<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 5.5.2653.12">
<TITLE>RE: Media Type "text/csv": new draft (-02) and Last Call</TITLE>
</HEAD>
<BODY>
<P><FONT SIZE=2>Yakov,</FONT>
</P>
<P><FONT SIZE=2>That draft wording sounds excellent.</FONT>
</P>
<P><FONT SIZE=2>Now, if you were to ALSO mandate the presence of the header record, that would mitigate my concern that computer applications should be able to exchange CSV files without worrying about same field order (simply because the header conveys the field names, rather than needing an out-of-band, bilateral agreement). And, of course, presence of a header record lets a computer application determine immediately whether the supplied fields are likely to be what it expects. All good aids for careful validation.</FONT></P>
<P><FONT SIZE=2>Thank-you,</FONT>
<BR><FONT SIZE=2>Clyde</FONT>
</P>
<P><FONT SIZE=2>-----Original Message-----</FONT>
<BR><FONT SIZE=2>From: Yakov Shafranovich [<A HREF="mailto:research@solidmatrix.com">mailto:research@solidmatrix.com</A>]</FONT>
<BR><FONT SIZE=2>Sent: Wednesday, March 30, 2005 8:08 PM</FONT>
<BR><FONT SIZE=2>To: clyde.ingram@edl.uk.eds.com</FONT>
<BR><FONT SIZE=2>Cc: GK-lists@ninebynine.org; ietf-types@alvestrand.no</FONT>
<BR><FONT SIZE=2>Subject: Re: Media Type "text/csv": new draft (-02) and Last Call</FONT>
</P>
<BR>
<P><FONT SIZE=2>clyde.ingram@edl.uk.eds.com wrote:</FONT>
<BR><FONT SIZE=2>> </FONT>
<BR><FONT SIZE=2>> But the same cannot be said of automated computer-based applications, </FONT>
<BR><FONT SIZE=2>> where maintaining a strict count of generated and expected </FONT>
<BR><FONT SIZE=2>> Comma-Separated-Values per record is not only easy, but also allows for </FONT>
<BR><FONT SIZE=2>> an extra level of data validation: namely that a received record is </FONT>
<BR><FONT SIZE=2>> corrupt if it has too few or too many fields. This is where </FONT>
<BR><FONT SIZE=2>> standardisation in the format of the CSV records becomes appropriate </FONT>
<BR><FONT SIZE=2>> material for an RFC.</FONT>
<BR><FONT SIZE=2>> </FONT>
</P>
<P><FONT SIZE=2>The draft as it is written now (-03) does not mandate that the same </FONT>
<BR><FONT SIZE=2>number of fields need to appear on each line, mainly due to the fact </FONT>
<BR><FONT SIZE=2>that the draft is focusing on the MIME type registration. Would the </FONT>
<BR><FONT SIZE=2>following change to section 2, subsection 4 be sufficient to address </FONT>
<BR><FONT SIZE=2>your concerns:</FONT>
</P>
<P><FONT SIZE=2>"Within the header and each record there may be one or more fields, </FONT>
<BR><FONT SIZE=2>separated by commas. Each line should contain the same number of fields </FONT>
<BR><FONT SIZE=2>throughout the file. The last field in the record may not be followed by </FONT>
<BR><FONT SIZE=2>a comma. For example:"</FONT>
</P>
<P><FONT SIZE=2>Yakov</FONT>
</P>
</BODY>
</HTML>