W3C home > Mailing lists > Public > public-csv-wg@w3.org > February 2016

Comments. W3C Editor's Draft 10 February 2016

From: Dave Pawson <dave.pawson@gmail.com>
Date: Wed, 10 Feb 2016 15:46:27 +0000
Message-ID: <CAEncD4ei_h-FUrys=4PTrofSAq6kOpHRcXgsafYT=A0gur+woQ@mail.gmail.com>
To: public-csv-wg@w3.org
[w3c](http://w3c.github.io/csvw/primer/#introduction)

"@context": "http://www.w3.org/ns/csvw",  context? Sort of see it, but surely
'format' (possibly include a version?) more reasonable, since that's what
it defines? And why insist on json? Restrictive. That's application layer?
Seems unecessarily complex compared to first line (custom and  practice?)

Seems to be drifting into semantics? Too much so? (section 3.1).
Are you equating this with an xml format? If so keep the 'schema' layer
apart from this spec.  3.4 drifting towards xml-schema hell?


Why is 'usage' in here? Transformation - section 4? Inappropriate IMHO.

6.4 skips over non 'comma' separated columns. Rather too glib? Why not
spec the separator in the metadata as per Ex 117? Other characters do
have fair reasoning.

This section is pretty key - why so late in the document?

6.6
"As a publisher, you can control where processors look for metadata
for your CSV files by listing the locations to look at within the
/.well-known/csvm file on your server. " Why assume all CSV is served
on the web? #weak


regards

-- 
Dave Pawson
XSLT XSL-FO FAQ.
Docbook FAQ.
http://www.dpawson.co.uk
Received on Wednesday, 10 February 2016 15:46:56 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 10 February 2016 15:46:56 UTC