relatively typeless system and having to go to documentation to figure out what operations have been written for a data structure, instead of just pushing dot and seeing a bunch of them, is quite the cultural shock. It's really about the strengths of the language being things your average developer is relatively uninterested in. You pay the price of S expressions to make it easier to do metaprogramming. But what if you don't want to spend your time metaprogramming? Then suddenly what seem like strengths to some suddenly turn into weaknesses. Clojure has its niche, probably as a better alternative to R in scientific environments where metaprogramming can give you tremendous gains, but it doesn't really fit what most of us have to do in a daily basis any more than writing it all in C.
jeremiep 12y ago
This was my initial struggle with Clojure too, trying to fit my existing world views into the language. It didn't take me long to realize my world views were completely wrong, not Clojure. Most developers don't want to learn, they want to ship. Even if this means reusing the same old school OOP+MVC approach and all the potential bugs and balls of muds it almost always creates over time. Yet this is what I like most about Clojure: the language makes it harder for you to write big balls of mud. If your design isn't simple to implement in Clojure, it's a pretty good sign it isn't such a good design to begin with. (I'm also someone who thinks Clojure is the best language right now)

by drcode 12y ago

bdavisx 12y ago

So it sounds like Clojure needs a really comprehensive guide/book on how to architect a large application.

drcode 12y ago
Agreed... but more than a book, more like a marketing campaign with some PR friendly buzzwords :-)
dominotw 12y ago
I think it needs a 'killer app' like rails/hibernate etc which are compelling enough for me to look at a new language.

dminor 12y ago

I think it probably has more to do with static typing than anything.

asdf1234 12y ago
Bingo. Clojure gets a lot of things right, but there's a lot of us that find dynamic typing absolutely unacceptable.

benjaminpv 12y ago

I find this to be true of a lot of promising new technologies. None of my day-to-day work is the kind of flashy/bleeding edge stuff that HN tends to cater to, but I'm always on the lookout for new tools that'll make things easier or more reliable. Oftentimes I'll go to evaluate something new only to find that their examples are the tried and true "get some tweets and do something with them" or "here's how you make yet another blog platform." Great. If my job consisted of making blogging platforms I might be interested. Show me how your product will free me from having to handle the drudgery of setting up yet another similar-yet-slightly-different form. Show me how your new framework makes it easier to do integration with several developers. Show me something that'll let me ease the technical debt we've accrued. That kind of thing.

hibikir 12y ago

While I have absolutely no problem with your opinions on programming languages (or anyone's, for that matter), I do not believe you are being very accurate when representing the views of the majority of developers. When you give most professional developers something like clojure, they ask: Where are my objects? where are my types? Going back to a relatively typeless system and having to go to documentation to figure out what operations have been written for a data structure, instead of just pushing dot and seeing a bunch of them, is quite the cultural shock. It's really about the strengths of the language being things your average developer is relatively uninterested in. You pay the price of S expressions to make it easier to do metaprogramming. But what if you don't want to spend your time metaprogramming? Then suddenly what seem like strengths to some suddenly turn into weaknesses. Clojure has its niche, probably as a better alternative to R in scientific environments where metaprogramming can give you tremendous gains, but it doesn't really fit what most of us have to do in a daily basis any more than writing it all in C.

jeremiep 12y ago

This was my initial struggle with Clojure too, trying to fit my existing world views into the language. It didn't take me long to realize my world views were completely wrong, not Clojure. Most developers don't want to learn, they want to ship. Even if this means reusing the same old school OOP+MVC approach and all the potential bugs and balls of muds it almost always creates over time. Yet this is what I like most about Clojure: the language makes it harder for you to write big balls of mud. If your design isn't simple to implement in Clojure, it's a pretty good sign it isn't such a good design to begin with. (I'm also someone who thinks Clojure is the best language right now)