<br><br><div class="gmail_quote">2009/9/12 Kent Karlsson <span dir="ltr"><<a href="mailto:kent.karlsson14@comhem.se">kent.karlsson14@comhem.se</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div class="im">
<font color="#800080"><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"><br>
</span></font></font><blockquote><blockquote><font color="#800080"><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">However, I'm not arguing for actually doing this.<br>
</span></font></font></blockquote></blockquote></div><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"><br>
<font color="#800000">...for BLEU scores.<br>
</font><br>
</span></font><blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;">I am relieved :)<br>
</span></font></blockquote><font face="Calibri, Verdana, Helvetica, Arial"><span style="font-size: 11pt;"><br>
<font color="#800000">I would be more supportive of a "translation status" extension (without BLEU, or other, scores).<br>
(And the RFC, if any, can very well be developed outside of the group on this mailing list...<br>
Formally it is subject to IETF review, not (formally) review here.)<br></font></span></font></div></blockquote><div><br><br>I am repeating myself, but I hope that the feedback from Yves Savourel makes clear that such information would be better in other metadata fields. If someone wants to develop such an RFC, they would need to co-ordinate a lot with others, to avoid a mess of relations between data categories in different fields. But it seems that nobody is really pushing hard into this direction anyway.<br>
<br>Felix<br></div></div>