![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
elfsJames Gosling, the "father of Java," has called on developers to stop using Emacs because his own company's IDE, Netbeans, "fits together components well" and has kept up with Moore's Law whereas Emacs has not.
As I see it, what he's saying is that one should tackle a steep learning curve and adopt a power-hungry monstrosity written in a non-cross-platform languange like Java, rather than use what used to laughingly referred to as "Eight Megs And Constantly Swapping" but has, over time, become lean, mean, and definitely user-friendly (although it's picky about who it's friends are).
And pimping a Java package is just absurd: Java, despite being less than 15 years old, already has a reputation similar to Cobol: it's a downslope language if ever there was one: you get paid big bucks to write in it because it's a boring language popular with big, faceless corporations with a lot of legacy work, not because it's an interesting and career-worthy language that excites developers. Every Java programmer I know would rather be working in something more spicy and productive like Python, Ruby, or (Gods help us all) Haskell, or getting down closer to the heavy metal and doing high-performance C/C++ stuff. Java is a hellish limbo with neither the "portable assembly" of C nor the developer productivity curve of modern languanges.
And as for IDEs, I find Emacs is simply more productive than anything else. Gosling's wrong about how "the only thing that's been added since 1984 is syntax highlighting"; he's never used the outlining, organizing, mind-mapping, revision-tracking and indexing tools. You can take Emacs out of my cold, dead hands only when you present me with something better.
Heck, my LJ entries are written with Emacs.
As I see it, what he's saying is that one should tackle a steep learning curve and adopt a power-hungry monstrosity written in a non-cross-platform languange like Java, rather than use what used to laughingly referred to as "Eight Megs And Constantly Swapping" but has, over time, become lean, mean, and definitely user-friendly (although it's picky about who it's friends are).
And pimping a Java package is just absurd: Java, despite being less than 15 years old, already has a reputation similar to Cobol: it's a downslope language if ever there was one: you get paid big bucks to write in it because it's a boring language popular with big, faceless corporations with a lot of legacy work, not because it's an interesting and career-worthy language that excites developers. Every Java programmer I know would rather be working in something more spicy and productive like Python, Ruby, or (Gods help us all) Haskell, or getting down closer to the heavy metal and doing high-performance C/C++ stuff. Java is a hellish limbo with neither the "portable assembly" of C nor the developer productivity curve of modern languanges.
And as for IDEs, I find Emacs is simply more productive than anything else. Gosling's wrong about how "the only thing that's been added since 1984 is syntax highlighting"; he's never used the outlining, organizing, mind-mapping, revision-tracking and indexing tools. You can take Emacs out of my cold, dead hands only when you present me with something better.
Heck, my LJ entries are written with Emacs.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2008-03-04 05:49 pm (UTC)For example, does it mean that Haskell is unusually spicy? unusually productive? Unusually obscure? A worse choice than Python or Ruby?
Bias: I know a little Haskell and Python, and they seem nice enough; I don't know any Ruby.
no subject
Date: 2008-03-04 08:28 pm (UTC)This all goes back to Perl, and Larry Wall's attempt to make a language that was as easy to manipulate as a shell script, as powerful as anything available in C, and as expressive as a linguist could make it.
Guido saw Perl and saw that it was... okay. But it was too big and ugly for the average human brain, and thus Guido created Python. Python fits your brain, as Pythonistas like to say. By taking "what you can do" and creating "one really good and obvious way to do it," Python cut down the cognitive overhead of debugging and maintainence, and that's a good thing.
Matz created Ruby because he wanted more expressive power, and he wanted the core of Scheme to be at the heart of it. So ruby has blocks and closures and currying and all the superpowers of Scheme, and it has a Python/Perlish look and feel. Ruby mostly fits your brain
Haskell, in my experience, forcibly makes room in your brain by shoving your cognition around until something clicks. My impression of Haskell has always been a painful one. Maybe someday I'll wrap my head around what Haskell is trying to express.