[TLUG]: ECMAScript ("Javascript") Version 4 - FALSE ALARM

[Mark Miller]

On 10/28/07, Mark Miller <erights at gmail.com> wrote:
> .fromJSONString() is inappropriate as a method of String.

Oops. I meant .parseJSON().



On 10/28/07, Robert Sayre <sayrer at gmail.com> wrote:
> On 10/28/07, Mark Miller <erights at gmail.com> wrote:
> >
> > However, Rebecca is related to the evil Allan of Microsoft, so perhaps
> > responsibility based design is part of some evil corporate plot? Or
> > maybe we should evaluate the logic of what people are saying
> > independent of their corporate affiliation?
>
> This reads like a flame to me. I think every message I've sent has
> suggested, you know, pointing out technical flaws when you criticize a
> proposal. [..]

Fair enough. So long as others discuss technical issues and do not
impugn the motives of others, I will do likewise. But there's one
issue which leaves me puzzled:

> * Temper tantrums about the name.
>   (there's really nothing to negotiate, aiui)

I don't understand your point. Both positions:
* advocating that the name stay the same
* advocating that the name be changed
cannot be resolved by technical arguments. The issue isn't technical.
When you say "there's really nothing to negotiate", I'm not sure what
you mean. Are you suggesting that we avoid deciding the name for the
new language, and confine ourselves instead to deciding only its
technical content? In any case, I agree that we can at least postpone
these non-technical issues for now.


> I understand that you want to explore more restrictive subsets, and
> maybe some conservative extensions. That's great. I think you should
> call it whatever you want, and it could work very well. Does ES4
> redefine your subset? Doesn't seem so, except in a few corner cases
> that are undeniable bugs in ES3. None of the scary new features are
> involved. Why is this a problem?

Regarding Caja, it might be no problem at all. I don't know yet. I
have now read the section of the overview document titled
"Compatibility" and every embedded bold "Note:". Congratulations.
Seriously. So far, I have spotted one thing which surprised me, but
nothing Caja can't live with and several things that are helpful. If
this text really does represent all the compatibility gotchas when
switching from ES3 to the proposed language whose name is not
negotiable, then the introduction of this language will not disrupt
Caja. It looks like the Caja translator can easily continue to
restrict Caja programs to an enforced subset of ES3, or close enough.

As time permits, over the next week, I will read the remainder of the
document carefully and let you know if I spot any other technical
issues that might be problematic for Caja.

-- 
Text by me above is hereby placed in the public domain

    Cheers,
    --MarkM