Add credit class schemas to LinkML model #28

paul121 · 2024-12-27T22:40:14Z

WIP PR to add schemas for BT01 credit class + project + batches in schemas/src/BT01 directory

Closes #14

paul121

@S4mmyb just opening a draft so we can use comments to chat discuss further changes to this

paul121 · 2024-12-27T22:45:01Z

schema/src/BT01/BT01CreditClass.yaml

@S4mmyb right now this is the BT01 credit class file you had started, except I have modified the ranges to be valid and work with linkml. The issue was we had ranges like schema:Duration and URL, but ranges need to be proper data types that linkml understands OR custom data classes.

I'm still not sure what our best option is for specifying a ISO 8601 Duration datatype. I wonder if this might be an easy contribution to LinkML core data types (they have other 8601 date/time types). Or, for now I have specified this in the description eg "Crediting term duration for the project. An ISO 8601 duration.".

paul121 · 2024-12-27T22:50:32Z

schema/src/BT01/BT01CreditClass.yaml

+      SDGs:
+        description: List of relevant Sustainable Development Goals for this impact.
+        range: SDG
+        multivalued: true


@S4mmyb currently SDGs as a slot name doesn't fit our linkml linting configuration (slots should be lower camel case). What about just SDG?

In the past I have tried to avoid using plural wording for technical properties and just stick with singular words everywhere, knowing that some singular properties might indeed have mutliple values, thinking that the schema will specify if they are multi-valued, not the naming itself. Curious what you think about this more generally.

I made this change to rename the slots to (lowercase) sdg

schema/src/BT01/BT01ProjectInfo.yaml

schema/src/BT01/BT01CreditClass.yaml

paul121

Things to pull out of BT01 Credit Class into a common credit class:

Methodology/Methodology list (added for all credit classes)
SDG

Used on both credit classes and projects

Primary impact
Project benefit

I'll start with these changes and start creating the new structure, then we can continue iterating off of that..

paul121 · 2025-02-22T03:12:46Z

schema/src/BT01/BT01CreditClass.yaml

+  MethodologyList:
+    class_uri: rfs:MethodologyList
+    description: List of methodologies for credit generation.
+    attributes:
+      itemListElement:
+        range: Methodology
+        multivalued: true
+        description: "List of methodology items."


I don't think we need this intermediary class. I see this was added to try and model https://schema.org/ItemList

Instead, I think we can just reference Methodology with multivalued: true

I've made this change

paul121

Okay, I've just rebased on previously merged PRs, these changes should be simpler now.

Unfortunately moving the BT01 Project + Credit class into a src/BT01 subdirectory is breaking the linkml lint and thus builds are failing ☹️ - but the gen-doc still works locally. It seems like a bug with the linkml lint command, so for now I'll just remove the BT01 subdirectory. Here is the error: https://github.com/regen-network/regen-data-standards/actions/runs/13501803621/job/37721823950#step:5:59

paul121 · 2025-02-24T15:41:02Z

schema/src/CreditClassInfo.yaml

+classes:
+  CreditClassInfo:
+    class_uri: rfs:CreditClassInfo
+    abstract: true
+    title: Credit Class info
+    description: Base class with common fields used to define credit class info.


I learned that there are additional common-metadata that we can use on all parts of the data model. The title is nice because we can give a more human-readable string that is rendered on top of markdown pages (Credit Class info instead of CreditClassInfo)

abstract is also useful to demonstrate that this is an abstract base class and should not be instantiated/used directly.

paul121 · 2025-02-24T15:41:25Z

schema/src/CreditClassInfo.yaml

+    slot_uri: schema:name
+    required: true
+    description: Name of the credit class.
+    rank: 1


Rank is another common metadata we can use to set the order of slots

paul121 · 2025-02-24T15:44:26Z

schema/src/CreditClassInfo.yaml

+  - SDG
+  - ProjectBenefit
+  - PrimaryImpact


I created these classes in separate yaml files because these classes might be used outside of credit classes. In each of these files I also defined slots that can be used to reference these classes. It's possible that a credit class might have multiple slots referencing the same class for different semantic purposes, in which case each class can define these specific slots themselves.

paul121 · 2025-02-24T15:45:37Z

schema/src/PrimaryImpact.yaml

Instead of "Primary Impact" would it make sense to rename this to a generic "Project Impact' class? That way projects, credit classes, etc could reference multiple impacts, not only a single primary impact ?

paul121 · 2025-02-24T15:49:20Z

schema/src/PrimaryImpact.yaml

+    attributes:
+      id:
+        identifier: true


Some of these classes have id attributes, I believe from @S4mmyb initial migration trying to match the existing schemas. I added the identifier property so that linkml treats it as such. But I'm curious if we actually want to have full identifiers impacts, benefits, etc? If we are using identifiers does that mean we actually should be pulling from a taxonomy and have a semantic enum?

I need to review existing data more, just leaving the comment while its on my mind cc @clevinson

I don't think it necessarily means that we need a semantic enum, but it should be something that is a full URI.

From the LinkML docs:

the range of an identifier can be any type, but it is a good idea to have these be of type Uriorcurie

A class must not have more than one identifier (asserted or derived). identifier marks the primary identifier.

I'm fine with us using app.regen.network/credit-class/C01 , or http://mainnet.regen.network:1317/regen/ecocredit/v1/class/C01 for these, or some other resource list that we can work on making fully resolvable later (e.g. registry.regen.network/C01 for any arbitrary resource that exists on-chain)

S4mmyb · 2025-02-28T19:17:07Z

Primary Impact and Secondary Benefits should just be combined into one impact data type. Within the Credit Class and Project Schemas we can add in a field to reference which is the primary, the rest are considered secondary an are unordered (i.e. no rank).
For now let's migrate the cobenefit list to a secondary benefit list, but we might make the decision to stick with the name co-benefits

schema/src/CreditClassInfo.yaml

S4mmyb · 2025-02-28T19:53:51Z

Source Registry should stay in
Things that I think are missing in the current Credit Class schema:
- program which refers to the program the protocol is using. This is optional
- creditProtocol is the root document of the credit protocol that defines the requirements and process to register and issue credits under a credit protocol. It could be registered under a program or independent. It could include the methods to measure and monitor (like ERA) or it could reference other methodology documents (like CarbonPlus or Ecometric).

clevinson

Few comments here from my call with @paul121 a few weeks ago. I'll go ahead and work on a PR to resolve these issues as well as address comments pointed out by @S4mmyb

schema/src/CreditClassInfo.yaml

schema/src/BT01CreditClassInfo.yaml

clevinson · 2025-03-13T22:25:20Z

schema/src/PrimaryImpact.yaml

+    attributes:
+      id:
+        identifier: true


I don't think it necessarily means that we need a semantic enum, but it should be something that is a full URI.

From the LinkML docs:

the range of an identifier can be any type, but it is a good idea to have these be of type Uriorcurie

A class must not have more than one identifier (asserted or derived). identifier marks the primary identifier.

I'm fine with us using app.regen.network/credit-class/C01 , or http://mainnet.regen.network:1317/regen/ecocredit/v1/class/C01 for these, or some other resource list that we can work on making fully resolvable later (e.g. registry.regen.network/C01 for any arbitrary resource that exists on-chain)

… credit class schemas

clevinson · 2025-03-15T01:15:05Z

Source Registry should stay in

Things that I think are missing in the current Credit Class schema:

program which refers to the program the protocol is using. This is optional

creditProtocol is the root document of the credit protocol that defines the requirements and process to register and issue credits under a credit protocol. It could be registered under a program or independent. It could include the methods to measure and monitor (like ERA) or it could reference other methodology documents (like CarbonPlus or Ecometric).

Added these in 07ccd43

clevinson

Notes from a live review with @S4mmyb today.

schema/src/ProjectImpact.yaml

clevinson · 2025-03-19T22:19:34Z

schema/src/CreditClassInfo.yaml

+        description: Eligible activities for registered projects.
+        multivalued: true
+    todos:
+      - What's the relationship between creditProtocol and approvedMethodologies, in terms of which is required and not? Should one be folded into the other, or atleast align on a single class for the fields' ranges?


Current descriptions are good enough. We would like to update the slot names to be better aligned with RDF style predicate names.

schema/src/CreditClassInfo.yaml

clevinson · 2025-03-20T17:36:50Z

Follows from this PR:

Reconcile usage of "Impact" class with taxonomic enum #34
Pull in credit-class datasets into a /datasets/v0 folder
Do a similar set of upgrades for project level schemas

paul121 commented Dec 27, 2024

View reviewed changes

paul121 commented Feb 13, 2025

View reviewed changes

paul121 force-pushed the semantic-enumerations branch 2 times, most recently from ee7e543 to 84ea3da Compare February 21, 2025 18:52

Base automatically changed from semantic-enumerations to main February 21, 2025 19:45

paul121 commented Feb 22, 2025

View reviewed changes

paul121 added 4 commits February 24, 2025 08:34

Add initial BT01CreditClass

f9623f8

Rename TerrasosProjectInfo to BT01ProjectInfo

d00a358

Add abstract base CreditClassInfo class

bd6f48a

Add BT01 directory in sidebar

f673a16

paul121 force-pushed the BT01 branch from 55f1def to f673a16 Compare February 24, 2025 15:36

paul121 added 2 commits February 24, 2025 08:54

use relative project info import

0d3eedd

Remove nested BT01 directory

f46ab05

paul121 commented Feb 24, 2025

View reviewed changes

github-actions bot deployed to regen-data-standards (Preview) February 24, 2025 15:57 View deployment

S4mmyb reviewed Feb 28, 2025

View reviewed changes

schema/src/CreditClassInfo.yaml Outdated Show resolved Hide resolved

clevinson reviewed Mar 13, 2025

View reviewed changes

remove unnecessary slot_uri and class_uri usage

f50078e

github-actions bot deployed to regen-data-standards (Preview) March 13, 2025 23:48 View deployment

clevinson added 5 commits March 13, 2025 17:17

update root schema/index.md to render properly in astro dev

9f8e430

reorganize slots, remove BT01CreditClassInfo, and add carbon & biodiv…

7ee3104

… credit class schemas

standardize ProjectImpact as single linkml class

fd7a55b

add program and creditProtocol to CreditClassInfo

07ccd43

fix bug where /schema/ was not loading properly in astro dev

fd258af

github-actions bot deployed to regen-data-standards (Preview) March 15, 2025 01:14 View deployment

clevinson reviewed Mar 19, 2025

View reviewed changes

refactor SDG to own file, change ProjectImpact -> Impact

ddb32f9

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 00:55 View deployment

clevinson added 2 commits March 19, 2025 17:59

updates to CreditingProgram and sourceRegistry

d220a8c

require name in registry

50712eb

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 01:02 View deployment

remove biodiversity credit class

a4ee32e

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 01:13 View deployment

drop geographicApplicability

4f65bde

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 01:23 View deployment

refactor eligible environment and activities in credit class info

0080ce3

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 01:30 View deployment

refactor files

e4ab90e

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 01:44 View deployment

clevinson added 2 commits March 20, 2025 10:00

updates on framework wg call

20c461c

cleanup environmentTypeTypes to singular

1f405e2

clevinson changed the title ~~Add BT01 schemas~~ Add credit class schemas to LinkML model Mar 20, 2025

final changes from wg call

7aea8c1

clevinson marked this pull request as ready for review March 20, 2025 17:23

github-actions bot deployed to regen-data-standards (Preview) March 20, 2025 17:24 View deployment

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add credit class schemas to LinkML model #28

Add credit class schemas to LinkML model #28

paul121 commented Dec 27, 2024 •

edited by clevinson

Loading

paul121 left a comment

paul121 Dec 27, 2024

paul121 Dec 27, 2024

paul121 Feb 24, 2025

paul121 left a comment

paul121 Feb 22, 2025

paul121 Feb 24, 2025

paul121 left a comment

paul121 Feb 24, 2025

paul121 Feb 24, 2025

paul121 Feb 24, 2025

paul121 Feb 24, 2025

paul121 Feb 24, 2025

clevinson Mar 13, 2025

S4mmyb commented Feb 28, 2025

S4mmyb commented Feb 28, 2025

clevinson left a comment

clevinson Mar 13, 2025

clevinson commented Mar 15, 2025

clevinson left a comment

clevinson Mar 19, 2025

clevinson commented Mar 20, 2025

Add credit class schemas to LinkML model #28

Are you sure you want to change the base?

Add credit class schemas to LinkML model #28

Conversation

paul121 commented Dec 27, 2024 • edited by clevinson Loading

paul121 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

paul121 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

paul121 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

S4mmyb commented Feb 28, 2025

S4mmyb commented Feb 28, 2025

clevinson left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

clevinson commented Mar 15, 2025

clevinson left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

clevinson commented Mar 20, 2025

paul121 commented Dec 27, 2024 •

edited by clevinson

Loading