[Javascript] Literal array declarations

[Paul Novitski]

At 10/29/2009 12:25 PM, Scott Reynen wrote:
>It's not preferred by everyone, but it is preferred by Douglas
>Crockford, the author of JSLint, and here's a quote from one of his
>books on why he prefers it:
>
>http://stackoverflow.com/questions/383402/is-javascript-s-new-keyword-considered-harmful/383423#383423
>
> > If you forget to include the new prefix when calling a constructor
> > function, then this will not be bound to the new object. Sadly, this
> > will be bound to the global object, so instead of augmenting your
> > new object, you will be clobbering global variables. That is really
> > bad. There is no compile warning, and there is no runtime warning


If omitting the "new" in a "new Array()" construction is bad form, 
why not check for that omission specifically instead of forbidding 
use of the syntax altogether? I don't mean this as a rhetorical 
question; I'd really like to understand Douglas' perspective.

There are many language features in JavaScript that would cause 
problems of one sort or another if entered incorrectly. We don't 
avoid them, we learn to code correctly. Jslint and validators in 
general can be a wonderful help by proofreading our work in detail. 
It's when jslint objects to the use of perfectly valid syntax & style 
not because the code is incorrect or causing a problem but because 
its misuse might cause a problem in some other script someday that 
jslint's value as a helper comes into question. I'm seeing jslint not 
as a validator, a proofreading assistant in everyday scripting, but 
rather as a set of training wheels whose mission is to teach solid 
coding practices but that is too strictly idiosyncratic to stay out 
from underfoot. I would welcome an argument that would soften this perspective.

Regards,

Paul
__________________________

Paul Novitski
Juniper Webcraft Ltd.
http://juniperwebcraft.com