• 0 Posts
  • 313 Comments
Joined 3 years ago
cake
Cake day: June 18th, 2023

help-circle
    1. It has gotten bigger. More active. More posts, more new content. When i first came over i would check the All page, sorted by eitber Active or Hot, and only find a couple of new posts per day. It is still nowhere near as active as Reddit was back then (probably a goos thing), but it has enough content to help me procrastinate at work now.

    2. LemmyNSFW died and has been replaced by FediNSFW recently. I am sure that it will be better in the long-term, but it still doesn’t seem to be back to where it was yet. I think a lot of the old posters were bots, largely re-posting from Reddit, and not all of those have been rebuilt yet. I have mixed feelings about that.

    3. The Connect app has gotten better and better. Love it.

    4. For the past year or so, Lemmy has been of a size big enougj for patterns to ripple and promulgate through it bht small enough to notice them. For example, almost immediately after New Years several different communities on different instances started to see a drastix influx of webcomic posts. Usually 4-panel ones. Usually low-fidelity ones (XKCD-style, not Girl Genius for example). And usually oned with some sort of error or controversy. Rage bait to get the comments going, but nothing controversial enough to get banned or removed.

    There would be new accounts made that just posted a handful of these comics quickly, and sometimes argue with people in the comments. Once people like me started pointing out the pattern they started deleting the posts and accounts after a couple days. I’m not sure when it stopped, but i have not noticed one for probably a month.




  • One of the most important pieces of Trump’s 2024 election was getting the Hispanic vote. Sure, he spent years before, during, and after his first term talking about cracking down on immigration on the southern border, talking about how all of the people coming from Mexico or further south were a bunch of criminals and drug dealers. BUT he courted the Cuban population in southern Florida - mostly the descendants of people who were wealthy enough to have their wealth redistributed by Castro. By calling Harris a communist he was able to get their votes, win Florida, and that would have made the difference in the election.

    The very concept of who counts as “white” changes depending on what the racists of the time want. In the not-too-distant past, Irish and Italian people were not considered white. I still don’t know if Bernie Sanders or other Jews are considered white or not.

    Even for data like this… Is it really a black women vs white women issue, or is it a rich person vs poor person issue? And yes, those economic division have deep roots in the history of racism, but that doesn’t explain the whole picture.

    Look at the people around Trump. Secretary of State Marco Rubio is one of those Cuban south Floridians. Secretary of Labor Lori Chavez-DeRemer is a Hispanic woman. Director of National Intelligence Tulsi Gabbard is part Samoan, born in Somoa. FBI Director Kash Patel is Indian - his family fled persecution in Uganada.

    And then there’s foreign policy. Trump seems to largely be pushing the USA away from predominantly white European countries that have historically been allies. Trying to break up NATO and undermine the EU, making threats against Canada and Greenland. Cozying up to Muhammed Bin Salman.

    The people in power love when the masses fight amongst themselves.






  • I’m comparing gender identity to other forms of protected identity: race, ethnicity, sexual orientation, etc.

    I’m not making any judgements about any group being superior or inferior to another. You’re the one doing that. Being “physically imposing” is subjective and variable. There is significant overlap between the largest women and smallest men, and that’s just staying withing the confines of binary cis people. Not to mention… Guns exist. Cars themselves are weapons far more dangerous than any human regardless of gender.

    Building a just society means we need to leave behind our biases and fears. To judge individuals not on the circumstances of their birth but the content of their character. That means discarding the luxury of pre-judging people, not just when it is easy, but also when it is hard.



  • It’s shocking to me how so many people I know who are woke as fuck and quick to shout down any bigotry are so quick to drop all of their principles to hate men.

    Segregation never works, and only serves to foster fear, hatred, and division within society You cannot define what a “woman” is in a way that excluded trans-women without also excluding some cis-women with them Your gender identity is what you say it is. Others should do their best to remember any pronouns or name changes Crime is too complicated and nuanced to be reduced to statistics, which are often used by racists and bigots to justify racist and bigoted policies It’s racist to cross to the other side of the street when a person of color comes walking towards you on the side you currently are on. It’s up to you to “Men” are evil monsters who cannot be trusted and need to be locked away

    It baffles me how so many people can have all of these ideas, including the last one, and not see the cognitive dissonance there. It’s succumbing to fear and hatred, the same methods of divisive propaganda that has harmed every other group.






  • For those who don’t want to read several pages of unnecessary text telling you what you probably already know:

    The math, while pretty involved, may tell a straightforward story (if you’re interested in the details of our analysis, see the Appendix). OpenAI has contracted 900K memory wafers per month from Samsung and SK Hynix. Partner commentary seems to indicate that’s a monthly number, so that represents 10.8 million wafers over 12 months. In terms of demand, a fully built-out 10GW Stargate cluster would require ~3 million GB200 Bianca Boards. Each board requires ~50% of a memory wafer in total; split between the HBM3e stacks embedded into its two B200 GPU (~30%) and its 480 GB of LPDDR5X system memory (~20%). That puts total wafer demand for the entire cluster at ~3 million wafers.

    Therefore, according to our best estimates, OpenAI likely needs less than 30% of the 10.8 million wafers it’s planning to buy

    So this is just putting some numbers to what a lot of people already guessed. The AI companies are not just buying a ton of RAM to build out their data centers. They aren’t buying enough other components to even use that RAM. They’re buying it so that no one else can.



  • An interesting theory, but I think the key question that needs to be answered is: why would OpenAI stakeholders be prepared to tank the company for Microsoft?

    I’m not saying it’s impossible. There could be some sort of tie between Altman and Microsoft, or the high level executives or shareholders of OpenAI.

    And who would be left holding the bag? NVIDIA? Is this a scheme to get them to make a ton of AI-centered GPU’s, so that when OpenAI goes down Microsoft would be able to buy that hardware super cheap? Maybe, but then again someone else could too, which would be easier if OpenAI’s assets were on the open market.

    I’m not buying this theory yet, but it’s kntriguing enough to keep an eye on at least.