Ontological question: | Services <> Benefits |

[Greg Bloom]

Folks: 

Nitty gritty question that I think I know the answer to but want to bounce off everyone here. 

An I&R system, as I understand it, connects people with 'community resources' -- as in services. Among other things, services can connect people with benefits. But benefits themselves would not be included in the domain of referral directory information. 

It would be: need --referred to--> services --enrolled in--> benefits

Is that right? 

I know that some 211s are engaged in enrolling people in benefits directly... but it seems like that's an expansion of 211 function into the service domain, and that benefits themselves are not a part of the domain that we're narrowly focused on.

Let me know if I have this correct. 

And please confirm whether this is an ontological issue or epistimological or ???

--
• gjb •

202.643.3648

[Eric Jahn]

I know that some 2-1-1s coordinate with One-e-App in many states (and
in my own county, as of next month or so as well), and One-e-App
brokers enrollment in local/state programs. It is an expansion, but I
think it's an inevitable expansion. I think a benefit is a
subcategory of "service event" which may fully/partially fulfill a
need. Automated submission of an application for enrollment to state
entitlement programs hasn't been a problem for One-e-App to implement
(using each state's different required methodology), but getting an
automated response from the agency has been difficult to achieve. In
Florida, they won't give an automated response to an application for
"client confidentiality" reasons. Instead, when the application is
submitted, the state follows up directly with the client, and leaves
One-e-App out of the loop. -Eric
Eric Jahn
Data Architect
Alexandria Consulting LLC
St. Petersburg, Florida
727.537.9474

> --
> You received this message because you are subscribed to the Google Groups
> "OpenInfoAndReferral" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to openinfoandrefe...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

[Sophia Parafina]

Aren't we splitting hairs here? Services provide or distribute benefits (in a broader use of the word), so I would think that they are a significant part of the domain. The way you have it framed is epistemological because it answers a how/what is type of question.

--

[Fitch, Dale K.]

In my experience the sequence stops prior to benefits, except as Eric pointed out. Actually it stops at the point of referral as the recipient agency does their own needs assessment to determine services. The 2-1-1 call center is supposed to follow-up on a random sample of callers to see if their 'needs' were met, not necessarily what services they received.

And, yes, this is an ontological issue as what constitutes services, benefits, referral and even 'need' is determined by the services domain, not necessarily 2-1-1 or the Open211. For example, there are various definitions of what constitutes 'homeless' and subvariants within those definitions. My personal favorite is the difference between hobos, tramps, and bums. The differences come out in the assessment (an epistemological process) which then determines the types of services that are needed.

Dale

---

From: openinfoa...@googlegroups.com [openinfoa...@googlegroups.com] on behalf of Greg Bloom [greg....@gmail.com]
Sent: Wednesday, November 13, 2013 8:36 PM
To: openinfoa...@googlegroups.com
Subject: Ontological question: | Services <> Benefits |

--

[Clive Jones]

As you say Greg, you do know the answer but we can confirm that in different ways!

Yes, some 211s are doing what I sometimes refer to as "enhanced 211" or "211 plus" whereby they have been asked/contracted to go beyond the assessment/referral to either complete an application (e.g. for SNAP) or set an appointment (e.g. for VITA) but these are  direct services tasks and certainly not data related ...

The contact is logged as the "problem/need" emerging from the client's situation and not as the 'benefit" which the client may or may not be finally approved for ...

It is certainly not part of the data set -- only in so far as most 211s use the Taxonomy indexing term assigned to the service/program that was referred to, as a convenient way to log the granular 'problem/need' of the client (e.g. Food Stamps/SNAP or Utility Service Payment Assistance).

Clive

--
Clive Jones
AIRS
www.airs.org
www.twitter.com/AIRSplace

[Eric Jahn]

I'm really enjoying this discussion.  I wish other human services forums could engage in this topic as candidly we've been.  Couple (okay more than that) points:  

Sophia, benefit is sort of a loaded term in human services.  Service event doesn't necessarily entail benefit.  An example non-benefit service event could be a "homeless outreach" which may not really benefit the homeless person initially in any real way, but may lead to benefits as the outreach leads to an intake assessment

.  A

nd then

maybe after that,

 transitional housing (real benefit)

is gi

ven to the client

.  Usually, benefits connote entitlement payments or something that usually the client would agree is a benefit, whereas some service events are

administrative tasks

 that the program has to do on behalf of the client

, and may get counted/reported to the funding source.

To expand upon what Dale

and Clive said about needs, some of the more forward thinking 2-1-1s see a key performance measure as the number of successful referrals which

resulted in eventual fulfillment of that

need identified during the brief phone screening.  But

fulfillment tracking

 requires feedback from the agency referred to, and lacking an automated need fulfillment response from that referred to agency, 2-1-1s

manually

perform follow-up calls to a percentage of clientele.  Follow-up calls are time consuming and difficult for high volume call centers, which is why I think

2-1-1s

 can't effectively follow up with everyone?  Clive would probably be able to fill in more details about follow-up. 

In current practice, when a 2-1-1 makes "warm transfer" of referral data to the referred to agency, it does so in the format that agency

referred to agency

expects.

<Eric's boring story time> When 2-1-1 Tampa Bay Cares makes a warm transfer of referral data to the County Behavioral Healthcare Network, it uses the data protocols of healthcare, since that's what their network's vendor supports (RHIO software).  So that means

using

SOAP + HL7

protocols

.  The 2-1-1 AIRS codes for

t

he client's

need

,

which is

passed to the referred to behavioral healthcare agency

,

are sort of taken with a grain of salt by the clinical professionals there

.  The clinicians

 do a much more rigorous assessment than the brief 2-1-1 screening could ever be expected to achieve.  The current effort for 2-1-1 Tampa Bay Cares is to get each of the behavioral healthcare providers to send back need fulfillment status messages.  This is for the original AIRS encoded need, not clinical ICD

,

 DSM

,

or whatever

code sets

they use in the behavioral healthcare world.  

2-1-1 Tampa Bay Cares will have to use

 "need fulfillment codes" associated with the AIRS code

,

like "need fully met", "need unmet", "need partially met", which is sort of vague, but really helps the 2-1-1 justify its existence to the community.  And that's part of the contract: the behavioral healthcare agencies get more clientele from the referrals, and the 2-1-1 gets the need fulfillment performance data in return. </Eric's boring story time>

Likewise, if they refer to the state's health and human services entitlement programs, the referral would have to be in the format that agency dictates, which is being specified by NIEM

H

uman

S

ervices as we speak. 

T

he

current objective

of this group

would

 be to try to turn that around and say, "We 2-1-1s don't have the resources to support each of your domain specific service provider formats; you're just going to have to learn how to talk open 2-1-1 with us.".

 

Alternatively, another way things could go is that NIEM Human Services creates adapters for health (HL7/IHE), like NIEM

had to also do

 for the OASIS Emergency Data Exchange Language. 

With HL7/IHE adapters

, any human service agency (not just 2-1-1s) can  connect point-to-point with any healthcare entity. 

But I see one short term problem in that NIEM Human Service's framework (NHSIA) may not meet all of 2-1-1's or other community (non HHS) requirements, and incorporating our extensions

and

changes may take time to advocate for.  What I recommend is defining an Open 2-1-1 standard based on NIEM Core data elements and trying to align as much as possible with NHSIA.  When NHSIA is finally ready and convinced to merge them in, each side will be minimally impacted.  This would still leave our task to that of defining the specific exchange payloads

,

API method signatures (IEPD exchange definition),

and improving the logic model.  B

ut

we'd use

 NIEM Core data type

s (https://www.acf.hhs.gov/sites/default/files/assets/i_datadictionaryandniemmapping_d01.xlsx)

and the NHSIA logical model

(

https://raw.github.com/hserv/open-human-services/master/doc/NHSIA_conceptual_data_model.jpeg), as opposed to reinventing an incompatible (logically and semantically) wheel. 

I'm hoping to also convince NIEM Human Services that a more community-centric (focusing on 2-1-1s, HUD HMIS, DOL-ETA, etc.) sister effort will be in their best interests also, and would merge gradually with their main domain, as they are ready. 

A third alternative, which is the current one we are embarking on, I think, is to make an Open 2-1-1 with a unique model and set of data types specific to the domain of 2-1-1.  Human services agencies would have to possess flexible logical models to support the concepts and structures of each silo they work with, and Open 2-1-1 would be added to the list of supported silos. 

 In a way, the addition of a new silo would continue the balkanization of human services, much like HMIS did in 2005 by creating their own data standard, instead of hashing one out with HHS.   

-Eric

[Eric Jahn]

Shoot, I apologize for the strange formatting again... it looked fine
before sent it. I'm editing in a pop-window in gmail , so I think I
need to look at my settings or something (like turn off html email).
I'll try to keep it from happening again... Sorry.

Eric Jahn
Data Architect
Alexandria Consulting LLC
St. Petersburg, Florida
727.537.9474

On Thu, Nov 14, 2013 at 10:57 AM, Eric Jahn
<er...@alexandriaconsulting.com> wrote:
> Thanks, you're too kind. -Eric

> Eric Jahn
> Data Architect
> Alexandria Consulting LLC
> St. Petersburg, Florida
> 727.537.9474
>
>

> On Thu, Nov 14, 2013 at 10:54 AM, Clive Jones <clive...@gmail.com> wrote:
>> Epic !!! (and not in the least bit boring ...)
>>
>> Clive

[Greg Bloom]

Really interesting discussion. Thanks y'all. 

I've copied key bits into a set of notes here (and organized that alongside key bits of other conversations all linked from hackpad as an ongoing set of notes). 

If I can try to clarify/recap and ask some followup questions here (feel free to answer in the hackpad; i'll do my best to consolidate there from email though).

Originally, I was referring to 'benefits' in a narrow sense, that of 'public benefits': food stamps, unemployment, Medicaid, disability, Social Security, etc. It seems like, ontologically speaking, that's a specific subset of benefits defined broadly as 'outcomes.' Meanwhile, that broad set of benefits is just outside of the domain of 'info-and-referral' -- in other words, the data about services does not include data about the benefits that the services might arrange. BUT a standard-setting initiative should be logically aligned with those other standards for data about those benefits, some of which are in development (through NIEM and others). 

Is that right? 

Some followup questions. 

Dale: you say “what constitutes services, benefits, referral and even 'need' is determined by the services domain, not necessarily 2-1-1 or the Open211.” — what is the ‘services domain’? Who is that? Is that the domain in a distributed sense? Or is there some kind of body? Or are you speaking… ontologically. 

Eric - 

Can you clarify what the three different alternatives are for an Open211 standard-setting project? I think you’re saying that one is ‘[an open211 standard] based on NIEM core data elements’  and another is a unique model based on the 211 domain. What’s the third? (And how could we base a referral data model on data elements from models describing other domains like benefits? Are there specific points that need to be aligned?

Thanks, y'all

Greg

• gjb •

202.643.3648

[Fitch, Dale K.]

Greg,

Yes, that would be domain in the distributive sense. What constitutes a service varies tremendously dependent upon the domain whether it be homeless, food, clothing, mental health, health, transportation, etc. Agencies from those domains know this and have no problem communicating with each other, at least at the people level. However, the typical consumer won't know those differences hence they may be confused thinking a service, is a service, is a service.

Two, fortunately, as you note, I&Rs can stop at the level of service and everything else will be outside the domain of I&Rs. At least from my perspective that would be a good place to draw a boundary, but I defer to the group.

Finally, I hate to quibble, but benefits are not 'outcomes.' There is a LONG, LONG discussion I could provide on that topic as outcomes in the human services have even more minutiae associated with them than services. In sum, it's best not to use the word 'outcomes' with I&Rs beyond our previous discussion of whether needs were met or not met.

Best,

Dale

---

From: openinfoa...@googlegroups.com [openinfoa...@googlegroups.com] on behalf of Greg Bloom [greg....@gmail.com]

Sent: Thursday, November 14, 2013 7:42 PM
To: openinfoa...@googlegroups.com
Subject: Re: Ontological question: | Services <> Benefits |

[Clive Jones]

I think Dale's long long discussion about human services outcomes
would be really fascinating but I can understand that it is a big
tangent!

Within I&R on the client side, it basically does stop at whether the
person gets the help they needed or not ... And if we find the "or
not", there should be some attempt to find 'why not' (and there are
about 20 potential reasons including we gave them the wrong
information) ...

There is some refinement on the concept of "help" -- sometimes folks
talk about their situation, get a referral and end up not contacting
it -- but just the fact they now have an option, gives them a bit of
confidence/perspective to get through a tough time. Also sometimes,
there are no referrals available and so the "need" technically is not
"met", but the I&R interaction explores family/personal strengths and
sources of "non-system" help. Plus the general intangible of
knowledge/empowerment.

At another level - and OK this is being self-serving - but I&R has an
outcome for the local community/social services networks. If (as John
Lennon almost sang "Imagine there's no I&R ..."), people with needs
are calling services on their own trying to get help and each call
ends with a suggestion to call someone else -- then those calls are
all being answered by staff at different agencies, plus individuals
get more frustrated and give up, then situations become problems,
problems escalate to crises, etc.

Clive

> here<https://hackpad.com/Open211-OpenIR-Ontology-iDO1F2BuFcX> (and organized

> that alongside key bits of other conversations all linked from hackpad as an
> ongoing set of

> notes<https://hackpad.com/Open-Information-Resource-and-Referral-Open211-Tixr2rEXQ0r>).

> <er...@alexandriaconsulting.com<mailto:er...@alexandriaconsulting.com>>

> wrote:
> Shoot, I apologize for the strange formatting again... it looked fine
> before sent it. I'm editing in a pop-window in gmail , so I think I
> need to look at my settings or something (like turn off html email).
> I'll try to keep it from happening again... Sorry.
> Eric Jahn
> Data Architect
> Alexandria Consulting LLC
> St. Petersburg, Florida

> 727.537.9474<tel:727.537.9474>

>
>
> On Thu, Nov 14, 2013 at 10:57 AM, Eric Jahn

> <er...@alexandriaconsulting.com<mailto:er...@alexandriaconsulting.com>>

> wrote:
>> Thanks, you're too kind. -Eric
>> Eric Jahn
>> Data Architect
>> Alexandria Consulting LLC
>> St. Petersburg, Florida

>> 727.537.9474<tel:727.537.9474>

>>
>>
>> On Thu, Nov 14, 2013 at 10:54 AM, Clive Jones
>> <clive...@gmail.com<mailto:clive...@gmail.com>> wrote:
>>> Epic !!! (and not in the least bit boring ...)
>>>
>>> Clive
>>>
>>>
>>> On 14 November 2013 07:38, Eric Jahn

>>> <er...@alexandriaconsulting.com<mailto:er...@alexandriaconsulting.com>>

>>>> <fit...@missouri.edu<mailto:fit...@missouri.edu>> wrote:
>>>>>
>>>>> In my experience the sequence stops prior to benefits, except as Eric
>>>>> pointed out. Actually it stops at the point of referral as the
>>>>> recipient
>>>>> agency does their own needs assessment to determine services. The 2-1-1
>>>>> call
>>>>> center is supposed to follow-up on a random sample of callers to see
>>>>> if
>>>>> their 'needs' were met, not necessarily what services they received.
>>>>>
>>>>> And, yes, this is an ontological issue as what constitutes services,
>>>>> benefits, referral and even 'need' is determined by the services
>>>>> domain, not
>>>>> necessarily 2-1-1 or the Open211. For example, there are various
>>>>> definitions
>>>>> of what constitutes 'homeless' and subvariants within those
>>>>> definitions. My
>>>>> personal favorite is the difference between hobos, tramps, and bums.
>>>>> The
>>>>> differences come out in the assessment (an epistemological process)
>>>>> which
>>>>> then determines the types of services that are needed.
>>>>>
>>>>> Dale
>>>>> ________________________________
>>>>> From:

>>>>> openinfoa...@googlegroups.com<mailto:openinfoa...@googlegroups.com>
>>>>> [openinfoa...@googlegroups.com<mailto:openinfoa...@googlegroups.com>]

>>>>> on behalf of Greg Bloom

>>>>> [greg....@gmail.com<mailto:greg....@gmail.com>]

>>>>> Sent: Wednesday, November 13, 2013 8:36 PM
>>>>> To:

>>>>> openinfoa...@googlegroups.com<mailto:openinfoa...@googlegroups.com>

>>>>> Subject: Ontological question: | Services <> Benefits |
>>>>>
>>>>> Folks:
>>>>>
>>>>> Nitty gritty question that I think I know the answer to but want to
>>>>> bounce off everyone here.
>>>>>
>>>>> An I&R system, as I understand it, connects people with 'community
>>>>> resources' -- as in services. Among other things, services can connect
>>>>> people with benefits. But benefits themselves would not be included in
>>>>> the
>>>>> domain of referral directory information.
>>>>>
>>>>> It would be: need --referred to--> services --enrolled in--> benefits
>>>>>
>>>>> Is that right?
>>>>>
>>>>> I know that some 211s are engaged in enrolling people in benefits
>>>>> directly... but it seems like that's an expansion of 211 function into
>>>>> the
>>>>> service domain, and that benefits themselves are not a part of the
>>>>> domain
>>>>> that we're narrowly focused on.
>>>>>
>>>>> Let me know if I have this correct.
>>>>>
>>>>> And please confirm whether this is an ontological issue or
>>>>> epistimological or ???
>>>>>
>>>>> --
>>>>> • gjb •

>>>>> 202.643.3648<tel:202.643.3648>

>>>>>
>>>>> --
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups
>>>>> "OpenInfoAndReferral" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an
>>>>> email to

>>>>> openinfoandrefe...@googlegroups.com<mailto:openinfoandreferral%2Bunsu...@googlegroups.com>.

>>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>>>
>>>>> --
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups
>>>>> "OpenInfoAndReferral" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an
>>>>> email to

>>>>> openinfoandrefe...@googlegroups.com<mailto:openinfoandreferral%2Bunsu...@googlegroups.com>.

>>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups
>>>> "OpenInfoAndReferral" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an
>>>> email to

>>>> openinfoandrefe...@googlegroups.com<mailto:openinfoandreferral%2Bunsu...@googlegroups.com>.

>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>
>>>
>>>
>>>
>>> --
>>> Clive Jones
>>> AIRS

>>> www.airs.org<http://www.airs.org>
>>> www.twitter.com/AIRSplace<http://www.twitter.com/AIRSplace>

>
> --
> You received this message because you are subscribed to the Google Groups
> "OpenInfoAndReferral" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to

> openinfoandrefe...@googlegroups.com<mailto:openinfoandreferral%2Bunsu...@googlegroups.com>.

> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>
> --

> • gjb<http://flavors.me/gjb> •

> 202.643.3648
>
> --
> You received this message because you are subscribed to the Google Groups
> "OpenInfoAndReferral" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to openinfoandrefe...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
> --
> You received this message because you are subscribed to the Google Groups
> "OpenInfoAndReferral" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to openinfoandrefe...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>


--

[Eric Jahn]

I'm glad Dale was the one to pounce on the "outcomes are different
than benefits" point, so I won't need to be the heavy on that one :)
Yeah, to give a concrete example (that strangely has little to do with
actual concrete): an actual benefit is, in HUD parlance, 1 month's
permanent supportive housing for a veteran, and the outcome which they
report is that the veteran is "stably housed". Good stuff. -Eric

Eric Jahn
Data Architect
Alexandria Consulting LLC
St. Petersburg, Florida
727.537.9474

[Sophia Parafina]

I think this discussion illustrates my concern about any effort called Open211. 

First, if 211 data is available in the AIRS schema, it's open. Putting on my coder hat, I can parse the XML, write the data into Java objects, and move on. It may not be as easy as using a CSV file but the format is published. 

The second concern is the term '211' itself, using it implies that it carries all the assumptions and intent of 211. Don't get me wrong, I think that all this is fascinating and would love to spend several days together with y'all discussing it. However, we should aim for a more broadly inclusive specification of which 211 is a subset of the larger body of data. Our intent with Ohana is to expose high fidelity, fine grained data. Let the user or implementor draw the boundary rather than impose a particular structure through the data. 

I am also wary of building a "kitchen sink" specification that purports to be all things to all everyone. However, I think that Eric's suggestion of aligning with NIEM and NHSIA is the first concrete step. 

Finally, the name should reflect a broader scope and I think Human Services or Civic Services would be a better name.

sophia

[Fitch, Dale K.]

I would concur with Sophia. Greg, your original notion of a ‘community resources data commons’ better encompasses what is being discussed because it does have a broader scope than 2-1-1. Any 2-1-1, whether Open or not, could certainly benefit from the data commons.

Dale

[Clive Jones]

Sophia,

With my AIRS hat on -- I am OK with that ... About 30% of our members are 211s but ergo-ish about 70% ain't. Some 45% of our agencies are engaged in providing I&R to older adults and persons with disabilities through Area Agencies on Aging (AAAs) and Aging and Disability Resource Centers (ADRCs) -- many of whom operate out of city, county and state governments. Plus there are folks providing comprehensive I&R in communities without the 211 banner, and other forms of specialized I&R.

In our best cases, an ADRC and a 211 covering the same area should have found ways of complementary data maintenance but this is not always the situation - maybe an API is ...

Fact is we are used to most folks focussing on 211 as it is higher profile and does have the 'larger' impact in terms of large databases and call volumes. "Civic" is not a term we use often but "Human Services" is more natural from our end -- it might also fit with NIEMs mindset (not that I would know that or not) ...

Clive

[Eric Jahn]

Greg Bloom wrote:
>Eric -
>Can you clarify what the three different alternatives
>are for an Open211 standard-setting project? I think
>you’re saying that one is ‘[an open211 standard]
>based on NIEM core data elements’ and another is a
>unique model based on the 211 domain. What’s the
>third? (And how could we base a referral data model
>on data elements from models describing other
>domains like benefits? Are there specific points that
>need to be aligned?

Greg,
I think the three options I see are:

1) Full NHSIA participation: Just to work within NIEM Human Services,
by defining exchanges based on NIEM Human Services data types,
extending them where needed, and submitting change requests for things
we need logically altered. Extensions are no problem:
-collaborative approach, support of NIEM Human Services domain
-slow to convince NIEM Human Services of our needed changes

2) Partial NHSIA participation: Use NIEM Core, but not make our work
completely NHSIA compliant, where our needs don't fit with NHSIA's
logical model. Merge as NHSIA is ready to accept changes. Full merge
eventually.
-independence/flexibility (no need to wait for HS domain approval of
every change)
-compatibility with NIEM and most of NHSIA

3) Separate Open 2-1-1 effort with our own data types (no NIEM/NHSIA)
-we get to build anew/flexibility
-need to crosswalk multiple logical models to achieve interoperability

My preference is to see how #1 goes. At least that will give us some
experience with the NIEM/NHSIA approach, and inform our own
requirements. If that proves cumbersome/problematic, try #2, then #3
as fall-back.

-Eric

[Greg Bloom]

Okay. This thread had a lot to sift through. Here's the tl;dr ('too long didn't read' version): How about the name OpenReferral as an alternative to (and/or complementary umbrella for) Open211? 

Dale: Outcomes - that was clumsy language on my part, it does point to the need for shared vocabulary. Let me try to rephrase. "We’re focusing explicitly on referral to services. After the point of referral, during the course of interactions between client and service, 'service events' occur, including for example enrollment [in public benefits] -- but this is the concern of a whole other (client-side) domain, with which we should work for alignment even though our domain for the most part does not include it." 

Eric: NIEM - your analysis seems on point to me. I’ve done my best to consolidate it here. It sounds like the first big move in this direction is mapping out the points of alignment (or misalignment) between NHSIA's model and the data spec that CfA has drafted. (Here it’s on github; I expect they’ll also have a google doc to share.) Is that right? If so, do you have a sense of how much time/energy that would take?

Clive: can you clarify, is the AIRS XML interoperable with MARC and ADRC systems? Is ADRC another point of alignment that should be considered in the design of a spec?

Sophia: you note that the Ohana team has been focused just on ‘exposing' data. That’s a helpful clarification. But the county you’ve been working for is relatively idiosyncratic in this field. Based on what we’ve heard in the last three or four months from this group and other city govs and 2-1-1 systems, the technical capacity to expose data is by and large a second order concern. From private referral systems, I’ve heard several times something along the lines of “sure, we could produce an API [they tend to estimate this will take them a few days or a week; I figure that’s probably a significant underestimation but still…] yet it’s hard for them to independently prioritize that when there isn’t a demand for it, and [more importantly] when it’s not clear how we could sustain the resources to produce the data in a world where anyone has access to it.” Meanwhile government-run systems (DC and New York) haven’t ever received the resources to produce high-quality data in the first place. And nobody in the field expects organizations to update the data themselves, unless there’s a broader alignment of incentives around the I&R system (which is a matter of policy and institutions as much as technology).

re: the name, as I said before, I’m also ambivalent about the name Open211. I think that the precedent set by Open311 can be illuminating, and as long as we have AIRS and some 211s on board, the initiative would be readily acknowledged within the field. I also recognize what Sophia and Clive point out: the ‘domain’ of service referrals is not precisely contained within 211 (some 211s do more than referrals; many referral systems are not 211s). 

As for another possible name, we should consider the simplest possible option: OpenReferral. It seems at first glance to me to be ontologically sound, semantically coherent, and politically neutral.

This having been said: does it have to be either/or?

I could imagine an standard-setting initiative (called OpenReferral or whatever) in which Open211 is the name of a project undertaken by participating 211 systems to develop and implement an API as part of a broader cycle of planning, engagement, and development. Meanwhile, the Ohana API is available for use in non-211 I&R systems. They’re all using the same internal logic as established by [OpenReferral], so it’s diversifying without balkanizing. And yet: my imagination is happily unburdened by technological realities. 

~greg

--
You received this message because you are subscribed to the Google Groups "OpenInfoAndReferral" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openinfoandrefe...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

--
• gjb •

202.643.3648

[Clive Jones]

Here's my bit:

Clive: can you clarify, is the AIRS XML interoperable with MARC and ADRC systems? Is ADRC another point of alignment that should be considered in the design of a spec?

I don't think it is "interoperable" with MARC but Eric might be able to clarify whether that is "just" a matter of coding XML elements into MARC language ...

As far as ADRCs are concerned - it depends what software they are using. Within the aging/ADRC sector, software is sometimes selected primarily for its case management capabilities with a resource database module tacked on -- some of those are less likely to use the AIRS XML. But "many/some" (I actually have no idea) ADRCs are using one of the same 5-6 softwares that 211s are using and these use the AIRS XML.

A couple of quick reflections about data updating -- when you send an email to an agency with a link and say "Hey, could you please update your record? If there are no changes could you please confirm this?" The data changes go to a holding area for review before going "live" ... a couple of reasons for this ...

1. The descriptions often need to be condensed into shorter simpler statements of what a program is or does as agencies tend to throw in their promotional language  

2. The descriptions may require clarification ("I noticed you changed the eligibility to x, does that mean ...?", "No that still applies, what has changed is that ...)

3. The descriptions may omit helpful information ("Could you go over the application process, please?")

Of course, not all agencies make changes at first or second or third request anyway ... which is another issue. 

There is also the issue of what is in the database and what is not in it (called 'inclusion/exclusion' in the field). Sometimes individual practitioners (e.g. counselors/therapists) are anxious to be included and I&Rs also have to be wary of programs such as "youth drop-ins" that pop up, look very convincing (too convincing is usually the tip-off) and might be a cult or another predatory organization targeting a certain population ....

Clive

[Fitch, Dale K.]

Greg,

If I may start at the end and then work back to the beginning, I agree, OpenReferral does avoid the conflation with 2-1-1. However, it seems like our primary focus has been on which taxonomy/XSD should be used to map the services. As such, have we not been spending most of our time talking about OpenServicesTaxonomy? The information that would be captured by this taxonomy could then be used by any entity (Open211 or otherwise) for different purposes. Apologies for the digression but I just felt a need to clarify purpose.

Thanks,

Dale

 

From: openinfoa...@googlegroups.com [mailto:openinfoa...@googlegroups.com] On Behalf Of Greg Bloom
Sent: Monday, November 18, 2013 8:25 AM
To: Eric Jahn
Cc: openinfoa...@googlegroups.com
Subject: Re: Ontological question: | Services <> Benefits |

 

Okay. This thread had a lot to sift through. Here's the tl;dr ('too long didn't read' version): How about the name OpenReferral as an alternative to (and/or complementary umbrella for) Open211? 

[Greg Bloom]

Dale - 

Not sure I follow. Are you clarifying because the potential use-cases would range far beyond referral? I see that, although not sure that it's a problem... 

if we were to create open/semantic/interoperability on the level of service taxonomy, paired with a new service directory data specification that aligns with AIRS and the new W3C, we could signal that the primary purpose of this is for the development of open referral systems, without (it seems?) discouraging other kinds of uses of that data. (Also the word 'referral' would seem also to apply in a broad knowledge-oriented sense, beyond just the service-specific context...)

Let me know if I'm missing your gist

[Fitch, Dale K.]

From the data I examined in Michigan, a client is referred for A, and ends up receiving A, B, and C because the assessment at the agency level uncovered additional needs. Sometimes a caller was referred for A, but actually needed D provided through a different agency. That pass off between those two agencies was unknown to 211. In sum, the broader purposes seems to be the services taxonomy of which referrals are one potential use.

 

Stated otherwise, today we have a conflated view of 211 as a provider of both referrals AND the taxonomy. It seems to me the purpose of the project is to apply new technologies to the taxonomy such that it can be used for referrals and other uses. Even the use of the term "referral" has a very specific meaning according to the AIRS Standards: http://www.airs.org/files/public/AIRS_StandardsVersion5_2.pdf

 

I know I'm splitting semantic hairs, but I think it may help explain some of the pushback you might experience from time to time.

[Clive Jones]

Thanks, Dale,

I don't think our definitions have changed but the Standards have changed. Here is the latest version: www.airs.org/standards ...

Clive

[Eric Jahn]

Greg, just wanted to quickly respond to your comments/questions (included below).  Thanks for creating the Hackpad entry. I'll try to review it soon. Checking alignment between NIEM's logical model and CfA's does seem like the next move, and shouldn't take more than a few hours, in my opinion.  Also, I think mapping AIRS schema elements to NIEM would take maybe 10 hours to document in a spreadsheet.  It could take fewer hours, if there is already decent alignment, and we don't have to define extended elements, based on existing ones.  We could use more hours later to more fully document the API in an IEPD  (Information Exchange Package Document) we'd submit to the NIEM repository. 

Also, OpenReferral sounds good, since it makes sense and describes unambiguously what we are working on.  Of course, it would require a taxonomy to be used within it (or allow for use of different taxonomies).

-Eric

[Sophia Parafina]

Ohana's model is based on AIRS so comparing AIRS and NIEM would be sufficient. We are planning to refactor Ohana in the coming year so it's a good time to align with NIEM/AIRS.

I like OpenReferral. We should start using that as the name for the thing we're doing, which from my perspective is developing an open data standard.

sophia

--

[Eric Jahn]

Sophia,
That's great news! I think we're really getting somewhere with this
group. Also, NIEM Human Services has agreed to meet with us
(virtually I guess). Today, I'll be trying to explain our purpose to
them/schedule it tentatively, hopefully. More soon, and I'll set up a
doodle poll with you all for the exact time. -Eric
Eric Jahn
Data Architect/IT Director

Alexandria Consulting LLC
St. Petersburg, Florida
727.537.9474

[Clive Jones]

Just to throw this thought/admission into the mix ... the basis of the AIRS XML is pretty good/firm but it would not suffer from further future work ... Eric Jahn worked with the main I&R software vendors and a group of I&R folks about 18 months ago to create a 3.1 version (fixing some glitches that had made data transfers difficult and making lots of little improvements along the way) but Eric did lay out some future work on the schema and the AIRS Board has that in its radar whenever it checks out how much is jangling in the coin jar ,..

Clive

[Greg Bloom]

Look at what I found: 

http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2014

I'm interested in pursuing this angle of exploration, but I'm not going to be terribly effective at that on my own. Anyone down to buddy up with me on a venture into this realm of ontologists?

subjectively,

greg

• gjb •

202.643.3648

[Greg Bloom]

Folks - see below regarding this Ontology network summit. I'm going to try to make the call, and Dale also expressed interest in pursuing this angle. Anyone else want to join us? 

See below: 

On Mon, Jan 13, 2014 at 12:17 PM, Obrst, Leo J. <lob...@mitre.org> wrote:

Re: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_16

 

 

Dear Colleagues,

 

 

We are pleased to announce that the new Ontology Summit season will be

launching this Thursday (Jan-16.) You are cordially invited to

participate at the upcoming Ontology Summit 2014. Our theme for the

Summit this year is: "Big Data and Semantic Web Meet Applied

Ontology."

 

 

****************** Headlines ***********************

 

Please join us this Thursday at the Launch Event. Thereafter,

activities will be rolling out every week, until ending in April, 2014. In

particular, please join us every Thursday, when we will be featuring

virtual panel sessions as part of the Ontology Summit 2014 program:

 

 

*Thu 16 Jan 2014* - OntologySummit2014: "Big Data and Semantic Web

Meet Applied Ontology" - Launch Event - Co-chairs: Professor Michael

Gruninger & Dr. LeoObrst - see details on the session page at:

http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_16

... *RSVP now*!

 

*Thu 23 Jan 2014* - OntologySummit2014 session-02 - Track A: Common

Reusable Semantic Content - I - Co-Champions: MikeBennett,

AndreaWesterinen, GaryBergCross - see developing details on the

session page at:

http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_23

 

*Thu 30 Jan 2014* - OntologySummit2014 session-03 - Track B: Making

use of Ontologies: Tools, Services, and Techniques - I - Co-Champions:

ChristophLange, AlanRector, ...  - see developing details on the

session page at:

http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_30

 

* between now and end Apr-2014, we are planning to run virtual panel

sessions every Thursday at the same time window, ~ 2 Hrs, starting

9:30am PST / 12:30pm EST / 6:30pm CET / 17:30 UTC ... please mark your

calendars.

 

*Register your attendance* by emailing <pete...@cim3.com> off-line

or register yourself directly to the wiki session page. Please specify

the date of name of the session(s) you are registering for. ... RSVP!

 

 

********************* Details ************************

 

This is our 9th OntologySummit, a joint initiative by NIST, Ontolog,

NCOR, NCBO, IAOA & NCO_NITRD with the support of our co-sponsors. The

theme adopted for this Ontology Summit is: "Big Data and Semantic Web

Meet Applied Ontology." In an earlier planning sessions late last

year, the community and subsequently, the summit organizing team

,brainstormed on how best to frame the issues. These inputs were

carefully reviewed and synthesized by the Ontology Summit Organizing

Committee and its co-champions. It is this plan and program that we

will be discussing with everyone during the launch event.

 

Since the beginnings of the Semantic Web, ontologies have played key

roles in the design and deployment of new semantic technologies. Yet

over the years, the level of collaboration between the Semantic Web

and Applied Ontology communities has been much less than expected.

Within Big Data applications, ontologies appear to have had little

impact.

 

This year's Ontology Summit is an opportunity for building bridges

between the Semantic Web, Linked Data, Big Data, and Applied Ontology

communities. On the one hand, the Semantic Web, Linked Data, and Big

Data communities can bring a wide array of real problems (such as

performance and scalability challenges and the variety problem in Big

Data) and technologies (automated reasoning tools) that can make use

of ontologies. On the other hand, the Applied Ontology community can

bring a large body of common reusable content (ontologies) and

ontological analysis techniques. Identifying and overcoming ontology

engineering bottlenecks is critical for all communities.

 

OntologySummit2014 will pose and address the primary challenges in

these areas of interaction among the different communities. The Summit

activities will bring together insights and methods from these

different communities, synthesize new insights, and disseminate

knowledge across field boundaries.

 

Following earlier Ontology Summit practice, the synthesized results of

this season's discourse will be published as a Communique.

 

See developing details at: '''OntologySummit2014''' (homepage for this summit)

 

---

 

These sessions, like all Ontolog virtual events, are open and free of

charge. Anyone who is interested, or (better still) who may have

something to contribute, is welcome. Please refer to event details on

the session pages, to which the hyperlink is given above, where you

will find session agenda, conference call dial-in, slides and other

pertinent information. Feel free to pass this invitation along to

colleagues who may also find these sessions to be of interest.

 

*RSVP* by emailing Peter Yim at <pete...@cim3.com> offline (or add

yourself directly to the session page if you are already an Ontolog or

Ontology Summit community member) so that we can prepare enough

resources to support everyone. [Please state clearly the date of the

session you are registering for in your email.] ... All respondents

will be subscribed (if they aren't already so) to the

[ontology-summit] mailing list so they are participate in the virtual

discourse.

 

These sessions will be recorded and made available in a publicly

accessible archive. Therefore, before participating, please make sure

you are cognizant of our IPR policy (ref:

http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid32).

 

 

Regards.  =ppy

 

For and on behalf of the

 

OntologySummit2014 - General Co-chairs,

Michael Gruninger & Leo Obrst

 

and Members of the

OntologySummit2014 Organizing Committee

 

http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2014

_________________________________________________________________
Join this list: mailto:ontolog-inv...@ontolog.cim3.net
Leave this list: mailto:ontolog-invi...@ontolog.cim3.net
Msg Archives: http://ontolog.cim3.net/forum/ontolog-invitation/
Community Wiki: http://ontolog.cim3.net/wiki/
Community Portal: http://ontolog.cim3.net/

-- 
• gjb •

202.643.3648

--
• gjb •

202.643.3648