- From: Marc Blanchet <marc.blanchet@viagenie.ca>
- Date: Fri, 31 Mar 2023 08:15:17 -0400
- To: Martin Thompson <mt@lowentropy.net>
- Cc: Mark Nottingham <mnot@mnot.net>, HTTP Working Group <ietf-http-wg@w3.org>, Tommy Pauly <tpauly@apple.com>
> Le 31 mars 2023 à 04:46, Martin Thomson <mt@lowentropy.net> a écrit : > > Hey Marc, > > I wonder why RFC 9292 is not appropriate here. That format is designed to be compact, which would very much suit a simple mapping. Your suggestion was also raised by a participant of the DTN wg. I was waiting this week to make the presentation on the current draft to see where we are going and then take into account all comments such as this one. It seems to fit, but I haven’t looked closely. So potentially yes. > > On the other hand, if space efficiency[1] is important, some amount of compression might be appropriate. However, I'd suggest that cutting back on what is included in the message is better than compression and - for this domain - that seems like a better approach given likely endpoint capacity constraints, robustness requirements, and overall simplicity. Yeah. While size is obviously important, energy is also. In the primary use case of dtn, compression do happen at the lower layers of the stack (where there is hardware acceleration), so we don’t need to do anything at this layer. Marc. > > [1] ...or size efficiency, but you know, puns > > On Fri, Mar 31, 2023, at 17:20, Mark Nottingham wrote: >> HTTP folks, >> >> Marc Blanchet (CC:ed) has a draft which we should be aware of: >> https://datatracker.ietf.org/doc/draft-blanchet-dtn-http-over-bp/ >> >> In a nutshell, this is a mapping of HTTP semantics onto the "bundle >> protocol" as an alternative transport protocol. Bundle is designed for >> operation in high-delay networks (hence the WG name, Delay Tolerant >> Networks) such as space. >> >> Because of the types of organisations that do procurement for things >> like space travel, this is likely to need to be standards-track. >> >> (Marc, anything to add?) >> >> From a HTTP WG perspective, we should come to agreement about how >> involved we want to be in this work. At a minimum, we'll do at least >> one HTTPDIR review. Do people feel more is needed -- e.g., closer >> coordination after this is adopted by DTN? >> >> Cheers, >> >> -- >> Mark Nottingham https://www.mnot.net/
Received on Friday, 31 March 2023 12:28:04 UTC