Protocol Action: 'Right-to-left scripts for IDNA' to Proposed Standard

[Abdulrahman I. ALGhadir]

Cleaned.
>> 3.  The requirement set for the BIDI rule
>> (...)
>> Label Uniqueness: No two labels, when presented in display order
>>      in the same paragraph, should have the same sequence of characters
>>      without also having the same sequence of characters in network
>>      order, both when the paragraph has LTR direction and when the
>>      paragraph has RTL direction.
>
>let ABC 3 different labels in network order A != B !=C
>let JKL their LTR display order J != K != L
>let MNO their RTL display order M != N != O
>
>given the requirement above :
>
>J = M           J = M = A
>and       =>
>K = N           K = N = B           JKL = ABC
>                                      =>
>K = N           K = N = B           MNO = ABC
>and       =>
>L = O           L = O = C
>
>                             L = N => L = K which is absurd
>suppose L != O =>  or
>                             L = M => L = J which is absurd

if ABC 3 different labels in network order A != B !=C
Then  ABC will be A.B.C LTR context
If you read these labels in RTL context then you'll start reading from right.(right?) 
A.B.C<-start from here 
Result = C.B.A(display order for RTL context draw C 1st then B ... etc)
So the assumption will be like this 
let JKL their LTR display order J != K != L
let ONM* their RTL display order M != N != O
and rest will be like this:

J = O           J = O = A
and       =>
K = N           K = N = B           JKL = ABC
                                      =>
K = N           K = N = B           ONM = ABC
and       =>
L = M           L = M = C

Now ~

From above JKL = ABC and ONM = ABC
This will result 
JKL=ONM (J=O=A , K=N=B, L=M=C)

I might be wrong thought.

-----Original Message-----
From: slim.amamou at gmail.com [mailto:slim.amamou at gmail.com] On Behalf Of Slim Amamou
Sent: 16/Feb/2010 4:52 PM
To: Patrik Fältström
Cc: John C Klensin; Amr Zaki; Shawn Steele; Aharon \(Vladimir\) Lanin; Michel Suignard; idna-update at alvestrand.no; Mark Davis ☕; Abdulrahman I. ALGhadir; Vint Cerf
Subject: Re: Protocol Action: 'Right-to-left scripts for IDNA' to Proposed Standard

On Mon, Feb 15, 2010 at 8:16 PM, Patrik Fältström <patrik at frobbit.se> wrote:
>
> On 15 feb 2010, at 18.54, Slim Amamou wrote:
> This is not possible to achieve as we have a requirement that the domain names are to always be passed around in logical order, as we do not know what (in free flowing text for example) what is the beginning and end of a domain name.

This is nice. It looks like I can mathematically demonstrate that
labels can only be in network order given that assertion.

> 3.  The requirement set for the BIDI rule
> (...)
> Label Uniqueness: No two labels, when presented in display order
>      in the same paragraph, should have the same sequence of characters
>      without also having the same sequence of characters in network
>      order, both when the paragraph has LTR direction and when the
>      paragraph has RTL direction.

let ABC 3 different labels in network order A != B !=C
let JKL their LTR display order J != K != L
let MNO their RTL display order M != N != O

given the requirement above :

J = M           J = M = A
and       =>
K = N           K = N = B           JKL = ABC
                                      =>
K = N           K = N = B           MNO = ABC
and       =>
L = O           L = O = C

                             L = N => L = K which is absurd
suppose L != O =>  or
                             L = M => L = J which is absurd


Note that this is a recursive demonstration : I demonstrated for N+1
labels which means that a domain name with any number of labels must
be displayed in network order to satisfy the requirement.

Please review this demonstration carefully, as I did not do maths
since 1996 (and I was not really good at it).

> And, this is one of the reasons various recommendations say that one should not mix RTL and LTR in domain names, labels etc. If you do not mix directionality, you will not get any problems, right?

Right. But this will be problematic since current CC-TLDs are RTL
including those of countries using RTL languages.

-- 
Slim Amamou | سليم عمامو
http://alixsys.com

-----------------------------------------------------------------------------------
Disclaimer:
This message and its attachment, if any, are confidential and may contain legally
privileged information. If you are not the intended recipient, please contact the
sender immediately and delete this message and its attachment, if any, from your
system. You should not copy this message or disclose its contents to any other
person or use it for any purpose. Statements and opinions expressed in this e-mail
are those of the sender, and do not necessarily reflect those of the Communications
and Information Technology Commission (CITC). CITC accepts no liability for damage
caused by this email.