exploration report on encryption in IPLD #348
Conversation
|
Prototype code for some of this already is pushed and reviewable in multiformats/js-multiformats#59 , for anyone who likes a more literal and fully worked thing to grind over. |
| and works in the natural way an IPLD codec is expected to work. | ||
|
|
||
| (This is a big change in stance. | ||
| Previously, we've considered it unclear whether codecs) |
There was a problem hiding this comment.
whether codecs ..... "are the right place to do this"?
| and so this document might be nearly the first description of them: | ||
|
|
||
|
|
||
| ### encryption codecs use multicodec indicators |
There was a problem hiding this comment.
this section might want to be updated - the current approach is to have a single new ipld codec which encodes the cipher as a multicodec integer, so we have one new codec (for now) and a bunch of ciphers in the table.
warpfork
force-pushed
the
encryption-report
branch
from
7ef4620 to
32669e4
Compare
|
I updated this to end with a note that discussion is ongoing in the other PR. I'd like to freeze the rest as "exploration report". (The filename has January in it; it's an accurate statement about where thoughts were in January. I'm unlikely to further revisit this anytime soon unless there's a fresh impetus.) I'll merge this tomorrow if there's no objections, just to get it out of the list of work that looks like it's in-progress. |
I've written a short (sorta) informal (relatively) exploration report on recent conversations we've had about encryption in IPLD.
Like any exploration report, this is probably incomplete, it's single author, it's subject to change, it's not a totally definitive stance of the entire IPLD team yet, etc. But... we might take this content and move it towards being something more ratified and serious, and perhaps eventually into a spec.
There's some other conversations going on about encryption right now in various venues, so I wanted to scrape this together to maybe give a home to accumulating some shared vocabulary about the more advanced concepts being discussed. We've also had some interesting new ideas recently about how to connect encryption primitives to the rest of the IPLD universe, so those are discussed here.
Comments welcome on this, of all sorts. Usually an exploration report is meant to stay a single author thing and go on the record that way, but in this case, I'd love lots of feedback of all kinds, and will also probably edit this to include notes from some meetings over the next couple of day.