- From: Owen Ambur <Owen.Ambur@verizon.net>
- Date: Thu, 30 Oct 2008 10:41:51 -0400
- To: <public-egov-ig@w3.org>
- Message-id: <000101c93a9d$a5c5cd10$f1516730$@Ambur@verizon.net>
Vassilios, per Jose’s request, I am addressing this to the listserv so that
others can join in our dialog if they wish.
You have touched on a key point, i.e., distinguishing between what should be
confined to the government domain versus that which cuts across all
organizations. In the U.S. government relevant policy includes OMB Circular
A-76 and the FAIR Act.
http://www.whitehouse.gov/omb/circulars/a076/a076.html &
http://www.whitehouse.gov/omb/procurement/fair-index.html
In my view, there is nothing “inherently governmental” about software
programming logic. In general the only thing that distinguishes
governmental activities from commerce and industry is the kinds of *data*
(types of records) required to conduct “We the People’s” business (“in Order
to create a more perfect Union”).
http://en.wikipedia.org/wiki/United_States_Constitution#Preamble:_Statement_
of_purpose
That’s why I believe the Data Reference Model (DRM) is the only one of the
U.S. Federal Enterprise Architecture (FEA) “models” that really matters. I
don’t believe we can truly understand what our “business” is without
understanding the information required to conduct it.
Even though I worked for the top-ranked enterprise architecture program in
the U.S. government, I did not see any value in the FEA Business Reference
Model (BRM) … until I rendered it as a “strategic plan” in StratML format:
http://xml.gov/stratml/index.htm#FEAPMO Viewed as a set of goals and
objectives, I can see the potential value of the BRM – but only if actual
performance against those objectives is measured and reported to
stakeholders (i.e., citizens).
On the other hand, since .gov agencies are wasting so much of the taxpayers’
money on software applications and systems that are “proprietary” in nature
and cannot efficiently and effectively share records, I do also believe that
the FEA Technical Reference Model (TRM) is important – to the degree that
the commercial sector lacks incentives &/or cannot be trusted to specify,
implement, and enforce compliance with standards for interoperability.
Again, however, the TRM is only meaningful to the degree that it is used to
measure and report actual performance of agencies in implementing and using
those standards.
The FEA Service Component Reference Model (SRM) comes closer to modeling the
“internal parts and components of a public service.”
http://xml.gov/stratml/FEASRM.xml However, from my perspective, few, if any
of those “components” are inherently governmental in nature. So I don’t
believe government should waste the taxpayers’ money competing with the
private sector to reinvent them. On the other hand, I do believe the
government has a role on behalf of citizens and taxpayers to ensure that
commercial enterprises are implementing the relevant *technical* standards
for interoperability among those software components and services. In the
U.S. federal government, relevant guidance includes OMB Circular A-119 and
the National Technology Transfer and Advancement Act (NTTAA).
http://www.whitehouse.gov/omb/circulars/a119/a119.html &
http://ts.nist.gov/Standards/Conformity/nttaa.cfm
Owen Ambur
Co-Chair Emeritus, <http://xml.gov/index.asp> xmlCoP
Co-Chair, AIIM <http://xml.gov/stratml/index.htm> StratML Committee
Member, AIIM <http://www.aiim.org/Standards/article.aspx?ID=29284> iECM
Committee
Invited Expert, W3C <http://www.w3.org/2007/eGov/IG/> eGov IG
Membership Director, <http://firmcouncil.org/id5.html> FIRM Board
Former Project Manager, <http://et.gov/> ET.gov
<http://ambur.net/bio.htm> Brief Bio
From: Peristeras, Vassilios [mailto:vassilios.peristeras@deri.org]
Sent: Thursday, October 30, 2008 4:55 AM
To: Owen Ambur; Jose M. Alonso
Cc: John Sheridan; Kevin Novak; Rick Murphy; Óscar Azañón
Subject: RE: types of use cases? -- Re: regrets for today's meeting
Hello Owen,
>>I believe we already have enough (too many) large "conceptual models."
This is true ...but two observations:
a) Yes, we have (too) many eGov conceptual models modeling different
things. However, we are still not sure what each model presents. E.g. in my
understanding in FEA there is no representation of what a public service is
and how it operates. BRM rather proposes a functional clustering of service
identifying public sectors (The Business Reference Model provides a
framework that facilitates a functional (as opposed to organizational) view
of the federal government’s Lines of Businesses.” [1]) But this has nothing
to do with modeling the internal parts and components of a public service.
So I think we have to be careful when we say that we have many conceptual
models. Conceptual models of what?
b) “Large” is not what we are looking for, as we need simple, small and
extensible models as you already mentioned. I agree XML is a good tool for
creating such models. I prefer RDF but this is another discussion.
>>While the IG doesn't have the means to *populate* such a model, it should
>>not be a large undertaking to specify an XML schema (XSD) that others
could
>>use to contribute toward the population of a worldwide model for
government
>>services
I agree that the IG won’t populate such a model. I also agree with Jose
previous remarks that perhaps this is something for another IG or WG. But I
thought that this is one of the main goals for this IG: to identify areas
where future work could go on towards more detailed specs and not to create
any spec in any specific area. Perhaps I misunderstood the IG mission.
>>The XSD for the U.S. Federal Enterprise Architecture (FEA)
>>Business Reference Model (BRM) could be taken as a starting point:
>>http://www.whitehouse.gov/omb/egov/documents/FEA_XML_Schema_Rev_2.3.xsd
As already explained, the BRM and other FEA models could provide insights
towards this goal. However, there are some other initiatives that are closer
to what is discussed here (e.g. UK CGIM, OASIS Reference Model for SOA and
more).
>>However, I believe the IG's time would be better spent highlighting the
>>necessity of identifying, implementing, and measuring/reporting
conformance
>>to *technical standards* for interoperability.
A general observation from a modeling perspective: the more we move towards
technical standards, the more communality we may find with other domains.
For example I still can not see why the FEA DRM model is an eGovernment data
model as the larger DRM part might have come from other domains, too. I
think that W3C at this technical level could help to identify existing low
level technical (thus globally applicable) standards and then try to
disseminate and persuade governments to use them (e.g. Web Services, XML,
RDFa, microformats etc). The more we try to speak of the government business
per se the more we move to non-technical issues. The FEA BMR is clearly such
a model, very useful to be used for creating a map and categorization of the
administrative space (or jungle;-)… but not for modeling a public service
operation as explained beforehand.
My personal stand is that we can achieve many things with the usage of low
level standards but if our goal is a) to really promote semantics and
relevant technologies and b) we really believe in the business value these
technologies may bring then you definitely need to create our own small yet
extensible domain-specific models, i.e. we need specific and commonly agreed
XML schemas that will capture the semantics (conceptual models) not just the
use of XML (technical standards). All these may be marked as “conceptual
models” and ideally should also be independent of technology to be reusable.
Towards this direction modeling in a generic way the public service is an
interesting exercise and could provide one quite useful of the many
different “conceptual models” we may need, i.e. in the context of developing
semantic SOAs, and/or annotating public services descriptions as currently
available through numerous national and federal eGov portals with RDFa on
the web.
Best regards,
Vassilios
[1] http://www.whitehouse.gov/omb/egov/documents/CRM.PDF
-----Original Message-----
From: Owen Ambur [mailto:Owen.Ambur@verizon.net]
Sent: 29 October 2008 15:30
To: 'Jose M. Alonso'
Cc: Peristeras, Vassilios; 'John Sheridan'; 'Kevin Novak'; 'Rick Murphy';
'Óscar Azañón'
Subject: RE: types of use cases? -- Re: regrets for today's meeting
I believe we already have enough (too many) large "conceptual models." What
I believe is needed are more and better *implementations* of relatively
small components that constitute measurable objectives toward the
realization of larger goals upon which some degree of consensus exists.
On the other hand, to the degree that folks may wish to collaborate on the
specification of a common model for describing public services, it seems to
me that the differences Jose references could easily be addressed as
*extensions* to the core set that are deemed common across most, if not all
governments. ("Extensible" is XML's first name.)
While the IG doesn't have the means to *populate* such a model, it should
not be a large undertaking to specify an XML schema (XSD) that others could
use to contribute toward the population of a worldwide model for government
services. The XSD for the U.S. Federal Enterprise Architecture (FEA)
Business Reference Model (BRM) could be taken as a starting point:
http://www.whitehouse.gov/omb/egov/documents/FEA_XML_Schema_Rev_2.3.xsd
However, I believe the IG's time would be better spent highlighting the
necessity of identifying, implementing, and measuring/reporting conformance
to *technical standards* for interoperability.
Owen
-----Original Message-----
From: Jose M. Alonso [mailto:josema@w3.org]
Sent: Wednesday, October 29, 2008 7:34 AM
To: Owen Ambur
Cc: Vassilios Peristeras; John Sheridan; Kevin Novak; Rick Murphy; Óscar
Azañón
Subject: Re: types of use cases? -- Re: regrets for today's meeting
Well, I see two sides of "common specification for describing public
services": technical and business/conceptual.
The one you refer to below would fall in the technical side, and would
be doable. The other, I'm not that sure. I'm just involved very
slightly on R&D and found some papers on the subject and organizations
working on this but, every time I think of it, looks like something
not possible to me.
I don't see how there could be just one single conceptual model of
eGov services that could be applicable to numerous countries. The
differences are too large. I see that Vassilios mentions a pilot in
Cyprus in his reply. This seems doable (though yet costly) to me, but
I don't think that model would work in say Angola, Argentina, Ghana...
even the US or Spain.
I'd like to be proven wrong :)
I'm also copying Oscar. My institute has been working with the State
government to open source their Interoperability Framework (this was
finally announced last week, see http://www.asturias.es/openfwpa). I
remember hearing Oscar and some of my colleagues at CTIC about the
underlying conceptual model, and that the FW is much more than code.
I'm wondering how he envisions others will use it and implement it,
and I'd appreciate his thoughts on this.
Vassilios, if you believe that reference conceptual model is doable
and that you could find interested organizations, why not talk to them
and start an XG? If chartered appropriately, it could coordinate with
the IG and become something else (maybe IG or WG) when re-chartering
the eGov Activity, who knows...
Jose.
El 23/10/2008, a las 5:00, Owen Ambur escribió:
> Hey, guys, point b) below makes me think of the U.S. Federal
> Enterprise
> Architecture (FEA) Business Reference Model (BRM) and Service
> Component
> Reference Model (SRM), both of which are available in StratML format
> at
> http://xml.gov/stratml/index.htm#FEAPMO or, more specifically,
> http://xml.gov/stratml/FEABRM.xml & http://xml.gov/stratml/FEASRM.xml
>
> It also reminds me of GSA's OSERA:
> https://osera.gov/c/portal/layout?p_l_id=PUB.1.38 I'm copying Rick
> Murphy in
> the event he may have comments or suggestions to offer in that regard.
>
> However, as per my exchange with Mike Daconta, I believe it is more
> important to focus on finalizing the XML schema for the FEA Data
> Reference
> Model (DRM). See http://xml.gov/index.asp#DRM and particularly
> http://xml.gov/draft/drm20060105.xsd
>
> Owen
>
> -----Original Message-----
> From: public-egov-ig-request@w3.org [mailto:public-egov-ig-request@w3.org
> ]
> On Behalf Of Jose M. Alonso
> Sent: Saturday, October 18, 2008 1:23 PM
> To: Peristeras, Vassilios
> Cc: John Sheridan; Kevin Novak; public-egov-ig@w3.org
> Subject: types of use cases? -- Re: regrets for today's meeting
>
>
> Hi Vassilios,
>
> [+Group, this needs discussion]
>
> I added your two ideas to the wiki so we could discuss at the F2F.
>
> Anyway, I believe these are a bit different from what we have
> discussed so far. Have any of those been implemented and deployed by a
> government already?
>
> I'm asking because I believe this shows also another issue for the
> Group: what kind of use cases do we want? Real and implemented
> solutions or also ideas that could address some eGov issues?
>
> We might want both, but I have a slight preference for the former. As
> we recently discussed several times on "web basics" even basic things
> like persistence are not being correctly implemented or government
> don't provide any data in reusable open formats although they have
> been around for quite some time already.
>
> Would love to hear the opinion of the Group so we could focus the F2F
> discussion even better.
>
> Thanks much again for sending those!
> Jose.
>
>
> El 15/10/2008, a las 13:53, Peristeras, Vassilios escribió:
>> John,
>>
>> Thanks for your message.
>>
>> I have two areas in my mind.
>>
>> a) The first could be actually related to an existing W3C
>> submission, namely SIOC (Semantically Interlinked On-line
>> Communities) which is an attempt to create a simple ontology for
>> linking the blog sphere and web-based discussions. Recently adopted
>> by Yahoo [1]. This may fit to “Transparency and Participation” and
>> more specifically to
>> Engage: Using the web to interact with citizens and businesses, 1.
>> Government to citizen - e.g. blogs by ministers and officials, or,
>> publishing consultation documents as wikis or with a "comment on
>> this" facilities.
>>
>> b) The second could be linked with a common specification for
>> describing public services. The approach is similar to SIOC: a
>> simple and easy to be used ontology which could give unifying
>> semantics (annotations) to service descriptions based on different
>> perspectives by service providers. We already have an RDFa based
>> prototype on this that uses an (eGov) extension of SA-REST as the
>> annotation vocabulary.
>> I think this fits to both Provide: Using the web to deliver public
>> services and Enable: Using the web as a platform to deliver data for
>> re-use
>>
>>
>> Simplicity and clear value of the approach is key for future
>> community/governmental adoption thus the use cases will try to
>> highlight on these.
>>
>> Please let me know if you find these ideas relevant.
>>
>> Best regards,
>> Vassilios
>>
>> [1] http://developer.yahoo.com/searchmonkey/smguide/SIOC.html
>> [2] http://www.w3.org/2007/eGov/IG/wiki/Use_Cases
>>
>>
>
>
>
Received on Thursday, 30 October 2008 14:43:34 UTC