Rendered at 12:46:49 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
greg_dc 31 seconds ago [-]
Huh! Fellow Edinburgh resident here. I had no idea there was a rifle range here.
xg15 6 hours ago [-]
> I wanted to cook venison from scratch, which meant learning to shoot, which meant keeping track of my progress, which meant porting a 2012 OpenCV paper and training a state-of-the-art computer vision model, which meant the dinner took a bit longer than expected.
Procrastination level: Ultimate
Tangurena2 12 minutes ago [-]
Or a Carl Sagan fan:
> If you wish to make an apple pie from scratch, you must first invent the universe.
Yup. Or deer shaving, in this case. The punchline is he never actually got round to shooting a deer.
I do think he understated the difficulty of the hunt itself. He's planning to use the "supervision" rule to avoid needing his own firearm license, and male deer are indeed unlicensed for shooting (but not female deer!). Then you have to find one. He's right that they have reached "pest" status, since humans killed off the wolves. Every now and again someone suggests reintroducing the wolves, to cull the deer (and occasional tourists).
The open terrain (because the deer eat saplings) can make it easier. I have a great photo somewhere of a single majestic deer which I just happened to see from the road when I had my telephoto lens with me and mounted on the camera. I've even once seen a deer in Edinburgh itself, along a railway cutting.
f055 2 hours ago [-]
This read is amazing and the development work is very impressive, great job and congrats! That said, my 20-30yo self would end at that. However, my 40yo+ self has a piece of wisdom here: the brass plugs are there for a reason: they slow things down. Technocracy (screens, apps, automation) is not good for our mental health. Human minds need small, calm, slow, manual processes. Like plugging the brass plugs.
thisumang 6 hours ago [-]
The title is so bad, can't understand what's it about. Atleast around other topics here.
saaaaaam 5 hours ago [-]
No, it’s a great title. It was curious enough that I clicked and then I read the whole thing.
It’s unlikely I would have done that if it had been something like “Using computer vision to score rifle shooting cards”.
duozerk 3 hours ago [-]
In other words it's a great clickbait title, yet still a bad title.
greg_dc 1 minutes ago [-]
Clickbait is where the title leads you to believe the article is more interesting than it is. I'd argue this was the opposite of that.
seszett 2 hours ago [-]
I wouldn't say that. The current title indicates that the article is likely written in a less clinical manner than an article called "Using computer vision to score rifle shooting cards" would have been.
RyJones 16 hours ago [-]
My USPSA rank is public: I'm terrible with pistols. I haven't shot in competition for over a decade. This is the kind of project that tickles a couple of my nerves and might get me back to the range.
spockz 6 hours ago [-]
A very nice read and surprising ending. Love the perseverance of the research and the cleanliness of the app.
jfengel 15 hours ago [-]
Scoring is based on the outermost ring, rather than the innermost ring?
Huh. I'd have expected it to be based on the center, but I guess the goal is "it must be entirely within this ring to count" rather than just "I hit this ring".
valzevul 12 hours ago [-]
I think it depends on the discipline, NSRA .22 in the post uses the outermost edge, but ISSF (Olympic rifle/pistol, for example) uses the innermost edge.
donglebix 15 hours ago [-]
This ... Is beautiful
_carbyau_ 13 hours ago [-]
For comment reading edification, there are already electronic scoring targets for shooting.[0]
They use wave detection from each corner - either air/sound or via the target backing - to triangulate and with modern electronics can be quite accurate.
It's nice from an audience point of view to be able to see the results of each shot almost immediately. Kinda like watching snooker championships.
This approach is novel however and has other pros and cons.
There are also targets with fluorescent backgrounds and special black paint that flakes off near the holes. There is a limit to how many holes you can see in the target but it is way better than plain paper.
adammarples 1 hours ago [-]
If you always count borderline shots as in, or out, then you will be consistent enough to help you track the only thing that matters, your progress.
HoldOnAMinute 15 hours ago [-]
Wasn't sure what to expect when I clicked this link.
jmpman 12 hours ago [-]
I've been building a similar piece of software but with vibe coding. It's to the point that I'm using gauge blocks to measure the precise scoring ring dimensions and then using various warping techniques to get the photo to map precisely. In a weekend I've been able to get it to sub pixel accuracy.
ErroneousBosh 4 hours ago [-]
> Best gourmet pastries this side of the pond (and they also do doughnuts!).
Haha no, not even close. Saddler's Forfar Bridies take that prize.
pjc50 4 hours ago [-]
That has to have been an intentional joke; Greggs' aren't good, just cheap and ubiquitous, and almost any proper hand-made-on-site bakery will beat them.
sandworm101 13 hours ago [-]
>> .22 bullet is 0.22" across (duh)
Um... No. An american 22 can be very slightly smaller. American-invented calibers are measured to the depth of the grooves in a rifled barrel. The rest of the world measures to the flat parts between the grooves. So no, it is not obvious how wide a bullet is.
And beware the plural. If someone (usually a salty navy person) says that a gun is "50 calibers" he means something completely different than a "50 caliber".
Wow, I had no idea. The ones we had at the range were 0.22LR and the boxes are marked 5.7 mm which is also not precisely 0.22".
gweinberg 12 hours ago [-]
Yeah, given the nominal precision it's surprising how far off some of the numbers are. A .38 is the same diameter as a .357.
qball 10 hours ago [-]
Legacy and marketing have as much to do with it as local variations in how bores are actually measured do.
All the .38s and 9mms of the world are just slight variations on .36" round ball, .44 caliber pistols are generally .429", there's a .45" pistol caliber labelled .460 (.454 also counts), .50 BMG is actually .510", calibers claimed to be "7.62mm" use either a .308" or .311" projectile depending on the country of origin and sometimes not even then (France and Switzerland call this size 7.5mm, Argentina called this 7.65mm, Japan called it 7.7mm, the British called it .303), "8mm" can be either a .318" or .323" projectile, .32s are all .312" diameter, but one cartridge that uses this same projectile labels it as .30 and another .327.
The same 5.7mm projectile (.224") is used in cartridges that claim to have a diameter of .220, .221, .222, .223, .224, .225, 5.6mm, 5.56mm, and 5.7mm.
.277" projectiles are used in cartridges that call themselves 6.8mm, .270, .277; same thing with .284" projectiles used in cartridges that call themselves 7mm and .280.
rootusrootus 10 hours ago [-]
> A .38 is the same diameter as a .357.
Not just diameter, the bullet itself is identical. The cartridge is longer with more powder in a .357, which makes it a good bit more powerful in practice (2 to 3 times as much energy).
The explanation for the caliber discrepancy is halfway interesting:
"Despite its name, the caliber of the .38 Special cartridge is actually .357 inches (36 caliber/9.07 mm), with the ".38" referring to the approximate diameter of the loaded brass case. This came about because the original 38-caliber cartridge, the .38 Short Colt, was designed for use in converted .36-caliber cap-and-ball Navy revolvers, which had untapered cylindrical firing chambers of approximately 0.374-inch (9.5 mm) diameter that required heeled bullets, the exposed portion of which was the same diameter as the cartridge case."
Won't a fired .22 bullet be sized to the depth of the grooves due to obturation?
sandworm101 5 hours ago [-]
It is more complex. A softer bullet can get wider durong firing. And the barrel can expand slight also. This is why caliber invention isnt as simple as just picking a size.
beto_carreto 11 hours ago [-]
[flagged]
teiferer 15 hours ago [-]
Of all the things one can automate in this whole journey - he chose the ring counting on the shooting range? I don't get it.
I totally see the programming challenge there, but it's in no substantial way making the journey any easier. Any somewhat working human brain can count this quite quickly and then move on with other things.
Really, I don't get it.
rmunn 4 hours ago [-]
Because he kept hitting his head on the low ceiling beams as he walked over to look at his target.
If he had been shooting at an outdoor range, or even an indoor range with a higher ceiling, he probably wouldn't have been pushed to automate the process.
valzevul 15 hours ago [-]
Counting rings is easy indeed, but scoring borderline shots without a scoring gauge is not, because the visible bullet hole is often smaller than the bullet itself.
teiferer 7 hours ago [-]
But why would he care about this millimeter precision? His objective is not to participate in the Olympics but to shoot deer. He wants to improve general shooting abilities, not sub-millimeter accuracy. If he now and then counts a ring wrong, then what's the problem? That's what I don't get.
Ancapistani 6 hours ago [-]
> His objective is not to participate in the Olympics but to shoot deer.
Where do you see that?
The article is about someone in Scotland who took up marksmanship as a hobby.
krisoft 4 hours ago [-]
> Where do you see that?
There are multiple mentions of him being motivated by wanting to shoot deer for meat. It is a through line via the article.
> The article is about someone in Scotland who took up marksmanship as a hobby.
I wish it were so. With a bit more self awareness the author could have said “initially picked up a rifle to learn to hunt deer, but doing so i learned how targets are scored and become interested in automating that process.” There is nothing wrong with that. But pretending that someone is doing all this coding to get better charcuterie is where it becomes frustrating yak shaving.
teiferer 6 hours ago [-]
The article literally starts with "I wanted to cook venison from scratch, which meant learning to shoot"
jagged-chisel 15 hours ago [-]
Now that the software exists, one can use it from a mounted camera and provide immediate scoring. No need to wait for the human and the target to be in proximity.
Procrastination level: Ultimate
> If you wish to make an apple pie from scratch, you must first invent the universe.
From episode 9 of his Cosmos TV series.
https://kottke.org/23/11/if-you-wish-to-make-an-apple-pie-fr...
https://en.wikipedia.org/wiki/Cosmos:_A_Personal_Voyage
There were a number of "history of technology & invention" TV series, all inspired by Connections.
https://en.wikipedia.org/wiki/Connections_(British_TV_series...
I do think he understated the difficulty of the hunt itself. He's planning to use the "supervision" rule to avoid needing his own firearm license, and male deer are indeed unlicensed for shooting (but not female deer!). Then you have to find one. He's right that they have reached "pest" status, since humans killed off the wolves. Every now and again someone suggests reintroducing the wolves, to cull the deer (and occasional tourists).
The open terrain (because the deer eat saplings) can make it easier. I have a great photo somewhere of a single majestic deer which I just happened to see from the road when I had my telephoto lens with me and mounted on the camera. I've even once seen a deer in Edinburgh itself, along a railway cutting.
It’s unlikely I would have done that if it had been something like “Using computer vision to score rifle shooting cards”.
Huh. I'd have expected it to be based on the center, but I guess the goal is "it must be entirely within this ring to count" rather than just "I hit this ring".
They use wave detection from each corner - either air/sound or via the target backing - to triangulate and with modern electronics can be quite accurate.
It's nice from an audience point of view to be able to see the results of each shot almost immediately. Kinda like watching snooker championships.
This approach is novel however and has other pros and cons.
[0] https://en.wikipedia.org/wiki/Electronic_scoring_system
Haha no, not even close. Saddler's Forfar Bridies take that prize.
Um... No. An american 22 can be very slightly smaller. American-invented calibers are measured to the depth of the grooves in a rifled barrel. The rest of the world measures to the flat parts between the grooves. So no, it is not obvious how wide a bullet is.
And beware the plural. If someone (usually a salty navy person) says that a gun is "50 calibers" he means something completely different than a "50 caliber".
https://en.wikipedia.org/wiki/Caliber
All the .38s and 9mms of the world are just slight variations on .36" round ball, .44 caliber pistols are generally .429", there's a .45" pistol caliber labelled .460 (.454 also counts), .50 BMG is actually .510", calibers claimed to be "7.62mm" use either a .308" or .311" projectile depending on the country of origin and sometimes not even then (France and Switzerland call this size 7.5mm, Argentina called this 7.65mm, Japan called it 7.7mm, the British called it .303), "8mm" can be either a .318" or .323" projectile, .32s are all .312" diameter, but one cartridge that uses this same projectile labels it as .30 and another .327.
The same 5.7mm projectile (.224") is used in cartridges that claim to have a diameter of .220, .221, .222, .223, .224, .225, 5.6mm, 5.56mm, and 5.7mm.
.277" projectiles are used in cartridges that call themselves 6.8mm, .270, .277; same thing with .284" projectiles used in cartridges that call themselves 7mm and .280.
Not just diameter, the bullet itself is identical. The cartridge is longer with more powder in a .357, which makes it a good bit more powerful in practice (2 to 3 times as much energy).
The explanation for the caliber discrepancy is halfway interesting:
"Despite its name, the caliber of the .38 Special cartridge is actually .357 inches (36 caliber/9.07 mm), with the ".38" referring to the approximate diameter of the loaded brass case. This came about because the original 38-caliber cartridge, the .38 Short Colt, was designed for use in converted .36-caliber cap-and-ball Navy revolvers, which had untapered cylindrical firing chambers of approximately 0.374-inch (9.5 mm) diameter that required heeled bullets, the exposed portion of which was the same diameter as the cartridge case."
https://en.wikipedia.org/wiki/.38_Special
I totally see the programming challenge there, but it's in no substantial way making the journey any easier. Any somewhat working human brain can count this quite quickly and then move on with other things.
Really, I don't get it.
If he had been shooting at an outdoor range, or even an indoor range with a higher ceiling, he probably wouldn't have been pushed to automate the process.
Where do you see that?
The article is about someone in Scotland who took up marksmanship as a hobby.
There are multiple mentions of him being motivated by wanting to shoot deer for meat. It is a through line via the article.
> The article is about someone in Scotland who took up marksmanship as a hobby.
I wish it were so. With a bit more self awareness the author could have said “initially picked up a rifle to learn to hunt deer, but doing so i learned how targets are scored and become interested in automating that process.” There is nothing wrong with that. But pretending that someone is doing all this coding to get better charcuterie is where it becomes frustrating yak shaving.