<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Dec 4, 2015, at 11:16 AM, Larry press <<a href="mailto:larrypress@gmail.com" class="">larrypress@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div class="">Folks,<br class=""><br class="">The NSF Web site says "In March 1991, the NSFNET acceptable use policy<br class="">was altered to allow commercial traffic:"<br class=""><br class=""><a href="http://www.nsf.gov/od/lpa/nsf50/nsfoutreach/htm/n50_z2/pages_z3/28_pg.htm" class="">http://www.nsf.gov/od/lpa/nsf50/nsfoutreach/htm/n50_z2/pages_z3/28_pg.htm</a><br class=""><br class="">But this statement of the AUP, dated June 1992, contradicts that:<br class=""><br class="">https://w2.eff.org/Net_culture/Net_info/Technical/Policy/nsfnet.policy<br class=""><br class="">When was the AUP changed to allow commercial traffic?</div></div></blockquote><br class=""></div><div>Mumble. There were significant changes in 1991 and 1992 which affected</div><div>commercial traffic on the NSFNET.</div><div><br class=""></div><div>(This is actually an example of where the wikipedia article is fairly complete:</div><div> <<font color="#4787ff" class=""><u class=""><a href="https://en.wikipedia.org/wiki/National_Science_Foundation_Network#Commercial_traffic" class="">https://en.wikipedia.org/wiki/National_Science_Foundation_Network#Commercial_traffic</a></u></font>>)</div><div><br class=""></div><div><div>Prior to the NSFNSF AUP change in 1991, organizations connecting that were clearly </div><div>commercial in nature would be asked (either by NSF, Merit, or the NNSC at BBN) </div><div>whether their usage was going directly in support of research (e.g. DECWRL, HP </div><div>Labs, etc.), as it was generally not assumed to be the case.</div><div><br class=""></div><div>Several months after the 1991 NSFNET AUP change, I know that the NNSC was </div><div>told that interconnection of commercial entities could be assumed to be in </div><div>support of research and educational purposes (I presume the Merit folks were </div><div>told the same.)</div></div><div><div><div><br class=""></div><div>Rick Adam’s reflects on this change in the state of affairs (pre-1991 vs post-1991 </div><div>NSFNET connection) in an interesting-people email here: <<a href="http://archive.is/vTrkS" class="">http://archive.is/vTrkS</a>></div><div><br class=""></div></div></div><div>Even after the NSFNSF AUP change, a purely commercial network was not supposed </div><div>to be routed over the NSFNET, and this was something that the newly formed ANS CO+RE </div><div>folks took care to remind NSFNET regional networks about… (all while coincidentally </div><div>reminding them of the availability of their commercial packet routing over the same </div><div>infrastructure, aka, COMBits <<a href="ftp://ftp.cs.toronto.edu/doc/net.policy/ans-plan.txt" class="">ftp://ftp.cs.toronto.edu/doc/net.policy/ans-plan.txt</a>>).</div><div><br class=""></div><div>While there was some purely commercial interconnection with the NSFNET, this</div><div>was predominantly via the brokered interconnection with the CIX and commercial</div><div>networks not participating in the CIX were not routed on the NSFNET infrastructure</div><div>in general (e.g. <<a href="http://www.merit.edu/mail.archives/mjts/1992-06/msg00015.html" class="">http://www.merit.edu/mail.archives/mjts/1992-06/msg00015.html</a>>).</div><div><br class=""></div><div>Minor routing issues related to purely commercial use of the NSFNET continued to </div><div>come up from time to time until the transition over to 5 commercial backbones </div><div>[UUNET, PSI, BBB, MCI, Sprintlink] architecture (aka “vBNS/RA/NAP”) in 1994.</div><div><br class=""></div><div>/John</div><div><br class=""></div></body></html>