Hi! > -----Original Message----- > From: Russ Housley <housley@xxxxxxxxxxxx> > Sent: Saturday, November 21, 2020 3:43 PM > To: IETF <ietf@xxxxxxxx> > Cc: Roman Danyliw <rdd@xxxxxxxx> > Subject: Re: TLS access Re: Call for Community Feedback: Retiring IETF FTP > Service > > > > > On Nov 20, 2020, at 12:10 PM, Salz, Rich > <rsalz=40akamai.com@xxxxxxxxxxxxxx> wrote: > > > >> Separate question: we were told for the newcomers slides to use > >> rfc-editor.org for official RFC references. How do we resolve that difference > of views? > > > >> I'm not knowledgeable enough to be the authority on this. > > > > Fair enough. Let's chase this down tho and figure out what it should be. > "There can be only one" definitive source. > > The definitive source is rfc-editor.org. The files are copied from there for > serving by ietf.org. To contextualize access to this non-authoritative data set via FTP, I've provided another chart to the "12 Days in the life of an IETF FTP server" at https://docs.google.com/document/d/1JAXspeaMWFl8ML3hSezFSM0VsJsHI4uyDlQ2dHip8jo/edit# Access to RFCs is 27% of all FTP traffic and represents 37% of all unique IP addresses. This new chart in the "What information are the users requesting?" section provides a distribution of requests and IP address across the top level directories. A few highlights: * 64% of all requests are for either an I-D or RFC (or 60% of all unique IP addresses, using no aggregation) * the next most popular category (28%) is /ietf which contains historic charter and minute information of WGs; most of the usage is syncing the directory * all other directories constitute < 8% of usage * multiple top-level directories (4) had only a single IP address access it in the sample period Regards, Roman
