The Slovakian peasants used to be famous for the shawls they made. These shawls were wonderfully colored and patterned, woven of yarns which had been dipped in homemade dyes. Early in the twentieth century aniline dyes were made available to them. And at once the glory of the shawls was spoiled; they were now no longer delicate and subtle, but crude. This change cannot have come about because the new dyes were somehow inferior. They were as brilliant, and the variety of colors was much greater than before. Yet somehow the new shawls turned out vulgar and uninteresting.
Now if, as it is so pleasant to suppose, the shawlmakers had had some innate artistry, had been so gifted that they were simply "able" to make beautiful shawls, it would be almost impossible to explain their later clumsiness. But if we look at the situation differently, it is very easy to explain. The shawlmakers were simply able, as many of us are, to recognize bad shawls, and their own mistakes. […]
So we do not need to pretend that these craftsmen had special ability. They made beautiful shawls by standing in a long tradition, and by making minor changes whenever some thing seemed to need improvement. But once presented with more complicated choices, their apparent mastery and judgment disappeared. Faced with the complex unfamiliar task of actually inventing forms from scratch, they were unsuccessful.
— Christopher Alexander, Notes on the Synthesis of Form
Top-ranked discussion points in Silicon Valley’s break rooms this week:
DeepSeek / Did you buy the dip?
Severance
RTO and layoffs
Predicted top-ranked discussion points this coming week:
So did you buy the dip?
Grammys / Oscars
Tariffs
Tech workers mostly laughed at Wall Street this week for shorting chip stocks in apparent response to DeepSeek’s claim of a 10x cheaper GPU usage opportunity to train a suite of models rivaling OpenAI’s. NVIDIA stock dropped 16% within the first trading window (from $142 to $118 a share), recovered 7% of its original value the next day, then stabilized back down to its current share price of $120 (a 13% overall drop in January) where it’s currently trading. (One of my favorite trackers to watch is the Nancy Pelosi Portfolio.)
President Trump’s tariffs are already seeing retaliatory tariffs from Canada and Mexico (jury’s still out on China) which throws another wrench into increasingly tumultuous market trading behaviors. Last week I praised the DeepSeek paper for its beauty and inventiveness, but this week I want to highlight the brewing societal and economic shifts it has brought to the forefront of global attention and speculation. For this analysis I will use the highly scientific and precise method of reflecting on what sentiments has been algorithmically served to me via X over the past few weeks and attempting to interpret and distill them here in my own terms.
First, we are seeing a heightened collective social desire and nostalgia for organic experiences and technologies in the way we were accustomed to before the current wave of online-first assistants, chatbots, large-scale automation. Flipside (2023) provided a pretty good picture of vinyl culture; record shops were once a natural community-building and music discovery third place now largely phased out by streaming services, leaving a dwindling and largely aging population of record collectors in their wake. Our work as a society is being pushed so close (and in many instances past) the limits of what can be automated that it’s humanly impossible to understand what we have really created. Writing, image, audio and video generation are becoming so good and so indistinguishable from even those of highly-trained subject matter experts that a friend recently wondered aloud if the script of the film we were watching had been generated by ChatGPT. (Definitely could have been written by DeepSeek.)
I recently skimmed Christopher Alexander’s Notes on the Synthesis of Form (from which I pulled the passage that opens this blog post) which also reminded me to link to the paper Transformers Learn In-Context by Gradient Descent (Oswald et al., 2023).
I’ll end the blog post here because I’ve been staring at this screen all afternoon and I think it’s time to officially call it, but thanks for reading and may you get through the Sunday scaries in peace tonight.
