6 ms·

Programming languages can't all be picked up that quickly, since semantics may differ a lot. You can't say "Give me any programming task and I'll pick it up".

by ramchip 16y ago

Programming languages can't all be picked up that quickly, since semantics may differ a lot. You can't say "Give me any programming task and I'll pick it up". What if you're a Web guy and the task is a C++ medical application? What if we're a finance company and the language is Haskell or ML? There's no way someone with no experience in a related language could become proficient without a fairly long training period in the company. What if I'm a hardware IP company and the language is Verilog? To pretend that what takes many years and domain knowledge to become proficient in, you could just "pick up" on the go, is what sounds a bit entitled to me.

BenSS 16y ago
Once you know one language well, many of the skills transfer. Besides, he states he is looking for an ENTRY level position and fully realizes he's not a guru.
- DisposaBoy 16y ago
  Yes many skills transfer. Like how one ane write C++ from readin the docs without ever learning C++ or how one can write C++ having only ever learned C. The part you're missing is that it doesn't make you any good in C++, in fact I'd argue that it makes you terrible if you're going to try to write X code in Y language. The fact that it's entry-level doesn't mean anything. A lot of times these entry-level jobs aren't something you can just go and start pasting code together.
- Hubbert 16y ago
  Yes. Experience is great, but everyone has to start somewhere. The unemployment rate for under 25 is extremely high. Society doesn't focus much attention these days on "upward mobility." But but an overreliance on "experience" (generally "paid, professional experience"), in extremis, will permanently shut out an entire generation from the world of middle-class employment. Is that really a good idea? "Resolved: Individuals who did not have a good (preferably government) job on July 31, 2007 and a record of unblemished professional success with progressive experience and responsibility and no failures or false starts should abandon all hope." Is this really what we want to promote?
jemfinch 16y ago
> Programming languages can't all be picked up that quickly, since semantics may differ a lot. The second day with my new employer I was tasked with writing an analysis project in a language I'd never even heard of before joining. Within two weeks, I was writing code in that language every bit as good as my coworkers'. Your statement may hold true for people who know one or two languages (like most CS graduates these days), but once you know a wide array of languages (my list is Python, C and C++, OCaml and SML, Lisp and Scheme, Perl, and the semantics but not the libraries of Lua/Ruby/Java/Erlang/Haskell) picking up a new language really isn't a challenge at all: the semantics almost certainly fit somewhere in the space of the languages you know, and what small bits do not are easily learned. (As a side note, that's why I still put OCaml, Erlang, and Lisp on my resume; not that I imagine very many employers would be interested in working in those languages, but because I want them to know that whatever they do their work in likely lies within the range of my existing semantic knowledge.)
- jbooth 16y ago
  >Within two weeks, I was writing code in that language every bit as good as my coworkers'. Just cause your coworkers sucked doesn't invalidate the point. I challenge anyone to write good code in a new language after 2 weeks of learning it. You can hack, sure, but I've never seen anyone write anything respectable in a language they've only been in for 2 weeks. In the worst case they think it's good and it's hard to teach them how to do it right because they think they know it all already. Also, if you claim to know the semantics but not the library of a language, you probably don't actually know the semantics. Corner cases matter.
  - jemfinch 16y ago
    EDIT: An HNer says without evidence that my coworkers suck and gets upvoted; I come to their defense and get downvoted? What strange world is this in which baseless accusations are encouraged? > Just cause your coworkers sucked doesn't invalidate the point. My coworkers don't suck. In fact, I've yet to meet a technical person here that I can confidently say I'm smarter than. > I challenge anyone to write good code in a new language after 2 weeks of learning it. I've taken your challenge and passed. Are you calling me a liar? > Also, if you claim to know the semantics but not the library of a language, you probably don't actually know the semantics. Corner cases matter. Libraries are molded by semantics, and not vice versa. It's certainly necessary to know the libraries in order to be proficient in a language, but it's not at all necessary to know the libraries in order to know its semantics well.
Natsu 16y ago
> Programming languages can't all be picked up that quickly, since semantics may differ a lot. You can't say "Give me any programming task and I'll pick it up". Speak for yourself. I learned AutoLisp with one bit of sample code someone left behind and a bit of Googling in about a week, starting from zero. Nobody else where I worked knows it at all. But from that, I coded a full set of functions that did everything we need. I reverse-engineered the semantics of some crazy internal scripting language to some of our industrial machines. Nobody bothered to document that thing at all and it was custom to my employer. I learned to use a different testing tool from nothing more than a bit of sample code and being told that it was based on VBS. I then improved the code I was given. Some of us really can figure out the semantics of something just from reading a few examples. This also works with foreign languages. Particularly if I can get parallel translations (e.g. subtitles), I can start working backwards to extract meaning and solve for unknown words. In short, you'd be surprised about how quickly we can figure out things, especially if we've seen something like it before. Sure, there are pitfalls, but anywhere where you have nice syntax aware IDEs and whatnot, it's not nearly as much of a barrier as you seem to think. The more languages you've used, the easier it all is. This goes for both human and computer languages.

by ramchip 16y ago

BenSS 16y ago

Once you know one language well, many of the skills transfer. Besides, he states he is looking for an ENTRY level position and fully realizes he's not a guru.

DisposaBoy 16y ago
Yes many skills transfer. Like how one ane write C++ from readin the docs without ever learning C++ or how one can write C++ having only ever learned C. The part you're missing is that it doesn't make you any good in C++, in fact I'd argue that it makes you terrible if you're going to try to write X code in Y language. The fact that it's entry-level doesn't mean anything. A lot of times these entry-level jobs aren't something you can just go and start pasting code together.
Hubbert 16y ago
Yes. Experience is great, but everyone has to start somewhere. The unemployment rate for under 25 is extremely high. Society doesn't focus much attention these days on "upward mobility." But but an overreliance on "experience" (generally "paid, professional experience"), in extremis, will permanently shut out an entire generation from the world of middle-class employment. Is that really a good idea? "Resolved: Individuals who did not have a good (preferably government) job on July 31, 2007 and a record of unblemished professional success with progressive experience and responsibility and no failures or false starts should abandon all hope." Is this really what we want to promote?

jemfinch 16y ago

> Programming languages can't all be picked up that quickly, since semantics may differ a lot. The second day with my new employer I was tasked with writing an analysis project in a language I'd never even heard of before joining. Within two weeks, I was writing code in that language every bit as good as my coworkers'. Your statement may hold true for people who know one or two languages (like most CS graduates these days), but once you know a wide array of languages (my list is Python, C and C++, OCaml and SML, Lisp and Scheme, Perl, and the semantics but not the libraries of Lua/Ruby/Java/Erlang/Haskell) picking up a new language really isn't a challenge at all: the semantics almost certainly fit somewhere in the space of the languages you know, and what small bits do not are easily learned. (As a side note, that's why I still put OCaml, Erlang, and Lisp on my resume; not that I imagine very many employers would be interested in working in those languages, but because I want them to know that whatever they do their work in likely lies within the range of my existing semantic knowledge.)

jbooth 16y ago
>Within two weeks, I was writing code in that language every bit as good as my coworkers'. Just cause your coworkers sucked doesn't invalidate the point. I challenge anyone to write good code in a new language after 2 weeks of learning it. You can hack, sure, but I've never seen anyone write anything respectable in a language they've only been in for 2 weeks. In the worst case they think it's good and it's hard to teach them how to do it right because they think they know it all already. Also, if you claim to know the semantics but not the library of a language, you probably don't actually know the semantics. Corner cases matter.
- jemfinch 16y ago
  EDIT: An HNer says without evidence that my coworkers suck and gets upvoted; I come to their defense and get downvoted? What strange world is this in which baseless accusations are encouraged? > Just cause your coworkers sucked doesn't invalidate the point. My coworkers don't suck. In fact, I've yet to meet a technical person here that I can confidently say I'm smarter than. > I challenge anyone to write good code in a new language after 2 weeks of learning it. I've taken your challenge and passed. Are you calling me a liar? > Also, if you claim to know the semantics but not the library of a language, you probably don't actually know the semantics. Corner cases matter. Libraries are molded by semantics, and not vice versa. It's certainly necessary to know the libraries in order to be proficient in a language, but it's not at all necessary to know the libraries in order to know its semantics well.

Natsu 16y ago

> Programming languages can't all be picked up that quickly, since semantics may differ a lot. You can't say "Give me any programming task and I'll pick it up". Speak for yourself. I learned AutoLisp with one bit of sample code someone left behind and a bit of Googling in about a week, starting from zero. Nobody else where I worked knows it at all. But from that, I coded a full set of functions that did everything we need. I reverse-engineered the semantics of some crazy internal scripting language to some of our industrial machines. Nobody bothered to document that thing at all and it was custom to my employer. I learned to use a different testing tool from nothing more than a bit of sample code and being told that it was based on VBS. I then improved the code I was given. Some of us really can figure out the semantics of something just from reading a few examples. This also works with foreign languages. Particularly if I can get parallel translations (e.g. subtitles), I can start working backwards to extract meaning and solve for unknown words. In short, you'd be surprised about how quickly we can figure out things, especially if we've seen something like it before. Sure, there are pitfalls, but anywhere where you have nice syntax aware IDEs and whatnot, it's not nearly as much of a barrier as you seem to think. The more languages you've used, the easier it all is. This goes for both human and computer languages.