For everyone

Browsing by domain and by type

Lists of every kind of thing, and everything a domain answers for.

Open the list of one type

The sidebar groups types, for example Business, Logical and Physical. Each type shows how many there are. A small square beside a type means it is harvested from a source system rather than written by people.

Open a group in the sidebar.
Press a type, for example Tables.

The list shows Name and Definition or description. A Key column appears when keys differ from names. The box above the list narrows it by name or key as you type.

For types that people write, such as Business Terms, the list also offers Load from Excel and, if you may create assets, a New button such as New business term. See Editing an asset and Loading from Excel.

Choose the columns you see

Press Choose columns.
Tick what you want to see. Columns are grouped under Attributes, Responsible, Linked to, Harvested and The record.
Press Done choosing.

The list becomes a table with one column for each thing you ticked. An inherited responsibility is marked inherited, and hovering over it shows where it comes from. Linked assets are links. An empty cell shows a dash.

Drag the edge of a column heading to resize it. Double-click the edge to fit the column to its contents.

A list always opens as the plain list, on its usual columns. Columns you chose, and a saved view you opened, last until you leave. To keep an arrangement, save it as a view and pick it from the views list next time. Press clear columns to go back to the plain list.

Filter the list

The Filter button appears once you have chosen columns.

Press Filter.
Pick a column, then how to compare it, such as contains, is empty or is inherited. The choices depend on the kind of column.
Type or pick a value if one is needed, and press Add filter.

Every filter in force appears after showing only, with a × to remove it. clear all removes all of them. A line under the filters says how many assets match, for example "12 of 40 tables match".

Note: You can only filter on a column that is showing. If you remove a column, its filters are removed with it.

A date filter takes today, or today + 30 days, as well as a date you pick — worked out when you open the list, not when you saved it.

Sort and filter any table

Every table in KRISID sorts and filters on every column, not only this list: the metamodel, people and teams, runs, and the rest.

Click a column heading to sort by it. Click it again to reverse the order. The arrow beside the heading shows which column is sorted and which way.
On a table of more than five rows, type in the filter box under a heading to keep only the rows whose value in that column contains what you typed. Boxes in several columns combine. The last column shows how many rows are left, for example 5 of 23. Empty the box to see every row again.
A table with a Name column opens sorted on it, A to Z. Other tables keep their own order until you sort them, for example newest run first.

Lists arrive a page at a time — 100 rows by default, and Rows under the table changes that to 200, 500 or 1,000. The size is remembered in your browser and used by every list in KRISID.

Long text is cut to one line, so every row is one line and the eye runs down the names. Hover a cut cell to read the whole of it, or press Show full text beside Export… to let the paragraphs out; Trim long text puts them back. Like the page size, the choice is remembered in your browser and used by every list, including lenses. An export is never trimmed — the file is built from the rows themselves and carries every value whole, whatever the screen is showing.

Sorting and quick filters work on the rows on the page in front of you, and are forgotten when you leave it. To narrow the whole list rather than the page, add a condition with Filter above and save the view.

Save and export a view

A saved view is a named set of columns and filters for one type.

Press Save as… and give it a name. Everyone can see a saved view.
Pick a saved view from the drop-down at the start of the bar. Press Save to update it, Rename… to give it a new name, or Delete view to remove it. Renaming keeps its columns, filters, levels and sharing. Only the person who saved it, or someone with Change the metamodel, can rename it.
Press Export… to download. What this view shows gives you the columns on screen, as Excel or CSV. A workbook to edit and load back gives you an Excel workbook in the form Load from Excel reads.

Only the person who saved a view, or someone who holds METAMODEL_MANAGE, can change or delete it.

Show a view as a hierarchy

A view can show what sits under each row, level by level. For example, Subject Areas, then the Data Entities each one contains, then each entity's Data Attributes.

Open the list, for example Subject Areas, and press Hierarchy.
Press Add a level. Choose the relation that leads down from the level above, such as contains Data Entities. Every relation the metamodel has for that type is offered, in either direction.
Tick the columns you want for that level. Each level has its own columns. Level 1 uses the view's own columns and filters.
Press Add a level under that to go further, for example has Data Attributes. Press × on a level to remove it and the levels under it.
For a relation from a type to itself, such as broader than between Business Terms, tick as deep as it goes to follow it down every level.
Press Done with levels, then Save as… or Save. The levels are saved with the view.

A hierarchy never becomes the list's default. Next time you open the list it shows the plain list, and you open the hierarchy by picking its view.

If you leave before saving, the levels are kept as a draft in this browser. When you come back to the list, a notice says You have an unsaved hierarchy here. Press Resume it to carry on, or Discard to drop it. Saving the view clears the draft.

On screen the hierarchy is laid out like the workbook, without the indent:

One row per asset, directly under its parent. Level-1 names are bold.
Name, Relation (such as Subject Area contains Data Entity, empty for level 1), Asset Type, then every level's columns, each once. A row fills only its own level's columns.
▾ before a name folds what is under it, and ▸ unfolds it. Everything shows when the table opens.

Something with nothing under it, such as a subject area with no entities, still gets a row.

To extract it, press Export… and Download. A hierarchy always comes out as Excel, as an outline:

One row per asset, directly under its parent. A subject area, then each of its entities, then each entity's attributes.
The name is indented by its level, and level-1 names are bold.
The rows are grouped, so the + and − beside the rows, or the 1 2 3 buttons at the top left, fold and unfold each branch. The workbook opens with everything showing.
Relation says in full how the row was reached from the one above it, such as Subject Area contains Data Entity. It is empty for level 1.
Asset Type names the row's type, such as Data Entity. It is filled in for you, with the type's current name from the metamodel.
One header row. After Name, Relation and Asset Type come every level's columns, each once. A column two levels share, such as Key, is a single column. Each row fills only the columns chosen for its own level, and the rest stay empty.

Something with nothing under it is still a row. Excel groups up to seven levels, so rows deeper than that are still indented but fold with the seventh level.

See everything a domain answers for

By Data Domain is in the Overview group of the sidebar. It lists every domain with the number of assets it covers. That includes the assets it owns directly and everything those assets contain. A table in a schema that a domain owns is counted in that domain.

A co-owned chip means some of the domain's assets are also owned by another domain.
Press a domain's name to see everything in it, grouped by type. How says owned or inherited. Via names what the asset inherited from, and Also owned by names any other domain.

Open the assets behind a count

Each domain card shows counts by type, such as 18 Tables. Press a count to open those assets.

The list opens on a view named for the type and the grouping, such as Tables by Data Domain. KRISID makes that view the first time anyone presses a count for the type and uses it for every domain. The domain you pressed is applied as a condition that reads In Data Domain and names the domain. It is not saved into the view. The list always shows as many rows as the count said.

Press ← and the domain's name at the top to return to By Data Domain.
Remove the condition with its × to see every asset of that type.
You can change the view's columns like any other view. Once someone has changed them, their columns are the ones that open.

Browse the physical estate

Browse the physical estate shows harvested assets nested the way they really are. It is in the sidebar group that holds your source systems, and takes that group's name.

Press Browse the physical estate. The screen is headed The physical estate and groups systems by technology, such as Snowflake or Power BI.
Press ▶ beside a row to open the level below it: system, database, schema, table or view, column.
Press a name to open that asset.

To find one system, type in Filter systems or technologies… at the top. It keeps the systems whose name or key contains what you typed, or every system of a technology whose name does, and says how many it shows, for example 2 of 7. Only the top level is filtered: open a system and everything under it is there. The filter is remembered in this browser when you come back.

Each system is shown by a name people use. Where the source only gives an account or host (a Snowflake account locator, a Databricks host), the system takes the name of its connection on Sources; hold the pointer over it to see the account. A name the source does give, such as the name typed into a Power BI source's What to call this tenant, is kept. Each row shows its type, a data type or row count where the source gave one, and the start of its description. If nothing has been harvested yet, the screen says so. See Connections and harvesting.

Browse the logical estate

Browse the logical estate is in the Logical group of the sidebar. It opens on your Subject Areas. Press ▶ to see the Data Entities in one, and again to see an entity's Data Attributes.

The nesting is read from the metamodel: a relation where one thing holds many of the next and each of those sits in exactly one. Type in the box at the top to narrow the top level by name, the same way as on the physical estate. Something that sits under nothing, such as a Data Entity in no Subject Area, is shown at the top under its own heading, for example Data Entities not under any Subject Area, so it isn't lost. Any other layer whose types nest the same way gets its own button too, such as Browse the business estate for broader and narrower Business Terms.

Missing in Source

Missing in Source appears beside a harvested asset in lists, in the estate browser and in Ask results. It means the last harvest that looked for the asset did not find it. Hover over the mark to see when it was first missed.

Nothing has been deleted. The asset changes back to In Source when the source presents it again. Whoever owns the source decides whether it should return or be removed.

When enough of them have built up that nobody reads the mark any more, they can be cleared in bulk: open the thing they sit under, or the list for that kind, and use Clear what has gone…. Only what is marked is deleted, only after it has been missing for longer than the cut-off, and only what you are allowed to delete. See Clearing up what has gone.