jm + ietf   4

Paul Vixie's answer to "was DNS intentionally designed to be insecure?"

no. nor ip itself, or ncp which preceded it, or tcp, or udp, or icmp, or smtp,
ot http. it was insecure because it evolved in a safe, germ free academic
bubble. absolutely none of it was designed with billions of people in mind, or
the full cross section of humanity which would include criminals and national
intelligence services. the world of the internet in 2019 would have been seen
as a total freak show by the community who deployed dns in the 1980's.

nothing that can be abused won't be. you may or may not believe this; it's
considered controversial, and there are arguments being had about it today.

but noone considered that now-controversial near-truism at all when the core
internet protocols were first designed and implemented. the idea of abuse was
considered novel in the 1990's when commercialization and privatization
brought abuse into the internet world and burst the academic bubble. a lot of
old timers blamed AOL and MSN and even Usenet for the problems, but in
actuality, it's what humans _always_ do at scale. putting the full spectrum of
human culture atop a technology platform designed for academic and
professional culture should have been understood to be a recipe for disaster.
ietf  computers  abuse  internet  security  dns  paul-vixie  history  scale  culture 
4 weeks ago by jm
Can HTTP/2 Replace MQTT?
MQTT definitely has a smaller size on the wire. It’s also simpler to parse (let’s face it, Huffman isn’t that easy to implement) and provides guaranteed delivery to cater to shaky wireless networks. On the other hand, it’s also not terribly extensible. There aren’t a whole lot of headers and options available, and there’s no way to make custom ones without touching the payload of the message.

It seems that HTTP/2 could definitely serve as a reasonable replacement for MQTT. It’s reasonably small, supports multiple paradigms (pub/sub & request/response) and is extensible. Its also supported by the IETF (whereas MQTT is hosted by OASIS). From conversations I’ve had with industry leaders in the embedded software and chip manufacturing, they only want to support standards from the IETF. Many of them are still planning to support MQTT, but they’re not happy about it.

I think MQTT is better at many of the things it was designed for, but I’m interested to see over time if those advantages are enough to outweigh the benefits of HTTP. Regardless, MQTT has been gaining a lot of traction in the past year or two, so you may be forced into using it while HTTP/2 catches up.
http2  mqtt  iot  pub-sub  protocols  ietf  embedded  push  http 
february 2015 by jm
IAB Statement on Internet Confidentiality
Newly designed protocols should prefer encryption to cleartext operation. There may be exceptions to this default, but it is important to recognize that protocols do not operate in isolation.  Information leaked by one protocol can be made part of a more substantial body of information by cross-correlation of traffic observation.  There are protocols which may as a result require encryption on the Internet even when it would not be a requirement for that protocol operating in isolation.

We recommend that encryption be deployed throughout the protocol stack since there is not a single place within the stack where all kinds of communication can be protected.

Wow. so much for IPSec
ipsec  iab  ietf  snowden  surveillance  crypto  protocols  internet 
november 2014 by jm
IETF expedited publication of RFC6449 before J.D. Falk passed away
I had no idea JD was sick. Very saddened to hear about this, he was a nice guy and a great member of the anti-spam community :(
jd-falk  death  cancer  rfcs  ietf  anti-spam  people 
november 2011 by jm

Copy this bookmark: