sharcoal.bsky.social
fka sharcoal (/sharcoal2 /sharcoal69) on the bird app
106 posts
80 followers
131 following
Regular Contributor
Active Commenter
comment in response to
post
It's safer to conceive of it that way than to think about how, but for wealth, this would be you. It's evil that the ruling classes have dug a huge pit in our society, chucked the working class in, and the media is gleefully complicit in blaming them for not climbing out with ease.
comment in response to
post
So many of the people I know are choosing the Maslows--housing and food and survival--by necessity. They get demonised for the very choices they're forced to make, as though poverty weirdly only happens to parents who are shitty and neglectful.
comment in response to
post
if children are given screens to keep them busy while a parent chops reduced-section vegetables and nukes something beige, that parent has prioritised correctly. Food over cursive is always the correct decision.
The vast majority of people are doing their best against bad odds.
comment in response to
post
If all available caregivers are working, including grandparents and extended family, where they can even afford to live nearby, and childcare is £££££, where do you expect people to find the time to teach their child much beyond the basics?
comment in response to
post
omfg no I didn't, good for him
comment in response to
post
now Rumplestiltskin's here to terrorise Colm Meaney
comment in response to
post
if their line doesn't go up it's all our fault :(
comment in response to
post
climate scientists only revise their estimates in one direction, and it's not the good one!
AMOC collapse went from 'unlikely in the next 70 years' to 'likely in the next few decades' to 'could be why Jan's been warmer than expected' oceanographicmagazine.com/news/amoc-at...
comment in response to
post
and today, posting a headline which implies january's warmth shows climate scientists have made a mistake and are stupid, rather than the warmth showing climate scientists are right and we're all fucked, just faster than expected!
comment in response to
post
'me? yeah, i'm uniting millionaires at the Crypto dot com arena. you're welcome, society'
comment in response to
post
you guys need to read the whold job advert
they’re essentially looking for an experienced anarchist organiser who is a chartered accountant on the side
pinkpeacock.gay/jobs/
comment in response to
post
yeah, everyone i know in academia finds it a terrifying and precarious grind. It's really sad
comment in response to
post
ahhh you have such a treat ahead. Incredible show
comment in response to
post
As I said earlier, scaling was always more about business than science. Scientifically, there’s no law of physics that says AI advancements must come from scaling rather than approaches using the same or fewer resources. Scaling is just an incredibly easy-to-follow formula. 17/
comment in response to
post
But DeepSeek shows - in the same moment as the Stargate announcement - that the trade off that OpenAI & co frame as wholly necessary is actually not. 15/
comment in response to
post
Should Stargate follow through with $500B, it'll be the largest infrastructure spend in history and significantly accelerate the already startling pace of data center development post-ChatGPT. In turn that will significantly accelerate the aforementioned consequences. 13/
comment in response to
post
There is empirical evidence that scaling AI models can lead to better performance. For businesses, such an approach lends itself to predictable quarterly planning cycles and offers a clear path for beating competition: Amass more chips. 5/
comment in response to
post
The problem is there are myriad huge negative externalities of taking this approach - not least of which is that you need to keep building massive data centers, which require the consumption of extraordinary amounts of resources. 6/
comment in response to
post
Much of the coverage has been focused on US-China tech competition. That misses a bigger story: DeepSeek has demonstrated that scaling up AI models relentlessly, a paradigm OpenAI introduced & champions, is not the only, and far from the best, way to develop AI. 3/
comment in response to
post
it's so funny. i can't tweet publicly about much of it but oh my god lol, lmao etc..
comment in response to
post
Remember: they reopened coal plants to fuel this "revolution" and now there's some dramatically cheaper thing that does mostly the same stuff without the need for the biggest data centers and newest chips. OpenAI has no real moat! Neither does Anthropic, or Google, or anyone really!
comment in response to
post
[Lwaxana voice] My beloved Mr. Homn, Carel Struycken, lost his home in the LA fires, and I’m certain we can get Starfleet’s finest to help him and his family to rebuild their lives. Now be a dear, and boost. 💋✨