On Friday, September 2, 2011 5:26:58 AM UTC-7, xavierm02 wrote:
>
> I'll digg into it and see if I can make my API so that it is possible
> to get from one to the other. This way I would allow writing code with
> that API and it would be compiled to my API so that old browsers get
> it.
>
Good luck! The problem with back-porting features like this is that the 
syntax is foreign to outdated JavaScript engines. A bridged api that 
normalizes both would be welcome, so perhaps you can experiment with an 
improvement to the ES5 syntax? Believe me, folks will forgo a programming 
feature if the syntax is too difficult... just look at CCS gradients! The 
only way to use them is thru toolkits and generators! One caveat: though 
widely accepted, the ES5 construct is still a draft and you-know-who-soft 
could roll the wrong way.

> Even the ES5 Object constructor lets you handle non-mapped properties.
> I didn't understand what you mean by non-mapped...
> You mean having a default getter?
>
>  Yes. The ES5 Object exposes the low-level handling of undefined (or 
non-mapped) object members. So, when you call "foo.bar" and "bar" isn't a 
member of "foo", you could invoke a function that handles this 
non-mapped/undefined property. JavaScript engines have done this internally 
since the beginning. The ES5 syntax merely exposes this rigging for more 
control.

> > One last note: I'm phasing out my own use of GSet, as it's numerous 
> closures actually degrade performance.
>
> Obviously but unless you loop on the object, i can't be that much.
> You just need to store the value in a local variable when using the
> object from outside.
>
I'm glad you're experimenting with how to make smarter objects. See my 
response to Scott a moment ago, regarding performance. I would only add 
this: as a base for other code, "small everything" is key to libraries like 
yours. You're right to worry about optimization later, but don't assume a 
tank engine can fit in a Porsche.

- best,

bemson

-- 
To view archived discussions from the original JSMentors Mailman list: 
http://www.mail-archive.com/[email protected]/

To search via a non-Google archive, visit here: 
http://www.mail-archive.com/[email protected]/

To unsubscribe from this group, send email to
[email protected]

Reply via email to