Becky
@beckvalle.bsky.social
Background in geography, archaeology, & programming - agent-based evacuation models - high performance computing - geographic information systems - learning Bangla - interested in many things - UIUC CIGI Lab & CyberGIS Center - She/they
Reposted by Becky
I turned off Gmail’s new default setting that uses your inbox to train AI, which removed all sorts of functionality that never required AI before, including inbox sorting. Went from 5 unread to like 10,000 instantly lol.
I guess I’ll be unsubscribing from literally everything! Enshittification.
I guess I’ll be unsubscribing from literally everything! Enshittification.
November 22, 2025 at 12:59 AM
I turned off Gmail’s new default setting that uses your inbox to train AI, which removed all sorts of functionality that never required AI before, including inbox sorting. Went from 5 unread to like 10,000 instantly lol.
I guess I’ll be unsubscribing from literally everything! Enshittification.
I guess I’ll be unsubscribing from literally everything! Enshittification.
Reposted by Becky
Oh wait this is actually a good idea
November 22, 2025 at 4:27 PM
Oh wait this is actually a good idea
Reposted by Becky
It’s widely known (and, I think, pretty uncontroversial) that learning requires effort — specifically, if you don’t have to work at getting the knowledge, it won’t stick.
Even if an LLM could be trusted to give you correct information 100% of the time, it would be an inferior method of learning it.
Even if an LLM could be trusted to give you correct information 100% of the time, it would be an inferior method of learning it.
Relying on ChatGPT to teach you about a topic leaves you with shallower knowledge than Googling and reading about it, according to new research that compared what more than 10,000 people knew after using one method or the other.
Shared by @gizmodo.com: buff.ly/yAAHtHq
Shared by @gizmodo.com: buff.ly/yAAHtHq
November 21, 2025 at 12:49 PM
It’s widely known (and, I think, pretty uncontroversial) that learning requires effort — specifically, if you don’t have to work at getting the knowledge, it won’t stick.
Even if an LLM could be trusted to give you correct information 100% of the time, it would be an inferior method of learning it.
Even if an LLM could be trusted to give you correct information 100% of the time, it would be an inferior method of learning it.
Reposted by Becky
15 years ago @jevinwest.bsky.social and I talked about how we needed a data scientists’ code of ethics like the Hippocratic oath.
We still do.
We still do.
November 15, 2025 at 7:41 PM
15 years ago @jevinwest.bsky.social and I talked about how we needed a data scientists’ code of ethics like the Hippocratic oath.
We still do.
We still do.
Reposted by Becky
“Don’t let the bastards grind you down. I love you all.”
November 15, 2025 at 6:15 AM
“Don’t let the bastards grind you down. I love you all.”
Reposted by Becky
No no no begs every archivist. You are never going to be able to find anything. Please don’t start using emojis in file names. Who asked for this? What fresh hell is next?
November 12, 2025 at 10:38 AM
No no no begs every archivist. You are never going to be able to find anything. Please don’t start using emojis in file names. Who asked for this? What fresh hell is next?
A minor yet frustrating thing that AI has ruined for me is food recipes on the internet. So many food blogs I would regularly visit several years ago are impossible to find under pages and pages of worthless text or just gone.
November 11, 2025 at 5:34 PM
A minor yet frustrating thing that AI has ruined for me is food recipes on the internet. So many food blogs I would regularly visit several years ago are impossible to find under pages and pages of worthless text or just gone.
Reposted by Becky
your search box should not be so shit that I have to slap together a web scraper
November 10, 2025 at 4:35 AM
your search box should not be so shit that I have to slap together a web scraper
Reposted by Becky
Can AI simulations of human research participants advance cognitive science? In @cp-trendscognsci.bsky.social, @lmesseri.bsky.social & I analyze this vision. We show how “AI Surrogates” entrench practices that limit the generalizability of cognitive science while aspiring to do the opposite. 1/
AI Surrogates and illusions of generalizability in cognitive science
Recent advances in artificial intelligence (AI) have generated enthusiasm for using AI simulations of human research participants to generate new know…
www.sciencedirect.com
October 21, 2025 at 8:24 PM
Can AI simulations of human research participants advance cognitive science? In @cp-trendscognsci.bsky.social, @lmesseri.bsky.social & I analyze this vision. We show how “AI Surrogates” entrench practices that limit the generalizability of cognitive science while aspiring to do the opposite. 1/
Reposted by Becky
Chatbots — LLMs — do not know facts and are not designed to be able to accurately answer factual questions. They are designed to find and mimic patterns of words, probabilistically. When they’re “right” it’s because correct things are often written down, so those patterns are frequent. That’s all.
June 19, 2025 at 11:21 AM
Chatbots — LLMs — do not know facts and are not designed to be able to accurately answer factual questions. They are designed to find and mimic patterns of words, probabilistically. When they’re “right” it’s because correct things are often written down, so those patterns are frequent. That’s all.
Reposted by Becky
So many amazing elements from vintage maps, scanned & cleaned up for re-use by @kmalexander.bsky.social
kmalexander.com/free-stuff/f...
via Evan Applegate
kmalexander.com/free-stuff/f...
via Evan Applegate
April 2, 2025 at 5:04 PM
So many amazing elements from vintage maps, scanned & cleaned up for re-use by @kmalexander.bsky.social
kmalexander.com/free-stuff/f...
via Evan Applegate
kmalexander.com/free-stuff/f...
via Evan Applegate
Reposted by Becky
Refresh the page
If you like this cartoon, I can draw one for you. www.worldofmoose.com/products/ref...
If you like this cartoon, I can draw one for you. www.worldofmoose.com/products/ref...
April 7, 2025 at 12:33 PM
Refresh the page
If you like this cartoon, I can draw one for you. www.worldofmoose.com/products/ref...
If you like this cartoon, I can draw one for you. www.worldofmoose.com/products/ref...
Reposted by Becky
Historians are going to have better records from the 19th century than the 21st, is my most medievalist futurist prediction.
We won't just lose the archives they want us to lose but also the ones they couldn't back up properly.
YOU DID WHAT?
April 7, 2025 at 2:09 PM
Historians are going to have better records from the 19th century than the 21st, is my most medievalist futurist prediction.
Reposted by Becky
@yaelrice.bsky.social and I co-authored this piece precisely to combat misguided work like this so people don't have to constantly rehearse the arguments about why it's specious. We laid it all out here for you!
hyperallergic.com/604897/how-s...
hyperallergic.com/604897/how-s...
April 5, 2025 at 5:54 PM
@yaelrice.bsky.social and I co-authored this piece precisely to combat misguided work like this so people don't have to constantly rehearse the arguments about why it's specious. We laid it all out here for you!
hyperallergic.com/604897/how-s...
hyperallergic.com/604897/how-s...
Reposted by Becky
I wanted to quickly explain why, school kids should learn markdown instead of MS Office, and ended up writing another major Epos:
ia.net/topics/markd...
This is, I kid you not, about 1/6th of what I wrote. I'll publish the rest later.
ia.net/topics/markd...
This is, I kid you not, about 1/6th of what I wrote. I'll publish the rest later.
Markdown and the slow fade of the formatting fetish
Notes on a revolution in slow motion.
ia.net
April 1, 2025 at 7:08 PM
I wanted to quickly explain why, school kids should learn markdown instead of MS Office, and ended up writing another major Epos:
ia.net/topics/markd...
This is, I kid you not, about 1/6th of what I wrote. I'll publish the rest later.
ia.net/topics/markd...
This is, I kid you not, about 1/6th of what I wrote. I'll publish the rest later.
Reposted by Becky
Wow! This is LEATHER!
Rob Goodwin responded to me in IG and said that there is a thermoplastic base but it’s mostly tooled leather. Absolutely incredible.
#TheWheelOfTime
Rob Goodwin responded to me in IG and said that there is a thermoplastic base but it’s mostly tooled leather. Absolutely incredible.
#TheWheelOfTime
January 23, 2025 at 10:53 PM
Wow! This is LEATHER!
Rob Goodwin responded to me in IG and said that there is a thermoplastic base but it’s mostly tooled leather. Absolutely incredible.
#TheWheelOfTime
Rob Goodwin responded to me in IG and said that there is a thermoplastic base but it’s mostly tooled leather. Absolutely incredible.
#TheWheelOfTime
Reposted by Becky
Spread the word for the Summer School 2025! i-guide.io/summer-schoo...
This year's topic is "Spatial AI for Extreme Events and Disaster Resilience" and will take place in Boulder, CO on Aug 4-8th, 2025!
Participant applications are due Monday April 14th.
This year's topic is "Spatial AI for Extreme Events and Disaster Resilience" and will take place in Boulder, CO on Aug 4-8th, 2025!
Participant applications are due Monday April 14th.
Summer School 2025 : I-GUIDE
i-guide.io
April 1, 2025 at 3:02 PM
Spread the word for the Summer School 2025! i-guide.io/summer-schoo...
This year's topic is "Spatial AI for Extreme Events and Disaster Resilience" and will take place in Boulder, CO on Aug 4-8th, 2025!
Participant applications are due Monday April 14th.
This year's topic is "Spatial AI for Extreme Events and Disaster Resilience" and will take place in Boulder, CO on Aug 4-8th, 2025!
Participant applications are due Monday April 14th.
Reposted by Becky
LAST CALL for I-GUIDE Forum 2025!
We can't wait to see your posters, papers, workshops, etc. in June, but you need to get your submissions in today! i-guide.io/forum/forum-...
The I-GUIDE Forum 2025 is co-located with the Sustainability Research and Innovation Congress in Chicago!
We can't wait to see your posters, papers, workshops, etc. in June, but you need to get your submissions in today! i-guide.io/forum/forum-...
The I-GUIDE Forum 2025 is co-located with the Sustainability Research and Innovation Congress in Chicago!
Forum 2025: Call for Participation : I-GUIDE
i-guide.io
March 31, 2025 at 5:07 PM
LAST CALL for I-GUIDE Forum 2025!
We can't wait to see your posters, papers, workshops, etc. in June, but you need to get your submissions in today! i-guide.io/forum/forum-...
The I-GUIDE Forum 2025 is co-located with the Sustainability Research and Innovation Congress in Chicago!
We can't wait to see your posters, papers, workshops, etc. in June, but you need to get your submissions in today! i-guide.io/forum/forum-...
The I-GUIDE Forum 2025 is co-located with the Sustainability Research and Innovation Congress in Chicago!
Reposted by Becky
The easy way to get non programmers to understand why old systems are more robust than newer ones, and can't be easily or quickly rebuilt from scratch:
"what is a valid name a person can have?"
In 50 year old code, each edge case has already been seen, fixed.
www.kalzumeus.com/2010/06/17/f...
"what is a valid name a person can have?"
In 50 year old code, each edge case has already been seen, fixed.
www.kalzumeus.com/2010/06/17/f...
Falsehoods Programmers Believe About Names
|
Kalzumeus Software
Classic essay about how software routinely bumbles human names.
www.kalzumeus.com
March 30, 2025 at 10:08 PM
The easy way to get non programmers to understand why old systems are more robust than newer ones, and can't be easily or quickly rebuilt from scratch:
"what is a valid name a person can have?"
In 50 year old code, each edge case has already been seen, fixed.
www.kalzumeus.com/2010/06/17/f...
"what is a valid name a person can have?"
In 50 year old code, each edge case has already been seen, fixed.
www.kalzumeus.com/2010/06/17/f...
Reposted by Becky
Our big social media platforms have been *intentionally designed* to make it near impossible to find (and post) useful, timely information about disasters, as this article points out - this has been angering me for years.
We need “most recent” and functional search!
Which Bluesky provides.
We need “most recent” and functional search!
Which Bluesky provides.
Social media platforms are not built for this
We have simultaneously too much and too little information.
www.theverge.com
January 20, 2025 at 11:42 PM
Our big social media platforms have been *intentionally designed* to make it near impossible to find (and post) useful, timely information about disasters, as this article points out - this has been angering me for years.
We need “most recent” and functional search!
Which Bluesky provides.
We need “most recent” and functional search!
Which Bluesky provides.
Reposted by Becky
Please enjoy part 1 of a series of blog posts written by myself and @brettfavaro.bsky.social offering our advice on how scientists can effectively speak with jouranlists about our areas of expertise. Part 2 is coming next week.
Please read and share!
www.southernfriedscience.com/advice-for-t... 🧪
Please read and share!
www.southernfriedscience.com/advice-for-t... 🧪
Advice for talking to the media as a conservation scientist: Part 1
By David Shiffman and Brett Favaro. Many environmental scientists understand that there is value in communicating about their work through the media, as publicity can help raise public awareness of…
www.southernfriedscience.com
March 13, 2025 at 5:49 PM
Please enjoy part 1 of a series of blog posts written by myself and @brettfavaro.bsky.social offering our advice on how scientists can effectively speak with jouranlists about our areas of expertise. Part 2 is coming next week.
Please read and share!
www.southernfriedscience.com/advice-for-t... 🧪
Please read and share!
www.southernfriedscience.com/advice-for-t... 🧪
Reposted by Becky
I am somehow for the first time ever reviewing for PNAS and their reviewer instructions include this BANGER of a line:
"The purpose of peer review is not to demonstrate proficiency in identifying flaws"
Print it in eleventy point font and hang it from the hillside. Scream it from the rooftops.
"The purpose of peer review is not to demonstrate proficiency in identifying flaws"
Print it in eleventy point font and hang it from the hillside. Scream it from the rooftops.
March 13, 2025 at 2:59 PM
I am somehow for the first time ever reviewing for PNAS and their reviewer instructions include this BANGER of a line:
"The purpose of peer review is not to demonstrate proficiency in identifying flaws"
Print it in eleventy point font and hang it from the hillside. Scream it from the rooftops.
"The purpose of peer review is not to demonstrate proficiency in identifying flaws"
Print it in eleventy point font and hang it from the hillside. Scream it from the rooftops.
It is easier for me to find local news from the 70s and 80s (archived in a library) than from my own childhood.
It continues to bother me that it’s possible for me to read newspapers from 100+ years ago and not data journalism from the 10 years ago (when I wrote about this back then, the main cause was technical breakage and not corporate malevolence)
ABC News has now fully taken down the old 538 website, including all interactive projects since 2014. Aside from erasing history this prevents access to publicly released data, including raw polls, averages, model estimates & story dta. Totally unacceptable for a company (allegedly) doing journalism
March 8, 2025 at 6:53 PM
It is easier for me to find local news from the 70s and 80s (archived in a library) than from my own childhood.
Reposted by Becky
obviously i'm biased but i think this post is a very good summation of how you should treat c going forward
faultlore.com/blah/c-isnt-...
faultlore.com/blah/c-isnt-...
C Isn't A Programming Language Anymore - Faultlore
faultlore.com
March 8, 2025 at 4:32 PM
obviously i'm biased but i think this post is a very good summation of how you should treat c going forward
faultlore.com/blah/c-isnt-...
faultlore.com/blah/c-isnt-...
Reposted by Becky
Reminder: if you haven’t yet read “The Mythical Man Month,” buy two copies so you can read it faster.
March 6, 2025 at 6:57 AM
Reminder: if you haven’t yet read “The Mythical Man Month,” buy two copies so you can read it faster.