Hacker News Reader

Comment by s5300 | Hacker News Reader

I wonder if there’s money to be made in the application of ML/NN’s for an app that uses your phone camera to tell if snowfall on a slope is within avalanche conditions, along with an unmanned drone that’s able to drop a cannonball or whatever to induce the avalanche pingable from the app…

View on HN

Maybe of some interest is my project https://openavalancheproject.org which is an attempt to improve/automate some of the forecasting of this with DL.

My goal isn't really a money making one though.

We already know the slope angles on which avalanches can occur (23-45deg irrc), and that data is published and accessible. Avalanche conditions are much more complex, sometimes varying from slope to slope.

As for mitigation, it’s not uncommon for an avalanche to trigger after the 2nd+ rider to travel over the same area to trigger a slide, so to effectively mitigate a combination of techniques are required (see inbound any control on ski resorts using bombs, ski cuts, etc).

The best way to avoid an avalanche is to avoid avalanche terrain and understand the current conditions. It’s a risk/reward evaluation every backcountry rider needs to balance.

Best advice I've heard from people who have been doing backcountry for decades: Don't. Hunt. Powder. Don't storm chase. What might make sense for in-bounds at a resort doesn't make sense in the uncontrolled backcountry.

The YouTube videos and magazine shots make it look like that's what it's all about. Foot after foot of fresh snow and face shots. But fresh snow, unconsolidated, after a storm is deadly. Especially early season. The people I know who have been doing it for years prefer late season spring skiing in the backcountry, even if it's icey and crusty.

I don't think a camera would be good enough. You need to be able to investigate the layers involved. Maybe some kind of probe that measures temperature/composition as it penetrates the snowpack. And then take regular sampling along a ridge, etc. I suppose if you were to spend a lot of money you could have a drone drop the probe, too. Or maybe there's some kind of radar or whatnot that could penetrate the snowpack and measure density layers and so on.

Hey cmrdporcupine - I’m a Saab 900 classic person - have daily driven them for several years. I’m very interested in EV conversion; I would honestly like to talk if you have an 84’ spg in good physical shape that you’d potentially like fully rejuvenated and teslaabified! Is there a good way to message you or can I give you my e-mail or something? Thanks. Dan. An 84 spg? 85 maybe? Would be an excellent opportunity.

I have a vision, if I were to meet someone with an 84’ spg in their garage that wanted to be part of that vision in the early going - could work well for both of us! Thanks.

Oh, hai? I mean, I have not actually attempted to EV convert. I do have an 84 SPG in half-disassembled condition in my garage. Brake lines corroded out, so I have to bend new ones and am too lazy to do it so it's sat for a few years. Like all old 900s in the northeast it has rust. Fender arches. But unlike the other 900 I had, the floorpan isn't corroded out. Engine ran at one point, probably would again. No idea about transmission (manual), might be ok. Interior is all pulled out so I could get at the brake lines.

Oh, and I'm in Ontario.

You can email me at ryan . daum [at] gmail.com

and the usual headliner sagging and cracked dash

Reply on news.ycombinator.com