When a Mind names a table, a model, a workflow or a person in an answer, the chat can show the name as a chip: a small button with the object's name and a badge for its kind. Click a chip to open that object from the chat. To start a chat, see Ask a Mind.
| Badge | Where the object opens |
|---|---|
| Table, View, Dictionary, Static, Transformation, Logical Model, Source System | The object's popup window, the same window you open from Warehouse. |
| Data source | The Lake screen, focused on that data source. |
| Lake table, Lake column | The object's popup window. |
| Semantic model | The model's page in DataPrism. |
| Workflow | The workflow and its runs. |
| Workflow run | The run as a diagram. |
From outside Minds, workflow and run chips open the Workflows screen instead.
While DataMind looks the object up, the chip shows a short loading state.
A Mind refers to people by id. The chat turns each id into a chip with the person's avatar and name, or their email when the account has no name. The Mind itself sees ids, not names.
A chip that cannot be matched shows a dashed outline, and its tooltip reads Not found. It does not open anything when clicked. This happens when the object was deleted, or when it is not available to your account. DataMind does not guess between objects that share a name, so an unmatched reference is marked instead of opening the wrong object.
Chips show in the flow of the answer. A reference inside a code block stays exactly as written, in plain text.
The object behind the reference was deleted, or it is not available to your account. DataMind does not guess between objects that share a name, so an unmatched reference is marked instead of opening something else.
The chat could not match that id to a person, so it shows the id in place of a name. The Mind works with ids anyway, and a name is added when the id matches a person.
No. A role chip shows a role's name, or Role #7 when the role is no longer active, but a role has nothing to open.
No. A Mind refers to people by id. Your chat adds the name and avatar for display, so the Mind never sees them.
A workflow chip keeps the id until its name loads, and also when the name cannot be loaded. A run chip always shows an id, because a single run has no name of its own.
A reference inside a code block stays exactly as written, so the code reads as generated. Chips appear in the flow of the answer.
It opens the model's page in DataPrism. To learn what a semantic model is, see Semantic models and BI tables. To build a semantic model from a chat, see Model Builder.
For a table reference, the badge shows the table's actual type from Warehouse, such as Fact, in place of the word the Mind used.
DataMind focuses the window you already have open, or restores it if it was minimised, instead of opening a second copy.