The chaos of travel has died down. We've been all over the country and gone for weeks and weekends at a time. The snow is starting to hit the mountains, though, signalling a shift in our outdoor activities. Here's a summary of what's been going on with Plastic Friday.
1. We traveled to Rainy Lake and Isle Royale National Parks as part of our vacation. I brought my trash picker-upper with me in my hiking backpack and some trash bags. Much to our surprise, though, there wasn't much trash to pick up! We hike so many trails and do so much back-country camping that our baseline is fairly established so the deviance from the norm was clear. We had a few theories, such as Isle Royale requiring such effort to get to (we took the 6-hour, one-way ferry) that it discourages the casual explorer who is more likely to drop trash and leave it. Both of these parks are also far up north and you've got to want to visit them as there's not much else around there. We still picked up the odd granola bar wrapper, but even these were more likely to be the corner of the wrapper that fell out of someone's pocket rather than the whole wrapper. One thing we didn't see: cigarette butts! Awesome!
2. We spent a three-day weekend in the Collegiate Peaks area of Colorado. This is a popular area for ATV's and Jeeps. Our hike up into the valley north of Mt. Antero was frequently punctuated by loud, oil-burning ATV's. It's always ironic when they power past you, blasting exhaust everywhere and they smile and wave. (To be very clear, I support their access and wish them no ill will, but I can still be annoyed about breathing in oil-rich fumes.) Anyway, as expected there was quite a bit of trash everywhere. Enough that I stopped trying to collect it all. The area is also a popular mining region so it gets some industrial traffic. (We found aquamarine right on the road.) We spent some time talking about how you could address some of the trash and pollution effectively. There was no clear answer because there's no clear way to enforce any policy. It's too remote, too vast, and too trafficked.
3. Google officialy released their 2.0 version of their TensorFlow software. I've been spending time going through their tutorials. Some of their documentation and tutorials are still catching up with their functionality. Still, I've been able to do simple projects.
In conjunction, I've been spending a lot of time doing web development with Python and JavaScript. It's taken about 6 months, but I'm starting to feel that I could develop just about anything I can think of. The issue is the costs of hosting. Each website that uses a SQL database costs around $70 a month, which is more than I can afford right now. Somehow, I need to find a very cheap hosting option that still has easy access to Google's compute capabilities. Or I need to get sponsored in some way.
4. I've refined the trash and documentation process. In previous iterations, I was collecting trash and documenting it in situ with various backgrounds. This was too slow, especially in trash dense areas. My new process is to take a picture of the trash in place, remove it and take a picture of where the trash was, then move on to the next litter. Then, at home, I dump out the trash and run it through the gauntlet. This ends up being much faster. One drawback is that I don't have a good way to associate the images of the trash in the environment with the trash measurements and green-screened backgrounds. I'm still collecting statistics, but I lose some of the link between the two. Short of coming up with a custom mount/system, though, I don't think it's tenable to run the analysis in-field in real-time.
5. Kind of interesting: our HOA got a new maintenance person in the last couple of weeks. Ever since he started, the amount of trash in our community has gone down quite a bit. I used to just be able to walk around and collect trash from my neighborhood. His efficiency and attention to trash has made it harder for me! Ha!
As I mentioned above, winter is starting to make an appearance. Most of my work is likely to turn towards software/analysis work rather than in-field collecting. Our area does tend to stay snow-free on the ground so there might still be an opportunity here and there do verify what I'm up to with new data.
1. We traveled to Rainy Lake and Isle Royale National Parks as part of our vacation. I brought my trash picker-upper with me in my hiking backpack and some trash bags. Much to our surprise, though, there wasn't much trash to pick up! We hike so many trails and do so much back-country camping that our baseline is fairly established so the deviance from the norm was clear. We had a few theories, such as Isle Royale requiring such effort to get to (we took the 6-hour, one-way ferry) that it discourages the casual explorer who is more likely to drop trash and leave it. Both of these parks are also far up north and you've got to want to visit them as there's not much else around there. We still picked up the odd granola bar wrapper, but even these were more likely to be the corner of the wrapper that fell out of someone's pocket rather than the whole wrapper. One thing we didn't see: cigarette butts! Awesome!
2. We spent a three-day weekend in the Collegiate Peaks area of Colorado. This is a popular area for ATV's and Jeeps. Our hike up into the valley north of Mt. Antero was frequently punctuated by loud, oil-burning ATV's. It's always ironic when they power past you, blasting exhaust everywhere and they smile and wave. (To be very clear, I support their access and wish them no ill will, but I can still be annoyed about breathing in oil-rich fumes.) Anyway, as expected there was quite a bit of trash everywhere. Enough that I stopped trying to collect it all. The area is also a popular mining region so it gets some industrial traffic. (We found aquamarine right on the road.) We spent some time talking about how you could address some of the trash and pollution effectively. There was no clear answer because there's no clear way to enforce any policy. It's too remote, too vast, and too trafficked.
3. Google officialy released their 2.0 version of their TensorFlow software. I've been spending time going through their tutorials. Some of their documentation and tutorials are still catching up with their functionality. Still, I've been able to do simple projects.
In conjunction, I've been spending a lot of time doing web development with Python and JavaScript. It's taken about 6 months, but I'm starting to feel that I could develop just about anything I can think of. The issue is the costs of hosting. Each website that uses a SQL database costs around $70 a month, which is more than I can afford right now. Somehow, I need to find a very cheap hosting option that still has easy access to Google's compute capabilities. Or I need to get sponsored in some way.
4. I've refined the trash and documentation process. In previous iterations, I was collecting trash and documenting it in situ with various backgrounds. This was too slow, especially in trash dense areas. My new process is to take a picture of the trash in place, remove it and take a picture of where the trash was, then move on to the next litter. Then, at home, I dump out the trash and run it through the gauntlet. This ends up being much faster. One drawback is that I don't have a good way to associate the images of the trash in the environment with the trash measurements and green-screened backgrounds. I'm still collecting statistics, but I lose some of the link between the two. Short of coming up with a custom mount/system, though, I don't think it's tenable to run the analysis in-field in real-time.
5. Kind of interesting: our HOA got a new maintenance person in the last couple of weeks. Ever since he started, the amount of trash in our community has gone down quite a bit. I used to just be able to walk around and collect trash from my neighborhood. His efficiency and attention to trash has made it harder for me! Ha!
As I mentioned above, winter is starting to make an appearance. Most of my work is likely to turn towards software/analysis work rather than in-field collecting. Our area does tend to stay snow-free on the ground so there might still be an opportunity here and there do verify what I'm up to with new data.
Comments