Re: focus on the functions, not whether it's an "emergency", "crisis" or "disaster" [was] Re: Proposal for a domain

Hello CH

Thanks for the insights, my quick replies to main issues (apols for top
posting)
- proposed definitions are useful, and surely add wider dimensions to the
discourse
please do add your stuff to the wiki, for future reference, (just try not to
wipe everything else in the process). I think this would be a good
opportunity to start a glossary, or some kind of related resource. Shall we
keep the original links also? where did you obtain those nice definitions
from?

- your three proposed 'subdomains' point to the beginning of categorization,
which is
the natural thing that happens when defining a new domain. you say pandemic,
hurricane, and conflict.  I read categories (disease, weather, social) -
note that I am not sure the names for these categories are appropriate at
this stage, lets think

- From my experience, the names that we give to things are very important in
knowledge representation. I will follow up the importance of the distinction
between the three names as per the subject header in the original thread

- I can see how different people represent different point of views, and
follow different paths to the 'truth' (read: true representation). So maybe
if some members of this group feel more inclined to start working on a
subdomain, or on cluster of subdomains, I think they should go ahead, just
keeping in mind that it would be useful to attempt to define the boundaries
of the top level domain (humanitarian crisis, whatever you want to call it)
to follow as general guidelines to frame their work, and avoiding ending up
with lots of small pieces that do not fit together (this could result in the
failure of this mission, which is a believe to create a global conceptual
and semantic framework that can be referenced by systms developers and
implementers). Btw, let me take a look at our mission statement, we do have
one somewhere dont' we?


Paola Di Maio



On 6/9/07, C H <craighubleyca@yahoo.com> wrote:
>
> I agree the name is poorly thought out but there are
> other options.
>
> --- paola.dimaio@gmail.com wrote:
> > 1. In an earlier thread, I wonder whether 'disaster
> > management' is the best
> > definition. Others whom I work with on this issue,
> > support the view that
> > 'disaster' is not the best word that we should use
>
> Agreed.  However the UN terminology leans toward
> "local resilience", for reasons you can read here:
>   http://openpolitics.ca/resilient+community
>
> Following that (UN RCCI) terminology, loosely:
>
> "Relief" is a function that takes place after
> disasters but can often be prepared for well in
> advance.  A more general term is "response" with
> "relief" reserved for the very worst cases.
>
> "Resilience" is a function, both before and after the
> major clusters of emergencies which we call disasters,
> which minimizes the need for relief or any response
> beyond local capacity.  That is, resilience increases
> effective local capacity.  Because resilience means a
> lot of design and attitude changes, "preparation" is
> too narrow a word.
>
> "Recovery" and "reconstruction" are efforts to restore
> and reinforce resilience after it has proven wanting
> during a disaster/crisis.  Since what goes in is never
> what existed before (and failed), "restoration" isn't
> a good word.
>
> > for a number of reasons.
> > examples: disaster is the last stage of a crisis,
> > also it is a very generic
> > term, etc. We prefer 'emergency mangemnet'.
> > Therefore we propose that the
> > name of this list, and all realted issues are
> > changed to 'emergency management'
>
> Please add this proposal to the other proposals for
> names of the W3C working group.  Personally I don't
> think "emergency management" is correct since it is
> implying that (a) the emergency is what is being
> managed rather than the local systems that must deal
> with it long term (b) management somehow becomes a
> very different thing in an emergency (c) an emergency
> is defined as something requiring help from outside in
> the form of "management", when that's often not what's
> needed (it may be just supplies, just instructions, or
> just additional personnel to fill in gaps in the local
> capacity).
>
> The worst mistake of "disaster relief" or "emergency
> management" is to assume that everything has failed
> and that there is no local capacity worth bolstering
> nor even investigating.  This leads to serious errors
> like, in the Kashmir earthquake, overlooking the fact
> that the Boy Scouts (of Pakistan) could get to remote
> mountain villages with more useful supplies and more
> applicable skills than any of the outside aid
> agencies.
>
> In 99.9 per cent of all emergencies, it's handled by
> local capacity and institutions not by an intervening
> outside agency.  If the intent is to make very clear
> what information the intervening agencies must pick up
> from local/field operatives, what they must
> investigate and add to and share, and what they must
> pass off to the local authorities after the fact, it
> may make sense to focus more clearly on the intent.
>
> In which case naming the phases (building resilience,
> relief, recovery/reconstruction) makes more sense than
> the blanket term "emergency" which means near-nothing.
>
> > I can forward to this list the discussions that have
> > already taken place if you are interested
>
> More important to update the wiki page at W3C with the
> new names proposed.  Then consider them all at once so
> their implications for scope can be properly debated
> among all, not just a few people who got there first.
>
> > 2. The risk with starting with one domain, is that
> > we do not sufficiently
> > generalize the problem definition, therefore we will
> > end up with a knowledge
> > representation that only applies to 'fly pandemic'
> > or either flu, or either
> > pandemic, ie a narrow problem
>
> The danger is that to prepare for all is to prepare
> for none.  The solution is a mix of scenarios that are
> each very specific.
>
> > Starting from one domain is a possible approach, but
> > we should explore alternatives
>
> I'd propose three, as follows:
>
> - a pandemic because it presents maximum challenge to
> personnel and institutions that must operate with
> fewer people than usual, and with some only able to
> telework, and occurs in many places at once minimizing
> the help available from outside and maximizing the
> need to rely on local personnel guided by information
> from the best in the world, but does not affect the
> physical infrastructure except insofar as personnel
> problems lead to outages and stress problems, and it
> can last a long time - an analysis of effects is at
> http://openpolitics.ca/pandemic+influenza
>
> - a hurricane because it presents a very short term
> challenge that quickly fades but has many follow on
> problems (mudslides, floods, lack of water and food)
> that can only be alleviated by very quick responses;
> it typically affects a whole region that is relatively
> easy to get to (by boat if nothing else) and damage to
> physical infrastructure is probably most widespread in
> this case but deaths tend to be few, evacuations many.
>
> - a conflict between factions in a contentious region
> because unlike the first two it's harder to anticipate
> and also unlike these the damage is done by
> intentional attacks and so can't be easily predicted;
> unlike the first two there will be serious hesitations
> to intervene due to security, and unlike the first two
> it is unlikely that a single authority can take over
> all responsibilities from the intervening aid agencies
>
> These three cases are all so different that to have a
> single model that deals well with the resilience,
> relief and recovery problems of each of them is to
> have at least solved the bulk of the problems.
>
> That said, the cases should be extremely specific, as
> in all scenario analysis, and violate some assumptions
> people hold.  If anyone thinks all three cases are
> well within the bounds of reality, then they are not
> extreme enough.  A perfect case for scenario analysis
> has 80 per cent of its features in common with known
> worst cases, 15 per cent or so in common with worst
> anticipated cases, and 5 per cent or so such a bad
> combination of factors (say a conflict occurring in
> the midst of a pandemic) that no one wants to believe
> it is realistic.  I've heard it said that unless at
> least one person leaves the room angry and certain
> that the case is worthless and badly biasing of the
> outcome, then the scenario is just not extreme enough.
>
> Personally I think three composite cases listing all
> the features of known disasters, plus two that are so
> bad that they exceed any known or anticipated event,
> is a minimum for a decent scenario analysis.
>
> Obviously if faced with choices of representation or
> detail, one weighs the cases and goes with the way
> that deals with the most-expected situation.
>
> > Another possibility, perhaps a bit more ambitious,
> > is to start defining a
> > 'top level' domain for emergency management, and
> > make sure that the
> > 'subdomain definitions'  all fit in there. This will
> > allow us to work the
> > foundations of  a framework that can be applied to
> > any domain.
>
> I don't believe the expertise to do this exists,
> though I am sure some will claim that it does.  If
> this is done then the scope and name of the group is
> absolutely critical.
>
> I think you'd have to have some breakdown of
> subdomains roughly as detailed as the following:
>
> emergency capacity
> - household/family level capacity (averages, examples)
> - village/neighbourhood level capacity (medical, fire)
> - (each) specialized facility unit capacity (cases
>   they can reasonably handle at once for long periods,
>   which will vary substantially between jurisdictions)
> - surge capacity for all of the above, degree to which
>   cases can be handled by non-specialists with
> training
>   and with only minimal briefing on forecare/aftercare
>
> local resilience
> - reliability of each utility (water, sewer, power) in
>   specific crises
> - reliability of each service (police, fire, medical)
>   in specific crises
> - supply chain / distribution resilience (if the local
>   services receive the supplies, will they be properly
>   deployed and used?  what help do they require to do
>   so?  one truck?  or a whole ship-borne chopper
> fleet)
> - communication and information/data fidelity (which
>   information can be reliably gathered by the local or
>   official authority, where do they need help for
> this)
> - recovery and reconstruction sophistication (high in
>   the US, low in Iraq, somewhere between in say
> Turkey)
> - degree to which sectarian/factional conflict
> inhibits
>   trust in a single local authority or representative
>
> damage minimization
> - when a disaster is anticipated, or imminent, what
>   steps can be taken to prevent injury or loss of a
>   vital piece of infrastructure like a road?
> - who holds lists of vulnerable persons and those
>   services they rely on, which may not be available
>   in a crisis
> - what movements of people (evacuations) are required?
> - what can be done to pre-deploy emergency responders?
>
> supply and service substitution
> - what can be used to replace what, in a pinch?  for
>   how long?  how can makeshift solutions in the field
>   be compiled for use by others nearby?  what low
>   overhead methods, e.g. cell phone cameras, can be
>   used to share coping techniques
>
> etc.
>
> I'd keep the subdomains functional, that is, don't
> focus on who has what powers, focus on what needs to
> get done and don't worry about who needs to do it.  If
> required that's a third level.
>
> For instance, under "damage minimization" you have
> separate responsibilities for locals, national and
> regional authorities, and the intervening agencies
> that typically respond if the damage is very great.
> But unless you have made "damage minimization" the
> actual subdomain name, these three don't have the same
> stated goal.
>
> So the actual state you want to achieve, function you
> want to perform, constraint you have to respect, is a
> subdomain.
>
> > Just thoughts
> >
> > Cheers
> > Paola Di Maio
> >
> >
> >
> > On 6/8/07, Xu, Wei <Wei.Xu@tudelft.nl> wrote:
> > >
> > >
> > > Dear Soenke,
> > >
> > > First of all, I would like to introduce myself.
> > I'm a PhD student working
> > > on the topic "Geo-information and formal semantic
> > for disaster management",
> > > OTB, Delft University of Technology, the
> > Netherlands.
> > >
> > > I submitted to the mail-list of W3C Incubator
> > Group and read your letter
> > > just now. You suggested we start by a specific
> > domain or a sapmle crisis,
> > > which I fully agree. However, do you think in the
> > future we will model
> > > different types of disaster? Or do you think it is
> > better if we define
> > > different types of disaster by means of the
> > management processes? Do you
> > > think if it is better if we abstract all the basic
> > processes of managing a
> > > disaster. For instance, the process of traffice
> > control, which is
> > > independent of any types of disaster --- it
> > doesn't matter if the disaster
> > > is caused by hurricane, by an exploration or
> > whatever (in the Netherlands,
> > > at the moment, the processes of managing disaster
> > is well-studied and
> > > modeled).  I think it is difficult to define a
> > certain type of disaster,
> > > because if we define the types of disaster, there
> > are more things we need to
> > > think about, for instance, the damage levels of
> > disasters, the people
> > > involved in the management work of disasters and
> > ect.
> > >
> > > It to me, how to model disaster management,
> > remains unclear. But I do
> > > agree with you, we need to start by a specific
> > domain and a sample.
> > >
> > > Thank you for reading.
> > >
> > > Kind Regards
> > > Wei XU
> > >
> > > -----Original Message-----
> > > From:
> > public-disaster-management-ont-request@w3.org
> > [mailto:
> > > public-disaster-management-ont-request@w3.org] On
> > Behalf Of Soenke Ziesche
> > > Sent: donderdag 7 juni 2007 12:40
> > > To: public-disaster-management-ont@w3.org
> > > Subject: Proposal for a domain
> > >
> > >
> > > Dear all,
> > >
> > > Since I'm joining this group slightly later I
> > would like to briefly
> > > introduce myself: I have been working for the UN
> > for several years in
> > > information management, currently  as the Head of
> > the Resident Coordinator's
> > > office in Blue Nile State, Sudan. I am very
> > interested in applying Semantic
> > > Web approaches to humanitarian information
> > management and have published an
> > > article on xml.com in this regard:
> > >
> > >
> > >
> >
>
> http://www.xml.com/pub/a/2006/12/13/semantic-wikis-and-disaster-relief-operations.html
> > >
> > > I'm very keen on contributing to the Incubator
> > Group. But since I haven't
> > > worked with the W3C before, I thought it's maybe
> > better for me to get an
> > > overview first and to wait until specific tasks
> > will be distributed.
> > >
> > > However, one proposal I'd like to make at this
> > stage: While, obviously,
> > > the deliverables of this project should be
> > applicable to any kind of
> > > disaster, I still believe it would be a good idea
> > to have a specific domain
> > > or sample crisis to work on and I'd like to
> > suggest to take "influenza
> > > pandemic" as such an example for the following
> > reasons: 1) This is an
> > > example for a complex and very large scale,
> > potentially even global
> > > disaster, thus a good test for the robustness of a
> > system. 2) As you
> > > probably know, there is indeed a risk of an
> > upcoming "influenza pandemic",
> > > potentially caused by a mutation of the Avian Flu
> > virus, and relevant
> > > departments of the UN are already working on the
> > preparation for such a
> > > pandemic.
> > >
> > > Looking much forward to cooperating with you.
> > >
> > > Warm regards,
> > > Soenke
> > >
> > >
> > > Humanitarian Affairs Liaison Officer
> > > Head of Resident Coordinator's Office
> > > Damazine, Blue Nile State, Sudan
> > >
> > > soenke.ziesche@undp.org
> > > +249-9-12178326
> > >
> > >
> > >
> > > --
> > > Der GMX SmartSurfer hilft bis zu 70% Ihrer
> > Onlinekosten zu sparen!
> > > Ideal für Modem und ISDN:
> > http://www.gmx.net/de/go/smartsurfer
> > >
> > >
> > >
> > >
> > >
> >
> >
> > --
> >
> >
> >
> > Paola Di Maio *****
> > School of Information Technology
> > Mae Fah Luang University
> > Chiang Rai - Thailand
> > *********************************************
> >
>
>
>
>
>
> ____________________________________________________________________________________
> Be a better Globetrotter. Get better travel answers from someone who
> knows. Yahoo! Answers - Check it out.
> http://answers.yahoo.com/dir/?link=list&sid=396545469
>



-- 



Paola Di Maio *****
School of Information Technology
Mae Fah Luang University
Chiang Rai - Thailand
*********************************************