Coming from KNIME
KNIME users feel at home quickly: nodes on a canvas, connected by ports, each with a status you can read at a glance. The differences are in how Daedalus handles variables, loops and reusable pieces, and that you never configure a node in a pop-up dialog.
How KNIME ideas map
| In KNIME | In Daedalus |
|---|---|
| Workflow | A workflow on the canvas, saved in a project folder on your machine. |
| Node repository | The add-node menu (right-click the canvas) and the ⌘K palette. KNIME node names work. |
| Configure dialog | The Attributes panel on the right shows the selected node's settings. No dialog to open and close. |
| Execute, node status lights | Run the workflow from the toolbar. Each node's badge says whether its output is up to date or needs a re-run. |
| Output port table, Node Monitor | Select a node and its output shows in the dock under the canvas: Data, Metadata, and Map, Chart or Report on the nodes that draw them. |
| Metanodes and components | Named functions: a group of nodes that collapses into one box with its own inputs and outputs. |
| Loop Start / Loop End pairs | A loop is one frame drawn around the nodes. It runs them once per row or file of a table, or repeats until done. |
| Flow variables | No flow variables. Wire a one-row table into a node's parameter input instead. |
| Annotations | Comment regions (toolbar ⋯ → Draw → Region) and per-node notes. |
| KNIME Hub schedules and deployments | Cloud runs run a project on a schedule. A read-only share link shows someone your workflow. |
KNIME names in Daedalus
The most-used KNIME names and what to reach for here. You can type any of them into the add-node menu or the ⌘K palette: the matching node shows a chip with the KNIME name, so you learn the Daedalus name as you go. Where two nodes are listed, the first is the closest match.
Readers and writers
| KNIME name | In Daedalus |
|---|---|
| CSV Reader | CSV Reader |
| CSV Writer | Write CSV |
| Excel Reader | Excel Reader |
| Excel Writer | Write Excel |
| Parquet Reader | Parquet Reader |
| Parquet Writer | Write Parquet |
| JSON Reader | JSON Reader |
| List Files/Folders | Directory (Add data source, Folder); File then reads each path |
Databases and web
| KNIME name | In Daedalus |
|---|---|
| DB Reader | DB Reader |
| DB Writer | Write to DB |
| DB Query | DuckDB Query |
| PostgreSQL Connector | DB Reader or Write to DB |
| GET Request | REST Fetch |
| POST Request | HTTP Request |
| Send Email |
Rows
| KNIME name | In Daedalus |
|---|---|
| Row Filter | Filter |
| Row Splitter | Switch |
| Duplicate Row Filter | Distinct or Flag Duplicates |
| Row Sampler | Random Sample or Row Limit |
| Sorter | Sort |
| Top k Row Filter | Top / Bottom N |
| Concatenate | Union |
| RowID | Add Row Number |
Columns
| KNIME name | In Daedalus |
|---|---|
| Column Filter | Select Columns |
| Column Renamer | Rename |
| Column Auto Type Cast | Parse or Cast Types |
| String to Number | Cast Types or Parse |
| Expression | Add Column |
| Rule Engine | Add Column or Switch |
| String Manipulation | Text Operations or Add Column |
| String Replacer | Text Operations or Replace Values |
| Cell Splitter | Split to Columns or Split to Rows |
| Missing Value | Fill Missing or Interpolate NULLs |
Aggregation, joins and reshaping
| KNIME name | In Daedalus |
|---|---|
| GroupBy | Aggregate or Group Aggregate |
| Pivot | Pivot |
| Unpivot | Unpivot |
| Value Counter | Value Counts |
| Joiner | Join |
| Cross Joiner | Cross Join |
| Value Lookup | Lookup or Join |
| Lag Column | Shift / Lag |
| Moving Average | Rolling Window or Exponential Weighted Mean |
Date and time
| KNIME name | In Daedalus |
|---|---|
| String to Date&Time | Parse or Cast Types |
| Date&Time Difference | Date/Time Arithmetic |
| Date&Time Part Extractor | Extract Date/Time Parts |
| Date Shifter | Date/Time Arithmetic |
| Time Zone Modifier | Convert Timezone |
Statistics and data quality
| KNIME name | In Daedalus |
|---|---|
| Statistics | Describe |
| Data Explorer | Describe or Column Profile |
| Numeric Outliers | Outlier Flag |
| Table Validator | Assert |
| Linear Regression Learner | Linear Regression |
Flow control and scripting
| KNIME name | In Daedalus |
|---|---|
| IF Switch | Gate |
| CASE Switch Start | Gate |
| Empty Table Switch | Gate |
| Wait... | No timed or file waits; Wait For holds a branch until other steps finish |
| Chunk Loop Start | Use a Loop over a table of items (e.g. a Constant Table) |
| Table Row to Variable | No flow variables; wire a one-row table in as a parameter |
| Python Script | Code Gen |
Geospatial
| KNIME name | In Daedalus |
|---|---|
| Lat/Lon to Geometry | Make Point |
| Projection | Reproject |
| Buffer | Buffer |
| Spatial Join | Spatial Join |
| Dissolve | Dissolve, or Aggregate's Union (dissolve) to keep statistics per group |
| Geocoding | Geocode |
| Road Network Isochrone Map | Isochrone |
| Geospatial View | Map; the Map tab under the canvas |
Views and reports
| KNIME name | In Daedalus |
|---|---|
| Bar Chart | Chart; the Chart tab under the canvas |
| Line Plot | Chart; the Chart tab under the canvas |
| Scatter Plot | Chart; the Chart tab under the canvas |
| Table View | Table Section; the Data tab under the canvas |
| Report PDF Writer | Write Report |
Commands
| KNIME name | In Daedalus |
|---|---|
| Execute | Run: the toolbar's Run button, or ⌘↵ |
| Cancel | Stop run in the ⌘K palette |
| Node Monitor | the Data or Metadata tabs under the canvas |
| K-AI | Extend with AI opens the Assistant pane |
| Create Schedule | Deploy & schedule to cloud in the toolbar's ⋯ menu |
| Export Workflow | Export as Python library… in the toolbar's ⋯ menu |
| Share | Share… in the toolbar's ⋯ menu makes a read-only link |
Not in Daedalus yet
KNIME has a much larger node repository than Daedalus. The biggest gaps are machine-learning learners and predictors, statistical tests, widgets and file management. When you type one of these names, the menu shows the hint below instead of a node.
| KNIME name | What to do instead |
|---|---|
| Decision Tree Learner, Random Forest Learner, k-Means, Logistic Regression Learner | No built-in ML models yet; ask Code Gen to write one in Python |
| PMML Writer, PMML Reader, Model Writer, Model Reader | No PMML or model files; not supported |
| Single Selection Widget, String Widget, Refresh Button Widget | No interactive widgets or dashboards yet |
| String Configuration, Integer Configuration | No configuration dialogs; wire values from a Constant Table |
| Independent groups t-test, One-way ANOVA, Kruskal-Wallis Test | No statistical tests yet; ask Code Gen to write one in Python |
| Try (Data Ports), Catch Errors (Data Ports) | No try/catch; a failing node stops its branch. Inside a Loop, turn on Continue on error to skip failing items |
| Create Folder, Delete Files/Folders, Decompress Files | No file-management nodes yet; manage files outside Daedalus |
| XPath, XML to JSON | Read .xml files with XML Reader and set its Record path; for XML text in a column, use a Code Gen node |
| Microsoft SQL Server Connector, Oracle Connector | Only PostgreSQL, MySQL, SQLite and Snowflake are supported |
| DB Updater, DB Merger, DB Row Deleter | Not supported; only create, replace and append writes exist |
| Container Input (Table), Container Input (JSON) | Workflows cannot be called as services yet |
| SARIMA Learner (Labs) | Forecast (exponential smoothing or seasonal naive); for SARIMA itself, ask Code Gen to write it in Python |
| Global Moran's I, Local Moran's I | No spatial autocorrelation stats yet; try Code Gen |
Tips for KNIME users
- Type the KNIME name. GroupBy finds Aggregate, Joiner finds Join and Column Filter finds Select Columns, each with a KNIME: chip in the add-node menu and the ⌘K palette.
- Replace flow variables with a parameter table. Put the value in a Constant Table and wire it into the node's parameter input; Filter, Add Column and file names can all read from it.
- Loops have no start and end nodes. Select the nodes to repeat and draw a loop around them (toolbar ⋯ → Draw → Loop), then wire the table of items into it.
- Turn a metanode into a named function the same way: select the nodes and draw a named function around them. It collapses to one box you can run on its own.
- Outputs are cached for you. There is no Cache node and no Reset: Run re-runs only what changed, and the badges show what is stale.