W3C home > Mailing lists > Public > public-iri@w3.org > May 2010

Re: [bidi] Re: Special ordering for BIDI URLs

From: Matitiahu Allouche <matial@il.ibm.com>
Date: Mon, 31 May 2010 10:21:13 +0300
To: Adil Allawi <adil@diwan.com>
Cc: "aharon@google.com" <aharon@google.com>, "bidi@unicode.org" <bidi@unicode.org>, bidi-bounce@unicode.org, Mark Davis ☕ <mark@macchiato.com>, Murray Sargent <murrays@exchange.microsoft.com>, Nasser Kettani <Nasser.Kettani@microsoft.com>, "public-iri@w3.org" <public-iri@w3.org>, Shawn Steele <Shawn.Steele@microsoft.com>
Message-ID: <OFD7B2C3A9.676514C8-ONC2257734.00259E70-C2257734.00286569@il.ibm.com>
On May 31st, Adil Allawi wrote:

<quote>
Right now, in my iPhone, I have a Bidi-URL-aware user agent that lets me 
type and paste in Arabic URLs and even has a nice interface for switching 
the text direction of the URL to RTL.
</quote>
For those on the list who do not read Arabic, could you show us in 
pseudo-Bidi (upper case for Arabic letters) how the URL is displayed in 
LTR direction and in RTL direction?

<quote>
With this suggestion there is going to exist a number of years where older 
non-formatting agents will exist beside newer ones. If the newer agents 
order text differently to the older agents this has the potential to cause 
a significant amount of user confusion.
</quote>
Without the suggestion, the confusion already exists since URLs mixing LTR 
and RTL parts (which are all URLs in Arabic because of the presence of 
"http") are inherently confusing if a uniform direction convention is not 
adopted.
For instance, consider "http://123.CIBARA"

<quote>
There will also be no way at the server end of the URL process to know if 
a web request was directed from a potentially spoofed address.
</quote>
I don't think that servers try to detect any potentially spoofed address, 
Bidi-inspired or other.

<quote>
At some level the URLs will be stored and transferred from computer to 
computer with the control characters still embedded. Whether it is after a 
user copies a URL from the address bar or URLs are stored in systems that 
need the control characters to display correctly. Problems will occur in 
the transfer between aware and unaware applications. </quote>
An unaware application will transfer URLs without formatting characters. 
The aware application is supposed to recognize that a URL is involved and 
will add the formatting characters for presentation.
An aware application will transfer URLs with formatting characters.  When 
an unaware application uses the string for display, all is well (assuming 
the unaware application implements correctly the UBA, otherwise all bets 
are off).  When the unaware application wants to send the URL to a server, 
it may identify the formatting characters as invalid (from its point of 
view) and strip them, and all is well; it may flag the URL as invalid and 
notify the user, who will have to switch to an updated version of the 
application, which basically is a good thing; it may send the URL as is, 
and the server will respond with an error code, which should also alert 
the user that something is amiss and incite him/her to update the 
application.


Shalom (Regards),  Mati
           Bidi Architect
           Globalization Center Of Competency - Bidirectional Scripts
           IBM Israel
           Phone: +972 2 5888802    Fax: +972 2 5870333    Mobile: +972 52 
2554160
Received on Monday, 31 May 2010 07:21:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 April 2012 19:51:57 GMT