RFC-327
by Darius Kazemi, November 23 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.
Bye DTP
RFC-327 is titled “Data and File Transfer workshop notes”. It's authored by Abhay Bhushan of MIT Project MAC and dated April 27, 1972.
The technical content
This RFC collects notes from the data and file transfer workshop that was announced in RFC-309. The meeting was well-attended by Network Working Group members from both coasts.
The first day of the workshop featured a series of talks, which I will pithily summarize here:
- Steve Crocker (UCLA): network standardization is good and saves everyone lots of work, here is some mathematical proof thereof
- Richard Winter (Computer Corporation of America): the Datacomputer exists, and it's meant for other computer programs to use rather than individual human users
- Alex McKenzie (BBN): TIP users exist and have unique limitations, please do not forget them
- Bob Braden (UCLA CCN): 1% of our total computation load is from ARPANET users, and we promise there will be an official protocol for all of this very soon
- Ray Tomlinson (BBN): description of CPYNET
- Abhay Bhushan (MIT Project MAC): we need a way to evaluate which protocols are best, here is my set of criteria for doing just that
The second day was a big discussion and there was a lot of consensus reached on data formats, transfer modes, and restarting in case of connection loss. The big takeaway for me is that the Data Transfer Protocol of RFC-171 is no longer its own protocol, and its functionality can be duplicated by using certain combinations of FTP modes and procedures.
Analysis
As I said in my post on the first Data Transfer Protocol RFC:
This is a weird proposal that seems very design-by-committee, and each of the operating modes seems like it needs to be its own protocol. It really looks like half the RFC is just an abstraction of the FTP proposal in RFC-114, and then it has operating modes 1 and 2 (bit stream and block mode) kind of bolted on to it.
I'm glad that my analysis at the time wasn't off: they did in fact acknowledge that DTP was really just a couple extra FTP-like operating modes in its own “protocol”.
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 ActivityPub, including a Node.js reference implementation, an RSS-to-ActivityPub converter, and a fork of Mastodon, called Hometown. You can support my work via my Patreon.