Lien

@liendata.bsky.social

518 Followers 865 Following

Data. Semiconductor. Process mining. Homemade memes. Occasional artist. Skills of a data engineer and education of a data scientist... but I enjoy working as an overqualified data analyst instead 🦄

Screenshot that shows 6 objects called a variation of "Raad van State" having an id that is a variation of "0931.814.266".
Chart with title "2025 Goal Progress". Horizontal axis from January 1st to December 31st. Vertical axis from 0 to 100. Light blue area chart went up to about 20 around the end of March, plateaued until mid August, climbed up to 50 by end of September and climbed further to 100 by beginning of December.
Success kid meme. Mentally preparing myself for another day of recursive SQL. "they give you a copy of their database (your puzzle input)."
Screenshot of eProcurementOntology documentation (link at end). The text "TODO: add defintion" is highlighted. https://docs.ted.europa.eu/EPO/latest/_attachments/html_reports/ePO/EARoot/EA1/EA2/EA102.htm
Drawing of a wombat with clean, simple lines. The gray wombat is sitting and stacking three rounded cubes in front of it. The cubes are brown. To the left of the wombat, in light brown, the letters of PyStack't are stacked a bit jumbly in three rows: p y - s t a - c k 't.
Drake No/Yes Meme. No: Improving PyStack't. Yes: Creating a cute logo. The logo in question shows a black-and-white drawing of a wombat with clean, simple lines. The wombat is sitting and stacking three rounded cubes in front of it. To the left of the wombat, in light grey, the letters of PyStack't are stacked a bit jumbly in three rows: py - sta - ck't.
Mister Crabs (from Spongebob) lying in a pile of money. Caption reads "Software vendors after showing demo of AI going brrrrr to C-suite."
Screenshot of draw a fish app after you let the fish swim. My fish is orange/red and has a cute black eye with a white highlight.
Grey-banded Mining Bee. (Kruiskruidzandbij in Dutch.) Little bee sitting on a yellow flower.
Four panel comic about 2 blue aliens on a science fair. First panel: Young being standing next to a poster says "I have attempted science" and teacher asks "Please explain". Second panel: Young being replies "I formed an idea and then discovered I was wrong". Third panel: teacher looks closely at the poster and remarks "there are numerous diagrams" to which young being replies "I was wrong in numerous ways". Fourth panel: Young being says "I produced a detailed tribute to my wrongness" to which teacher replies "That is science".
Screenshot of terminal output. 18-06-2025 19:27 Extracting and mapping data for issue #17289 done ...66.5% (about 37679.9s remaining) 18-06-2025 19:27 Starting intermediate save process... 18-06-2025 19:27 Saving object-centric event data extracted from pandas-dev/pandas to DuckDB database file ./pandas_all_20250617.duckdb, schema stackt. Overwriting schema 'stackt' Table object_types (4 records) done. Table object_attributes (15 records) done. Table objects (214020 records) done. Table object_attribute_values (643050 records) done. Table event_types (37 records) done. Table events (836399 records) done. Table event_attributes (37 records) done. Table event_attribute_values (251858 records) done. Table relation_qualifiers (7 records) done. Table event_to_object (1837081 records) done. Table object_to_object (50698 records) done. Table event_to_object_attribute_value (0 records) done. 18-06-2025 19:32 Extracting and mapping data for issue #17089 done ...66.8% (about 37327.3s remaining) 18-06-2025 19:36 Extracting and mapping data for issue #16941 done ...67.0% (about 37056.2s remaining) 18-06-2025 19:37 Extracting and mapping data for issue #16903 done ...67.1% (about 36988.2s remaining).
Terminal output. Start extraction of object-centric event data from all issues of repository pandas-dev/pandas using GitHub REST API via PyGitHub library. Status updates will be displayed every 1% progress and every 5 minutes. While you wait, you can read about GitHub API rate limits here: https://docs.github.com/en/rest/using-the-rest-api/rate-limits-for-the-rest-api 17-06-2025 22:42 Starting data extraction for approximately 61672 issues ... 17-06-2025 22:42 Extracting and mapping data for issue #61672 done ...0.0% (about 78035.3s remaining) 17-06-2025 22:47 Extracting and mapping data for issue #61313 done ...0.4% (about 83959.7s remaining) 17-06-2025 22:52 Extracting and mapping data for issue #61045 done ...0.7% (about 85166.4s remaining) 17-06-2025 22:54 ⚠️ Skipping issue #60972 due to GitHub error: 404 {"message": "Not Found", "documentation_url": "https://docs.github.com/rest", "status": "404"} 17-06-2025 22:54 Starting intermediate save process... 17-06-2025 22:54 Saving object-centric event data extracted from pandas-dev/pandas to DuckDB database file ./pandas_all_20250617.duckdb, schema stackt. Overwriting schema 'stackt' Table object_types (4 records) done. Table object_attributes (15 records) done. Table objects (3012 records) done. Table object_attribute_values (8934 records) done. Table event_types (30 records) done. Table events (8423 records) done. Table event_attributes (30 records) done. Table event_attribute_values (2273 records) done. Table relation_qualifiers (7 records) done. Table event_to_object (19064 records) done. Table object_to_object (715 records) done. Table event_to_object_attribute_value (0 records) done. 17-06-2025 22:56 Extracting and mapping data for issue #60827 done ...1.0% (about 85049.9s remaining) 17-06-2025 22:57 Extracting and mapping data for issue #60791 done ...1.1% (about 84820.6s remaining).
Screenshot from githubstatus.com, showing Major Outage for API Requests.
Screenshot from Instagram post of twisteddoodles. Shows a nicely drawn sheep with a lanyard standing on a stylized conference floor (just some legs like she's talking to someone and a booth with 2 people in the background, drawn with simple light gray lines). Text reads "in the middle of the conference she suddenly realised she was in the wrong field.". Caption reads "But she still got the free pens".
Young Anna from the movie Frozen standing at Elsa's door, hands clasped and hopeful face. Top text: Me going to our data architects like. Bottom text: "Do you wanna build a DuckLake?".
Video showing a simple application. The right side shows a graph made up of event nodes and object nodes, and edges between them. On the left side are checkboxes that allow you to select or deselect the different object types and event types. The user starts with a few selected checkboxes. Any time she selects more, the graph updates to show additional nodes and edges. When all types are selectes, she zooms in so the label of the nodes become readable and explores the graph.
Same app, but now there are a lot more colors (edges are also colored) and no more overlapping labels, and even the edges are behaving nicely and not crossing each other constantly.
Screenshot of in-browser Dash app hosted at http://127.0.0.1:8050/. The title reads "Interactive object-centric process visualizations". Below is a card with title "Life cycle of a GitHub issue" that shows a zoomed out graph visualization of multi-colored nodes connected by grey edges. Nodes are clustered in light-gray rectangles.
Screenshot of in-browser Dash app hosted at http://127.0.0.1:8050/. The title reads "My First App with a Graph". Below is a card with title "Graph Title 101" and a square with two circles connected by a straight line. The first circle is labelled "Node 1", the second "Node 2". At the bottom of the page are some buttons: "Errors", "Callbacks", "v3.0.4", "Server".
Sunglass Doge Meme. Top text: When they ask to change something for a new edge case. Bottom text: But your code already covers it because this isn't your first data rodeo.
AI-generated coloring page of a twinkling unicorn jumping over a tree stump covered with star shaped moss.
Charcoal drawing of insect (Macrodactylus)
Charcoal drawing of an insect. The body seems done, but only some rough sketches of the legs are visible. The drawing is taped to a white board that is set on a wooden easel. The easel is placed against a white wall on which a small painting hangs of a landscape with a pink cloud. Next to the easel is a blue chair that holds some art supplies like charcoal and tape.
Gru's Plan Meme. Gru is standing next to a white flipchart explaining his big plan. The flipchart reads "Try to delete connection parameter." "Use error message to identify which tables still use the old data source." "Some tables use a hardcoded connection string" Gru is looking disappointed at his last flipchart reading "Some tables use a hardcoded connection string."
Meme of Patrick from SpongeBob explaining how to move stuff. get_github_log(). Get process data from GitHub repo. Store it in Stack't relational schema.
An unsolved slitherlink. Grid size is 10 by 18 cells.
A wide variety of Belgian beers, which are each served in their own distinct glass.
A happy snowman in the backyard. It has a black hat (reversed empty flower pot), a stick as nose, brown stones for its eyes and its coat buttons, black stones for a smiling mouth and two arms made from kids beach shovels.
Meme. Typos, typos everywhere.
Duck Imposter Game Leaderboard. First place: Lien with 100 points. Second place: winning-duck with 36 points. Third place: Pascal with 35 points. Fourth place: Ryan Wiley with 33 points.
Directly-follows graph of issue #9800 (from repo dbt-labs/dbt-core for those interested). It shows big brown nodes for all events that are linked to issue #9800 and small blue nodes for each snapshot of issue #9800. It also includes small red nodes for any users linked to the brown event nodes.
Surprised koala meme. Picture of a koala in a tree who is startled while eating eucalyptus leaves and now looks surprised with eyes wide open and 3 leaves halfway out of its mouth. Text reads: When you're busy solving bugs and the code suddenly runs.
Terminal output. Checking if DuckDB database file ../output/dbt-core.duckdb is available. New file will be created if it does not exist yet. IMPORTANT! Do not connect to DuckDB database file '../output/dbt-core.duckdb' while this script is running! (Or, if you like living on the edge, at least disconnect before all issues are extracted and the script tries writing to it.) Starting extraction of object-centric event data from 7 issues of repository dbt-labs/dbt-core using GitHub REST API via PyGitHub library. While you wait, you can read about GitHub API rate limits here: https://docs.github.com/en/rest/using-the-rest-api/rate-limits-for-the-rest-api Extracting and mapping data for issue #11175 done ...14.3% (about 12.2s remaining) Extracting and mapping data for issue #11173 done ...28.6% (about 8.0s remaining) Extracting and mapping data for issue #11169 done ...42.9% (about 6.1s remaining) Extracting and mapping data for issue #11168 done ...57.1% (about 4.3s remaining) Extracting and mapping data for issue #11167 done ...71.4% (about 2.8s remaining) Extracting and mapping data for issue #11166 done ...85.7% (about 1.4s remaining) Extracting and mapping data for issue #11163 done ...100.0% (about 0.0s remaining) Saving object-centric event data extracted from dbt-labs/dbt-core to DuckDB database file ../output/dbt-core.duckdb. Table object_types (3 records) done. Table object_attributes (7 records) done. Table objects (12 records) done. Table object_attribute_values (41 records) done. Table event_types (14 records) done. Table events (102 records) done. Table event_attributes (0 records) done. Table event_attribute_values (0 records) done. Table relation_qualifiers (2 records) done. Table event_to_object (109 records) done. Table object_to_object (7 records) done. Table event_to_object_attribute_value (0 records) done. All done!
Doge meme. Picture of a fat corgi dog. Different colored texts reading: "can be done in less than 1/3th of year", "100% progress bars", "challenge instead of chore", "activate completionist mode", "wow so dopamine".
Top 4 of scoreboard is occupied by Lien with 276, 211, 184 and 172 points.
Duck Imposter Game Leaderboard. First place is Lien with 191 points. Second place is Lien with 163 points. Third place is Alexandra BERGERE with 155 points. Fourth place is GJ with 121 points.
Scoreboard. First place is Lien with 163 points. Second place is GJ with 121 points. Third and fourth place is Lien with 101 and 88 points. Next places are Alexandre BERGERE with 80 points. Émilien with 67 points. Poopy butt with 55 points. Liv with 43 points and Liv the burglar with 41 points
Same leaderboard but one day later (scores were reset). Place 1-4 are all Lien with 119, 101, 90, 88 points. Place 5 is for Alexandre BERGERE with 80 points. Net scores are 67, 43, 32, 30 and 24 points.
Screenshot of top 10 leaderboard. First place "Mark Roddy" has 263 points. Second place is me with a score of 73 points. Next scores are 68, 48, 30, 27, 26, 23, 23, 22.
Two tables. First shows the original input data and the solution provided by Kshitji Aranke. Second table adds 3 extra dates to the input ('2025-03-10', '2025-04-22', '2025-07-20') and shows that one extra "chain" join is needed to obtain the correct solution.
Screenshot of a GitHub repo page. "About. Advent of SQL 2024, including set-up to run DuckDB in Docker and boiler plate code to get started." (Advent of SQL is highlighted) "Languages. Python 95.9% Dockerfile 4.1%" (Python 95.9% is highlighted)
Brown spotted kitten laying on its back on a light blue surface. Next to it is a big ball of bright red yarn. The kitten has unraveled a substantial amount of yarn which is surrounding and on top of the kitten. The kitten looks a bit surprised but doesn't seem to be in any harm.
Two hand-drawn pictures. Text above first is "Yes," and shows a white hand holding a smartphone. On the screen is a line graph that's shows first a strong upwards and afterwards a squiggly but high trend. Text above second is "but" and shows the same scene at a slighted tilted angle. This reveals that the smartphone has curved edges and that the last bit of the graph cannot be seen when you like straight at the screen. This final part of the graph has a very strong downwards trend.
Meme of a man on a balcony staring out to an asphalt road surrounded by trees and grass. Reads: I bring a sort of "yeah, you used the wrong data" vibe to corporate presentations that data scientists don't really like
Timelapse of a slitherlink puzzle being solved.
Hobbit Adventure Meme. Text at top: "How I feel when the data is not available yet". Text at bottom: "and I need to find it in the application database".
Meme. Top half shows a white stone fountain in a park with the text "What they expect" as title and below "pretty dashboards with clean & fresh data". Bottom half shows two pictures. The title above both is "What they forget". Left is a picture of an older toilet with text "How they produced the data". Right is a picture of shoddy plumbing with a lot of interconnected pipes with text "How much budget & time they gave the data team".