Hazel Weakly
@hazelweakly.me
I have thoughts. Lots of thoughts. They never stop thinking. Never stop thunking.
hachyderm.io/@hazelweakly
hachyderm.io/@hazelweakly
It’s just infuriating to me that there is zero amount of money I can pay to ensure that a family member does not get scammed or victimised or otherwise taken advantage of
Every other experience in life has a “pay more for safety” option except for the ONE thing we made necessary to access life care
Every other experience in life has a “pay more for safety” option except for the ONE thing we made necessary to access life care
It is surprising to me that none of the major platforms have actual _parental control_ features so that we can limit the amount of time that our parents can spend online, and which apps they’re allowed to use. I don’t worry about my teenager’s use of the internet at all, just my octogenarian dad’s.
December 26, 2025 at 8:57 PM
It’s just infuriating to me that there is zero amount of money I can pay to ensure that a family member does not get scammed or victimised or otherwise taken advantage of
Every other experience in life has a “pay more for safety” option except for the ONE thing we made necessary to access life care
Every other experience in life has a “pay more for safety” option except for the ONE thing we made necessary to access life care
Reposted by Hazel Weakly
Can't wait for the loosened up, non-perfect, but quickly shipped AI-enhanced updates to hit my banking and insurance services next year.
Im sure there will be no issues there at all.
Im sure there will be no issues there at all.
December 26, 2025 at 4:09 PM
New relationship level unlocked: girlfriend’s eyes are Hazel coloured
December 25, 2025 at 6:46 PM
New relationship level unlocked: girlfriend’s eyes are Hazel coloured
Reposted by Hazel Weakly
The two hardest problems in Computer Science are
1. Human communication
2. Getting people in tech to believe that human communication is important
1. Human communication
2. Getting people in tech to believe that human communication is important
December 24, 2025 at 12:42 PM
The two hardest problems in Computer Science are
1. Human communication
2. Getting people in tech to believe that human communication is important
1. Human communication
2. Getting people in tech to believe that human communication is important
Reposted by Hazel Weakly
An easy way to explain this:
Which is easier to do, bake a cake from scratch, or fix a cake that someone baked after they accidentally used salt instead of sugar?
Which is easier to do, bake a cake from scratch, or fix a cake that someone baked after they accidentally used salt instead of sugar?
Just a thing about this: a common statement in writing is that you can fix crap, but you can't fix a blank page.
I have heard AI apologists claim that this is why it's fine to generate a first draft.
Now, to be clear, fuck this.
But also: an AI draft is a fucking blank page.
I have heard AI apologists claim that this is why it's fine to generate a first draft.
Now, to be clear, fuck this.
But also: an AI draft is a fucking blank page.
I know I don’t talk about writing much. It is because I am writing so slowly.
But I want you to know that I am taking a part of this story that absolutely sucked and took me months to produce four of the worst chapters I have ever written and in the edit I figured out how to fix it.
But I want you to know that I am taking a part of this story that absolutely sucked and took me months to produce four of the worst chapters I have ever written and in the edit I figured out how to fix it.
December 24, 2025 at 3:41 PM
An easy way to explain this:
Which is easier to do, bake a cake from scratch, or fix a cake that someone baked after they accidentally used salt instead of sugar?
Which is easier to do, bake a cake from scratch, or fix a cake that someone baked after they accidentally used salt instead of sugar?
Reposted by Hazel Weakly
Just a thing about this: a common statement in writing is that you can fix crap, but you can't fix a blank page.
I have heard AI apologists claim that this is why it's fine to generate a first draft.
Now, to be clear, fuck this.
But also: an AI draft is a fucking blank page.
I have heard AI apologists claim that this is why it's fine to generate a first draft.
Now, to be clear, fuck this.
But also: an AI draft is a fucking blank page.
I know I don’t talk about writing much. It is because I am writing so slowly.
But I want you to know that I am taking a part of this story that absolutely sucked and took me months to produce four of the worst chapters I have ever written and in the edit I figured out how to fix it.
But I want you to know that I am taking a part of this story that absolutely sucked and took me months to produce four of the worst chapters I have ever written and in the edit I figured out how to fix it.
December 24, 2025 at 3:55 AM
Just a thing about this: a common statement in writing is that you can fix crap, but you can't fix a blank page.
I have heard AI apologists claim that this is why it's fine to generate a first draft.
Now, to be clear, fuck this.
But also: an AI draft is a fucking blank page.
I have heard AI apologists claim that this is why it's fine to generate a first draft.
Now, to be clear, fuck this.
But also: an AI draft is a fucking blank page.
The two hardest problems in Computer Science are
1. Human communication
2. Getting people in tech to believe that human communication is important
1. Human communication
2. Getting people in tech to believe that human communication is important
December 24, 2025 at 12:42 PM
The two hardest problems in Computer Science are
1. Human communication
2. Getting people in tech to believe that human communication is important
1. Human communication
2. Getting people in tech to believe that human communication is important
Reposted by Hazel Weakly
Safety Nerds and @ruthmalan.bsky.social, you might be interested in the paper that @kissane.myatproto.social refers to:
onlinelibrary.wiley.com/doi/epdf/10....
onlinelibrary.wiley.com/doi/epdf/10....
December 23, 2025 at 11:18 PM
Safety Nerds and @ruthmalan.bsky.social, you might be interested in the paper that @kissane.myatproto.social refers to:
onlinelibrary.wiley.com/doi/epdf/10....
onlinelibrary.wiley.com/doi/epdf/10....
Reposted by Hazel Weakly
Happy holidays, Bluesky! I got you a megathrust earthquake, soil liquefaction, spine-tingling papers about the way our networks confound knowledge, and a PDF in a pear tree. It's my wrap on a year of trying to make sense of how we make sense of what's happening to us.
www.wrecka.ge/landslide-a-...
www.wrecka.ge/landslide-a-...
Landslide; a ghost story
On March 27, 1964, a converted liberty ship named the SS Chena brought a shipment of supplies to the port of Valdez, Alaska. Valdez, which I need you to know is pronounced “valDEEZ,” sits at the end o...
www.wrecka.ge
December 23, 2025 at 10:19 PM
Happy holidays, Bluesky! I got you a megathrust earthquake, soil liquefaction, spine-tingling papers about the way our networks confound knowledge, and a PDF in a pear tree. It's my wrap on a year of trying to make sense of how we make sense of what's happening to us.
www.wrecka.ge/landslide-a-...
www.wrecka.ge/landslide-a-...
The lack of a top sheet in European bedding is just so outrageous to me. Like why?! Put a bunch of extra letters into our flavour of spelling English, sure, but no extra sheet between you and the duvet? Absolute scandalous behaviour
December 22, 2025 at 9:39 PM
The lack of a top sheet in European bedding is just so outrageous to me. Like why?! Put a bunch of extra letters into our flavour of spelling English, sure, but no extra sheet between you and the duvet? Absolute scandalous behaviour
Reposted by Hazel Weakly
Cat and Ashley work so hard to do justice to both:
- the scientific rigor of their fields
- the human duties of making those fields serve us equitably, honestly, without distortion or manipulation
You have been misled about girls and math ability. Don’t miss this one:
- the scientific rigor of their fields
- the human duties of making those fields serve us equitably, honestly, without distortion or manipulation
You have been misled about girls and math ability. Don’t miss this one:
December 22, 2025 at 4:56 PM
Cat and Ashley work so hard to do justice to both:
- the scientific rigor of their fields
- the human duties of making those fields serve us equitably, honestly, without distortion or manipulation
You have been misled about girls and math ability. Don’t miss this one:
- the scientific rigor of their fields
- the human duties of making those fields serve us equitably, honestly, without distortion or manipulation
You have been misled about girls and math ability. Don’t miss this one:
Reposted by Hazel Weakly
December 22, 2025 at 4:47 PM
Reposted by Hazel Weakly
When we say "no, everything hasn't been digitized," I need you to understand that we really mean is that virtually nothing has been digitized. This is because the realm of primary sources that historians use is incomprehensibly large.
Seems like it's worth posting this one again.
Why Do Historians Still Have To Go To Archives?
Why do historians go to archives? Hasn’t everything already been digitized?
contingentmagazine.org
December 22, 2025 at 1:40 AM
When we say "no, everything hasn't been digitized," I need you to understand that we really mean is that virtually nothing has been digitized. This is because the realm of primary sources that historians use is incomprehensibly large.
What’s the European equivalent of this monstrous concoction that is apparently a crime against both nature and the nervous system?
For science reasons, I need this
For science reasons, I need this
Between this post and this review I absolutely had to try this. Tracked some down. Wish me luck
December 22, 2025 at 9:54 AM
What’s the European equivalent of this monstrous concoction that is apparently a crime against both nature and the nervous system?
For science reasons, I need this
For science reasons, I need this
Reposted by Hazel Weakly
More people in software should learn to make constructive arguments from qualitative methods and data for researching information systems and user behavior.
December 20, 2025 at 8:27 PM
More people in software should learn to make constructive arguments from qualitative methods and data for researching information systems and user behavior.
I have a whole laundry list of beloved things in this industry that are actually pretty horrific and problematic and one of these days I’m gonna feel safe enough to actually write an article detailing out that list
I just hope people can separate their identity from the beloved yet flawed ideas :(
I just hope people can separate their identity from the beloved yet flawed ideas :(
"it seems reasonable" is the absolute curse of developer experience afaict
There's some techniques you shouldn't use that seem so reasonable! It's our responsibility to plug the ears of the next generation, lest the siren's song lead them to ruin.
Other "c'mon, it's gotta work!" techniques like SMOTE are in this class.
Other "c'mon, it's gotta work!" techniques like SMOTE are in this class.
December 20, 2025 at 8:31 PM
I have a whole laundry list of beloved things in this industry that are actually pretty horrific and problematic and one of these days I’m gonna feel safe enough to actually write an article detailing out that list
I just hope people can separate their identity from the beloved yet flawed ideas :(
I just hope people can separate their identity from the beloved yet flawed ideas :(
Reposted by Hazel Weakly
PLS-SEM considered harmful
Some days ago a student struggling with her master’s thesis e-mailed me, asking me about partial least squares structural equation modeling. I told her I can only tell her not to do that, and then she sent more details.>
December 20, 2025 at 8:14 PM
PLS-SEM considered harmful
Yay! I was so sad to see this not happening. I’m glad people have been able to take it on! Community work can be so difficult and draining without enough support. People often want to help! But making that possible is hard
It makes me overjoyed to see that the community was able to help after all ❤️
It makes me overjoyed to see that the community was able to help after all ❤️
Cloud Native Rejekts is continuing and the community is leading the way! 🚀💜
📣 CFP is now open: sessionize.com/cloud-native...
See you in AMS ❌
📣 CFP is now open: sessionize.com/cloud-native...
See you in AMS ❌
Cloud Native Rejekts EU 2026: Call for Speakers
Cloud Native Rejekts is the b-side conference giving a second chance to the many wonderful, but rejected talks leading to KubeCon + CloudNativeCon.Clo...
sessionize.com
December 20, 2025 at 3:23 PM
Yay! I was so sad to see this not happening. I’m glad people have been able to take it on! Community work can be so difficult and draining without enough support. People often want to help! But making that possible is hard
It makes me overjoyed to see that the community was able to help after all ❤️
It makes me overjoyed to see that the community was able to help after all ❤️
Reposted by Hazel Weakly
this genuinely keeps me up at night
like I don't think you go into learning science and education research as hard as I did without having this fundamental terror about how we're using humanity's potential and problem-solving
and it's REALLY HERE for me with software
like I don't think you go into learning science and education research as hard as I did without having this fundamental terror about how we're using humanity's potential and problem-solving
and it's REALLY HERE for me with software
December 19, 2025 at 7:33 PM
this genuinely keeps me up at night
like I don't think you go into learning science and education research as hard as I did without having this fundamental terror about how we're using humanity's potential and problem-solving
and it's REALLY HERE for me with software
like I don't think you go into learning science and education research as hard as I did without having this fundamental terror about how we're using humanity's potential and problem-solving
and it's REALLY HERE for me with software
For me there’s something I haven’t really seen expressed about AI that I like, and it’s where I see a ton of potential, but I do wanna clarify that I’m separating “AI” here from gen AI, the unethical issues, etc etc…
Anyways: I like the potential of ill defined and proximate program synthesis
Anyways: I like the potential of ill defined and proximate program synthesis
I hear this often, but if people were trying to replace programming languages why would they be trying to generate so much programming language output though?
I think it's pretty clear that people want multiple things here
I think it's pretty clear that people want multiple things here
What’s really amazing about vibe-coding is how people are replacing programming languages which are strictly deterministic with human speech which is highly ambiguous and expect programming to be faster and better.
December 19, 2025 at 4:09 PM
For me there’s something I haven’t really seen expressed about AI that I like, and it’s where I see a ton of potential, but I do wanna clarify that I’m separating “AI” here from gen AI, the unethical issues, etc etc…
Anyways: I like the potential of ill defined and proximate program synthesis
Anyways: I like the potential of ill defined and proximate program synthesis
Cluster analysis for developer productivity research is like that shape sorter meme. Every fucking shape just goes in the hole that you want it to go into
Which naturally should make you suspicious that the hole implies anything about the shape (it doesn’t!)
Which naturally should make you suspicious that the hole implies anything about the shape (it doesn’t!)
I have been beating this dead horse re: cluster analyses & I'm back. I regret to say I just played around with creating some clusters from my own developer research but one of the charts looks ......how do I put this ....... like something that would be drawn on a teenager's desk? so I cannot share
This is from the 2025 DORA report. As ever, the report is pretty obsessed with forcing clusters to emerge. I am quite skeptical.
December 19, 2025 at 3:17 PM
Cluster analysis for developer productivity research is like that shape sorter meme. Every fucking shape just goes in the hole that you want it to go into
Which naturally should make you suspicious that the hole implies anything about the shape (it doesn’t!)
Which naturally should make you suspicious that the hole implies anything about the shape (it doesn’t!)
Reposted by Hazel Weakly
It’s that time of the year where Home Alone is going to be on tv a bunch. As relatives go “they should have had a battery powered alarm” and make other judgments on the family’s lack of responsibility in forgetting their kid, get a better source to learn from what happened.
ferd.ca/home-alone-a...
ferd.ca/home-alone-a...
Home Alone: a Post-Incident Review
A post-incident review of the first Home Alone movie and how parents could leave Kevin behind.
ferd.ca
December 19, 2025 at 12:54 PM
It’s that time of the year where Home Alone is going to be on tv a bunch. As relatives go “they should have had a battery powered alarm” and make other judgments on the family’s lack of responsibility in forgetting their kid, get a better source to learn from what happened.
ferd.ca/home-alone-a...
ferd.ca/home-alone-a...
Doing worse than fucking 2008 is completely unhinged. Whew
December 19, 2025 at 3:08 PM
Doing worse than fucking 2008 is completely unhinged. Whew
Reposted by Hazel Weakly
Just gave my last talk of the year!
2025 was quite packed, I gave talks about:
- the age-period-cohort problem
- making rigorous causal inference more mainstream
- mediation analysis
- marginaleffects
- causal graphs (x10)
If you're curious, check out my slides here: juliarohrer.com/resources/
2025 was quite packed, I gave talks about:
- the age-period-cohort problem
- making rigorous causal inference more mainstream
- mediation analysis
- marginaleffects
- causal graphs (x10)
If you're curious, check out my slides here: juliarohrer.com/resources/
Resources
Here you can find a collection of things that may be helpful, including slide decks, a curated list of introductory papers and blog posts, as well as some infographics I have generated to explain v…
juliarohrer.com
December 18, 2025 at 12:38 PM
Just gave my last talk of the year!
2025 was quite packed, I gave talks about:
- the age-period-cohort problem
- making rigorous causal inference more mainstream
- mediation analysis
- marginaleffects
- causal graphs (x10)
If you're curious, check out my slides here: juliarohrer.com/resources/
2025 was quite packed, I gave talks about:
- the age-period-cohort problem
- making rigorous causal inference more mainstream
- mediation analysis
- marginaleffects
- causal graphs (x10)
If you're curious, check out my slides here: juliarohrer.com/resources/
Reposted by Hazel Weakly
Shame on the ACM. They should know better.
In fact, here's a fun exercise to the reader: How many points from the ACM's own code of ethics does this move violate?
www.acm.org/binaries/con...
In fact, here's a fun exercise to the reader: How many points from the ACM's own code of ethics does this move violate?
www.acm.org/binaries/con...
December 17, 2025 at 4:21 PM
Shame on the ACM. They should know better.
In fact, here's a fun exercise to the reader: How many points from the ACM's own code of ethics does this move violate?
www.acm.org/binaries/con...
In fact, here's a fun exercise to the reader: How many points from the ACM's own code of ethics does this move violate?
www.acm.org/binaries/con...