iiinova — where your digital self keeps evolving

Gregor Aisch @driven-by-data.net · 14/07/2026

The files come from digitized microfilms in the US National Archives and digitized records in the German Federal Archives. You’ll find lots of interesting stuff, from SA/SS membership cards to criminal and medical records to hundreds of pages of Heinrich Himmlers tax statements.

Screenshot of our search UI that lets you enter names, places and year of birth. The name field shows “z.B. Rudolf Hess” as placeholder

Screenshot of multiple historical records of Rudolf Hess in our database

Screenshot of a document labeled as curriculum vita of Rudolf Hess, zoomed in to am attached photography of Hess

051

Gregor Aisch @driven-by-data.net · 23/04/2026

Three weeks ago, we published an interactive story that lets our readers find their relatives in the Nazi Party membership archives. The response has been overwhelming, both in Germany and internationally. The story was covered by outlets such as Le Monde, CBC, Radio-Canada and CNN.

Screenshot of the story. The headline reads "Use Our Tool to Learn about Your Family's Nazi Party History"

1144

Gregor Aisch @driven-by-data.net · 08/04/2026

Really cool, thanks for sharing this. Tiny suggestion: I think there's one border-radius too many in the bubbles.

010

Gregor Aisch @driven-by-data.net · 02/04/2026

And another big project today: We've analyzed millions of member records from the Nazi-Party NSDAP and made them searchable and browsable. 🎁 www.zeit.de/wissen/2026-...

screenshot of the archive exploration tool.

32711

Gregor Aisch @driven-by-data.net · 13/03/2026

Say hello to the Raster mark in SveltePlot! Starting with v0.12 you can use it to create raster heatmaps from gridded and ungridded data! Again, this wouldn't be possible without @observablehq.com plot 💚. svelteplot.dev/marks/raster

Map of Canada showing weather stations distributed unevenly as dots, colored by maximum temperature measured.

Map of Canada showing a temperature heatmap derived from the weather station data using random-work interpolation.

0222

Gregor Aisch @driven-by-data.net · 07/03/2026

With Russian, Ukrainian, and now also Iranian airspace closed, only a small passage remains for international flights from Europe to Asia. My colleague @zacys.bsky.social wrote about how some airlines are profiting more from this than others, and I created this map. www.zeit.de/wirtschaft/2...

Globe of the Eastern hemisphere showing flight tracks between Frankfurt, Germany, and various destinations in Asia, including Mumbai, Bangkok, Hong Kong, Shanghai, and Tokyo. Russia, Ukraine and Iran are highlighted.

0214

Gregor Aisch @driven-by-data.net · 02/03/2026

📈 SveltePlot is on the DIE ZEIT homepage again! www.zeit.de/index

Screenshot of the ZEIT homepage on Monday, March 2. It features a line chart showing the surge of intraday crude oil price following the attacks on Iran.

2211

Gregor Aisch @driven-by-data.net · 21/02/2026

Sowohl SPIEGEL als auch ZEIT berichten über den Merkel-Besuch beim CDU-Parteitag. Dass Merz der Applaus zu lange gedauert hätte. Was ich als Leser jetzt vor allem sehen will: den Applaus in seiner vollen Länge. #Video ist leider noch nicht angekommen im deutschen Journalismus.

160

Gregor Aisch @driven-by-data.net · 19/12/2025

For a German, this is almost better than the actual 2025 review could've been 😅 #discord #privacy

040

Gregor Aisch @driven-by-data.net · 16/11/2025

SveltePlot 0.7 is out! Lots of new features since the last time I posted here: 🎉 jitter transform, 🐝 dodge transform (for beeswarm plots), 🎉 stackMarimekko, 📸 image mark, and most recently the 🧇 waffle mark! And the website now features a ton of examples. ➡️ svelteplot.dev

SveltePlot website featuring a grid of example plots

48914

Gregor Aisch @driven-by-data.net · 13/11/2025

I think this is pure charting gold from WaPo (credits to Szu Yu Chen, @emmakumer.bsky.social and @jeremybmerrill.com) www.washingtonpost.com/technology/i...

Line chart showing how the share of em dash uses in ChatGPT responses has increased from single-digit percentages in 2024 to almost 80% in July 2025. The chart cleverly used a dashed line resembling em dashes

Bar chart showing the most frequently used emojis in ChatGPT messages in 2025. The emojis are shown sitting on top of each bar. The winners are the brain emoji (26%) and the checkmark emoji (>30%)

0174

Gregor Aisch @driven-by-data.net · 15/09/2025

3. But why would you look at the distributions side by side (horizontally) when you can learn so much more if they're stacked vertically. You almost lose all of the benefits these 3 chart power-horses can offer us when not stacking them orthogonally to the main dimension.

The horizontal layout in a "raincloud plot" makes it actually harder to compare distributions...

A vertical layout makes it a lot easier to compare the distributions. Also a lot more mobile friendly.

131

Gregor Aisch @driven-by-data.net · 13/09/2025

For my latest Weekly Chart at @datawrapper.de I tried a different way of mapping destruction in Gaza using OSM buildings. www.datawrapper.de/blog/mapping...

OSM building footprints in Gaza overlaid with damage assessment by Corey Scher and Jan Van Den Hoek.

1213

Gregor Aisch @driven-by-data.net · 22/06/2025

I don't know how many climate scientists where involved in collecting all these samples and turning them into models of past surface temperatures, but we linked to all the related studies in the methodology.

Based on these studies
- Liesicki & Raymo, 2005 (https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2004PA001071)
- Jouzel et al., 2007 (https://www.science.org/doi/10.1126/science.1141038)
- Hansen et al., 2013 (https://royalsocietypublishing.org/doi/10.1098/rsta.2012.0294)
- Westerhold et al., 2020 (https://www.science.org/doi/10.1126/science.aba6853)
- Judd et al., 2024 (https://www.science.org/doi/10.1126/science.adk3705)

120

Gregor Aisch @driven-by-data.net · 20/06/2025

People tell me nobody wants to look at climate curves anymore, yet here we are!!! Please enjoy our interactive journey through 485 million years of climate history and dive into the fascinating and important work of paleoclimate scientists. 🎁 www.zeit.de/wissen/umwel...

911237

Gregor Aisch @driven-by-data.net · 13/05/2025

I made a website that documents all the features and shows a lot of examples that you can fork and edit on the Svelte playground. You can help improving this website, too! svelteplot.dev

Screenshot of the SveltePlot homepage at https://svelteplot.dev

1111

Gregor Aisch @driven-by-data.net · 13/05/2025

Last week I had the pleasure of finally releasing the #SveltePlot alpha version during my talk at the #SvelteSummit in Barcelona. It's a new visualization framework for @svelte.dev that I've been working on for the past 18 months, so it's about time for a 🧵

816535

Gregor Aisch @driven-by-data.net · 11/04/2025

Insider trading or not? We analyzed minutely stock data for the 20 most traded S&P 500 companies before, during and after Trumps announcement for pausing most tariffs and talked to financial experts about what this means www.zeit.de/geld/2025-04...

Intraday S&P 500 stock value and trade volume before and after Trumps announcement to pause tariffs on Wednesday. Curiously the trade volume went up 16 minutes before the announcement.

Minutely trade volumes for six S&P 500 countries before and after Trumps announcement to pause most tariffs. Each showing the same spike 16 minutes before the announcement.

292

Gregor Aisch @driven-by-data.net · 09/04/2025

Today is officially the day where intraday market charts have their time to shine in the media spotlight.

Intraday market chart of S&P 500, Nasdaq and DIJA indexes on WSJ.com homepage

Intraday market chart of S&P 500 index on nytimes.com homepage

1122

Gregor Aisch @driven-by-data.net · 27/02/2025

Oh, and the thread may sound a bit like we went straight for the solution, but of course there were a lot of uglier versions before that. One big leap forward was the decision to include the outer country shape, which is often missing in cartograms.

Earlier cartogram attempt without country border lines

Earlier cartogram attempt that tried to isolate states and used some weird blob like state borders

1241

Gregor Aisch @driven-by-data.net · 27/02/2025

Of course, no cartogram comes without distortion. The western states like North-Rhine-Westphalia are heavily populated and push the lesser populated districts to the East. But the final layout worked well during election night, I think.

A map using arrows to show how far the district circles have been moved away from their geographical position.

1471

Gregor Aisch @driven-by-data.net · 27/02/2025

To compute the final layout we use a force simulation with 3 forces: one of them keeps neighboring districts together while another pushes districts away from each other if they are too close. Coloring the nodes helped to identify problems during the process.

The final cartogram layout and the triangulation mesh. Nodes are colored by states

1440

Gregor Aisch @driven-by-data.net · 27/02/2025

Some of these fixed districts turned out to be too close to each other, so we defined a few exceptions for border districts we want to be free, and also a few custom positions that prevent the nodes to "leave" the country.

Close-up of the border districts, some are colored in red.

Close-up of the border districts, some are colored in red or yellow.

1410

Gregor Aisch @driven-by-data.net · 27/02/2025

Before running the network layout simulation on this, we marked all the outer districts as fixed. This allows the final layout to stay "in shape" of the country as a whole, which is important for readers to recognize their country. We used the TopoJSON arcs to automatically find these districts.

Map of the German election districts with the border districts highlighted in red.

1551

Gregor Aisch @driven-by-data.net · 27/02/2025

Due to Germany's convex outer shape we removed edges longer than a certain threshold and ended up with this final "network" of nodes and links.

The Delaunay triangulation of German election district centroids, with longer edges removed.

1410

Gregor Aisch @driven-by-data.net · 27/02/2025

Then we used a Delaunay tessellation of the district centroids. This later helps the districts to stay close to their geographical neighbors.

The Delaunay triangulation of German election district centroids.

1460

Gregor Aisch @driven-by-data.net · 27/02/2025

As input for the cartogram we use a pre-projected TopoJSON basemap of election districts with baked-in centroids (inner points). We used Mapshaper to generate the file.

A map of German election districts with their centroids shown as circles.

1511

Gregor Aisch @driven-by-data.net · 27/02/2025

For the German election we wanted to use a cartogram of German election districts but didn't find anything good, so we created our own. Here's a little nerd thread how the final algorithm worked 🤓 🧵 #gis #cartogram #election #datavis #cartography

Cartogram of 299 German election districts

18557179

Gregor Aisch @driven-by-data.net · 12/02/2025

POV: it’s 2025 and I’m still solving my problems with setTimeout()

Screenshot of Svelte code from a Typeahead component where setTimeout is used to prevent opening for 100ms after the element got closed

4200

Gregor Aisch @driven-by-data.net · 04/02/2025

The number of options for government coalitions in the upcoming German election heavily depends on how many smaller parties reach the five-percent threshold. We analyzed 50.000 elections to visualize this effect using animated kernel density plots 🤓 (text in 🇩🇪) www.zeit.de/politik/deut...

0265

Gregor Aisch @driven-by-data.net · 02/01/2025

Nice animation! I also like to show the Kneeling Curve split into decades as it makes it easier to see the acceleration of the carbon dioxide concentration.

Chart showing over six decades of CO2 concentration in the atmosphere. Each line represents one decade, from 1958 to November 2024. CO2 concentration is measured in parts per million* (ppm).

24112

Gregor Aisch @driven-by-data.net · 30/12/2024

One hundred! This is the first time I see someone outlive his obituary writer by seven years… #jimmycarter www.nytimes.com/2024/12/29/u...

NYT headline on the death of Jimmy Carter: “Jimmy Carter, Peacemaking President Amid Crises, Is Dead at 100“

The byline of that article says: “By Peter Baker and Roy Reed. Peter Baker is the chief White House correspondent for The Times; Roy Reed, who died in 2017, was a Times national correspondent who for many yars covered the South.”

1101

Gregor Aisch @driven-by-data.net · 05/12/2024

Und ja, die Grünen haben 2021 bei der Wahl weniger eingefahren als die Umfragen angedeutet haben, und die Union etwas mehr.

020

Gregor Aisch @driven-by-data.net · 01/12/2024

Das sind die Daten, die der SPIEGEL heranzieht. Sieht schon ein bisschen anders aus, oder?

110

Gregor Aisch @driven-by-data.net · 30/11/2024

To clarify, we're not "correcting" any polls. The polls we list on the page are exactly what the pollsters publish. We're just using the quality metrics to weight the polling average, shown as the line & numbers in the chart at the top.

120

Gregor Aisch @driven-by-data.net · 29/11/2024

Pollsters use different methods to conduct their polls. Some ask online, some by phone, others go from door to door! As a result some tend to over-estimate certain parties. In our newly added pollster-rating table you can now see these effects for yourself!

Table of German polling institutes and how we're rating their quality based on the average error (margin between final polls and the election result) and how biased the polls are towards individual parties.

272

Gregor Aisch @driven-by-data.net · 29/11/2024

For our German poll tracker we just added a similar chart, so you can see how close the average polls have been to the result in past elections. But another aspect of polling quality that is worth discussing further is how good the individual pollster have performed www.zeit.de/politik/deut...

Chart comparing pre-election polls from the 2017 and 2021 elections in Germany with the actual election result for each party.

180

Gregor Aisch @driven-by-data.net · 29/11/2024

Election polls are an important part of data journalism, but we often get comments from readers who are concerned about the quality of the polls. And rightfully so! That's why the US election we already published a chart comparing previous polls with election results as part of our poll tracker 🧵

A small multiple chart comparing national and swing-state pre-election polls with the 2020 election result, highlighting how far off the polls have been

1238

Gregor Aisch @driven-by-data.net · 27/11/2024

One pattern is striking: no party has lost as much and as fast in the polls while in government as the FDP (liberal democrats). In 2013 they even got kicked out of the parliament in effect, and the same could happen again in February.

174

Gregor Aisch @driven-by-data.net · 27/11/2024

We analyzed how political parties in Germany have tanked (or thrived) in the polls while they were in government. www.zeit.de/politik/deut...

17733

Gregor Aisch @driven-by-data.net · 25/11/2024

I love this subtle blame game going on here

screenshot of the before-posted maps saying "original page is of poor quality"

030

Gregor Aisch @driven-by-data.net · 22/11/2024

Yeah, because there are no maps that encode time, right? 😉

Charles Joseph Minard's Map of Napoleons March to Russia

A hurricane tracking map by the National Weather Service

Map of the Charles Darwin expedition on the HMS Beagle

New York Times map showing how the Camp Fire spread over time

110

Gregor Aisch @driven-by-data.net · 21/11/2024

We created a new dashboard for @zeit.de showing the current state of election polls in Germany ahead of the upcoming election in February. Also shows probabilities for govt. coalitions based on the polls. The full article version also features my favorite sparklines. www.zeit.de/politik/deut...

Screenshot of zeit.de showing an election polls dashboard (in German)

screenshot of the full article showing the election polls since 2021 in a big chart and the polls of the past 3 months as tiny sparklines below

58828

Gregor Aisch @driven-by-data.net · 13/11/2024

The other day we had an interesting discussion about election cartograms in our team that I think is share-worthy: At ZEIT Online went all-in using just a hexagon cartogram for the US election, and we even dismissed a geographical map as opt-in alternative..

Election cartogram for the US presidential election. I used the grey empty state not to trigger anyone accidentally (too soon, I know)

1184

Gregor Aisch @driven-by-data.net · 08/11/2024

I returned to news in May and since then we've covered 8 local elections in Germany, the election of the EU parliament, 3 state elections in Germany, national elections in France, UK, and Austria, and now the US election! Perfect timing for the German national government to collapse 🫣

screenshots of a ton of election maps we published since May

0182

Gregor Aisch @driven-by-data.net · 05/11/2024

Chris, I think you're missing a lot of context here. 1. Nate is not an "election forecaster". 2. The Times didn't publish an election forecast since the debacle in 2016. 3. There's a strike of 600 backend engineers at the NYT today that may affect the election infrastructure.

Quote from a NYT article by Nate Cohn about the Election Needle: "Publishing the Needle live on election night relies on computer systems maintained by engineers across the company, including some who are currently on strike. How we display our election forecast will depend on those systems, as well as incoming data feeds, and we will only publish a live version of the Needle if we are confident those systems are stable."

000

Gregor Aisch @driven-by-data.net · 02/11/2024

The United States according to Donald Trump, spinning globe scrolly with vintage state borders. www.nytimes.com/interactive/...

Map showing the Northeast of the US, with cities labeled as "hellhole" or "killing field"

110