ℍ𝕂-𝟞𝟝

  • 0 Posts
  • 31 Comments
Joined 2 years ago
cake
Cake day: July 14th, 2024

help-circle




  • Péter Magyar won the election in a blowout making Orbán’s party practically irrelevant. He is an ex-Fidesz member that broke off and is heading a broad grassroots party.

    He himself is old-school conservative coded, but his party is very broad and includes everyone from racial justice activists to members of the military to quite literally random people. Most of his party is brand new to politics and made up of working class people, most new representatives were like the town doctor or engineers before.

    The party’s policies are a working social net including investments into education and healthcare, an independent prosecutors office and EPPO membership, a special office to prosecute corrupt politicians, adopting the Euro and contributing to a stronger EU and stronger voices to the Eastern EU in Brussels, support for racial, sexual and other minorities and so on.



  • His job is not to reform the EU, his job is to fix Hungary. For that, he needs to normalise relations with the EU, or the country is going bankrupt in months.

    The party which would have been with the social democrats in the EPP has managed to erode workers rights and introduce austerity while their politicians were getting rich off of public funds.

    I’ll take a renewables focused climate policy, less corporate welfare, and generally better social services funded by wealth taxes and reclaimed national wealth from the rich over saying the right things in Brussels.

    At least to me, workers should matter more to the left than party colours or performative bullshit.


  • I actually did an experiment on doing just that. For context, I’m an experienced software engineer, whose company buys him a tom of Claude usage so I had time to test out what it can actually do and I feel like I’m capable of judging where it’s good and where it falls short at.

    How Claude Code works is that there are actually multiple models involved, one for doign the coding, one “reasoning” model to keep the chain of thought and the context going, and a bunch of small specialized ones for odd jobs around the thing.

    The thing that doesn’t work yet is that the big reasoning model has to still be big, otherwise it will hallucinate frequently enough to break the workflow. If you could get one of the big models to run locally, you’d be there. However, with recent advances in quantization and MoE models, it’s actually getting nearer fast enough that I would expect it to be generally available in a year or two.

    Today the best I could do was a tool that could take 150 gigs of RAM, 24 gigs of VRAM and AMD’s top of the line card to take 30 minutes what takes Claude Code 1-2. But surprisingly, the output of the model was not bad at all.