Granularity of events (was Re: [PROPOSED WORK ITEM] Cryptographic Event Log)

I am happy to write up a more detailed evaluation comparing C2PA’s Content Credentials with CEL.

I think one key part – that would be good to discuss on its own merit, regardless of serialization - is about granularity of events.

Currently CEL only does one “event” per “entry”.  There is no model for grouped or related “events”.  I am not clear on all the use cases for CEL, but in ones that I am envisioning – especially in areas such as Agentic protocols – it is certainly a requirement to have multiple “events” as part of a single signed “entry” from a given “entity”.

@Manu Sporny<mailto:msporny@digitalbazaar.com> Is this something that the CEL team has discussed or considered?  If so, can you give insight into the chosen direction?

Leonard

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Saturday, July 19, 2025 at 11:36 PM
To: W3C Credentials CG <public-credentials@w3.org>
Subject: Re: [PROPOSED WORK ITEM] Cryptographic Event Log
EXTERNAL: Use caution when clicking on links or opening attachments.


On Sun, Jul 13, 2025 at 10:27 PM Michael Prorock <mprorock@mesur.io> wrote:
> Hate to ask you to spend more explanation time Manu, and appreciate the write up so far, but would you mind breaking out key differences you see from keytrans (and also SCITT from a more general digital notary perspective)?

Noting the feedback from Leonard, Daniel, Melvin, Christopher, Gabe,
and Dmitri regarding "Can you compare CEL to X", it's going to take
some significant effort to compare/contrast each of these
"cryptographic log" formats against each other. :)

I'll note that doing these comparisons was the very first issue raised
on the spec:

https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdigitalbazaar%2Fcel-spec%2Fissues%2F1&data=05%7C02%7Clrosenth%40adobe.com%7C0147651898ae4c19255e08ddc714bcfb%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C638885614069412491%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=ReazbTL7a8xyPwXSvzTsIwbknrKI80nmar9ASVH%2BcMw%3D&reserved=0<https://github.com/digitalbazaar/cel-spec/issues/1>

... and I've added the more recent mentions here:

https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdigitalbazaar%2Fcel-spec%2Fissues%2F1%23issuecomment-3092587006&data=05%7C02%7Clrosenth%40adobe.com%7C0147651898ae4c19255e08ddc714bcfb%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C638885614069428988%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=nKbKVGb8jHf4Sk4a%2B%2B4yRMhAdYnkC89tOuIeMfrQB28%3D&reserved=0<https://github.com/digitalbazaar/cel-spec/issues/1#issuecomment-3092587006>

All that to say that the work needs to be done for each technology
mentioned in issue #1. That will take time and that's what incubation
is partly about. Thankfully, Wolf has already done the comparison to
Provenance Marks.

So, in an attempt to provide at least a preliminary answer to each of
you in a single email:

Christopher and Wolf, Provenance Marks seem very close to what CEL is
trying to do. I'm most curious about the pre-commitment approach that
PMs use. It feels like we might be able to combine CEL/PMs or at least
re-use some of the functionality from PMs in CEL. Ideally, for me,
we'd join forces and merge the two but I'm saying that without
understanding every technical decision made in the PM spec.

Leonard, you described a way we could use a subset of C2PA to achieve
the same goals as CEL. I think the next step there is to try to
achieve each use case and example put forward in the CEL spec and see
what the C2PA solution looks like -- do some analysis on the two.

MikeP, there are a number of conceptual similarities between
SCITT/COSE Receipts, etc. The biggest set of differences seem to be
that SCITT seems to focus on CBOR/COSE, use binary merkle trees and
inclusion proofs, and then builds from there, which the other
approaches don't seem to take for whatever reason. Similar to C2PA,
we'd just have to mark up the examples in CEL and see what the log
looks like and how easy folks feel it would be to implement. SCITT
doesn't seem like an easy implementation lift whereas the did:webvh
folks have noted that their log mechanism, which is very similar to
CEL, only takes a week or two to implement fully. Again, it's going to
take time to do that analysis.

Daniel, similar response as to Leonard and MikeP -- we'll need to see
how the use cases covered via CEL compares to KERI when KERI is used
as the underlying technology.

Melvin, if you look section "2.5 Minimizing Event Logs", we might do
that compression using CBOR-LD, for which there has to be a JSON-LD
context. That said, we're trying really hard to not require JSON-LD at
all. You're right, we want to avoid friction; if the past several
years have been a lesson to us, introducing JSON-LD adds A LOT of
community friction. It's not clear it's mandatory for this particular
use case (though you can get benefit if JSON-LD is additive, but not
mandatory).

Dmitri and Brian, yes, of course we're going to learn from
did:webvh... and it might be that CEL moves closer to did:webvh
(possibly even just re-using the log format entirely). In this way,
perhaps the final solution will be a merge of did:webvh's log format,
PMs, and the CEL spec. Time will tell.

I'm sure those responses are not satisfying; they don't come close to
answering "How is CEL different from X". We acknowledge that those
questions need to be answered in time if CEL (or whatever variation of
it) is to be promoted beyond incubation. This email response is just
noting that we are tracking the "analysis of prior art" request as an
issue and still need to do the work to explore the alternatives at
depth.

If anyone has additional technologies to add to the list of "How is
CEL different from X", please add them here:

https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdigitalbazaar%2Fcel-spec%2Fissues%2F1&data=05%7C02%7Clrosenth%40adobe.com%7C0147651898ae4c19255e08ddc714bcfb%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C638885614069438832%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=UIo28%2BOMGDD1PQtSxhebLzoIe1HFBS0vIqMPl%2FyGw7Y%3D&reserved=0<https://github.com/digitalbazaar/cel-spec/issues/1>

-- manu

--
Manu Sporny - https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.linkedin.com%2Fin%2Fmanusporny%2F&data=05%7C02%7Clrosenth%40adobe.com%7C0147651898ae4c19255e08ddc714bcfb%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C638885614069448728%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=2Rk3Nd49lHHL3iffJEUcEXAs1MKSmm9vAePx5iinOp4%3D&reserved=0<https://www.linkedin.com/in/manusporny/>
Founder/CEO - Digital Bazaar, Inc.
https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.digitalbazaar.com%2F&data=05%7C02%7Clrosenth%40adobe.com%7C0147651898ae4c19255e08ddc714bcfb%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C638885614069458584%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=hZSKt7zMhmEB%2BCCIG8X2OEHXyuRdruuCqQ21mdv9fes%3D&reserved=0<https://www.digitalbazaar.com/>