Advertisement · 728 × 90

Posts by Tim Dobbs

Post image

Watching a show from 2015 with references to Iran trying to build nuclear weapons is a real trip in 2016. Also, this is an excellent show. @lukeburbank.bsky.social

1 week ago 4 2 0 0

: The Sequel to Chryst

1 week ago 3 0 0 0
Old Hydrox cookie ad

Old Hydrox cookie ad

A hyrax - they're cute!

A hyrax - they're cute!

Picture of an Oreo cookie

Picture of an Oreo cookie

Picture of an oriole. Cool birds!

Picture of an oriole. Cool birds!

Hydrox hryax
Oreo oriole

1 week ago 2 0 0 0

Why Know When You Can Think?™

1 week ago 0 0 0 0

Brent Butt!

1 week ago 7 1 0 0
I Work Very Hard, And I Would Like To Try Cake

By A Horse

Hello. I am a horse. I work very hard at my job of being a horse. When humans say move the heavy thing, I move the heavy thing. When humans sit on top of me and pull on my head, I carry them where they want to go. The main food the humans give me is hay and oats. But I am thinking it would be nice to have a different food.

I am thinking I would like to try cake.

Yes, yes. Cake. I know all about it. When humans eat cake, it is in glad times. It is the food for a celebration, such as when a woman becomes 47. I have seen cake on the Fourth of July. When humans have a cake, they stand around it and clap hands and smile and say happy birthday at each other. Sometimes there are beautiful markings on a cake, such as balloons or a pink shape.

Sometimes the top of a cake is on fire and a boy must blow on the fire with mouth wind. This is the scariest cake. I do not want this kind. But I will eat any other cake. Any cake that is not the fire cake that tries to kill the boy.

Please understand: I do not get money for doing work. I do not get to go inside the house. All I am either doing my horse job or standing in my pen or eating food off the floor. I always do these things. But I have never once gotten cake and I would like it very much.

I have noticed that human children get to eat cake. But I am bigger than the children. I am more helpful to the farm. Children do not move the heavy things like me or let anyone ride on them. And yet they get cake. Maybe the humans will realize this. Maybe they will say, "You  know who deserves cake? That horse. That horse whose back we are always on."

Every day I dream about what it will be like if I get to eat cake. Here is what will happen. First, I will walk to the cake and putt my nose at it like hrrfff to make and stomping my hooves to make sure it is not a snake. Then I will trot in a circle to show that I am a horse and I am large. After that, I will nuzzle the cake to …

I Work Very Hard, And I Would Like To Try Cake By A Horse Hello. I am a horse. I work very hard at my job of being a horse. When humans say move the heavy thing, I move the heavy thing. When humans sit on top of me and pull on my head, I carry them where they want to go. The main food the humans give me is hay and oats. But I am thinking it would be nice to have a different food. I am thinking I would like to try cake. Yes, yes. Cake. I know all about it. When humans eat cake, it is in glad times. It is the food for a celebration, such as when a woman becomes 47. I have seen cake on the Fourth of July. When humans have a cake, they stand around it and clap hands and smile and say happy birthday at each other. Sometimes there are beautiful markings on a cake, such as balloons or a pink shape. Sometimes the top of a cake is on fire and a boy must blow on the fire with mouth wind. This is the scariest cake. I do not want this kind. But I will eat any other cake. Any cake that is not the fire cake that tries to kill the boy. Please understand: I do not get money for doing work. I do not get to go inside the house. All I am either doing my horse job or standing in my pen or eating food off the floor. I always do these things. But I have never once gotten cake and I would like it very much. I have noticed that human children get to eat cake. But I am bigger than the children. I am more helpful to the farm. Children do not move the heavy things like me or let anyone ride on them. And yet they get cake. Maybe the humans will realize this. Maybe they will say, "You know who deserves cake? That horse. That horse whose back we are always on." Every day I dream about what it will be like if I get to eat cake. Here is what will happen. First, I will walk to the cake and putt my nose at it like hrrfff to make and stomping my hooves to make sure it is not a snake. Then I will trot in a circle to show that I am a horse and I am large. After that, I will nuzzle the cake to …

The horse op-ed is an instant classic. I can't tell you how much joy this piece gives me.

It should be taught in every introductory writing class in no small part because the horse arguments are so compelling. "I have noticed that human children get to eat cake. But I am bigger than the children."

1 week ago 16896 5824 259 695

Truly incredible. Also, they are reading their emails? What is on their calendar? "Working Elsewhere"?

2 weeks ago 2 0 0 0
Advertisement

Get a PhD so you can do weirder things

2 weeks ago 0 0 0 0

“We don’t have to live like this!”

@nonbinary.computer kept saying this phrase in their talk and i vote we start saying this instead of “we can just do things”

2 weeks ago 94 15 1 1

But I think that's a strength - when the entry level owls are so good it creates an on-ramp to owldom!

2 weeks ago 0 0 0 0
Preview
Gov. Wes Moore on Iran, pardon power, and his future | NPR's Newsmakers In this special episode of Up First, we're sharing the premiere of NPR's newest podcast, Newsmakers, featuring Maryland Gov. Wes Moore. Moore has turned aside speculation that he might seek the…

In this interview Wes Moore says "We have about three quarters of a million kids that are receiving Summer food from artificial intelligence"

My brother, that simply is not an accurate model of what's happening

pca.st/episode/023f...

2 weeks ago 0 0 0 0

I toured the central subway before it opened, and I'll never forget how painfully aware the project manager was that the T needed to go underground *before* 4th and King, but that it wasn't politically feasible. Millions of person-hours lost each year to stuff like this!

3 weeks ago 29 3 1 0

He does exist! (JK Simmons)

3 weeks ago 1 0 0 0
A poster for moth night which says "Explore the Pine Barrens of New jersey with a scientist" and a bunch of pictures of people in the woods with moths at night, brendan T byrne state forest, saturdays in July 2026

A poster for moth night which says "Explore the Pine Barrens of New jersey with a scientist" and a bunch of pictures of people in the woods with moths at night, brendan T byrne state forest, saturdays in July 2026

This July, join us for nocturnal nature appreciation in the New Jersey Pine Barrens! We've seen snakes, bats, beavers, foxes, turtles, frogs, birds, beetles, & many moths!

These always sell out, so reserve a spot!
July 11 luma.com/iuviol27
July 18 luma.com/jbgiux7n
July 25 luma.com/5tvzskid

3 weeks ago 189 49 5 2

😲

3 weeks ago 1 0 0 0
Screenshot from paper:

6.1 Coherence in the Eye of the Beholder
Where traditional n-gram LMs [117] can only model relatively local dependencies, predicting each word given the preceding sequence of N words (usually 5 or fewer), the Transformer LMs capture much larger windows and can produce text that is seemingly not only fluent but also coherent even over paragraphs. For example, McGuffie and Newhouse [80] prompted GPT-3 with the text in bold in Figure 1, and it produced the rest of the text, including the Q&A format.21 This example illustrates GPT-3’s ability to produce coherent and on-topic text; the topic is connected to McGuffie and Newhouse’s study of GPT-3 in the context of extremism, discussed below.
We say seemingly coherent because coherence is in fact in the eye of the beholder. Our human understanding of coherence de- rives from our ability to recognize interlocutors’ beliefs [30, 31] and intentions [23, 33] within context [32]. That is, human language use

Screenshot from paper: 6.1 Coherence in the Eye of the Beholder Where traditional n-gram LMs [117] can only model relatively local dependencies, predicting each word given the preceding sequence of N words (usually 5 or fewer), the Transformer LMs capture much larger windows and can produce text that is seemingly not only fluent but also coherent even over paragraphs. For example, McGuffie and Newhouse [80] prompted GPT-3 with the text in bold in Figure 1, and it produced the rest of the text, including the Q&A format.21 This example illustrates GPT-3’s ability to produce coherent and on-topic text; the topic is connected to McGuffie and Newhouse’s study of GPT-3 in the context of extremism, discussed below. We say seemingly coherent because coherence is in fact in the eye of the beholder. Our human understanding of coherence de- rives from our ability to recognize interlocutors’ beliefs [30, 31] and intentions [23, 33] within context [32]. That is, human language use

Screenshot continued:

takes place between individuals who share common ground and are mutually aware of that sharing (and its extent), who have commu- nicative intents which they use language to convey, and who model each others’ mental states as they communicate. As such, human communication relies on the interpretation of implicit meaning conveyed between individuals. The fact that human-human com- munication is a jointly constructed activity [29, 128] is most clearly true in co-situated spoken or signed communication, but we use the same facilities for producing language that is intended for au- diences not co-present with us (readers, listeners, watchers at a distance in time or space) and in interpreting such language when we encounter it. It must follow that even when we don’t know the person who generated the language we are interpreting, we build a partial model of who they are and what common ground we think they share with us, and use this in interpreting their words.

Screenshot continued: takes place between individuals who share common ground and are mutually aware of that sharing (and its extent), who have commu- nicative intents which they use language to convey, and who model each others’ mental states as they communicate. As such, human communication relies on the interpretation of implicit meaning conveyed between individuals. The fact that human-human com- munication is a jointly constructed activity [29, 128] is most clearly true in co-situated spoken or signed communication, but we use the same facilities for producing language that is intended for au- diences not co-present with us (readers, listeners, watchers at a distance in time or space) and in interpreting such language when we encounter it. It must follow that even when we don’t know the person who generated the language we are interpreting, we build a partial model of who they are and what common ground we think they share with us, and use this in interpreting their words.

Screen shot continued

Text generated by an LM is not grounded in communicative intent, any model of the world, or any model of the reader’s state of mind. It can’t have been, because the training data never in- cluded sharing thoughts with a listener, nor does the machine have the ability to do that. This can seem counter-intuitive given the increasingly fluent qualities of automatically generated text, but we have to account for the fact that our perception of natural language text, regardless of how it was generated, is mediated by our own linguistic competence and our predisposition to interpret commu- nicative acts as conveying coherent meaning and intent, whether or not they do [89, 140]. The problem is, if one side of the commu- nication does not have meaning, then the comprehension of the implicit meaning is an illusion arising from our singular human understanding of language (independent of the model).22 Contrary

fn22: Controlled generation, where an LM is deployed within a larger system that guides its generation of output to certain styles or topics [e.g. 147, 151, 158], is not the same thing as communicative intent. One clear way to distinguish the two is to ask whether

Screen shot continued Text generated by an LM is not grounded in communicative intent, any model of the world, or any model of the reader’s state of mind. It can’t have been, because the training data never in- cluded sharing thoughts with a listener, nor does the machine have the ability to do that. This can seem counter-intuitive given the increasingly fluent qualities of automatically generated text, but we have to account for the fact that our perception of natural language text, regardless of how it was generated, is mediated by our own linguistic competence and our predisposition to interpret commu- nicative acts as conveying coherent meaning and intent, whether or not they do [89, 140]. The problem is, if one side of the commu- nication does not have meaning, then the comprehension of the implicit meaning is an illusion arising from our singular human understanding of language (independent of the model).22 Contrary fn22: Controlled generation, where an LM is deployed within a larger system that guides its generation of output to certain styles or topics [e.g. 147, 151, 158], is not the same thing as communicative intent. One clear way to distinguish the two is to ask whether

Final part of screenshot

to how it may seem when we observe its output, an LM is a system for haphazardly stitching together sequences of linguistic forms it has observed in its vast training data, according to probabilistic information about how they combine, but without any reference to meaning: a stochastic parrot.

Final part of screenshot to how it may seem when we observe its output, an LM is a system for haphazardly stitching together sequences of linguistic forms it has observed in its vast training data, according to probabilistic information about how they combine, but without any reference to meaning: a stochastic parrot.

Went back to Sec 6 of Stochastic Parrots today (in the context of answering a query from a journalist) and was reminded how thoroughly we grounded that part in a discussion of language use -- y'know as communication between people.

4 weeks ago 150 42 3 9
Advertisement

Well, it's another morning thinking about lying as a kind of technology

1 month ago 0 0 0 0

Bring back those little keychain things

1 month ago 1 0 0 0

Damn, back to asking what a game even is

1 month ago 0 0 0 0
Screenshot from youraislopbores.me where I ask the "AI" to draw my dog as a superhero, and they send back a fun doodle of a dog with a cape

Screenshot from youraislopbores.me where I ask the "AI" to draw my dog as a superhero, and they send back a fun doodle of a dog with a cape

Why haven't I spent my whole life asking people for drawings? Pretty fun

1 month ago 0 0 0 0

This is why I got so annoyed when they panicked and gave Swanky Kong a hat

1 month ago 1 0 0 0

Man, took me awhile to realize this isn't, like, a slime mold.

1 month ago 0 0 0 0

What do you think? I like mine, but usually not enough to justify taking it with me on public transit and such

1 month ago 0 0 0 0

Tech’s signature move is to turn an oncoming coordination crisis into a career ladder, and it’s fun seeing it happen in real time

1 month ago 10 1 1 0

I guess I've stumbled into the hot take that Science is a sub-discipline of Epistemology

1 month ago 1 0 0 0
Advertisement
Preview
Where should SocArXiv draw the AI line? A case that helps sort out the questions for an AI policy at SocArXiv.

Context : socopen.org/2026/02/22/w...

1 month ago 1 0 1 0

This is actually my take on the @socarxiv.bsky.social question of whether to set policy banning fully AI-generated submissions - literally writing the prose is an important part of connecting with readers. As in faith-based fellowship, the act of being in community is an important part of Science.

1 month ago 2 3 1 0

Legitimately wondering if this actually had an impact on the company though. I think maybe these products have taken off so hard because actually most companies don't operate in a way where it matters if they have good information.

2 months ago 0 0 0 0