The chill of the early morning lent these closely-planted miniature irises the air of having huddled together against the cold. The colour and shape of them couldn’t help but brighten the morning, however.

Many things in It’s So Sad When Old People Romanticize Their Heydays, Also the 90s Were Objectively the Best Time to Be Alive reminded me of growing up. Overall a trip down memory lane, but also a reminder, similar to A truly great country is within reach, that there was a period between about 1990 and 2008 where things seemed to be, generally, on the up.

A record store was a place. And places demonstrate importance; sometimes they demonstrate devotion. You’d go in there and there would be a couple vaguely pretentious staff members and people pawing through racks of CDs and a wall of t-shirts and posters. And they’d play cool shit that you hadn’t heard before, which was one way to discover new stuff. So was flipping endlessly through every CD in a row. When you were there you were Doing Music.

Probably there were a couple of proper record stores in Chester, UK where I grew up, but I was not the kind of confident kid that’d hang out under the watchful eye of any vaguely pretentious staff regardless of the shop type.

A recent piece reminded me of the positivity that we had in the early 2000s.

I still retain an optimism, but it’s a little scary to look how much the country has changed in the last 15 years. In the 2000s, after Labour’s landslide in 1997, the atmosphere was far more positive than it is now. And when we look to things like the quality of schooling and, particularly, the ratings of the NHS, we can see that feels justified.

Nesrine Malik recalls arriving in the UK in the mid-2000s and finding a welcoming country that allowed her to build a life from very little.

Alan Jacobs takes the time to look at one writer who thinks about how good it would be to write without writing, and a second who looks forward to a time when they can write about something without researching it. How do they imagine doing it? Using AI that mocks their writing or researching style.

Alan captures why this is likely to result in utterly flawed pieces:

As I was walking this morning I suddenly understood the most fundamental thing that’s wrong with the way Smith and Thompson think about these matters: Smith assumes that at the outset of a writing project he already knows what he wants to say and just has to get it said; Thompson assumes at the outset of a writing project that he understands what he needs to know and just has to find a way to know it. But for me writing isn’t anything like that.

I find this too, that in the process of writing, I often end up somewhere different than where I expected to arrive when I started out.

A GPT language model can only use about 4,000 words of context when generating its next words. The resources required for generation in a GPT increase significantly as this “context window” expands. Unlike a normal Google search which takes milliseconds to run, generating responses in ChatGPT, for example, takes on the order of seconds. That’s expensive.

I’d not thought through the implications of this limited number of tokens a GPT large language model can use when generating its output. As Allen Pike explains in A 175-Billion-Parameter Goldfish, as well as dollar-cost, this has deep effects.