eGovernment Interest Group Meeting

welcome and intros

[intros]

<josema> scribeNick: beng

john:agenda ok?
... take input from outside work and kick on with use cases at 9H45
... the group & jose have had a number of conversations
... talk about semic
... renke to introduce SEMIC.EU

Input from outside work

[martin points out he worked with the previous project]

[SEMIC.EU slides: HTML, PPT]

Benjamin: will the asset be fixed once agreed?

Amit: this becomes a facilitating body, is this your role ?

Renke:not fixed, the idea is not to change the asset but the connection to the asset from the Member State
... we have another community in ehealth
... they just use semic to publish the assets
... there is not just hosting, also improve the coaching

martin: have you looked at the mothodology used by XBRL
... in particular localization

renke: we use components from UN/cefact
... semic is not just spread these concepts, the communities have to decide which methodology they want to use
... and we publish them, and advise them and indicate envisionned problems

<josema> we had a presentation yesterday about XBRL + SW at the Plenary (potential XG coming?)

john: for eJustice, can you explain the public policy objective this community is trying to achieve?

renke: in the ejustice space there are 27 countries. Why do they want to share this information ?
... find the names of companies
... how can one handle the different semantics each country has to store company data
... the gov organisation must know if the company really exists in order to validate giving them information

martin: Services directive to be effective end 2009 to have information on any activity related to a municipality
... will mean there is only one point of contact where all companies can get information

renke: the solution from ejustice can be reused in another context
... collect assets and modelize them, build relations between concepts and rules
... also a lot of communication, not just technical.

martin: [explain how europe works for people from other countries]

john: generalize out
... there is a set of ppolicy drivers that want to improve the flow of judicial information
... i.e., get a parking ticket in france and have to pay it in UK
... we need information sharing
... problem : standardazing the judicial system is difficult
... other route : achieve the policy outcome without going into massive standardization

jeff: are there legal barriers to exchange information accross borders?

all: yes

martin: subsidiarity principle : only do centrally what has to be done centrally
... respect what is already done at local level
... can only be change on a volontary basis
... europe can not impose anything locally except if its done everywhere
... semic approach : each of the 27 members states is a black box

renke: 5200 different sources just in Germany
... impossible to build a standard in all issues

john: this is an instance of goverments solving G2G information via government action
... govs have an objective they are not outsourcing
... need to achieve some sort of degree of information sharing between entities without imposing a degree of standardization
... and finaly need a mechanism for interoperability : in europe this is Semic

Martin: in social security

[this was another example]

amit: intereaction between superannuation

[retirement fund]

john: every nation has a different regime in europe
... this is more interoperability than standardization
... q : why chose interoperability over standardization or outsourcing?

oscar: in the long run, if you have common assets, it is easier to achieve internal interoperability and external also
... stick to what can be done in 12/18 months

john: from apoliticy of sociological point of view standardizaion can be viewed as bad

jeff: how hard to agree on ontologies

martin: difficult

john: tough when there are deep differences
... i.e., common law vs civil law

martin: semantic interoperability : common ontology but not an ontology shared by all organizations

renke: maybe having a large ontology is good, but we wanted to start with practical examples
... there are already problems inside a state
... italy, germany : problem of the concept of name
... find the solution on specific problems, such as "the name concept"

amit: look at http://www.agimo.gov.au & http://www.govdex.gov.au
... contact is Brian Stonebridge

john: the specifics lie in the data
... do not model the data, look at "note" fields
... its easier to construct the ontology after that
... viable approach in a data rich environment such as ours

<josema> forensics on existing data... interesting

amit: we use SAP and people have found ways of cheating it already

john: as assets are identified and constructed to capture some part of the process that led to their creation

benjamin: was it analysis of the data or of the model

renke: 2 different views
... 1) how to solve technical problems
... 2) look at what we are doing to find a way of improving the process
... I could report in 3-4 months

martin: see where a bottom up method is useful when creating an ontology
... and find where/when a top down approach is prefered
... make a comparison and make a decision tree

john: interesting discussion

jeff: is there any citizen resistance to information sharing?
... and how is this dealt with?

renke: maybe none really knows what is happening

john: very important in the UK
... public trust is low at the moment
... due to data loss
... problems with companies contracted to resolve the issues
... data is not secure
... 3 reasons to think about sharing information
... identified in the UK Wallport/Thomas review
... 1) requirement from the social sciences community for longitudinal data sets
... vast amount of personnal information, to be used to understand what is happening in your society
... possible to anonymize
... 2) bring information together to deliver a better service
... a citizen might want the data access level to change during time i.e., just after getting knocked over by a car
... 3) national securityscribe: consent is no longer needed
... bill going through parliament to keep all emails ever sent
... in the Web context how can one achieve a degree of data sharing ?
... anonymized data is less problematic.

benjamin: very important in the UK
... working on PDS, Personal Data Server
... secured usb with personal information that gets anonymized
... but you need some trust identity
... question: is there a way for an individual to retain the info
... in order to send it as needed while preserving privacy?

john: use case framework next

Use Cases Framework

<josema> see also http://lists.w3.org/Archives/Public/public-egov-ig/2008Oct/0034

john: how do we address the vast number of issues in the egov space
... what framework to put forward to look at the use cases ?
... motivation about the simple approach
... governments role on the Web: 3 main areas
... this seems to be general everywhere
... 1) using the web as a mechanism for delivering public services
... historically : in the 90's not sure what to do, and little by little more services have been added
... there was no real interaction with the citizen
... the state is simply throwing out services with various levels of success
... 2) engage with people
... rise of social computing, creation of big communities and patterns of behaviour between these communities
... i.e., parenting advice
... if one question is "how to interact with this part of the government service"
... shouldn't the person responsible of this service also be present in this forum
... design public services, receive comments and improve them
... politician get excited by the management of people's experience of their use of services
... citizen 2 citizen & citizen 2 gov interacction
... 3) governments sit on a heap of data
... this is a monopoly
... govs can send this information in order to create new services
... question is : what is the role of the state as a "data provider"

amit: this is focused on gov 2 citizen
... this has policy implications that need to be driven top down
... the G2G use cases are also challenging
... interoperability at all levels : city to state, etc
... in AUS, interested in looking at what has happened around the world

vagner: agree with the 3 areas in general but most govs are only in the first area
... few in 2nd area, and 3rd is only a possibility !
... we need to be ahead on technology, so need to find use cases in 2nd and 3rd, maybe even focus on them

john: govs are better at 1 and G2G

amit: disagree
... people are interested but it is not a success

<jeffs> +1

amit: most gov departments are hierarchical

amit: a lot of innovation needs to happen in order to see interoperability appear

martin: add to the 1st use case
... there is no real technical obstacle to impementation
... govs want to move forwards, they start at technical point, but to move on to a larger scale project,
... this creates a conflict betwee, technicians and lawyers
... when making standards mandatory:
... 1- each aspect of the spec is translated into legal terms
... 2- say these standards are mandatory and then add them to the law

martin: it is clear neither from legal nor technical point of view what needs to be done
... proposition : have an intermediate level
... divide into technical and non technical issues
... creating such a model would be beneficial

[BREAK]

<josema> scribeNick: amit

John: adding government to government as fourth strand
... engage is new, enable is also new, not much activity in these areas in terms of what gov's do right now
... the main focus is develop use cases in these two days, and need to figure out what the use cases will be for
... how detail the use cases need to be? we need to capture successes and failures....

John: how do we capture things that have happen, or new things that few people have implemented?

<josema> +1 to identify the target audience

oscar: we need to identify target audience and consumers of information,
... discuss what we are going to produce as a group and
... what they are going to use the products (outputs of the group) for

john: who is the audience?
... our audience is primarily working in gov and also people who are working in gov projects
... so that they lead the web to its full potential
... assume that these people will not be aware of technology knowledge,
... helping gov people making better decisions about the Web

martin: policy makers, legal expert, political, who are they?

oscar: need to discern between non-tech as well as tech people...

john: policy makers (including web policy makers) we should target them...

jeff: there is a second audience, ie citizens,
... we need to focus on citizens being able to check up their own personal data that the gov holds...
... this has long run implications for society...

oscar: need to distinguish between local, regional and national level, and should we have targeted messages for each?

beng: is it about data correction or is it about citizen having impact on gov policy

jeff: both

martin: who owns the information about you? are you or the gov the owner?

oscar: there is also statistical information, eg demographics...

beng: who owns the data? with a long example that i cant capture

martin: if one has registries and information is being stored,
... and one needs to use that information, then from the perspective of both gov and citizen,
... that this information has updated and managed for it's life
... through technology and process

john: there are many categories, citizen to gov, gov to citizen, gov to gov,
... we should try and capture one of each of those in our use cases,
... hence giving us a broad range of unique problems...

martin: there are cases where the gov might be involved or where the community will not be wanted

john: back to audiences...
... we have different types of audiences, for each of the above areas
... do these people interact or are they a long way away?
... there are policy people and practioners in each area
... do we focus both?

amit: both, in my experience setting up and Special Data Infrastucture (SDI)
... I had to communicate to both communities
... in Australia we have the Office of the CIO, which has a board where all
... gov orgs come and meet
... type of authority we should target?
... then we have AGIMO, sets up technology policy, standards, etc.
... also to be considered
... Web is majority component, eGov is broader, they do more than Web wrt technology

martin: is Web goal or means?
... eGov defines more integration of Services than Web alone

oscar: provide case studies to policy makers...
... pick case studies at the technical level, and then provide methods,
... standards and pro's and cons about standards
... also provide areas where new work in standards is required

<josema> now that we discuss about it, I have to remind you of charter scope: "The eGovernment Interest Group (eGov IG) is designed as a forum to support researchers, developers, solution providers, and users of government services that use the Web as the delivery channel."

amit: standards "per se" is not enough, how do I make them useful for a given government?

oscar: need information on how to coordinate between standards
... we need to highlight scenarios and the use of standards

josema: points to the e-gov standards...
... gov does not need a list of standards, but they way they should integrate them...

oscar: perhaps we can focus our work with a usecase format applied here... and then demonstrate how standards satisfy the usecase and also problems with implementation

martin: what about trying to focus on the migration to WCAG 2?
... it's very difficult to understand why WCAG 2 is the way it is...

amit: my opinion as former developer, I only want a superficial knowledge
... not why a Group made a specific decision

josema: the idea behind this group is to identify egov specific issues
... and then coordinate with other W3C groups to improve other outputs...
... eg WCAG 2

renke: agrees with josema, an example is SEMIC's interest and influence W3C standards via this IG

josema: this is a better process, and something that W3C is geared to do...

john: do we try and construct usecases centric to egov or most digestable to W3C?

josema: we need to focus on egov use cases...

john: which comes back to audiences
... egov/web policy makers, practitioners and W3C people
... do we need to complete the loop between all three?

jeff: we are talking about process,
... one way to do this is to minimise discussing thing groups,
... and focus on what groupsthinks that we should focus on

john: this is helpful, it also means that the use cases that we constructed
... on our own knowledge and understanding, rather than abstract problems

vagner: concerned about framework,
... we need to be concerned about what makes this group unique

amit: that brings us to usecases..

<Vagner> list of some relevant observatories in Portuguese http://www.observe.org.br/observegov/Default.aspx?idPagina=6330

oscar: we need to understand the deliverables are and by when they will be required...
... we need as a group to agree on scope

josema: lets focus on developing usecases... some real examples...

<Beng> http://www.w3.org/2007/eGov/IG/wiki/Use_Cases

josema: persistence use case
... gov publishes, URI's, etc, and then something may get lost due to change in CMS or some tech change

john: in the record of Parliamentary proceeding in the UK, only 50% of the of the questions and answers posed are working... (ie hansard)
... ie the links that document the Q/A of parliament...
... gov and non-gov citations persistency is about the same, ie gov is doing no better...
... in 2005 UK gov tested URL references in hansard, and more failed than worked
... so persistence is a problem...
... and rate of citations is increasing, and this happens because,
... 1. new CMS, 2. structure of gov is not stable ie departments changing url all the time, aggregation and dis-aggregation, at all levels of gov,
... 3. either a change of minister or political adminsitration
.... and political influence to remove and show information....

amit: provenance, temporal and persistance of web information from government (consistency)

martin: maybe the focus should be on provenance of information, and persistence
... provenance, temporal and persistance of web information from government (consistency)

john: gov is irritated by the broken link problem
... and we have a web archiving tool kit, that can be archived centrally
... but we did not know what web sites we had
... but even getting access to web sites does not help in understanding who own the web sites
... we ramped up the archiving and enabled a push mechanism
... and we mandated a behaviour, when someone requests a web resource,
... the user can be served the latest version on the department web site or to the uk gov archive
... this was the only way solve the problem centrally
... allowing gov web people to play fast and loose, and let them do what they needed
... using 301 was a great idea...ie resource has permanently moved...
... costing UK gov 500,000 Pounds per annum...
... archived on quarterly basis or an important event basis
... also using sitemaps protocol to improve discovery of information on archive

jeff: perhaps we should focus on good practice as well as standards
... not just standards
... using this case study and explain archival practice... in gov...
... people are trying to do a good job, but unable to do this
... due to lack of knowledge...

john: example is accessible web sites...
... people developing new accessible web sites and throwing the old inaccessible one away...

jeff: take the 4 interaction models, gov 2 gov, gov 2 citizen, citizen 2 citizen, citizen to gov
... and develop use cases in these areas

martin: develop usecases and then approach other W3C working groups for feedback and comment...

john: we need to influence w3c of the realities of gov and how they are continously changing, and not part of the cool URI thinking...
... we have 4 areas, and articulate simple minded ambitions of gov, this needs to given within the business context and the public policy context...
... and develop usecases and good practice for policy and practioners within gov....
... and on this side influence W3C working groups...

amit: good summary

[starting after lunch]

<josema> scribeNick:josema

john: recap
... four different areas mentioned before in terms of interactions
... and considered our role and interactions with other W3C Groups
... in order to identify and pose questions to them to help ground their work
... in the reality of eGovernment
... how to communicate with government officials and practitioners
... if we can identify few cases that could show this, we could succeed
... (and make govs save some money)
... discussed one particular use case
... "cool URIs don't change vs. government does change"
... is that a fair reflection?

renke: I'd add from SEMIC that some of the projects we have listed there
... could help the eGov IG as case studies
... helping them to improve their solutions
... we are developing a scorecard idea to improve those
... we are discussing with several projects to find the best solution
... what could help us is a kind of template to describe this
... feedback from W3C would be useful in this sense

john: some kind of template to capture it?

renke: yes

john: in this case our role could be to capture your feedback
... identify W3C areas that are addressed there and act as a channel to discuss
... with the relevant Groups at W3C
... then collect that back and document
... send back to SEMIC/publish it

renke: beyond feedback, give context to other projects
... so other states working on similar projects could see
... maybe also bring people to W3C to discuss them

Lauri Hirvonen (Nokia): EU very interested in developing this and funding pilot projects
... any of you involved?
... big discussion on eID, eg. you need it to make your tax declaration online
... is that within scope?

john: don't think we could add much value to that
... given our scope and focus

martin: I can reverse the question
... and there is relationship with what is going on in EU
... eg. cross-participation, liaisons

lauri: several levels of government in Finland, are you take care of all?

john: yes, we have no choice, given the diversity the government has

[people agree]

john: if we can make that flow of dialogue happen, that would be good
... would it help SEMIC?

renke: sounds good to me, I'm happy to go back to IDABC and tell them

john: this could be an approach we could bring to other organizations

[Charter dependencies: http://www.w3.org/2008/02/eGov/ig-charter#coordination]

[and some more at: http://www.w3.org/2001/11/StdLiaison]

josema: two concerns
... other Group's charters and W3C Membership of participants

[FAQ http://www.w3.org/2007/eGov/IG/faq]

john: I can also share how I justified the business decision

ACTION jsherida to share how he justified the business decision of becoming W3C Members

<trackbot> Created ACTION-14 - Share how he justified the business decision of becoming W3C Members [on John Sheridan - due 2008-10-30].

josema: very difficult to convince governments because of time commitments issues

john: this model mentioned above could be of use
... connect initiatives this way
... we need to make progress on the use cases
... before adjourning today
... I can share some of my work

[john introduces Show Us a Better Way]

john: 120+ ideas that are very good and need more review
... two key observations:
... only 20 data sets to make these ideas work
... huge overlap, in particular: education, health, transport
... about how the gov can support the citizen to make a better choice
... public housing, best transportation to go from A to B
... best place to send kid to school
... there's also around performance information to get to a service ??
... stroke me the disconnection between the citizen and the information the government produces
... if we could surface performance information in a useful way
... we can support citizens much better
... we can now assure based on the competition that much more people had the same idea
... around this topic
... a data reuse use case on this topic could be very useful

martin: if you use the approach of asking the citizens what they want
... similar to what jeffs proposed this morning, then how to present this to the Groups ??
... you try to identify sth that is a real problem, using a real problem as an example
... you need action to solve it, if you get only academic questions then that's not useful

oscar: what is the process of prioritization of services the government puts online?
... put up there what is possible, what it seems is useful for citizens, what others have done
... but this topic itself is something not usual

john: why is easier to pick a restaurant or plan your holidays
... than to pick a school for your kids?
... if we surface performance data, people will use it in different useful ways
... will allow to have a better view of the information
... one clear example, two big Web sites on health where people discuss
... one run by gov (@@) and one by charity (@@@)
... different type of engagement, discussions, and big debate is who is doing better job
... some people feel less moderated in a non-gov forum
... making the performance information available make this services work better, too

[scribe got disconnected from network for 3-4 minutes]

john: I see many potential benefits
... it's all about data reuse
... how do you write the business case of having the gov exposing data but not charging for its reuse
... even when the benefit is clearly connected to what the government is
... this is a big concern to policy makers

oscar: what's the cost of not doing so?

john: back to the stds point of view
... lots of data published already but hidden by proprietary format

oscar: example: eurostat in excel format

martin: some kind of adapter for relational DBs to expose data on the Web?

oscar: DBPedia

martin: information has been probably lost in the process if that's in Excel format

<john> for an example see: http://www.justice.gov.uk/publications/companywindingupandbankruptcy.htm

martin: how it can be benefitial for govs to present data in a way that is useful for citizens, public servants,
... things like statistical data
... how do we create a useful environment in which standards can help achieve this goal
... probably most out there, needs just to be put together
... this could be discussed with specific use cases with other Groups

john: that's one example, best example could be something like
... school performance data available in proprietary format, not easy to reuse
... from W3C side say that open formats is better choice
... from govs, if I do so, that will help make my citizens make the choice easier

<john> another example: http://www.dcsf.gov.uk/performancetables/

[john recaps]

john: for tomorrow: focus on use cases on one side, good practices on the other and match them
... and agree on having some of those written up and done by the Group Members

martin: we should settle that we want to be the connector between government organizations and W3C Groups
... we act as intermediate, too

oscar: propose to make a list of deliverables and plan
... roadmap for the next months and action people

[all agree]

martin: see if we have enough for the 4 subject areas mentioned today
... focus on creating the roadmap to create the deliverables
... more important than the details

john: we'll start with PLING joint meeting, then work on this as proposed

[ALL to think about these areas and prepare for the discussion tomorrow]

[ADJOURNED]