smilekzs

> the GPUs are consumables on a timeline of ~3-5 years

When newer GPUs with clearly better perf/$ are released at a yearly cadence, then older GPUs might indeed become _obsolete_ at this pace, meaning that it makes more economic sense to swap them out because datacenter occupancy is extremely limited. Many of these GPUs might themselves still be functional.

On a related note: Would love to see cyclomatic complexity as a supporting metric for readability. I increasingly find AI-adjacent and AI-written code suffering from this.

My personal yardsticks:

How often does a conceptually incremental change correspond to a proportionally incremental code diff, vs. a surprisingly whole-world-upside-down rearchitect?

Do the code patterns naturally leave enough room for you to incrementally shift/transpose how things are organized?

When the answers to these are unfavorable, I often find what effects to "coloring" in the framework, library, or language, sometimes all at once.

Sigma-Delta Modulation

The legal guardian is responsible for gatekeeping what their minor sees or does not see.

Last time I tried, a lot of CAD/CAE packages still won't operate under WINE, at least not out of the box.

A key consideration in favor of running your local LLM despite all the trouble: The commercial serving endpoint may not exist tomorrow, or at least not at the same price.

+1.

Brought up in a low-trust env and this became habitual.

Google TPUs are VLIW machines on a circuit-switched network.

Or, more neutrally, a different tradeoff point between mass transit and personal cars.

Off-roading aspirations and 3rd row legroom (S1) seem to be major differentiators from Rivian.

As for autonomy, Waymos have LIDARs which at least provides more redundancy.

I see these as different design tradeoffs so no judgment implied.