W3C home > Mailing lists > Public > public-csv-wg@w3.org > October 2014

Important note on default charset in "text/csv" (RFC 4180)

From: Yakov Shafranovich <yakov-ietf@shaftek.org>
Date: Wed, 29 Oct 2014 21:16:11 -0400
Message-ID: <CAPQd5oQYer6T1yOzdPeL3iS7wnwZm7tqMZZuhnZFrykPQf4EOg@mail.gmail.com>
To: W3C CSV on the Web Working Group <public-csv-wg@w3.org>
I noticed the github issue Jeni posted earlier (#44) as well as issue #8 in
the model document (
http://www.w3.org/TR/2014/WD-tabular-data-model-20140327/). The issue is
that we want the default character set to be UTF-8 while RFC4180 when I
wrote it defines it as plain ASCII.

While going through the documents, it turns out that the default character
set for "text/csv" is actually now UTF-8. This change took effect when
RFC7111 which defines CSV fragments was approved. The CSV mime type now
consists of RFC 4180, RFC 7111 with a combined registration appearing here:


This means that while RFC 4180 does mandate ASCII, for standards purposes
on the IETF side, this has been changed and the default now is in fact

Received on Thursday, 30 October 2014 01:17:12 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:21:42 UTC