@bro's banner p

bro


				

				

				
0 followers   follows 3 users  
joined 2022 September 05 20:25:55 UTC

				

User ID: 700

bro


				
				
				

				
0 followers   follows 3 users   joined 2022 September 05 20:25:55 UTC

					

No bio...


					

User ID: 700

biggest fan

My first gut reaction whenever I encounter him after having not for a while is "this guy's so fucking annoying, just stop", and the degree to which I ever read any further is out of a desire to see him slip up badly enough I can dismiss him altogether or see him definitively put in his place by someone who clearly knows better.

The problem is I haven't seen that yet.

Here's hoping I don't get so entangled this or any time that I can't return to not paying attention. There's enough other and more consequential shit where the same general dynamic that his claims point towards is ruining everything, that it would already be several full time jobs' worth of time and misery to attempt to follow it all.

But yes, unless he's been properly owned and continued anyway in purely ego-driven contrarianism-poisoned bad faith, god bless him for not dropping it despite it making him This Fucking Guy? Again?

I'm also not 100% certain about "Scott's disappearance". Do you mean immediately following the doxxing threat until he reemerged a few months later, or do you mean ongoing since the doxxing (ie, metaphorically, old Scott never came back).

Karlin (and Hanania) definitely know who Scott Alexander is, it's probably just not immediately salient to them that he went dark for a few months 4 years ago.

It's hard to tell, because there are a very few tiny fragmentary stories that come up in searches and I haven't been checking front pages for it every day to see if they've been highlighted, but it seems like effectively no general-interest major mainstream news organization has really said much/anything about the rollout of ChatGPT's vision capabilities yet.

LLM vision was already obviously 100% set in the pipeline so its arrival should not be that surprising to anybody who's been paying sufficient attention for the last year-and-a-half, but that's almost nobody. Given how much the next several normie-tiers down from that, including normies who matter, do seem to be relatively engaged with ChatGPT and its consequences, and they almost certainly had (and from this still have) no idea this was coming, shouldn't it be at least some degree of international frontpage news that it's even been announced?

I guess they're waiting until it gets actual wide user release over the next few weeks, but it's not like the media here are sober conservatives about not jumping the gun on things that will obviously eventually be enormous consequential news to their readership. What's going on here?

How many tokens are in Twitter? How many new ones are generated per day? What proportion are they of humanity's total output? What are they worth?

What the fuck was this responding to:

Yo Shavit @yonashav 7h
Replying to @rharang

We very intentionally are not sharing it widely yet - the hope is that a mini public demo kicks a social response into gear

I can't see the tweet it's replying to. Yonadav Shavit is "Head of Preparedness" at OpenAI.

Surely just about actual use access to Sora, right?

Could you expand on this at all?

This is incredible, thank you so much for this.

(also thank you OP, I am eagerly following your series)

Regarding the viral Planet of the Bass, it was surprisingly difficult to dig this half-remembered reference out of the thick sludge of my brain, and Google and LLMs were not helping when I wrongly told them it was Balkan; so in case others are experiencing the same struggle: Stop The War - Speak The Hungarian Rapper

Whack the bottom with the heel of your hand a few times then try again.

If that doesn't do it, the never-fail solution is to put some dents in the edge of the lid by tapping around its circumference with a utensil. The best option for this is holding a butter knife by the blade and hitting with the handle.

Not sure how bad this is for the seal if you're intending to reseal it long term, but this always works.

Why don't you believe Musk knows or cares about that? There's extensive evidence that he's extremely AI pilled.

I haven't seen enough evidence or skepticism on display to fully buy into the Google Cloud claim (or any of the other just-desserts infrastructure stories), especially against the motivations and track records at play.

This seems to be the only source that ever reported anything about it:

Weekend scoop with @ZoeSchiffer for subscribers: Twitter is stiffing Google on its payments for Google Cloud, and significant parts of the company’s trust and safety infrastructure could collapse by the end of the month

I understand the impulse, but this is a sour grapes, "actually by losing we won" take.

This is a low-key minor disaster for everybody, the world is poorer, nobody wins, vs a world where this did not happen (or if it turns out to be "necessary", in a world where it was not "necessary").

But yes, if this holds, Twitter delenda est.

Is it just me, or did Twitter just completely nuke all content from being viewable without logging in?

my pick whenever I'm not using Mac

How are the bathhouses in Istanbul, I hear good things

The furthest I've been able to trace this has been to an Intel Slava Z telegram post of the only screenshot of it that seems to exist (2 minutes before the Eva Bartlett tweet of the same screenshot, that Musk later responded to), and I'm not fluent enough in the russian side of the internet to look any further than that.

There are no public archives of it:

https://archive.ph/https://myrotvorets.center/*

https://web.archive.org/web/*/http://myrotvorets.center/*

Off-topic for subthread but since you are here:

Was Elon Musk actually on Myrotvorets or is the screenshot fabricated?

For the above comment: again, did not appear in incognito at first, then did within a few seconds of self-reporting the comment.

I will leave this one un-self-reported for a few hours as an experiment.


20 minutes: this comment still does not appear publicly.


45 minutes: comment appears publicly after Zorba's attention

If you're in a 1970s British TV WWII documentary mood, also give The Secret War a try (vimeo, wikipedia)

I can't easily find a reference for it, but I think Scott asked for his full comments to be inlined as part of his deal for moving his blog there (which was a big deal for Substack at the time).

I'm not sure if there's an option for it that just nobody else uses, or he's given a special case.

I guess since the comments perform fine in the tiny paginations everyone else has, and he's no longer as important to them since they've grown by orders of magnitude, they've never bothered fixing it.

I think this would be more "people need to freak out about this and have it widely disseminated that this is now possible and generate some antibodies" than directly about the explicit concrete program you're imagining.

Though that will probably also happen, if it's not already, and that is almost certainly somewhere on this guy's mind, and all insiders to this generally, to greater or lesser degrees, whether in favor of it or hoping to stave it off.

If you want to read some more particular entanglement directly into that tweet, there'd be good odds on "yet" being mostly understood to mean "before the election".

I don't think the Titanic is in a trench. Though I'd also kinda dimly thought it was, probably just Mandela-Effect-style confusion with the Mariana Trench. From either half-remembered childhood Deep Ocean Facts, or their more recent James Cameron connection.

Very low-hanging fruit, but here are Ian Miles Cheong and Tim Pool beclowning themselves in the face of this, and Elon Musk getting real close to it.

Is the new season of Star Trek: Picard worth watching?

For the type of curmudgeonly old trekkie who had to stop watching Discovery after 1 season because it was genuinely too hurtful to see that done to something he loves, and has only experienced any of Picard through the noble sacrifice of the Red Letter Media reviews?

I stopped watching the RLM reviews halfway through this season because it seemed like it might actually end up worth watching the show, but the problem is now I can't check in to see their final verdict without having it spoiled.

Eigenrobot loves his illegibility, but I think if you were to really pin him down "far right" would be a nontrivial mischaracterization.

He's not especially anti-immigration, for example. And he considers "right wing" an outgroup. Among other things.