W3C home > Mailing lists > Public > uri@w3.org > August 2002

Fwd: Re: Deliciously pedantic: what's the plural of URI?

From: Al Gilman <asgilman@iamdigex.net>
Date: Tue, 13 Aug 2002 20:45:15 -0400
Message-Id: <>
To: uri@w3.org

[non-architectural thread followed on this list per Roy's suggestion.]

>Resent-Date: Fri, 9 Aug 2002 22:11:07 -0400 (EDT)
>Date: Fri, 9 Aug 2002 14:47:36 -0700
>Cc: www-tag@w3.org
>To: Tim Bray <tbray@textuality.com>
>From: "Roy T. Fielding" <fielding@apache.org>
>X-Mailer: Apple Mail (2.482)
>Subject: Re: Deliciously pedantic: what's the plural of URI?
>Resent-From: www-tag@w3.org
>X-Mailing-List: <www-tag@w3.org> archive/latest/2013
>X-Loop: www-tag@w3.org
>Sender: www-tag-request@w3.org
>Resent-Sender: www-tag-request@w3.org
>List-Id: <www-tag.w3.org>
>List-Help: <http://www.w3.org/Mail/>
>List-Unsubscribe: <mailto:www-tag-request@w3.org?subject=unsubscribe>
>>I note that Roy of late has been using URI as its own plural.
>Of late?  I've been doing that for over five years.  See RFC 2396.
>>   Elegant and defensible, but I prefer URIs as less surprising to the eye.
>>   Even more, I prefer consistency.  Clearly this is a subject on which 
>> consensus is not remotely possible... -Tim
>I prefer whichever one is easier to say while speaking, since I do not
>believe in the theory that people expand acronyms as they read.  But if
>others feel strongly about, they can always put it up as an issue on

** short

Use the 's.'

On this one, Roy, I think you lose.

In text-to-speech mediated delivery contexts (including screen readers) you 
should expand, or at least the user should have the option to force 
expansion.  In addition in this case you want to expand the singular and 
plural cases differently.

At least in terms of the current working estimate of the method by which 
you know how to expand something, this requires that the singular and 
plural symbols to be expanded be different strings.  Ergo add -s to get 
plural: URIs.

** long

In situations where the text is converted to speech automatically, there is 
a clear case for expanding the words of an initialism, if the initialism is 
not a known word in general use with a pronunciation in the dictionary of 
the text-to-speech system.  It is better to expand than simply to spell or 
to guess a pronunciation from natural language heuristics.
This is actually a pretty big act of chutzpah to assert this as obvious, 
but I want you to trust me on this.

The tradition is that this is done in the text at first use of the [acronym 
or other initialism, if you make a distinction] within the published work, 
but on the Web this
should be rewritten to the first time this symbol is encountered in a 
session reading a given corpus of content, which may be anywhere in said 
corpus (you don't necessarily first encounter the textually-first instance.

The following is piety in the sky -- nobody implements it yet -- but the 
suggestion of the moment in the accessibility guidelines is that one use 
the 'title' attribute on the 'acronym' elment to provide the 
expansions.  At the textually first instance of using the symbol.  Not all 
uses are marked as uses of a defined symbol, it is left to the tools to 
recognize this by lexing out the tokens and string matching against the 
known expandable tokens.  As I say, the status as of this moment is 
un-proven pious theory; but it is the consensus pious theory of those in 
the W3C who have been tasked to worry about this, the Web Content 
Accessibility Guidelines Working Group.

In this text-to-speech mediated mode of engagement, which has a definite, 
or may I say notorious role in accessibility for people with disabilities, 
there is a clear motivation to distinguish the expansions into singular and 
plural cases because it will throw comprehension off to put in language 
errors such as missing plurals in the synthesized speech.  Takes mind share 
from the user to double-check and repair the grok, which forces them to 
operate the general rate of the TTS slower.  A drag.

So to keep the rules for the TTS delivery path simple, please let us 
distinguish the singular and plural of URI with different spellings, as is 
the regular thing (most commonly followed rule) in English which is the 
native language of the documents in which URIs are defined.

<end of speech/>


Received on Tuesday, 13 August 2002 20:45:26 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:04 UTC