Well, of course, there's a unicode consortium description of what
'standard' means in the context of unicode:

https://unicode.org/standard/standard.html

But really, I was thinking more about the distinctions drawn here:

http://www.unicode.org/reports/about-reports.html#Types

That said, since we're talking about errors, it's perhaps more
relevant to refer unicode documents on that issue. Like, perhaps,

http://unicode.org/L2/L2002/02298-j1n6815.pdf

(and then try to assign J specific meanings to its terminology. For
example, a unicode 'octet' might be a J 'literal' or more
specifically, an item of a. or something which maps to one of those
values, while a unicode 'byte' might be a  J 'atom'.  And, so on, ...
but  there's a lot to plow through... anyways, the point I was trying
to get at was that a lot of what it says about errors can fit onto J
in a variety of ways.)

Thanks,

-- 
Raul

On Fri, Sep 13, 2019 at 2:01 PM 'robert therriault' via Programming
<[email protected]> wrote:
>
> Raul,
>
> I am pretty sure that the Unicode Technical Committee think it is a standard, 
> although it has developed over a number of versions as characters are added 
> or moved from private encoding to publicly defined encodings.
>
> https://www.unicode.org/versions/Unicode12.0.0/ch01.pdf
>
> Is that what you mean by overlapping standards?
>
> It is pretty clear that UTF-32 encoding would never accept surrogate pairs as 
> code units in any version.
>
> Cheers, bob
>
> > On Sep 13, 2019, at 10:50 AM, Raul Miller <[email protected]> wrote:
> >
> > On Fri, Sep 13, 2019 at 1:08 PM 'robert therriault' via Programming
> > <[email protected]> wrote:
> >> So, where does this leave us? Well, we are kind of... sort of... doing 
> >> unicode, but in the process of making the process convenient, we have 
> >> drifted from the actual unicode standard.
> >
> > Unicode is not a single standard, but a bunch of overlapping standards
> > and recommendations.
> >
> > (J itself also has a variety of standards and expectations, typically
> > representing refinements of ideas used elsewhere.)
> >
> > Anyways, it doesn't fit the design of Unicode for the core language to
> > implement all of the unicode standards. It's perfectly alright to have
> > library code implement errors, if that makes the underlying
> > implementation simpler. For example. And, if someone sees enough need
> > to implement those errors, of course.
> >
> > Or have I overlooked something important?
> >
> > Thanks,
> >
> > --
> > Raul
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to