cybersim's blog

Katharsis im Kino

check my website (cybersim) — Sat, 08 Aug 2026 07:53:54 +0200

cybersim's blog https://cybersim.ch/posts/katharsis-im-kino/ -

In einem geisteswissentschaftlichen Haushalt aufzuwachsen bietet gewisse Vorteile. Odysseus war spätestens als Teenager ein alter Bekannter. “Die rosenfingerige Eos” eine vertraute Wendung. Als Erwachsene hatte sich der “Odyssee”-Comic von Frieda Bünzli längst in mein Gehirn eingebrannt. Und ja, nach Kirke tragen die Gefährten noch einige Panels lang Säulischwänzli.

Nun also: “Die Odyssee” von Christopher Nolan, gedreht auf IMAX für IMAX. Ein Film in sagenhafter Länge, mit epischem Aufwand und historischem Format.

Ich klatschte in die Hände, als ich die ersten Schafe sah. Polyphem!

Ich schauderte in Anbetracht der Schweinefleischsuppe.

Die Schiffe, so klein – das Meer, so riesig. Ich hob die Hände und sah die ebenfalls erhobenen Hände meines Sitznachbarn. Wir schauten uns an. Das, das ist Kino!

Vor kurzem fing ich die zweite Staffel von “Wednesday” an. Nach zwei Folgen brach ich ab. Zu künstlich, zu wenig real. Berührungen fühlten sich seltsam losgelöst an, das Licht aufgeklatscht, die Figuren bewegten sich wie unverankert im Raum. Wie wenn sich zwei Figuren in Games umarmen: Eine Simulation des Realen, immer nur fast, nicht ganz. Heutiges Filmemachen ist ein Simulakrum geworden bis zum Punkt, dass ausser dem Boden unter den Füssen der Schauspieler:innen nichts mehr echt sein muss.

In Nolan’s Odyssee ist hingegen alles greifbar: Der Zyklop ist ein mechatronisches Konstrukt. Ich rieche den Gestank der Griechen im trojanischen Pferd. Das Mittelmeerlicht ist echt. Die Laistrygonen riesig.

Das ist Kino.

Mit Tricks das Unmögliche möglich machen. Das Unmögliche glauben. Und gleichzeitig als Zuschauerin wissen, mit welchem Risiko der Trick vollführt wurde. “Pass auf!”, möchten wir immer noch Buster Keaton zurufen. Wir spüren die Enge im trojanischen Pferd, weil sich tatsächlich eine Gruppe Menschen samt Kameraleuten in ein echtes Holzwerk gequetscht und Stunden ausgeharrt haben. Und selbst wenn ich dem Film nie verzeihen werde, dass die Frage nach dem Ehebett nicht vorkommt – wir fühlen nach, was Penelope und Odysseus fühlen, als sie endlich wieder zusammenfinden.

Heutige Technologien ermöglichen das Unmögliche derart real zu machen, dass es perfekter als die Realität ist, perfekter als unsere Sehgewohnheiten. Kino ist hyperreal geworden. Und irgendwo vor Greenscreens und im Post-Editing hat das Kino (und Streaming) den Bezug zur Realität verloren, die haptische Bildsprache. Nichts verankert uns mehr in der Realität, und suspension of disbelief benötigt keine suspension mehr, denn alles ist so künstlich, dass es nie real sein kann.

Nach drei pausenlosen Stunden klatsche ich mit dem Rest des Publikums und stehe ich mit steifen Gelenken auf. Das war ein Epos für ein Epos! Passend zum antiken Stoff fühle ich cineastische Katharsis: Genau darum gehe ich ins Kino. Genau darum schaue ich gerne ältere Filme, als Computerkunst noch teuer war und die Schauspieler:innen Blut und Wasser schwitzten. Das ist real. Das ist echtes, greifbares Kino.

- https://cybersim.ch/posts/katharsis-im-kino/ -

Millenial Travelling

check my website (cybersim) — Sat, 25 Jul 2026 06:14:12 +0200

cybersim's blog https://cybersim.ch/posts/millenial-travelling/ -

Millennials are hybrid people. We are the offline and the online generation, those who grew up with and without The Mighty Internet, with and without perpetual connectivity. I remember Lonely Planet books as the one trusty guide that promised to carry me in and out of a place (“Getting There & Away: bus 888 to the airport, 45 min”), highlighting safe places to sleep, even more places to visit. A street lamp in the darkness of the faraway unknown. Travelling then meant being hooked up to a Swedish public pay phone so I could spell out my credit card number to a youth hostel in the next city.

Then the Internet made a sound most people today would not recognize anymore: we consciously went online. Typing away on our parents’ computer and blocking the landline for hours, ideally during the night. For the first time, we travelled virtually. Physical travelling still meant “Getting There & Away”, but also relying on lists of Internet cafés: Places that offered an Internet connection to “go online”. On grimy keyboards, we checked hotels, train schedules, and the next stop, while someone was probably stealing your credentials. A virtual street lamp to orient you in a foreign place and connect you back home.

Then the sound of the Internet was lost to the smartphone. The magic hand-held in your pocket, always there, always online. Train schedules now in your hand. The route to and from the airport. Physical and virtual travelling converged. Our eyes are now glued to our phones, looking for directions instead of asking locals. Our faces light up from endless notifications, shitstorms, and textual quarrels that could be resolved over a cup of coffee. But the smartphone also brought possibilities, convenience, and flexibility. We can now travel virtually before we do it physically. It meant booking a hotel room at 2 a.m. because the communal authority was not happy to see a white foreigner sleeping in a friend’s home. The smartphone changed everything.

Then the big P happened: trapped at home, we only travelled virtually via our smartphones, via our computers to faraway places; we doomscrolled until fake news, fake outrage, endless virtue signaling made us sick of social media and the ubiquitous connection. It was the time I uninstalled all my social media apps from my phone. I was nauseous from the endless connectivity that tied me to some virtual non-place where attention is the only currency that matters, where social connections are thin and fleeting.

The Pandemic held lessons for all of us. I also realised that we had reached a stage where, after years of being non-stop online, we had to re-learn how to be consciously offline. We need to learn to put the phone away, to switch off the computer and look at flowers. Feel the haptics of our presence by making something with our hands. The post-pandemic years also announced the Age of the Stochastic Parrots, and with it the knowledge that whatever we put online would be regurgitated by some AI model, stripped of its original voice and author. All your online actions will be monetized – while this was already true before the Pandemic, it has now reached every website via a deluge of bot traffic (hello, bot).

This is the age when we have to put our phones away and look again at our hands and see what’s in front of us. If you’re at your sewing machine, your Zucks, Musks, Altmans of the world do not earn a single dollar. If your phone stays in your pocket, you will actually travel. You will talk to the locals. So use your phone to get there and away, but once you’re there, put it away and look at the world with your own eyes rather than through your camera lens.

Final note: The observation that when you use your sewing machine, you’re not feeding Silicon Valley billionaires, comes from Heather Lou, owner of Closet Core Patterns. She mentions it in episode 7 of “A Handbuilt Life” in conversation with Gretchen Jones.

- https://cybersim.ch/posts/millenial-travelling/ -

Brief Blog Update

check my website (cybersim) — Thu, 02 Jul 2026 06:24:17 +0200

cybersim's blog https://cybersim.ch/posts/blog-update-2026/ -

You may have noticed the color of the blog has slightly changed, and in the background, some issues with the original theme have been fixed that broke with new Hugo releases. The update happened with the support of an LLM: Frontend web development has never been my forté, and as my days are filled with computer work, I limit it to fun tasks in my spare time, and writing (S)CSS is not one of them. This is what an LLM is exactly for: Support in some tech tasks that I’m vaguely familiar with and can verify, but would cost me days to write from scratch.

I always felt compelled to add a picture to every blog post, a force majeure from my journalist days. However, the images rarely add to the content and the blog does not serve any journalistic purpose. Back when the blog started, image generation with LLMs was all the rage, nowadays I’m just tired of the repetitive blandness. There’s a blog post in the back of my mind about the visual and semantic blandness of LLM-generated content, and while it’s fun to see the slot machine churn out whatever, I’ve been increasingly bored by it. Additionally, I became more aware of all the issues surrounding LLM use and apply it more deliberately now. Future blog posts will have less images, so I can focus on content and not on generating the perfect illustration for it. Finally, the LLM was able to locate some CSS issues for posts without an illustration – again something that would have cost me days to figure out.

Here’s again my LLM usage policy for the blog as TLDR: A real human writes this blog; LLMs serve as sidekick editors to fix my non-native speaker errors. Some images are LLM-generated, and I always indicate the source if they’re not by myself (like a good journalist). In case you wonder where I fall on the LLM-hypetrain-craze-bubble-Luddite spectrum, the not-so-serious “AI Compass” identified me as The Journalist. Some things apparently never change.

- https://cybersim.ch/posts/blog-update-2026/ -

My Chinese Culinary Heatwave Hit

check my website (cybersim) — Sat, 27 Jun 2026 17:28:48 +0200

cybersim's blog https://cybersim.ch/posts/cold-strange-flavor-noodles/ -

For many people with multicultural experiences, cooking is one way to evoke a culture that’s absent. Unsurprisingly, there are a couple of dishes that immediately transport me back to places in China; and in summertime, one bowl of cold noodles lightens up my Sinologist heart, bringing back hot, sweaty summer days in some backwater provincial capital.

One recipe that I make every summer is “怪味凉面” (guaiwei liangmian, “cold strange flavor noodles”). The closest to match my memories¹ comes from a YouTube cooking channel: Two chefs cook on camera, commenting in a heavy Beijing accent, and even livestream events like their New Year’s banquet at home.

Interestingly, I haven’t found any well-translated authentic recipes of “怪味凉面” yet, and as we are very much headed towards more scorching summers, people here in Europe might enjoy this dish too, minus the cultural flashback. Here’s a translated version of the video – note that a lot of Chinese cooking follows the rule of thumb and rarely follows exact measurements. The proportions given are something I figured out through other recipes.

Ingredients

Makes about 2-3 portions.

2 tbsp Chinese sesame paste
2 tbsp soy sauce
1 tbsp dark rice vinegar
Sesame oil, to taste
Seasoning: Salt, MSG, sugar, ground Sichuan pepper, white sesame seeds, all to taste
1 tsp grated ginger
Small scallion, chopped
1/3 cucumber
Wheat noodles, per person ca. 60-70g

Notes:

Use Chinese sesame paste, as it’s prepared differently from tahini. If you can’t find it, tahini works too, but might affect consistency when combined with the soy sauce and rice vinegar.
MSG is not needed, but mentioned in the video. Switzerland uses Aromat instead, but please don’t use that.
Noodles are a topic in and of themselves. You can use ramen noodles, even spaghetti or soba noodles, but the taste will be different and the sauce sticks differently to each type.
The base paste without any fresh ingredients (no ginger, no scallions) lasts about 10-15 days in the fridge according to the video.

Procedure

Cook the noodles until al-dente, rinse them in cold water and drain. Arrange in 2-3 bowls.
Chop the cucumber into thin strips (here’s a cutting technique).
Slowly add the soy sauce to the sesame paste, continuously stirring. The sauce will change consistency; once it starts turning creamy, add the rice vinegar. In the end, the sauce should have a creamy consistency.
Add the seasoning, mix well.
Add ginger, scallions and the sesame oil, mix well.
Arrange cucumber on top of noodles, cover with sauce.
To taste: mix some sesame oil with chili oil, drizzle over.
Enjoy!

When my sauce gets too thick, I add it to the bottom of the bowl. The lukewarm noodles will turn the sauce creamy and easy to mix.

More Recipes!

Before YouTube and the Internet, there were recipe books, so I carried armfuls of them back home to recreate all the flavors I loved. However, Europe is not the place to get donkey meat or some fancy plant; and I always struggled to translate these recipes into my European context.

Here are two websites that my Chinese cooking is based on, as they come close to all the tastes that I miss:

Red House Spice: Wei Guo is based in England and is my go-to source for Chinese cooking. She manages to keep all the delicious flavors while also working within the limits of our supermarkets.

The Woks of Life: The Leung family is based in the US and is my other important source. Some of their recipes are only possible because the US has many more Asian grocery stores with a wider selection, especially when it comes to fresh produce. Some other recipes are influenced by US culture (“Chinese take-out”), written with a US audience in mind.

The title is forcibly auto-translated by YouTube, as Silicon Valley thinks people usually just speak English. The whole channel is in Chinese. ↩︎

- https://cybersim.ch/posts/cold-strange-flavor-noodles/ -

APTs and Elephants

check my website (cybersim) — Sun, 22 Mar 2026 09:20:33 +0100

cybersim's blog https://cybersim.ch/posts/apts-and-elephants/ -

The threat landscape is scattered with animals: pandas, bears, and even kittens — all dangerous cyber creatures. There’s even sandworms and sandstorms. These names are commonly used by vendors, researchers and threat intelligence companies to designate APTs, Advanced Persistent Threats. Each company has its own nomenclature: Microsoft uses a weather system, Palo Alto’s Unit 42 references star constellations, and in 2013, Mandiant (now Google) published a report on the initial APT1, which is associated with the Chinese government. MITRE maintains an extensive list of commonly known threat actor groups; threat intelligence researchers usually keep their own “Spreadsheet of Doom”, a Who’s Who of threat actors.

Newcomers to cyber threat intelligence (CTI) examine such a list and observe that Magic Hound is also called TA453, COBALT ILLUSION, Charming Kitten, ITG18, Phosphorus, Newscaster, APT35, or Mint Sandstorm. They are then tempted to draw a big equals sign between all these names. Which name to use is left to the author or their employer; sometimes, it is simply what initially stuck in public perception. For example, I read more about “Charming Kitten” and less about “Magic Hound”.

However, APTs are a construct.

They are not clear-cut entities.

Rather, these names are similar to blind people who touch an elephant for the first time and describe their impression of what an elephant is (Sinology side note: 盲人摸像). Each company requires data to form any threat intelligence. For those who sit at the intelligence source, this means customer data, such as agents running on customer devices, or employees responding to customer incidents. Sophos serves many SMEs, which means their data is based on small companies who are often easily compromised. Google Mandiant responds to incidents of Fortune 500 companies, who operate on a global scale and usually have a decent security posture. Kaspersky has lost a lot of Western customers since the war in Ukraine started, which means their customers are now in other parts of the world, with different IT environments than Europe. Consequently, the data that formed “Magic Hound” won’t be identical to the one that constructed “Charming Kitten”. Based on some similarities however, they are commonly grouped together. As a result, analysts see the various names of a group on MITRE’s web site and use their names interchangeably, which is imprecise and misses this key background knowledge.

And what distinguishes your run-of-the-mill script kiddie from an APT? What turns a UNC (UNCagetorized threat actor, Mandiant) into a Kitten (threat actor linked to Iran, CrowdStrike)? Most vendors have their “secret magic sauce” that marks the birth of a threat group. As an open-source, vendor-neutral rule of thumb: An activity group can be formed when two vertices of the Diamond Model overlap across intrusions or campaigns, for example overlaps in infrastructure and victims. Note that this must not be necessarily an “APT”; and all companies will have their own methodology of what constitutes “their” APTs.

Finally, APTs change over time. China’s PLA and general governmental system have been overhauled multiple times under Xi Jinping, and the TTPs of the Lazarus Group has surely not stayed identical since 2009.

Which name to use, then? CTI is, like its big aunt Journalism, about who consumes the intelligence. Most of your audiences won’t be aware of the minutiae detailed in this blog post. They likely don’t even care. So: Use the threat actor name that fits your audience, but try to point out somewhere whose nomenclature it is and provide one, two alternative names. If you say “ELECTRUM”, my brain will draw a blank, but if you say “Sandworm”, I’ll see the elephant.

- https://cybersim.ch/posts/apts-and-elephants/ -

It's Been a While

check my website (cybersim) — Fri, 05 Dec 2025 10:00:24 +0100

cybersim's blog https://cybersim.ch/posts/its-been-a-while-infosec/ -

“You know a lot”, remarked an apprentice recently while I was looking at some tickets from the Stone Age and explaining long-gone threats. The more younger people join infosec, the more I realise: Yes, in fact, I have been around for a while. Surely, Hackers (1995) was my gateway drug, but attending my first Chaos Communication Congress in 2010 was true First Contact. Back then, if you weren’t already working in tech, this was one of the few opportunities to dive into infosec topics. Live-hacking CTFs on YouTube were not really a thing yet, and if you wanted access to the scene, this was the place to go. Most other conferences — even today — have price tags clearly meant to be paid by an employer and not by yourself. And if it’s one-third of your part-time salary, why even bother? During the 2010s, I became a regular C3 attendee while also connecting with more and more peers online (thanks, 2010s Twitter). And even though I’m only a few years into an actual infosec job, the knowledge has accumulated: Confs, meet-ups, more confs, workshops, years of tech-journalism, a computer science degree.

A lot has changed. My first Blackhoodie workshop was a watershed moment (eternal thanks, Marion): For the first time in my life I was in a room full of other women keen on learning reverse engineering. They exist! The event grew from 30 to 80 women within a couple of years and is now a movement of its own, with self-organised events all over the world. Infosec has become more accessible. Gatekeeping in offensive security and “hacking” is still the modus operandi, but CTFs have proliferated, most conferences stream to YouTube (even some pricey ones), tutorials and materials are everywhere.

A lot has not changed. Fifteen years later, I still enter a room full of strangers and no one takes it for granted that I can actually handle a computer, let alone a command line. You can have a conference for women in cybersecurity, but once you subtract all non-technical people, the number left is your representation in today’s teams. Women’s toilets at infosec events still have next to no queues (except for said conference type). Non-tech people are amazed to hear that, yes, the percentage of women (and other non-cis-het-White-male identities) in my field has remained low. Please do not ask me why.

But: Over the last 15 years, I’ve grown to become part of communities — especially local communities. Our tiny, casual meet-up before the Pandemic has grown into a sizeable member group. Once a month I get to sit at a table with like-minded FINTA people; once a year I am in a room full of good old friends and peers who accompanied me for the last 10+ years (thanks!). “Nevertheless, she persisted”, goes the popular saying, and hopefully, when I retire, everyone who enters a room will at least be considered a potential next Acid Burn.

- https://cybersim.ch/posts/its-been-a-while-infosec/ -

The Birthday Dress

check my website (cybersim) — Sat, 13 Sep 2025 11:01:09 +0200

cybersim's blog https://cybersim.ch/posts/the-birthday-dress/ -

Some birthdays are more special than others, and some projects are more challenging than others. My “Joan Wiggle Dress” represents both. The pattern was the reason why I joined Gertie’s Patreon in the first place, but the striped, stretchy Ponte that is required for the chevron pattern was surprisingly hard to find. After several twists and turns my research led to a mauve stretch lace with a wave pattern (of all things) in an online shop. Several months later I found the perfect backdrop in a brick-and-mortar shop: silver faux leather. Never have I sewn lace or faux leather — this project was going to match the birthday in uniqueness.

While the pattern is simple and comes together quickly, the materials were a challenge: Lace is tricky, faux leather too thick to handle, and together they spell trouble. I did a lot of research on sewing techniques and received savvy advice in a local haberdashery shop that has sadly closed since. Without that expert lady there, this project would have failed spectacularly.

Here’s some notes on the process:

Three prototypes
Test-wash everything
One-shot sewing only, no ripping open of seams. Clippers are your best friends – every pin, every needle leaves permanent holes.
Adjusted the height (easy) and orientation of the bust dart (hard). Many finished dresses show a “V”-shaped bust dart instead of a straight line, and I wanted a straight one too.
So much cold sweat lost over matching up the lace on the vertical front seam
Thank you, shop lady: Faux leather dulls the needle in the blink of an eye, so do not use the overlocker or its knife. Changed Jeans needles twice on my regular sewing machine.
Use the lace like interlining and sew it onto the underlying faux leather, then treat as one
To achieve this, use wash-away stay tape to fix the lace onto the seam allowance of the feaux leather, then zigzag the two together
Feaux leather does not need to be finished and can be left untreated
Thank you, shop lady: A wash-away film called “Avalon” from Madeira is used for quilting, but I used it to sew together the bust darts. Sew the film just inside the bust dart seams onto the right side of the interlined fabric, then close the bust dart. That way, the lace does not shift on the slippery feaux leather and you get precise seams.
Hand sewing. So much hand sewing. I sewed all the seams by hand and finished the hem with a lace band.
Best worn with an underdress

The amount of XP gained during this project definitely unlocked a new sewing skill level: The dress turned into an absolute success! This marvelous Joan Wiggle Dress is the first glamour item in my wardrobe, and it shall be worn on many fancy occasions to come. Also the faux leather feels like a second skin and still the dress is comfortable to wear for hours. If ever I can find striped Ponte, there shall be a twin for less glamorous events.

- https://cybersim.ch/posts/the-birthday-dress/ -

Die neue Mitbewohnerin

check my website (cybersim) — Sun, 07 Sep 2025 18:44:48 +0200

cybersim's blog https://cybersim.ch/posts/dressform/ -

Verrenkter Nacken, zerstochene Finger, ungelenke Selfies: Irgendeinmal war genug. Eine Schneiderbüste musste her. Also ein Torso auf einem Ständer, woran eine Hobbyschneiderin alle Änderungen und Drapierungen anbringt, um sie später auf das Schnittmuster zu übertragen. Statt sich verzweifelt vor dem Badezimmerspiegel zu drehen und Nadeln in den ~~Körper~~ Stoff zu stechen, arbeitet man sozusagen mit einem Ersatzkörper. Doch nicht irgendeinen – ich wollte meinen. Denn egal, ob Ready-to-Wear-Kleider oder Off-the-Shelf-Büsten, mein Körper ist jenseits jeglicher Zielgruppenvorstellungen.

Vor Jahren hatte ich bereits ein kleines, kanadisches Unternehmen im Auge. Dessen Geschäftsmodell ist ähnlich wie die (gescheiterte) Idee der Kleidermarke Zozo: Die Kundin erhält einen lustigen, hautengen Anzug nach Hause (Stichwort: Mocap), scannt den eigenen Körper mit dem Smartphone, und das Unternehmen fertigt daraus eine Schneiderbüste. Damals ging die Idee von Zozo ein, weil die Messungen viel zu ungenau waren. Zufällig entdeckte ich ein Schweizer Unternehmen, das ebenfalls Schneiderbüsten auf Mass anfertigt; und da ich gerne lokale Geschäfte unterstütze, unternahm ich einen Roadtrip in die Heimatregion von Wilhelm Tell.

Am Dorfeingang begrüssten mich bereits die Statuen von Tell und Walter, doch bei “Zimmermann Büsten” war es keine Frau Zimmermann, sondern Jenny Curtins: Die gelernte Damen- und Theaterschneiderin folgte während Corona dem Ruf ihrer heimatlichen Berge und kehrte nach Schwyz zurück. Dort kam sie in Kontakt mit der bereits älteren Inhaberin von “Zimmermann Büsten”, einem Betrieb, der seit 1937 Büsten von Hand herstellt. Von ihr lernte sie das Handwerk, übernahm den Betrieb und beschloss, den Fertigungsprozess um einen Twist zu erweitern.

Und hier kommt das wundersame Venn-Diagramm “Hobbyschneiderin” und “Computernerd” ins Spiel. Treffen sich nämlich klassisches Handwerk und moderne Technologie, ist meine Neugierde geweckt. In einem iterativen Prozess entwickelte Jenny Curtins eine neuartige Schneiderbüste auf Mass, wobei sie das gelernte, traditionelle Büstenhandwerk um moderne Technologien erweiterte. Ihre Innovation traf genau meine Schnittmenge, und als Journalistin wäre daraus sofort eine Reportage geworden. Nun arbeite ich aber in der Infosec, und mein kleiner Blog muss für die Story herhalten.

Ich betrat also das Obergeschoss des Chalets, wo sie zusammen mit einer Freundin ihr Atelier beherbergt. In einem holzgetäferten Zimmer mit niedriger Decke zwängte ich mich zuerst in Shapewear, um allfällige Unebenheiten auszugleichen. Dann betrat ich eine kleine Plattform ohne mir den Kopf an der niedrigen Decke anzustossen, und erstarrte in der Pose einer Schaufensterpuppe. Während die Plattform rotierte und ich sie mit Fragen löcherte, nahm Jenny Curtins mit einem 3D-Handscanner meine Körperform digital auf. Zu meiner Überraschung hat die Technologie in den letzten Jahren enorme Fortschritte gemacht und sich von einem hochpreisigen Industrieprodukt zu kommerzialisierter Massenware entwickelt. Massenware, die Innovation im Kleinen ermöglicht. Schliesslich bestaunte ich mein 3D-Rohbild im Handscanner – eine Punktewolke mit noch vielen Unebenheiten, die später am Computer ausgebessert werden müssen.

Danach folgt der Schritt von digital zu analog: Das finale Modell wird im 3D-Drucker gedruckt und dann in klassischer Fertigungsweise zu einer Büste verarbeitet. Der Torso wird aussen gekleistert, mit einer Wattierung versehen und danach mit Stoff satt überzogen. Same-Day-Delivery ist ausgeschlossen – der ganze Prozess dauert mehrere Monate. Und wenn schon lokal, dann richtig: Der Holzständer kam gleich vom ansässigen Drechsler.

Drei Monate später kam sie endlich, die Schneiderbüste, die verdrehte Nacken, zerstochene Finger und ungelenke Selfies erübrigt. Ein Meilenstein nach bald zehn Jahren Hobbyschneiderei: Check! Und last but not least: Bessere Bilder! Denn die Büste ermöglicht, den Blog hier mit besseren Aufnahmen meiner Werke zu zieren. Zwei ältere Blog-Posts haben deswegen gleich bessere Bilder erhalten:

Schlussnotiz 1: Es gibt bewusst keine Links - der Post ist keine gesponserte Werbung, sondern eine Würdigung des Venn-Diagramms “klassisches Handwerk trifft moderne Technologie”. Wer mehr zum Unternehmen wissen möchte, weiss wie googeln.

Schlussnotiz 2: Der Blog-Post entstand zuerst auf Englisch, aber irgendwie hing ich fest. Der Text wollte einfach nicht von der Hand gehen. Wechsel auf Deutsch, und siehe da – die Worte flossen. Ein Fall von lokalem Einkaufen, lokalem Schreiben.

- https://cybersim.ch/posts/dressform/ -

Erasure of Expertise

check my website (cybersim) — Fri, 30 May 2025 08:32:47 +0200

cybersim's blog https://cybersim.ch/posts/erasure-of-expertise/ -

There’s a shop in my town that’s about to close. I usually don’t care about such announcements, as a lot of my shopping happens online anyway. Except when.

Except I’m a hobby sewist with no formal training, who entered the craft by way of Burda without Google or YouTube (1/10 would not recommend). I still learn by failing and improve with each garment, but sometimes I am just stuck. Stuck as in unable to measure myself properly, stuck as in unable to do certain fitting adjustments. Stuck, because of lacking expertise. Of course, there’s plenty of expertise online, yet also “expertise”. This is when human contact helps you to become unstuck. And I kid you not, such humans can also be found in real stores.

Many bigger towns have at least one brick-and-mortar haberdashery selling just that one thing that you need to succeed. That button in the perfect matching color. Top-quality zippers. Horsehair canvas. But these stores not only sell things, they have staff who know their stock down to the last tiny brooch and are often skilled crafters and sewists themselves. Elderly women who have sewn all their lives, know the tricks of the trade, and how to complete the most intricate projects – and who are more than happy to share their knowledge.

This is why I love and need physical shops: They not only sell goods, they also provide expertise. A recent project (to be blogged) had so many firsts that I was at a loss as how to tackle it. Instead of finding questionable opinions online, I went to my local haberdashery and the employee there knew exactly what I needed, provided options, and I succeeded thanks to her help.

When I went back with my finished pride, I spotted a note of doom in the shop window: The original owner retires and closes the store. It will likely be replaced by some chain, whose employees answer every question with “We only have what’s here”. I presented my project to the employee to thank her and share my joy – and expressed my grief about the shop’s closure. Its end erases the best haberdashery in my vicinity, and all the expertise that helped my project to completion. “We are looking for a successor”, she said politely. Her eyes were watery.

Image source: Nightcafé/Google Imagen 3.0

- https://cybersim.ch/posts/erasure-of-expertise/ -

Reflections on Writing and AI

check my website (cybersim) — Sun, 13 Apr 2025 20:00:51 +0200

cybersim's blog https://cybersim.ch/posts/ai-writing-reflections/ -

The most frequent answer you hear in IT: “It depends”. Using AI to write texts? “It depends”. End of story.

Just kidding.

The answer I sent to my colleague was longer — basically a mini-blog post that evolved into this one.¹

First, the good news: Asking GenAI to write texts is useful. My native language is Swiss German; an AI that fixes my non-native English errors is a game changer. I will always miss some meanings, idioms, gerunds — things the AI can spot and explain.

Other use cases resemble the idea of “Gebrauchsmusik”, which in this case would be “Gebrauchstexte” in a wider sense: Boilerplate texts with specific purposes that I find tedious to write, as it is about formalities, less about creativity. Let AI handle the boring stuff.

That’s basically it.

If you look at a football field or a golf course, all you see is grass, neatly trimmed to the same height, conforming to some standard. It’s boring as hell. AI texts are a trimmed sports lawn. Sometimes, uniformly dull with lots of empty phrases.² Sometimes, they are repetitive and lackluster, void of any rhetorical device or sense of rhythm.³ It’s like walking through a concrete industrial space instead of walking through a lively, colorful street. But writing should use all the colors to play and paint a picture that echoes your inner voice.

Text is also like clay: It can be easily reshaped into any form. Following this metaphor, I am rarely attached to my texts.⁴ If you ask me to scrap them, edit them, change them in any way and give me a good reason to do so, I am happy to oblige. Especially if you are an editor who does these things daily. Always trust the editor. The editor will improve your text, while keeping its origins. On the other hand, the AI lawn mower alters your texts ignoring both their essence and your voice.⁵

Which brings up the next point: Editing is not the same as writing. Me reviewing AI output requires a different skill set than writing. It requires fact-checking details, looking for cohesion and what’s missing. Writing, however, is something completely different. As a journalist I learned the phrase that you truly understood a topic if you’ve reported on it. Not by reading, but by creating a story and writing it, no matter if it’s for ears or eyes. Thinking is often a mess, jumbled, chaotic. Writing means putting elements in a sequential order, considering the how and why, what’s left in and out. What’s at the core of this problem? This is the major issue. Then the minor one. Finally, what’s not relevant and left out. AI will save you the trouble, but also shave down your skills.

Coincidentally, two recent blog posts even highlight this: Luciano Nooijen writes how he ditched most code editors as he lost his “Fingerspitzengefühl” when coding (another German word!). Nico Dekens writes how CTI analysts lose their critical thinking because they rely too much on AI output. Writing, coding, critical thinking all turned into passive skills when they are in fact a craft.⁶

I neither want to lose my voice nor my craft.

Addendum: While writing this, I realized that my criticism is actually bigger than the listed advantages. That is not my final verdict though. My final verdict is that we still learn to use AI for all sorts of tasks, while figuring out what helps and what harms us in the long run. We need to try and experiment — and have discussions about it.

Post addendum: Of course I asked GenAI to fix my English. It changed "It depends". to "It depends.". Thanks, AI, but this is a decision that usually precedes eons of newsroom editor fights, and differs across English variants. You just changed it.

The AI also rewrote Sometimes, uniformly dull with lots of empty phrases. Sometimes, repetitive and lackluster, devoid of any rhetorical device or sense of rhythm. to a Sometimes uniformly dull, filled with empty phrases. At other times, they’re repetitive and lifeless, lacking rhetorical devices or any sense of rhythm. It completely missed the intentional repetition for exactly the reason it was criticized.

Image source: Nightcafé/Dreamshaper XL Lightning

Sorry, dear colleague, for the wall of text that may have surprised you, not sorry, because it led to an intellectual exchange that sparked my thinking and writing. Thank you :) ↩︎
Many models have been trained on the sales pitches and ads that pollute the open web. ↩︎
I won’t link that infosec blog post from a big vendor. It was a desert made of concrete. ↩︎
Why else would I publish them here? But: Stealing my texts and “repurposing” them as yours is not clay, that’s plagiarism. ↩︎
This is where “Gebrauchstexte” as a term is handy: They do not need to carry my voice. If I have to write Gebrauchstexte, let the AI do it – no problem. ↩︎
Many, many years ago I read Richard Sennett’s “The Craftsman”. I dimly remember it’s basically about exactly this topic and I should give it another read. ↩︎

- https://cybersim.ch/posts/ai-writing-reflections/ -