Open question
Profiles: a usability idea → a governance question
2025 Living Data
Mikhail Podolskiy proposes profiles/templates as manageable subsets of the model; the paleo example shown is explicitly made up.
Watch 2025 at 55:40 ↗Read the source captions
Implementing the proposed DwC-DP format in GBIF: data publishing tools
solve this issue. Uh one of the ideas to solve this thing is uh introduce new concept of a profile or template which would be a subset of the model with uh uh some tables and fields their fields with the comprehensive description of it and some keywords for search. They can be public or local. they can be downloadable online or not or yeah um you can create resources from these templates you can edit those templates and yeah basically that's the idea so here's the example for for example for pentological data set which is a madeup example it's it's not real data data set profile so just just the idea it can be many many profiles Um sorry. So for implementing profiles
2026 This talk
The 2026 discussion asks about registering and governing named profiles; John Wieczorek says profiles are not yet established as standard products and could be community-created and vetted.
Open on Vimeo ↗Read the source captions
SYM42 discussion: DwC-DP profiles, deriving occurrences, ABCD's future and system boundaries
The question is, John, we might anticipate versioned data package profiles. For example, GBIF requiring a Darwin Core data package with an occurrence table, among other tables of importance to its index, and so on. Might the Darwin Core maintenance group produce a registry of named profiles? If so, what might a governance structure look like for DP profiles? Checking. Yep, testing. A bit of background. There are no profiles now because our original pre-feedback was, "Please don't have more than one data model," and we took that seriously. So there is only one right now. But once people started seeing it, especially after Columbia with all those boxes on the screen, they started backpedaling and saying, "Can you make one for me and my stuff?" So I think we might be heading in that direction, but we haven't talked about how such production would be proposed and managed. If I were to imagine, I would say that these would have to be incorporated as additional products for the standard. I don't know if everyone's familiar with what's in Darwin Core, what's not in Darwin Core. But recently, with our latest public review, schemas, some existing schemas, have been incorporated as part of the standard. This includes all of the GBIF-produced cores, occurrence event and taxon, plus several of the extensions that are obviously just simple Darwin Core type things. So measurement or fact, identification history, alternative identifiers, but not much beyond those things. Humble extension and chronometric age. But not, for example, the OBIS extended measurements or facts. Why? It seems obvious. Because we haven't had a conversation with OBIS to say if they want us to. Also, none of the GGBN extensions. Why? There's been no conversation about whether they want us to. Here, if there is a community desire to create profiles, that's a reasonable thing, but they should be created by the community, and they should be community-vetted in the same way that everything else is, if they're going to be part of the standard. So I think that the door is open to try to do that, but right now we have no infrastructure for it, no plan for it. We need to discuss it. Okay. Yeah. Thanks a lot. I hope this answers the question. We have a second Q&A question here online in Vova. It actually affects the presentation that I gave on Dina and occurrences and material entities, and the question is: "If GBIF
The problem recurs at a different level: making the model usable versus recognising and governing shared profiles.
About this connection
Local templates and draft community mappings are not the same thing as registered, standardised profiles. Do not read “no profiles” as proof that no prototypes or drafts existed.
