by Darius Kazemi, April 25 2019

In 2019 I'm reading one RFC a day in chronological order starting from the very first one. More on this project here. There is a table of contents for all my RFC posts.


RFC-115 is titled “Some Network Information Center Policies on Handling Documents”. It's authored by Richard Watson and Jeanne B. North (aka Reddy Dively and dated April 16, 1971.

The technical content

This document is about the Network Information Center, which is essentially a library (containing both physical and electronic records) held at Stanford Research Institute.

This RFC describes what a “NIC Number” is. If you've been reading original RFCs as you read along with this blog, you probably have seen that a document like RFC-90 is not just numbered “90” but is also “NIC 5707”. In short: when the NIC collects a new document that it decides is of use to the general network community, they apply a unique number to it for their filing system. Documents that the NIC might care about include but are not limited to “technical reports, RFC's, brief network memos, journal articles, and letters”.

The document describes a rather complex encoding system for records in their computerized catalog, but fortunately this is mostly just informational and most people on the network don't have to know any of this stuff. But it's basically a database schema before database schemas were… a thing. They have ways to ask a computer “tell me all the NIC documents that are stored at this physical location that were authored before a certain date.”

They go on to talk about having written programs that query this data and produce print catalogs that are sorted and filtered by various criteria.

One thing that the author of a document can do is call the NIC on the phone and ask them to pre-assign a NIC number to the document, which streamlines the whole process since that means the very first versions of a document that get distributed already have a NIC number. That means that every copy of the document in existence will have this reference number somewhere.

They also describe “functional documents”, which today we'd call a document with some kind of version control attached to it, for instance a wiki page. They describe it as “a document whose title and function remain constant, but whose contents can change.” The NIC does not currently support these kinds of documents as fully as they would like, but has plans in place to do so.

For substantial revisions, they will create a new document with a new NIC number, and there will be text that indicates that this is the updated version of a particular older document. There will be a document attached to it that lists every revision between versions.


My takeaway from this whole thing is: manual version control is hard and librarians are heroes.

Also: the NIC “strongly recommends” avoiding errata-style revision of documents, which I think is ironic since the modern RFC series uses an errata system.

Further reading

The NIC eventually became InterNIC, the main organization that allocated domain names until 1998 when ICANN took over.

The NIC records are currently stored at the Computer History Museum and you can view the guide to their full collection online. Some of it is scanned and available online, much of it is in paper boxes in their warehouse where it's available to researchers. (You can also pay a scanning fee and point them to a document in the guide and they will scan and put it online for the world!)

This interview with Elizabeth “Jake” Feinler, who ran the NIC for decades, is a great read.

How to follow this blog

You can subscribe to this blog's RSS feed or if you're on a federated ActivityPub social network like Mastodon or Pleroma you can search for the user “@365-rfcs@write.as” and follow it there.

About me

I'm Darius Kazemi. I'm an independent technologist and artist. I do a lot of work on the decentralized web with both ActivityPub and the Dat Project. You can support my work via my Patreon.