<html><head></head><BODY bgcolor="#FFFFFF"><p><font size=2 color="#000000" face="Fixedsys">Kenneth,<br><br>I have removed the 421 response code and use an<br>empty response. This new draft will be posted<br>when I gather up a few more revisions.<br><br>I would like to get others comments on your XOVER<br>suggestion - what do others on the mailing list<br>think about this suggestion?<br><br>Nat<br><br>----------<br>> From: Kenneth M. Sternberg <<font color="#0000FF"><u>elf@spry.com</u><font color="#000000">><br>> To: <font color="#0000FF"><u>natba@microsoft.com</u><font color="#000000"><br>> Subject: draft-ballou-nntpsrch-03.txt<br>> Date: Thursday, May 29, 1997 2:33 PM<br>> <br>> <br>> 	I have been reading up on all the latest developments in NNTP and<br>> I was fascinated by the above draft. However, in reading it I came across<br>> a specified behavior that is not consistent with RFC 977 and most modern<br>> implementations as covered in draft-barber-nntp-imp-07.txt. <br>> <br>> 	Under section 5.1 you state that "A response of 421 indicates that<br>> there are no articles that match the search criteria." However, popular<br>> implementations documented in Barber (2.1.6, et. al.) specify that "If<br>> nothing is matched an empty list is returned, not an error." Writers of<br>> clients would probably prefer to receive a consistent behavior overall.<br>> <br>> 	I'm also concerned about the response of a SEARCH OVER command.<br>> Since some articles could have enormous cross-postings it might be<br>> preferable just to return the Message-ID. The newsgroup is just meta-<br>> information about the article, not a unique key to it, after all.<br>> <br>> 	Elf<br>> 	<br>> -- <br>> Elf M. Sternberg, Unix Systems Engineer <font color="#0000FF"><u>elf@spry.com</u><font color="#000000"><br>> Spry Consulting Group, CompuServe Internet Division 425.957.8000<br>> 3535 128th Ave SE FAX: 425.957.6000<br>> Bellevue, WA 98006, USA <font color="#0000FF"><u>http://www.compuserve.com</u><font color="#000000"><br>> PGP 2.6.2 Public Key Available <br>> <br>> For the 50 billionth week, life continues to be unfair. Death and<br>> suffering continue to be randomly distributed among the planet's<br>> life forms with such mitigating factors as solid community standing,<br>> good genetic background, and previous good works in no way taken into<br>> account. Despite the efforts of the Code of Hammurabi, the U.S. Bill<br>> of Rights, and People for the Ethical Treatment of Animals, life is<br>> expected to remain unfair into the foreseeable future.</p>
</font></font></font></font></font></font></font></font></font></body></html>