SV_MEETING_TITLE -- 01 Nov 2010

<sandro> dd: 2005 network of offices

<sandro> dd: PSI directive in EU

<sandro> dd: JoseMa ctic had funding to start this activity

<sandro> dd: document sept 2005, me and rigo

<sandro> dd: 2007 asturias workshop to launch activity/group

<sandro> dd: internet governance forum, PSI directive, a lot of resources

<sandro> dd: workship in Vilnius, 3 weeks ago

<sandro> dd: culture of some countries, to keep info closed.

<sandro> dd: tried to get bicycle accident info for my town in France

<sandro> dd: they're sending it on paper

<sandro> dd: !

<sandro> karen: in W3C we're normally pulled; in this case it seems like push -- govts are often very resistant.

<sandro> karen: A few weeks ago, Tim was invited to speek at Harvard, Kennedy School

<sandro> karen: Mostly older professors, steeped in govt history

<sandro> karen: ... but they didn't even know basic things, eg twitter

<sandro> karen: ... some folks had a nice openness to understand

<sandro> karen: ... push that has to get done

<sandro> karen: Where can W3C focus?

<scribe> scribenick: Karen

Sandro review eGov IG Note published

Organized in five main topics

1. Participation and Cititzen Engagement

key point to allow public servants to use public Web sites

http://www.w3.org/TR/egov-improving/

<sandro> these slides: http://www.w3.org/2010/Talks/1101-egov/

Sandro: this document points out things that planners need to understand

may be obious to some, but not to those who are new

Rigo: tax declaration system in France
... challenge with massive scaling in gov't sector
... can overwhelm your servers

Roger Cutler: you should talk to the "Dancing with the Stars" show

scribe: they had public voting that overwhelmed the system
... not sure what they did to overcome it

Rigo: Cloud computing is so important to eGov
... so you can start small and scale quickly
... private industry support will be important

Sandro: traditional mechanisms of getting comments from the public have not scaled in the past
... Web makes it easier for public to comment on public policies

2. Participation and Engagement Issues

Sandro: challenges for what gov't employees can and cannot do; what roles and persona they may take on
... how to use social networking
... what are fora for asking authoritative questions
... data portability; archiving
... lots of big questions

Second big issues was Open Gov't Data

scribe: this section was written 1.5 years ago
... not particularly focused about various technologies

Karen: one reason for that was main message was "please use standard technologies in gov't

and also many different levels of experience regarding use of Semantic Web technologies, XML, HTML, etc.

Sandro: other issues are limitations of gov't employees to participate

Third Part was Interoperability

scribe: concept of interoperability in time
... big thing is using open standards

Rigo: EU Interoperability Framework uses a definition
... that fits W3C pretty well
... but other groups did not
... took lobbyists a couple years to get that off the agenda
... We fit the EIF1 definition of Open Standards

Sandro: another technique is to use open source to promote interoperability

Roger: Why? I don't understand it

Rigo: let me explain that from a larger perspective
... EU point of view since I was involved in work here
... For example one village acquired some development software
... then made it open source
... the mayors of various cities would meet and describe same problem
... and suggested they use open source solution
... this was a strong trend
... so they realized they could tailor things more effectively
... so it built the rumor that open source is good with eGov

Roger: so it has nothing to do with interoperability

Rigo: correct; there is a large understanding that open source meant to these towns and villages to get things done

Jim: non-technical people don't understand difference
... between open standards and open source
... some of non-technical people include the high-level policy makers
... don't know difference between open standards and open source
... As Roger pointed out, it's a myth
... if you allow customization, it's a virtue
... but it pushes against interoperability
... it's a complicated question
... has to be looked at on a case by case basis
... We should make the case about open standards

Sandro: in this report there is a page and a half on open standards and a little on open source

Roger: I think W3C should not endorse open source vs. vendor products
... there are real problems with open source

Jim: really not a question about open source
... I am responsible for open source and open standards across HP
... no gov't could decide only to buy proprietary
... it's a development methodology, not an alternative to open standarads

s/standards

Rigo: the big fight is about what is open
... after EIF shifted their definition
... like example of CRM being relabeled identity management
... then everything became "open" term
... "open" now meaningless because it's the latest cool term
... everybody claims to be open
... but maybe we should develop criteria for that

<sandro> rigo: Everybody claims to be open now

Roger: and we have something called "Open Spirit" a proprietary software

Vagner Diniz: it's not wrong to say we can find more friendly people

scribe: among open standards rather than proprietary community
... idea is not to advocate open source
... but we should advocate open standards
... in terms of awareness we can find more friendly people in the open source community; they are more helpful for us

Daniel: open source and open standards are not the same, yes
... Open source people do like us because we give them open standards which are free
... whether or not open source is good in and of itself
... as a way to develop software
... free software
... is a philosophy
... sharing programming is hard
... some W3C people come from this community
... saying it's just a development methodology may offend some

Gautier Poupeau: different question with exposed data

scribe: we can make interoperability with data
... and open source data
... but that is not the key point

Roger: I was just commenting on promoting interoperability and open source

Rigo: It's a how on promoting eGov and open data
... open source is often used as a pre-cursor with low investment development
... before doing a commercial product
... there is this aspect that leads to this kind of assertion, but we have to disambiguate the assertion

Jim: I think this is a good topic; we are not wasting time
... It is true the open source and open standards community overlap
... maybe a stepping pointt
... is what people really want is royalty free
... RAND is the default rather than W3C RF
... open source can be misleading
... it is under a license
... HP supports 600 licensees

s/licenses

scribe: to really follow all those and administrate them is big
... cannot just take your program and hand it to the people in the next village
... In terms of Linked Data
... there have been contraints
... one of major ones is Privacy
... EU has stricter rules
... Different and sometimes conflicting rules about tracking, destroying storing data
... Identity for some is heart of security
... different ways to protect identity will show up
... many issues will face on open source side

Roger: Open source is also not risk free
... why organizations like mine feel good about use of W3C standards because they are relatively risk fee
... they have been rung out and vetted
... there are may patent trolls

[Roger describes example of patent vulnerability in open source]

scribe: I think open source is big risk for a company like ours
... risk factors from W3C are very different from Open Source

Harry Halpin: I am working in social and identity space for W3C

scribe: we are trying to follow US gov't work
... I would be interested in hearing from people
... Identity is important to eGov
... and it's historically an area of patents
... Going back to 1998, P3P
... importance of RF standards
... we need to identify areas that are crucial
... there are vastly different requirements
... and make sure baseline core standards are RF
... and can be implemented
... a bit of a weakness; so I'm interested to hear what eGov people think of the identity space

Daniel: what I have been looking at for eGov
... is about access
... you want what you have paid for back
... I don't want my identity in this process
... I want to make use of the data I paid for
... not interaction with gov't that I paid my tax
... but another eGov platform area

Harry: there is another realm of applications for clinical health records that needs identity assurance

Roger: there is something here
... thinking about the person who asked Daniel for his physical address
... perhaps people are unwilling to send something that could be altered electronically
... our company still sends me PDFs to sign and send back
... because they don't trust it
... there is a huge element of not just making technology work
... but getting people to understand that it does work and get them to trust

Rigo: I have been involved in EU since 1995
... today you can do ID electronically
... but IT guys saw business opportunity
... but it doesn't take into account the user

Jim: coming back to patents
... W3C is not immune from patent trolls
... trolls can also attack W3C standards
... but you want someone to back it up
... and Open Source community is not set up to provide that guarantee
... on the other side
... Open Source benefits from thorough W3C vetting
... and many eyes on code are best way to find flaws on code or algorim
... so even though Open Source is exposed, it may be more secure due to number of people using it

Sandro: Last topic in report was interoperability frameworks
... I was not sure about this

Rigo: I can give a short presentation about that

Sandro: how about two sentences

Rigo: We have a large discussion about this topic with EU
... they developed a framework
... EU is not like US Federal gov't
... from inside EU , it's still 27 countries cooperating
... focus on how those gov'ts send messages to each other
... it is a big area needing procedures
... needed to understand data
... They started to exchange XML Schema
... each gov't had its own schema store
... Questions about which models to use, how to find them
... For example, we may give local, regional or nat'l gov'ts information
... about what others due
... European Interoperabilty Framework (EIF)
... was developed
... same people who developed the open standards definition
... I did an interoperabilty analysis for egov for them
... I came up with three levels we take into account
... First is syntactical; exchange of XML schema
... second one is semantic
... still have procedural constraints
... so has to be procedural interoperability
... and this is hardest to do
... Current frameworks only touch on schema interoperability
... Linked Open data is a large test on whether we can achieve interoperability on intra and inter-gov't data
... It's for them a concrete problem how to exchange data between them
... have constraints
... just cannot send things from one dept to another
... gov't has to constrain itself
... there are walls
... this interoperability frameworks touches on them
... also touches on rules for procurement
... Procurement is an important level for gov'ts to achieve what they want
... versus what the vendors want

Jim: I think the term interoperability frameworks is a high-level term, like APIs
... would be a mistake to read too much into it
... somehow exchange data
... I am chair of board of the Open Group
... 50 states had data set up differently for example
... they got together to develop a common framework
... so applications sold to schools could use that same framewok
... in that case it's an XML framework
... simple, does the job
... for good or ill, it has been put in legislation as a matter of law
... works well enough that other places are starting to implement it as well
... Schools Interoperability Framework (SIF)

<sandro> "Schools Interoperability Framework"

Jim: We can talk about them, but need to speak more generally

Roger: So I am hearing that interoperabilty farmework is a rationalization of open standards

Jim: developed from Microsoft

Karen Burns: I am queen of interoperabilty framework for New Zealand gov't

scribe: from end to end, procurement, how we use tech
... I have to transform siloed agencies to collaborate
... globally these things have been focused on technical standards
... if it's open it reduces my cost

Roger: I would call it information architecture

Karen: for us it wraps everything together

Sandro: What I heard IF is the way you cross the silos

Karen: policy, standards, processes
... things are maturing now
... I am playing now in the transformational side
... legal and policies to change things

Gautier: I think there are different levels for interoperability
... Semantic Web is an interoperabilty framework
... Rigo explained that it's more semantic global
... we have to explain this different lower level
... of frameworks
... for example, first you have RDF, the OWL

Sandro: There are different ways of breaking that down

Daniel: so regarding interoperability
... I found a license I would have to sign
... to use this software
... license is asking me, if I have an application
... based on data the gov't gave me
... that whenever the gov't updates the data
... then I have to update my site
... so that's how far we go into interoperability
... if you operate something you have to do it well
... It does not prevent that you have to use open standards

Rigo: We had that problem
... for a very long time I was lobbying to have laws

free online

scribe: From Journal Officiel
... they got commercial entitty
... prepared ready to go database on a military service
... and just by doing this they earned 6 million francs per year
... If your application doesn't reflect the law
... so they must force you to update your stuff

Daniel: if not interoperabilty through open standard
... one day for another they could switch XML syntax

Sandro: So one question
... what is the difference between interoperability and open gov't data?

Rigo: There is a large overlap

Karen: that is a whole other discussion

Rigo: Take the UK example of when they released tons of data
... they correlated the crime and geolocation data
... also complaint data
... local mayors and villages
... in areas where you had an idiot who complained a lot
... the price of your house was half what it was before
... because people were thinking it was a bad area
... from complaints about noise or theft

Roger: in a highly regulated industry, we must provide gov't with a lot of data
... mixed up with gov't standards

Sandro: A clear definition for interoperabilty

Emanuelle: I see it around domains
... a certain level of interoperabilty to exchange in a bounded domain
... have to define constraints that are domain specific

Gautier: there are other issues
... like archiving
... Private data
... you have to have data to think about archiving it, even if not public
... yet have to think about archiving issues
... have to think about interoperability of data in data preservation

Yosueke: We are talking about Interoperabilty Frameworks and Interoperabilty
... IF is broader term
... covers mapping and procedural aspects
... there is also possibility to extend interoperabilty
... semantics, culture are different
... can exchange data
... IF is something will be the premise for
... cooperatively working with gov't

Daniel: What I have seen in past couple of years
... with work TimBL is doing
... more on policy side
... gov't making data open for free
... assumption of geeks is we will make something out of the data
... there is a change in spirit
... do something useful and we will make sense of it
... there is a change I think

Vagner: interoperability is more realted to policy, rules, architecture that provide conditions to data exchange or application exchagne
... Open data is one way to faiclitate interoperability; not the only way

Jim: Looking down stream, I think the term open data

<sandro> Vagner-br: open data as one way to get interoperablity

Jim: is going to lead to problems
... may interpret it as public data
... in Norway, open public data is what their individual income and taxes are
... may not be the same with other gov'ts
... we will need to have terminolofy that people will understand clearly
... if people around this table are having trouble with distinctions
... the legislators and public will have trouble, too

Karen: second that

Sandro: In the IG note they define open gov't data and public sector info as synonymous

Rigo: yes
... I would not overload the term
... what we mean is this is data that you can get from the gov't without forcing them by court
... we can extend in light of Daniel's experience
... we may mean it's in digital format
... and avoid the paper route
... which they are capable of doing
... so I think it's what is available publicly

Daniel: public sector info is not the same thing as open data

Roger: I just did some Google searches
... I convinced myself that interoperability framework
... does not have a widely accepted defnition
... first reference is to the Estonian Interoperability Framework

Rigo: That is not surprising; the Estonians and Austrians were the first in this area

Roger: suggest that you define it clearly

Karen: it means nothing right now

Rigo: a catch-all

Daniel: One remark is that eGov activity
... reminds of beginning of WAI
... it was clear that we needed several groups
... Education and Outreach to agree on terms
... technical groups, one of access
... keep in mind that the education and outreach was one of the first ones
... we need to have some kind of group to deal with that task force

Sandro: so about five minutes before break
... one more slide on interoperability
... we have raised many of them
... Go quickly through next two

4. Multi-channel delivery

scribe: providing services through multiple channels
... from f2f and phone to web and mobile
... issues were about really understanding the requirements of the consumer
... mobile devices are limited
... coordination of the different channels
... so you don't provide different levels of service on different channels
... and Accessibilty

Fifth Item is Identification and Authentication

scribe: allows confidence in transactions
... but "lack of coherent analogies to the forms and protocols that have endured for centuries"
... notion of fair rsk allocation and risk
... gov't may think it controls id authentication but cannot
... Finally, next steps
... It was an active group at the time document was written
... Spirit of enthusiasm to help gov't transition
... Focus on Linked Data
... and look at data provenance issues
... The actual next steps for IG
... unless something changes
... will run as more low-level
... meeting 1 to 2x a month
... with speakers; more like a seminar
... coming out of these topics
... Hoping to get your input on topics
... of interest to you; or ones you may be willing to speak about
... maybe that will draw a new community and do a new veresion of these reports
... Let's take a break in 30 minutes

- DRAFT -

SV_MEETING_TITLE

01 Nov 2010

Attendees

Contents

Summary of Action Items

Scribe.perl diagnostic output