daniyeg [he/him]

  • 13 Posts
  • 806 Comments
Joined 2 years ago
cake
Cake day: August 18th, 2024

help-circle


  • time is not really on anybody’s side, but the US elite can afford to have the strait closed for a few more weeks in order to extend the war of attrition on iran’s economy. the blockade is still trump’s preferred tactic and he doesn’t want to negotiate, only to dictate terms he dug himself into by publicizing them. china is still choosing to release its reserves and there will be a temporary oil boost when the tankers that left during MoU reach their destination.

    i think the hope within the US administration is that iran’s liquidity problem or another specific mechanism blows up into another internal crisis. there had been discussions about liberalizing the electricity prices which would almost certainly lead to protests, but i don’t know if those plans were postponed or not because of the war. it could also be that after the summer the inflation is simply too much and without wages keeping up there is a second round of protests. overall it’s not great and US keeps exploiting iran’s willingness to negotiate in order to delay the oil doomsday deadline.







  • fans, preferably ceiling fans but desk fans and standing fans also work. for ceiling fans make sure the fan is spinning the correct way. for desk fans you can do a poor man’s version of water AC and put a thin damp rag (not dripping otherwise the water might damage the fan) on the cage so it blows the air through the rag. it probably only works in dry climates though.

    if your building doesn’t trap heat well, open the windows and doors to have air circulation but have your curtains closed to minimise the amount of light entering the building.










  • the ML field is sustainable and productive… LLMs and LLM derivatives are not. they have an inherent use case which is why they’ll stay around (unlike NFTs), but there are two factors that lead me to not be concerned about their climate impact for now:

    1. all these big numbers you see about the AI buildup and its impact will probably never happen because i think the enthusiasm runs out and kills the bubble before oversupply takes over and pops it.

    2. chinese models are far more efficient and cheaper to run while being open weight, i think they will absolutely dominate unless US companies also make cheaper models. regardless in both scenarios the compute needed falls drastically. granted they are still not sustainable, but they are better in that regard.

    i think the available infrastructure right now can perfectly handle the natural demand for this tech at this point. the issue is that they keep selling this idea of “AGI” and total AI domination, which is not going to happen, at least with transformers and LLMs. there’s also the structural issues with ML approaches, like past is not a reliable indicator of the future and not everything can be optimised and represented by an objective function.