Re: [gaia] disaster relief communication

[Rex Buddenberg <[e-mail]>]

Kurtis,

I'm reminded of a metaphor I used to use (OK, maybe a parable) about
the blind men an the elephant.

'Disaster relief' is the elephant.  Any web site that tried to ground
on the state of the art would sprawl all over the place.  Just what are
you asking for?  A taxonomy, please?

Internet technology is unquestionably a large (but not exactly
organized) set of tools at least potentially useful for disaster relief
-- I've had students raid my lab (I'm retired now, so this was former
life) and go set up a quick-erect extension of the internet.  We found
that it spontaneously got used (especially if there were WiFi
appendages to it).  
     Required some discipline to keep a fairly narrow capacity
available for those that needed it.  We had one church group show up in
Hurricane Katrina where one member was photographing the group's soup
kitchen and then planning to send the footage back to home church over
a rather slender thread.  
     As an example of the elephant, almost any disaster relief internet
scenario requires emergency power.  That's not exactly internet
technology (but an SNMP agent on the generator is, so we go full
circle).  Does it or does it not belong in your proposed compendium?
Which parts of the elephant do we try to get organized and presented?  
     Don't get me wrong, good idea.  But exactly where we go from here
isn't always clear.


Further (another feel of the elephant).  Pre-disaster preparation has a
whole lot to do with disaster relief.  I've seen a lot of
communications systems (both in and out of formal emergency services
organizations) that were not built with disaster relief in mind
(internet into schools, for example).  One (supporting a police
department!) had a hut at the base of a tower with the radio
electronics in it.  One fine day, the air conditioner quit (I got an
unconfirmed report that it had literally fallen off the wall). Hut
overheated and the radio equipment all quit.  
     Because of lack of redundancy, the tower was a single point of
failure so there was a large footprint in the city out of emergency
comms.
     Because there was no management (e.g. an SNMP agent in the a/c and
an SNMP console to watch such things), it took hours to figure out what
was wrong (and then another hour for the repairman to actually fix it).
BTW, once the hut cooled down, all the radio electronics reset and
started working again.
     Because the emergency services comms was a segregated stovepipe
from any other comms (either POTS phone or internet), there was no
crossover available.  
     Flunks all three of the principles of high availability
engineering.
     This happened on a 'normal day'.  But the location was in a city
where the Hayward Fault runs through (Californians will recognize this
-- recall Loma Prieta and San Francisco earthquakes).  




On Tue, 2017-10-03 at 11:14 -0700, Kurtis Heimerl wrote:
> I want to support Steve's request here; as someone who has dabbled in
> Disaster Relief it feels like there's an opportunity to do impactful
> work in the space but I don't know of any good places to get grounded
> in the current state of the art. Can we have any part of the upcoming
> GAIA meeting be focused on exploring this topic? Any domain experts
> in Singapore we can invite?
> 
> On Tue, Oct 3, 2017 at 11:04 AM, Rex Buddenberg <[e-mail]
> m> wrote:
> > Suggest that there are two (at least) genres that need to be merged
> > --
> > treated together.  Emergency services (reach to fire/ ambulance/
> > police/ ...) is the other genre.  In a disaster, expect a push to
> > build
> > out both.
> > 
> > Emergency services communications is one of the bastions of non-IP
> > technologies.  P25 is an example of a protocol heavily pushed by
> > various emergency services agencies. But it's non-routable.  Much
> > of
> > the development has been colored by the perceived need to jam
> > whatever
> > comms link is concocted into the narrowband Land Mobile Radio
> > channels
> > (25kHz and less).  
> > 
> > The economics is that the two genres end up costing twice for the
> > infrastructure.  This is true both for permanent infrastructure and
> > quick-build into disaster areas.  
> > 
> > Warning: this is an area of acrimonious debate, often sadly lacking
> > in
> > facts.  But it is a debate that needs to be joined.
> > 
> > 
> > 
> > On Tue, 2017-10-03 at 17:40 +0100, Arjuna Sathiaseelan wrote:
> > > Hello Steve,
> > >
> > > the IEEE global humanitarian technology conference is a good
> > venue to
> > > look at for the latest research/deployment experience papers:
> > >
> > > last year: http://sites.ieee.org/ghtc/event-2016/call-for-papers-
> > 2016
> > > /
> > >
> > >
> > > this looks like a good journal to keep an eye on when the papers
> > get
> > > published: http://ieeeaccess.ieee.org/special-sections-closed/mis
> > sion
> > > -critical-public-safety-communications-architectures-enabling-
> > > technologies-future-applications/
> > > regards
> > >
> > > a decent survey paper:
> > > http://eprints.whiterose.ac.uk/87438/5/Survey_of_wireless_communi
> > cati
> > > on_technologies_for_public_safety.pdf
> > >
> > > regards
> > >
> > > On 3 October 2017 at 17:25, Steve Song <[e-mail]>
> > wrote:
> > > >
> > > > Hi all,
> > > >
> > > > Are there any particularly good web resources and/or academic
> > > > papers that
> > > > profile the range of disaster relief technologies / solutions
> > both
> > > > planned
> > > > and currently in use?
> > > >
> > > > Many thanks... Steve
> > > >
> > > > --
> > > > [phone]
> > > > [e-mail]
> > > > http://nsrc.org
> > > >
> > > >
> > > > _______________________________________________
> > > > gaia mailing list
> > > > [e-mail]
> > > > https://www.irtf.org/mailman/listinfo/gaia
> > > >
> > >
> > >
> > 
> > _______________________________________________
> > gaia mailing list
> > [e-mail]
> > https://www.irtf.org/mailman/listinfo/gaia
> >