@tomfinnigan.bsky.social
“Why is DNS so hard? It’s only naming things and cache invalidation”
November 18, 2025 at 2:51 PM
“Why is DNS so hard? It’s only naming things and cache invalidation”
I’d be tempted to build a system that minimizes some energy and solve a linear system
November 8, 2025 at 3:34 AM
I’d be tempted to build a system that minimizes some energy and solve a linear system
I thought this could be avoided by pressing ctrl-opt-b?
October 26, 2025 at 10:51 PM
I thought this could be avoided by pressing ctrl-opt-b?
You could make the exact same arguments about crypto, and many did.
Another way of saying the same thing is that either you’re deeply invested in the topic and it is difficult to admit the investment wasn’t worth it, or you didn’t see value early and didn’t waste your time / effort / money
Another way of saying the same thing is that either you’re deeply invested in the topic and it is difficult to admit the investment wasn’t worth it, or you didn’t see value early and didn’t waste your time / effort / money
October 17, 2025 at 3:51 PM
You could make the exact same arguments about crypto, and many did.
Another way of saying the same thing is that either you’re deeply invested in the topic and it is difficult to admit the investment wasn’t worth it, or you didn’t see value early and didn’t waste your time / effort / money
Another way of saying the same thing is that either you’re deeply invested in the topic and it is difficult to admit the investment wasn’t worth it, or you didn’t see value early and didn’t waste your time / effort / money
Manifold is extremely robust
October 9, 2025 at 2:00 PM
Manifold is extremely robust
This is generally not due to a slicing issue, it might be that an x-axis belt is slipping, and needs to be tightened
September 30, 2025 at 2:31 PM
This is generally not due to a slicing issue, it might be that an x-axis belt is slipping, and needs to be tightened
It seems like writing what is essentially a C++ to C++ FFI should be doable. I once worked in a project that would compile C++ code on the fly and load it as a dll
September 27, 2025 at 5:24 PM
It seems like writing what is essentially a C++ to C++ FFI should be doable. I once worked in a project that would compile C++ code on the fly and load it as a dll
I thought it was this bit
September 20, 2025 at 12:49 PM
I thought it was this bit
Was talking with someone who was going to go for cataract surgery, and they told him he wouldn’t need glasses after, because they would replace his lens with a trifocal
September 17, 2025 at 3:40 PM
Was talking with someone who was going to go for cataract surgery, and they told him he wouldn’t need glasses after, because they would replace his lens with a trifocal
For rendering / animation, glTF is fine. If you’re doing modeling / subdivision, there’s not really a great option.
FBX and USD can represent most of what you want, but both require a large 3rd party library.
OBJ is useful because it is simple and easy to implement, but can’t represent skinning.
FBX and USD can represent most of what you want, but both require a large 3rd party library.
OBJ is useful because it is simple and easy to implement, but can’t represent skinning.
September 6, 2025 at 1:19 PM
For rendering / animation, glTF is fine. If you’re doing modeling / subdivision, there’s not really a great option.
FBX and USD can represent most of what you want, but both require a large 3rd party library.
OBJ is useful because it is simple and easy to implement, but can’t represent skinning.
FBX and USD can represent most of what you want, but both require a large 3rd party library.
OBJ is useful because it is simple and easy to implement, but can’t represent skinning.
I mean, 10% increase is better than 40% increase, I guess
September 3, 2025 at 8:33 PM
I mean, 10% increase is better than 40% increase, I guess
That data shows a fairly consistent 10% increase over 2015-2019, which is not great.
September 3, 2025 at 8:30 PM
That data shows a fairly consistent 10% increase over 2015-2019, which is not great.
The pushback you get is probably mostly from people who still consider it a risk, and feel abandoned / marginalized for that.
September 3, 2025 at 5:25 PM
The pushback you get is probably mostly from people who still consider it a risk, and feel abandoned / marginalized for that.
At a certain level, risk is determined societally - if most of society has decided that something isn't risky, then it's very hard for individuals to feel like it is risky, and vice versa
I think that's what most people mean when they say 'During Covid' - when society considered covid a risk.
I think that's what most people mean when they say 'During Covid' - when society considered covid a risk.
September 3, 2025 at 5:25 PM
At a certain level, risk is determined societally - if most of society has decided that something isn't risky, then it's very hard for individuals to feel like it is risky, and vice versa
I think that's what most people mean when they say 'During Covid' - when society considered covid a risk.
I think that's what most people mean when they say 'During Covid' - when society considered covid a risk.
They have pros and cons. For example, when my neighbor has a barbecue, my purpleair sensor spikes. Which is useful for me, but less useful for everyone else around me that might be getting information from my sensor
September 1, 2025 at 3:43 PM
They have pros and cons. For example, when my neighbor has a barbecue, my purpleair sensor spikes. Which is useful for me, but less useful for everyone else around me that might be getting information from my sensor
If you’re trying to answer the question of how many people have died sooner due to covid, those numbers are a massive undercount (and have been since the beginning). For example, we know that covid dramatically increases the risk and severity of cardiac events www.nih.gov/news-events/...
First wave of COVID-19 increased risk of heart attack, stroke up to three years later
NIH-funded study focused on original virus strain, unvaccinated participants during pandemic.
www.nih.gov
August 26, 2025 at 11:30 PM
If you’re trying to answer the question of how many people have died sooner due to covid, those numbers are a massive undercount (and have been since the beginning). For example, we know that covid dramatically increases the risk and severity of cardiac events www.nih.gov/news-events/...
While there are anecdotal reports of underreporting acute covid as a cause of death (to help family cope or political reasons), I think it’s also safe to assume that that’s been happening at roughly the same rates, so if you’re mainly interested in comparing time periods, that seems reasonable.
www.nih
August 26, 2025 at 11:29 PM
While there are anecdotal reports of underreporting acute covid as a cause of death (to help family cope or political reasons), I think it’s also safe to assume that that’s been happening at roughly the same rates, so if you’re mainly interested in comparing time periods, that seems reasonable.
I think you can rely on those numbers for what they are - hospital-reported deaths of people with acute covid.
August 26, 2025 at 11:29 PM
I think you can rely on those numbers for what they are - hospital-reported deaths of people with acute covid.
So many people want covid to be over, it’s hard to trust any of the numbers except the wastewater at this point
August 26, 2025 at 3:54 PM
So many people want covid to be over, it’s hard to trust any of the numbers except the wastewater at this point
I bet if you look at who funded both companies you’ll find a significant amount of overlap
August 24, 2025 at 5:38 PM
I bet if you look at who funded both companies you’ll find a significant amount of overlap
Too much emphasis on spelling Mississippi, not enough focus given to spelling Massachusetts
August 23, 2025 at 12:51 PM
Too much emphasis on spelling Mississippi, not enough focus given to spelling Massachusetts