Re: [Cellar] [Matroska-Org/ebml-specification] IANA considerations for Element ID allocation (#179)

[Michael Richardson <[e-mail]>]

I am responding to comments in github by Dave Rice.

I have fixed the backtick for references and the , for thousands.
If you want to merge the paragraphs, I will do that once I'm finished
editing, because I find it annoying to edit that way (and it makes for
less understable diffs).

I have fixed the off-by-one on 268,435,456.

> Also perhaps these min/max values should be integrated into the table in
> the {{#element-id}} section

That's a good idea.

> Should define the ascii range specifically here. For instance the lower
> three bytes of the EBML Element are 0x45DFA3 which is extended ascii for
> Eߣ (~'ebml')

If we permit other than 7-bit US-ASCII to be reserved, then the discussion
will then be latin-1,latin-2,windows-XXXX, or UTF-8.  If you want 0x45DFA3,
then it's also available by First Come First Served, it's just not reserved.

> perhaps this paragraph and the prior could be merged

my opinion is that it more understandable this way.

> +Other Four Byte Element IDs may be allocated by First Come First Served.
> how?

Please see https://tools.ietf.org/html/rfc8126#section-4.4

>+Five Byte Element IDs (values from 268435457 upwards) are reserved for
>+Experimental use.
>
>that only leaves two available non-experimental five byte element ids.

I don't think it leaves any: it's all experimental.
Do we really need that any Five Byte Element IDs in specifications?







-- 
Michael Richardson <[e-mail]>, Sandelman Software Works
 -= IPv6 IoT consulting =-