W3C home > Mailing lists > Public > www-tag@w3.org > October 2013

Re: [Json] FYI ECMA, W3C, IETF coordination on JSON

From: Peter F. Patel-Schneider <pfpschneider@gmail.com>
Date: Tue, 08 Oct 2013 12:14:36 -0700
Message-ID: <5254599C.6090604@gmail.com>
To: Jorge Chamorro <jorge@jorgechamorro.com>
CC: John Cowan <cowan@mercury.ccil.org>, Allen Wirfs-Brock <allen@wirfs-brock.com>, JSON WG <json@ietf.org>, Tim Bray <tbray@textuality.com>, "www-tag@w3.org" <www-tag@w3.org>, Larry Masinter <masinter@adobe.com>
Well, for starters, interchanging non-integers can lead to loss of precision, 
which can be dangerous.  Further, interchanging numbers without some notion of 
the limits of the receiver is also dangerous.

Second, JSON does not only use sequences of digits to represent numbers, and 
there are many JSON numbers that cannot be represented as a sequence of 
base-ten digits.

Third, there are many numbers that humans interchange that cannot be 
represented as finite sequences of digits, even if you allow also a decimal 
point and a negative sign, for example 1/7, pi, and the square root of two.  
(Perhaps the wording in the introduction is meant to allow infinite sequences 
of digits, but then using JSON for interchange is a bit difficult.)  My 
understanding is that limiting the representation of numbers to something like 
the JSON syntax is more of a computer thing than a human thing.

peter

On 10/08/2013 11:43 AM, Jorge Chamorro wrote:
> On 08/10/2013, at 20:26, Peter F. Patel-Schneider wrote:
>
>> The paragraph on numbers, see below, seems rather dangerous, as well as being incorrect.
> Is it? Why?
>
Received on Tuesday, 8 October 2013 19:15:07 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:33:21 UTC