What should we call the Substrate subproject?

[AttilaMihaly]

Some of the maintainers have raised that "Substrate" doesn't tell you much about what the subproject actually is, so I'd rather open the naming question up properly than settle it on a call.

The subproject (PR #645) turns the specifications an organization already writes — regulations, domain models, runbooks — into plain markdown that tools can verify and an LLM can be trusted to act on. I called it Substrate because that is what the corpus is: the base layer everything else is built on, and the material the tooling and the model act upon. It sits underneath Morphir rather than beside it — knowledge gets captured and verified there, and Morphir's existing toolchain executes and compiles it. Fair warning that there is already a logo and a full design system built on that metaphor — the mark is a triangular lattice, and the palette, typography, and tokens all follow from it — which is honestly the only reason I have any attachment to the name.

Two other names came up along the way. Spectackle, from specification + tackle, though it reads as "spectacle" on first sight. And Understrand, from understanding plus an underlying strand, which unfortunately looks like a misspelling of "understand."

Neither of those convinces me either. What does everyone think, and what have we not thought of?

Let's timebox this

Naming discussions can run forever and we don't have the time. So: one round for alternatives, one round for voting, then we're done.

• Round one — ideas. Post any name you want considered in this thread, one week from today.
• Round two — vote. I'll put the shortlist into a poll, one week to vote.
• Then we conclude, and whatever wins is the name. No reopening it after the merge.

[AttilaMihaly]

Throwing one into the ring: Morphir Ground.

Three readings, and all of them are the message I want. It swaps the "common" out of common ground — which is exactly what the corpus is, the shared ground between the people who write the rules, the engineers who implement them, and the models working on both. It's ground truth, which is the claim the whole thing rests on. And it's grounding in the AI sense: tying what a model says back to a verified source.

Short, plain English, and nobody needs to know any chemistry to get it.

[stephengoldbaum]

What is the main thing a new viewer would associate with it? The spec aspect? The knowledge foundation?

[AttilaMihaly]

Knowledge foundation is the more important aspect in my opinion. And more importantly verifiable knowledge.

[stephengoldbaum]

How about Mophir Knowledge then?

…

On Wed, Aug 26, 2026, 10:28 Attila Mihaly ***@***.***> wrote: Knowledge foundation is the more important aspect in my opinion. And more importantly verifiable knowledge. — Reply to this email directly, view it on GitHub <#691?email_source=notifications&email_token=AAA7SPJGYVXIFDO6ENCP6JL5L3XYPA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOBRGYZTOMJYUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-18163718>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAA7SPIPFHMYE7AOHOWDF3T5L3XYPAVCNFSNUABIKJSXA33TNF2G64TZHMZDIMZSHA2DSNZWHNCGS43DOVZXG2LPNY5TCMBWHAZDSMRWUF3AE> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/AAA7SPO4LEGJAPXTHSUMO6D5L3XYPA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOBRGYZTOMJYUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSVGM33PORSXEX3JN5ZQ> and Android <https://github.com/notifications/mobile/android/AAA7SPML2I7UKYB77B66HET5L3XYPA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOBRGYZTOMJYUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSXGM33PORSXEX3BNZSHE33JMQ>. Download it today! You are receiving this because you commented.Message ID: ***@***.***>

[psmulovics]

"Morphir Verifiable Knowledge"?

[AttilaMihaly]

I like how descriptive these are. Maybe "Morphir Verifiable Knowledge" is a bit too long to say every time we mention it.

[DamianReeves]

To be quite honest, once I dug into and had the understanding of substrate, I actually was fine with the name. As I mentioned the naming was non-obvious, to me, but given the reasoning I can see it working, so I am okay with it.
I am also okay with morphir knowledge.

[stephengoldbaum]

If we stick with substrate as a concept, perhaps we can pick something catchier, like "base layer" or "foundation"

[DamianReeves]

So if I think of the parts of what I have in my head, we have:

• Knowledge - context used to aid understanding and to define vocabulary and a data dictionary. I think this portion can also incorporate some of the RDF and linked data concepts we have discussed throughout the years. It would help give context and meaning to words like "cusip" when we see it in a knowledgebase entry
• Intent - The core requirements and regulations our users are defining. In order to accurately define intent you need both knowledge and the logic and schema definition capabilities to model the domain that Morphir brings to the table
• Morphir Datamodel/IR - We all know what this is

So in my mind we have in addition to what we have always had a "knowledge sub-system" or a knowledge management system as well as an "intent management system".

This aligns with the attempts I've been making to incorporate tools like Morphir in the whole SDLC process.

That is an extremely long winded way to say I don't know that it is a "base layer", I think "base layer" is dependent upon the persona you are playing.

[DamianReeves]

A shorter summary of the above, or perhaps the implication:

Using substrate leaves enough generality for us to craft a backing story that I don't think foundation or base layer does.
So though I opened this Pandora's box, I am with @AttilaMihaly now and am good with substrate.

[AttilaMihaly]

Thanks for the nice breakdown. Yeah, I agree that "substrate" leaves more room for interpretation than "base layer" or "foundation". It is also increasingly familiar in the agent-tooling world. Just google "markdown as a substrate". This is what I got:

refers to the paradigm shift where plain-text Markdown files serve as the fundamental layer for information exchange, memory storage, and state management between humans and AI agents.

So it quite literally describes the paradigm shift that we are taking Morphir through without mentioning Markdown specifically.