Fwd: RFC 8369 on Internationalizing IPv6 Using 128-Bit Unicode

Philippe Verdy via Unicode unicode at unicode.org
Mon Apr 2 21:07:47 CDT 2018


For "ymojis" to be really private, they would have to be very secure, and
128 bit would not be enough:
"ymojis" should now be encoded with and least 512 bits and probably even
1024 bits (secure digital fingerprints), not 128 bits like MD5, or 160 bits
like SHA1.
So let's encode Unicode with 1024-bit codepoints to secure it !

2018-04-03 4:02 GMT+02:00 Philippe Verdy <verdy_p at wanadoo.fr>:

> Note: We're missing the definition of "ymojis", a safer alternatives of
> "umojis" (unknown), but that "you" can create yourself for use by yourself
> (i.e. private-use umojis) and whose meaning is not meant to be understood
> by anyone else than you, but which is warrantied to be understood by you,
> unlike umojis that are interchangeable and not so private (a common problem
> shared by PUAs in Unicode because of the "ConScript" registry which is a
> defacto standard for PUA!).
>
>
> 2018-04-03 3:49 GMT+02:00 Philippe Verdy <verdy_p at wanadoo.fr>:
>
>> It's fun to consider the introdroduction (after emojis) of imojis,
>> amojis, umojis and omojis for individual people (or named pets), alien
>> species (E.T. wants to be able to call home with his own language and
>> script !), unknown things, and obfuscated entities. Also fun for new
>> "trollface" characters. In fact you could represent every individual or
>> even single atom in the universe that has ever created since the BingBang !
>>
>> But unlike peoples and social entities, characters to encode don't grow
>> exponentially but still linearily at a slowing speed. Unicode characters
>> are not exploding like Internet addresses (organizations, users, computers,
>> phones: the IPv4 space expoloded only because the equipement rate of people
>> accelerated but now it is slowing down with high equipment replacement
>> rate, and only the explosion of IoT continues to drive some growth but it
>> will rapidly reach a cap, so even the IPv6 address space will never be
>> filled even if it will be much larger that the UCS encoding space; I can
>> expect a maximum in the range of ~300 billions devices at most with all
>> planet resources as global population will not be able to grow
>> exponentially and will necessarily cap, plus ~100 milions
>> services/organizations; all the rest will die and will be replaced and even
>> if we give a delay of 100 years before reusing addresses of died devices
>> and people in IPv6, this will leave lot of space, we'll never reach a small
>> fraction of the number of entities in the universe; we are also completely
>> unable to make any physical measurement with so many digits of precision:
>> even just 64 bit bit is really extremely large, but 128 bit was chosen in
>> IPv6 just to allow random allocation without needeing excessive centralized
>> management, an IPv6 address is even a good subtitute to the whole DNS
>> system and its overvalued black market of domain names: IPv6 is extremely
>> economic !)
>>
>>
>> 2018-04-03 3:06 GMT+02:00 Mark E. Shoulson via Unicode <
>> unicode at unicode.org>:
>>
>>> On 04/02/2018 08:52 PM, J Decker via Unicode wrote:
>>>
>>>
>>>
>>> On Mon, Apr 2, 2018 at 5:42 PM, Mark E. Shoulson via Unicode <
>>> unicode at unicode.org> wrote:
>>>
>>>> For unique identifiers for every person, place, thing, etc, consider
>>>> https://en.wikipedia.org/wiki/Universally_unique_identifier which are
>>>> indeed 128 bits.
>>>>
>>>> What makes you think a single "glyph" that represents one of these
>>>> 3.4⏨38 items could possibly be sensibly distinguishable at any sort of
>>>> glance (including long stares) from all the others?  I have an idea for
>>>> that: we can show the actual *digits* of some encoding of the 128-bit
>>>> number.  Then just inspecting for a different digit will do.
>>>>
>>>
>>> there's no restirction that it be one character cell in size... rendered
>>> glyphs could be thousands of pixels wide...
>>>
>>>
>>> Yes, but at that point it becomes a huge stretch to call it a
>>> "character".  It becomes more like a "picture" or "graphic" or something.
>>> And even then, considering the tremendohunormous number of them we're
>>> dealing with, can we really be sure each one can be uniquely recognized as
>>> the one it's *supposed* to be, by everyone?
>>>
>>> ~mark
>>>
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://unicode.org/pipermail/unicode/attachments/20180403/e9f2c987/attachment.html>


More information about the Unicode mailing list