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

Re: XHR vs JSON, was: Next Steps on JSON + Proposed TAG Resolution

From: Bjoern Hoehrmann <derhoermi@gmx.net>
Date: Fri, 18 Oct 2013 19:55:16 +0200
To: Anne van Kesteren <annevk@annevk.nl>
Cc: www-tag <www-tag@w3.org>
Message-ID: <hpp269h5a8orbv5e9ectfcv5els12ho5gs@hive.bjoern.hoehrmann.de>
* Anne van Kesteren wrote:
>On Fri, Oct 18, 2013 at 4:25 PM, Bjoern Hoehrmann <derhoermi@gmx.net> wrote:
>> There is only one factual claim above. Text formats that use Unicode
>> signatures are the exception and not the rule.
>Not for those produced and consumed by web platform clients.

I do not know what "web platform clients" are, and Google finds no hits
for that term on w3.org, neither in the singular nor the plural, but in
addition to JSON there are many formats that might qualify as counter-
example. A simple example is MHTML; text/plain is a rather problematic
one, robots.txt, any number of web browser configuration file formats,
some plain US-ASCII formats like various certificate formats and formats
like multipart/x-mixed-replace and application/x-www-form-urlencoded...

Not supporting Unicode signatures is a perfectly valid design choice; in
many cases they can be rather problematic, for instance, you would have
to have special code that ensures the first column header in CSV-style
formats cannot be mistaken for a Unicode signature which makes it hard
to use Unicode signatur-agnostic tools on them as is commonly done. Same
for other operations like concatenating two files where signatures would
have to be stripped. It is not uncommon that machine-generated scripts
and HTML documents include JSON-encoded data verbatim from a file or a
web service response where people "forget" to strip Unicode signatures,
unsurprisingly so, considering that JSON does not support them at all.

So, that JSON should support Unicode signatures since "All text formats
allow a BOM." is a bit far off. Be that as it may, the suggestion was
that the TAG works this out with W3C's Web Applications Working Group,
and other groups as necessary, so the JSON community suffers fewer JSON
variants. I have suggested the same to the IETF JSON Working Group.
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
Am Badedeich 7 · Telefon: +49(0)160/4415681 · http://www.bjoernsworld.de
25899 Dagebüll · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/ 
Received on Friday, 18 October 2013 17:55:38 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:33:22 UTC