Stefan Tilkov retweeted
Eventual consistency in Lotus Notes (1988). It all sounds very modern except for the bit about 2400 baud communication links! dl.acm.org/doi/pdf/10.1145/6…
7
16
2
173
Stefan Tilkov retweeted
BREAKING: we @citizenlab found signs of a #Pegasus spyware infection at the 🇬🇧Prime Minister's office, 10 Downing St.
We notified 🇬🇧.
We'd found other infections within the Gov.. THREAD 1/
Must-read by @RonanFarrow: newyorker.com/magazine/2022/…
71
2,387
307
4,017
Fascinating read, but for me it confirms my belief that the risk is not in AI actually becoming intelligent, but rather us humans mistaking its faking skills for understanding
"Perhaps the game of predict-the-next-word is what children unconsciously play when they are acquiring language themselves."
A brilliant piece by @stevenbjohnson.
nyti.ms/37SdY3M
2
2
18
I always considered it way too abstract and explicitly disliked the idea of schema validation adding information to it. Might have been very different if XML, namespaces and schema had all been designed at the same time, forming a consistent whole
1
1
Stefan Tilkov retweeted
posted: software professionals must take responsibility for the technology we produce. My @thoughtworks colleagues have come up with a playbook of techniques that help us do that
martinfowler.com/articles/20…
8
124
11
386
Having to use XSD to support backwards-compatible services, or trying to build modular components, or dealing with namespace names in attribute values, it all sucked
1
3
The horrible code generation approach led people to believe concepts applicable to local method calls could be transferred to network programming, and thus gave the world the horror that was WSDL
1
1
XSD relies on namespaces, which were added to XML too late, it mixes record-style structure definition into a markup language universe, and it led to the abominable infoset concept. I much preferred RNG or Schematron.
4
10
Stefan Tilkov retweeted
Digitale Gesundheitsanwendungen, die euch euer Arzt verschreiben darf. #diga
116
141
56
908
Stefan Tilkov retweeted
This is an absolutely fascinating and extremely lucid breakdown of a widespread statistical error that has fooled millions of people into thinking that something exists when it doesn't
The 'Dunning-Kruger effect' claims that unskilled people tend to overestimate their ability. The trouble is, the effect is a *statistical artifact*.
This gives new meaning to the title of Dunning and Kruger's paper: 'Unskilled and unaware of it'.
economicsfromthetopdown.com/…
18
117
11
345
Stefan Tilkov retweeted
Jedesmal, wenn Richard David Precht mit seinem aus den volkswirtschaftlichen Werken der letzten 100 Jahre zusammengewürfelten Thesenbrei zum öffentlich-rechtlichen Superintellektuellen hochstilisiert wird, löscht irgendwo ein echter Philosoph das ZDF aus seiner Senderliste.
151
363
20
5,957
Stefan Tilkov retweeted
"old music is so much better than what we have these days"
218
2,987
235
33,269
Das ist alles vielleicht nicht toll, aber unendlich viel besser als das absolut unerträgliche Discovery
1
5
Stefan Tilkov retweeted
The Twitter change where they switched everyone from their choice of timeline to ranked timeline reminds me of a question: why don't social media companies react to negative signals from users?
The change trained me to stop checking the app since it filled my feed with garbage.
11
79
29
489