7 ms·

Here's a long detailed comment from an HN user "embedded" with the rescue effort: https://news.ycombinator.com/item?id=15153567 https://news.ycombinator.com/it

Here's a long detailed comment from an HN user "embedded" with the rescue effort: https://news.ycombinator.com/item?id=15153567 https://news.ycombinator.com/item?id=15153567

JorgeGT 9y ago
That was a super interesting read that raised so many important points. Regarding the mentioned lack of maps of flooded areas, I wonder if volunteers were aware of the Copernicus EMS service activation [1]. This is a mapping team that, upon activation, constantly produces free maps of the emergency (fires, floodings, etc.) from satellite imaginery and radar data. The maps are constantly updates as soon as new satellite data is available. The most recent map [2], in fact, was uploaded just 15 minutes prior to this comment. More importantly, along each map a light vector package of the relevant GIS data [3] is published, so certainly this GIS data would be useful to add to an emergency app... As far as I know, the Hazards Data Distribution System (HDDS) of the USGS [4] also provides satellite imaginery, but I'm not sure if they also process this imaginery into freely-distributed, printable maps and vector GIS data. --- [1] http://emergency.copernicus.eu/mapping/list-of-components/EMSR229 http://emergency.copernicus.eu/mapping/list-of-components/EM... [2] http://emergency.copernicus.eu/mapping/system/files/components/EMSR229_07LAFAYETTE_01DELINEATION_MAP_v1_300dpi.pdf http://emergency.copernicus.eu/mapping/system/files/componen... [3] http://emergency.copernicus.eu/mapping/download/146034/EMSR229_07LAFAYETTE_01DELINEATION_MAP_v1_vector.zip http://emergency.copernicus.eu/mapping/download/146034/EMSR2... [4] https://hdds.usgs.gov/ https://hdds.usgs.gov/
- hackcasual 9y ago
  Seriously cool stuff. I've got relatives in Houston and was frustrated that I couldn't find an easy way to see if they were impacted
  - JorgeGT 9y ago
    I think their products are really great for the general public, because not everyone knows GIS and how to access satellite imagery. In some cases, such as fires, earthquakes, etc. they even estimate the impact into buildings (affected, destroyed, etc) which is really useful for people who may be wondering if their family/property is OK. Example: http://emergency.copernicus.eu/mapping/system/files/components/EMSR004_01FINALEEMILIA_GRADING_DETAIL02_v4_200dpi.jpg http://emergency.copernicus.eu/mapping/system/files/componen...
- maxerickson 9y ago
  NOAA gathered some post event aerial imagery: https://storms.ngs.noaa.gov/storms/harvey/index.html https://storms.ngs.noaa.gov/storms/harvey/index.html It looks like USGS has most of their activity here: https://www.usgs.gov/special-topic/hurricane-harvey https://www.usgs.gov/special-topic/hurricane-harvey
  - JorgeGT 9y ago
    Yep, I saw the really neat aerial imagery, and I know that in addition the USGS ingests enormous quantities of raw satellite imagery (https://twitter.com/USGS_HDDS https://twitter.com/USGS_HDDS), however most of that data is not freely available to the public, only to certain organizations. However, the main problem that I see (although I may be totally wrong!) is that all those raw data doesn't get condensed into ready-for-print maps of the affected areas, that can be distributed to volunteers on the field such as those made by Copernicus.
- yourapostasy 9y ago
  Thank you for bringing up Copernicus EMS and HDDS, I've added it to my observation notes. That data would be definitely be useful, but no one in the effort knew about these resources at the time. Water rescuers were given a general idea of an area that could be assumed flooded throughout, but boaters had to exercise caution at all times because of water hazards (both stationary and mobile) and flood waters could subside without warning, leaving some boats stranded on shallow mud flats. While it was somewhat life-threatening with water rescues, the lack of data could be worked around because the water rise to the point it was life-threatening was reckoned in 15-30 minute intervals, and people could coordinate rescues of rescuers around that timeframe. However, in a fire response version of such Net-based coordination, accurate, up-to-the-second assessment would be life-critical. Not sure how to address that without industrial-grade drones or even UAV's, and hella-expensive grade FLIR, and the FAA (and possibly DHS) would then be deeply involved. Not sure why UAV-based realtime GIS mapping wasn't deployed. The hardware tech exists, but perhaps the pipeline for data ingestion, cleanup, processing and publishing isn't up to realtime demands, even for a limited scope of "show me what is different from this satellite GIS map of the same area". That would be an interesting problem to solve, correcting for daylight shadows, transient objects like cars and planes, etc.; I'm sure the military intelligence folks in the IMINT sector have long since worked out the initial challenges of this problem space, but open algorithms for applying to this problem space (not to speak of code) don't seem readily available to me after searching around, so it might be a long slog up the learning curve. These disasters might be a good application of temporary air mobile vehicles supplying some form of wireless signal. If you're going to loft up a vehicle into the air to collect realtime GIS, you might as well blanket the area with an emergency wireless signal. Would have to solve problems around rate-limiting, dynamic prioritization based upon confirmed emergencies, etc. The flooding points out to me the lack of research into flood waters diversion [1]; there is still a lot we don't know about what happens when we try to mitigate excess flood waters. We know at least some flooding is necessary for managing topsoil erosion. But even the $26B estimate to bring Houston up to a more hardened flooding posture [2] seems like cheap insurance compared to the $100B estimates being thrown around now. I've wondered if it is possible, and what it would take to build a national water grid that intakes excess flood waters in one region of the nation, processes it so it is ecologically "clean" of invasive species, and shunts it to a drier region. The capacity of such an infrastructure within the channels themselves might lessen the need for designated floodwater storage plains, and simultaneously vastly increase land utilization. However, megascale infrastructure projects still seem out of our grasp in this era. [1] https://www.sciencedaily.com/releases/2016/01/160106114744.htm https://www.sciencedaily.com/releases/2016/01/160106114744.h... [2] https://www.theatlantic.com/technology/archive/2017/08/how-much-does-houston-spend-on-flood-control/538257/ https://www.theatlantic.com/technology/archive/2017/08/how-m...
posguy 9y ago
So Zello is a battery flattener? Perhaps a few Signal groups with a bot would have been better, as the other cited issues with it were around rescue request tracking and a breakdown in communication when Zello channels were too busy.
- technofiend 9y ago
  Zello has three problems: one is the afore-mentioned battery overhead, the second is cellphone push-to-talk adapters resold by the company are very expensive. You can find knockoff PTT devices on eBay but they don't have momentary switches, so you have to remember to click once to speak and again to clear the channel and any accidental brush against the toggle can end up blocking the channel indefinitely. And that's the third problem with Zello: its "only one person may speak at once" nature. As I mentioned previously, this really seems like the same problem a music festival faces just at a different scale. If someone could build an app that supports Glastonbury or Burning Man that could be the commercial underpinning that lets this hypothetical app donate services during a disaster. You'd already know how to add and allocate people to bands of channels based on their role and responsibilities. And how to deal with griefers who try to channel dominate or block for their own selfish or stupid ends. Heck maybe Zello (or this hypothetical app) add a more WhatsApp-like buffered playback mode that means people just queue for whose message is heard next but no one hears anything from you until your transmission ends, rather than trying to make your cellphone literally emulate a two way radio or Sprint/Nextel phone. That would also mean you could just see the previous stream of talk activity and play it back on joining or rejoining a channel. Add a way to tag conversations with rescue numbers and you can even make accessing data about a given rescue searchable. Sounds like a fun project. If anyone needs an SRE guy when they get ycombinator funding look me up. :-) One last note to potential investors: the Department of Homeland Security established CERT to let the general public be trained for assisting first responders. Included in that training is buying a go bag of your own for basic first aid. Adding a $10 waterproof PTT switch to your bag is not unreasonable. (Asking people to shell out $50-$175 which is what current Zello ptt adapters cost is too much IMHO). Anyway one could potentially sell this to the government although that would probably mean FEMA coordinating the onboarding of users and devices during a disaster for both the good and bad it implies.
paulmd 9y ago
I hate to be a wet blanket here but it's going to be difficult to impossible to maintain a large-scale voice commo system during a widespread natural disaster. This is stuff that Amateur Radio operators practice on a regular basis and it's difficult even with organized net operations schemes, this is not something people can manage just calling in. There is a very finite amount of bandwidth and during a time when normal commo links are down you need to maximize what bandwidth you have. If you switch to a TTY type interface you are going to be able to maintain comms a lot more efficiently for a given amount of battery life, and a much greater total communications volume. However from what you've said here, the voice is carrying a significant amount of contextual information and this might not be desirable. I recognize the merit of connecting rescuees directly to rescuers. However from what you've said, the "flat" organizational model is becoming a bottleneck and you don't think you'll be able to scale. My suggestion would be to look at the 9-1-1 model. You need to get people to local operators who can triage the most urgent calls according to certain categories (medical emergencies, trapped personnel, etc). You can record the 9-1-1 calls and push very compressed recordings to some kind of central system to maximize trunk bandwidth (assuming this is any kind of limitation vs operator time). During an emergency situation, it is probably not important that these "operators" are actually official emergency personnel, that is what you are doing when you connect rescuees directly to amateur rescuers anyway. I would just directly expose the triage category data to rescue personnel (volunteer or otherwise). Let people see the triaged location+severity of incidents in their area and communicate amongst themselves and they will respond intelligently to the most urgent situations. (again, this communication is much more energy-efficient in text form rather than in live voice) Please let me re-iterate the importance of rehearsing these scenarios. Amateurs regularly drill their net operation skills and since the "Cajun Navy" is now a thing it would be good to rehearse that as well. Nothing dangerous but imagine amateur radio operators calling into a net and requesting rescue at various (water/road-accessible) locations throughout a weekend for a real-world dispatch/recovery operations scenario. It's great that people stepped up but we absolutely need to drill these skills to prevent dumb mistakes. This is going to happen again, courtesy of global warming.

Here's a long detailed comment from an HN user "embedded" with the rescue effort: https://news.ycombinator.com/item?id=15153567 https://news.ycombinator.com/item?id=15153567

JorgeGT 9y ago

That was a super interesting read that raised so many important points. Regarding the mentioned lack of maps of flooded areas, I wonder if volunteers were aware of the Copernicus EMS service activation [1]. This is a mapping team that, upon activation, constantly produces free maps of the emergency (fires, floodings, etc.) from satellite imaginery and radar data. The maps are constantly updates as soon as new satellite data is available. The most recent map [2], in fact, was uploaded just 15 minutes prior to this comment. More importantly, along each map a light vector package of the relevant GIS data [3] is published, so certainly this GIS data would be useful to add to an emergency app... As far as I know, the Hazards Data Distribution System (HDDS) of the USGS [4] also provides satellite imaginery, but I'm not sure if they also process this imaginery into freely-distributed, printable maps and vector GIS data. --- [1] http://emergency.copernicus.eu/mapping/list-of-components/EMSR229 http://emergency.copernicus.eu/mapping/list-of-components/EM... [2] http://emergency.copernicus.eu/mapping/system/files/components/EMSR229_07LAFAYETTE_01DELINEATION_MAP_v1_300dpi.pdf http://emergency.copernicus.eu/mapping/system/files/componen... [3] http://emergency.copernicus.eu/mapping/download/146034/EMSR229_07LAFAYETTE_01DELINEATION_MAP_v1_vector.zip http://emergency.copernicus.eu/mapping/download/146034/EMSR2... [4] https://hdds.usgs.gov/ https://hdds.usgs.gov/

hackcasual 9y ago
Seriously cool stuff. I've got relatives in Houston and was frustrated that I couldn't find an easy way to see if they were impacted
- JorgeGT 9y ago
  I think their products are really great for the general public, because not everyone knows GIS and how to access satellite imagery. In some cases, such as fires, earthquakes, etc. they even estimate the impact into buildings (affected, destroyed, etc) which is really useful for people who may be wondering if their family/property is OK. Example: http://emergency.copernicus.eu/mapping/system/files/components/EMSR004_01FINALEEMILIA_GRADING_DETAIL02_v4_200dpi.jpg http://emergency.copernicus.eu/mapping/system/files/componen...
maxerickson 9y ago
NOAA gathered some post event aerial imagery: https://storms.ngs.noaa.gov/storms/harvey/index.html https://storms.ngs.noaa.gov/storms/harvey/index.html It looks like USGS has most of their activity here: https://www.usgs.gov/special-topic/hurricane-harvey https://www.usgs.gov/special-topic/hurricane-harvey
- JorgeGT 9y ago
  Yep, I saw the really neat aerial imagery, and I know that in addition the USGS ingests enormous quantities of raw satellite imagery (https://twitter.com/USGS_HDDS https://twitter.com/USGS_HDDS), however most of that data is not freely available to the public, only to certain organizations. However, the main problem that I see (although I may be totally wrong!) is that all those raw data doesn't get condensed into ready-for-print maps of the affected areas, that can be distributed to volunteers on the field such as those made by Copernicus.
yourapostasy 9y ago
Thank you for bringing up Copernicus EMS and HDDS, I've added it to my observation notes. That data would be definitely be useful, but no one in the effort knew about these resources at the time. Water rescuers were given a general idea of an area that could be assumed flooded throughout, but boaters had to exercise caution at all times because of water hazards (both stationary and mobile) and flood waters could subside without warning, leaving some boats stranded on shallow mud flats. While it was somewhat life-threatening with water rescues, the lack of data could be worked around because the water rise to the point it was life-threatening was reckoned in 15-30 minute intervals, and people could coordinate rescues of rescuers around that timeframe. However, in a fire response version of such Net-based coordination, accurate, up-to-the-second assessment would be life-critical. Not sure how to address that without industrial-grade drones or even UAV's, and hella-expensive grade FLIR, and the FAA (and possibly DHS) would then be deeply involved. Not sure why UAV-based realtime GIS mapping wasn't deployed. The hardware tech exists, but perhaps the pipeline for data ingestion, cleanup, processing and publishing isn't up to realtime demands, even for a limited scope of "show me what is different from this satellite GIS map of the same area". That would be an interesting problem to solve, correcting for daylight shadows, transient objects like cars and planes, etc.; I'm sure the military intelligence folks in the IMINT sector have long since worked out the initial challenges of this problem space, but open algorithms for applying to this problem space (not to speak of code) don't seem readily available to me after searching around, so it might be a long slog up the learning curve. These disasters might be a good application of temporary air mobile vehicles supplying some form of wireless signal. If you're going to loft up a vehicle into the air to collect realtime GIS, you might as well blanket the area with an emergency wireless signal. Would have to solve problems around rate-limiting, dynamic prioritization based upon confirmed emergencies, etc. The flooding points out to me the lack of research into flood waters diversion [1]; there is still a lot we don't know about what happens when we try to mitigate excess flood waters. We know at least some flooding is necessary for managing topsoil erosion. But even the $26B estimate to bring Houston up to a more hardened flooding posture [2] seems like cheap insurance compared to the $100B estimates being thrown around now. I've wondered if it is possible, and what it would take to build a national water grid that intakes excess flood waters in one region of the nation, processes it so it is ecologically "clean" of invasive species, and shunts it to a drier region. The capacity of such an infrastructure within the channels themselves might lessen the need for designated floodwater storage plains, and simultaneously vastly increase land utilization. However, megascale infrastructure projects still seem out of our grasp in this era. [1] https://www.sciencedaily.com/releases/2016/01/160106114744.htm https://www.sciencedaily.com/releases/2016/01/160106114744.h... [2] https://www.theatlantic.com/technology/archive/2017/08/how-much-does-houston-spend-on-flood-control/538257/ https://www.theatlantic.com/technology/archive/2017/08/how-m...

posguy 9y ago

So Zello is a battery flattener? Perhaps a few Signal groups with a bot would have been better, as the other cited issues with it were around rescue request tracking and a breakdown in communication when Zello channels were too busy.

technofiend 9y ago
Zello has three problems: one is the afore-mentioned battery overhead, the second is cellphone push-to-talk adapters resold by the company are very expensive. You can find knockoff PTT devices on eBay but they don't have momentary switches, so you have to remember to click once to speak and again to clear the channel and any accidental brush against the toggle can end up blocking the channel indefinitely. And that's the third problem with Zello: its "only one person may speak at once" nature. As I mentioned previously, this really seems like the same problem a music festival faces just at a different scale. If someone could build an app that supports Glastonbury or Burning Man that could be the commercial underpinning that lets this hypothetical app donate services during a disaster. You'd already know how to add and allocate people to bands of channels based on their role and responsibilities. And how to deal with griefers who try to channel dominate or block for their own selfish or stupid ends. Heck maybe Zello (or this hypothetical app) add a more WhatsApp-like buffered playback mode that means people just queue for whose message is heard next but no one hears anything from you until your transmission ends, rather than trying to make your cellphone literally emulate a two way radio or Sprint/Nextel phone. That would also mean you could just see the previous stream of talk activity and play it back on joining or rejoining a channel. Add a way to tag conversations with rescue numbers and you can even make accessing data about a given rescue searchable. Sounds like a fun project. If anyone needs an SRE guy when they get ycombinator funding look me up. :-) One last note to potential investors: the Department of Homeland Security established CERT to let the general public be trained for assisting first responders. Included in that training is buying a go bag of your own for basic first aid. Adding a $10 waterproof PTT switch to your bag is not unreasonable. (Asking people to shell out $50-$175 which is what current Zello ptt adapters cost is too much IMHO). Anyway one could potentially sell this to the government although that would probably mean FEMA coordinating the onboarding of users and devices during a disaster for both the good and bad it implies.

paulmd 9y ago

I hate to be a wet blanket here but it's going to be difficult to impossible to maintain a large-scale voice commo system during a widespread natural disaster. This is stuff that Amateur Radio operators practice on a regular basis and it's difficult even with organized net operations schemes, this is not something people can manage just calling in. There is a very finite amount of bandwidth and during a time when normal commo links are down you need to maximize what bandwidth you have. If you switch to a TTY type interface you are going to be able to maintain comms a lot more efficiently for a given amount of battery life, and a much greater total communications volume. However from what you've said here, the voice is carrying a significant amount of contextual information and this might not be desirable. I recognize the merit of connecting rescuees directly to rescuers. However from what you've said, the "flat" organizational model is becoming a bottleneck and you don't think you'll be able to scale. My suggestion would be to look at the 9-1-1 model. You need to get people to local operators who can triage the most urgent calls according to certain categories (medical emergencies, trapped personnel, etc). You can record the 9-1-1 calls and push very compressed recordings to some kind of central system to maximize trunk bandwidth (assuming this is any kind of limitation vs operator time). During an emergency situation, it is probably not important that these "operators" are actually official emergency personnel, that is what you are doing when you connect rescuees directly to amateur rescuers anyway. I would just directly expose the triage category data to rescue personnel (volunteer or otherwise). Let people see the triaged location+severity of incidents in their area and communicate amongst themselves and they will respond intelligently to the most urgent situations. (again, this communication is much more energy-efficient in text form rather than in live voice) Please let me re-iterate the importance of rehearsing these scenarios. Amateurs regularly drill their net operation skills and since the "Cajun Navy" is now a thing it would be good to rehearse that as well. Nothing dangerous but imagine amateur radio operators calling into a net and requesting rescue at various (water/road-accessible) locations throughout a weekend for a real-world dispatch/recovery operations scenario. It's great that people stepped up but we absolutely need to drill these skills to prevent dumb mistakes. This is going to happen again, courtesy of global warming.