How web bloat impacts users with slow devices

For cases when you don't agree with styles there is Reader Mode. Your way works also, but Reader Mode just simplier, it is just one click away.

https://en.wikipedia.org/wiki/Transistor_count has a table of transistor count over time. 2001 was Intel Pentium III with 45 million transistors and nVidia NV2A GPU with 60 million. 2023 has Apple M2 Ultra with 134 billion transistors and AMD Instinct CPU with 146 billion, and AMD Aqua Vanjaram CDNA3 GPU with 153 billion. That's some ~3,000x more, about a doubling every two years.

Core frequencies aren't going up, but amount of work per clock cycle is - SIMD instructions are up, memory access and peripheral access bandwidth is up, cache sizes are up, branch predictors are better, multi-core is better.

> "E.g., 1 TiB seems to be ~$50"

You can get a 12TB HDD from NewEgg for $99.99, Joel's blog said $0.0071 per megabyte and this is $0.0000083 per megabyte, ten thousand times cheaper in 23 years. Even after switching to more expensive SSDs 1TB for $50 is $0.00005 per megabyte, a hundred times cheaper than Joel mentioned - and that switch to SSDs likely reduced the investment in HDD tech. And as you say "I'm not sure I'd put more storage in a new machine than what I put in a 2011 build" few people need more storage unless they are video or gaming enthusiasts, or companies.

But TeX was designed to run on a single CPU-core, so no surprise here. I wonder what TeX could become if all Knuth had at the time a multicore machine with cores managing maybe 0.1 MIPS each (or even lower). Like what the world would become if we lived in a counterfactual world where Intel and its buddies starting in 1970s boosted not the frequency and instruction per second per core but number of cores?

My take we'd switched to functional-style programming at 1980s with immutable data, created tools to describe multistage pipelines with each stage issuing tasks into a queue, while cores concurrently picking tasks from the queue. TeX would probably have a simplified and extra fast parser that could cut input into chunks to feed them into a fullblown and slow parser which would be a first stage of a pipeline, and then these pipelines somehow would converge into an output stream. TeX probably would prefer to use more of lexical scoping, to reduce interaction between chunks, or maybe it would make some kind of a barrier for pipelines where they all stop and wait for propagation of things like `\it` from its occurrence to the end.

This counterfactual world seems much more exciting to me than the real one, though maybe I wouldn't be excited if I lived there.