Re: [election-services] RE: VIP/EML CONVERGENCE?

Peter - you beat me to it.

David - I think you are getting ahead of us and VIP. As Peter points out it is very unlikely that the States hold geospatial info in their EMS systems, and all we are trying to do for now is match what VIP want and that is to handle straight forward extracts of relationship data from the States. What you are proposing sounds great but is for the future I would suggest, maybe EML v7. From my experience of election officials they are very very slow to move to new ways of working so we need to be able to meet their ancient ways of working for now and hope to educate them for the future. Aaron Strauss made a very good point on the recent telecon which you missed. He said one of their main objectives of this phase of VIP was just to get the States to recognise the need to use standards. Any debate about which standard was for a future phase of their programme. So let's try and get over this hurdle in the easiest and quickest way and then hope to influence the future direction of VIP .....-)

Regards
John ( the UK Luddite!!)

M. +44 (0)7976 157745
Skype: gov3john

----- Original Message ----
From: "Zelechoski, Peter" <pzelechoski@essvote.com>
To: David RR Webber (XML) <david@drrw.info>
Cc: EML TC <election-services@lists.oasis-open.org>; John Borras <johnaborras@yahoo.co.uk>
Sent: Wednesday, 30 July, 2008 5:02:37 PM
Subject: [election-services] RE: VIP/EML CONVERGENCE?

David -

Most of the voter registration products seem to use street segment files to define their areas. Each range of numbers (e.g., 100-199) on a given street, and even or odd can be separated as well, is a segment. These segments are then placed into a set that defines a given precinct. That is the simple case and applies to a large percentage but not 100% of the time. You can find places where specific floors of a building are grouped, etc.

These products don't use geospatial information, so the likelihood that the jurisdictions will be able to provide that is not good.

- Peter

From: David RR Webber (XML) [mailto:david@drrw.info]
Sent: Wednesday, July 30, 2008 9:21 AM
To: John Borras
Cc: EML TC
Subject: RE: [election-services] Fw: VIP/EML CONVERGENCE?

John,

I think the one approach solves both. Frankly - the old approach was needed because they did NOT have the means to solve it easily - so it became a crazy morass of rules and amendments and notes and changes.

It's the same problem in essence - where is my qualifying address relative to the polling station and the geographic boundaries that define it?

I'm not trying to compete with Google - instead - I'm looking to use EML to empower their solutions (and all the other makers of mapping solutions).

They all need the raw geographic information in a simple and consistent way that is easy for the voting authorities to provide that's compatible with the existing mapping engines and standards.

So whether you provide street address boundaries and intersections - and / or geographic coordinates - the idea is the same - to spell out what the physical area looks like.

By using geospatial and consistent street addressing together (that's what its designed for!) - we can capture what that is in a coherent way - and the mapping software can read that XML in and visually show you it - so you know if you have it correct.

The last piece is then easy for the voter - provide their own street address - software shows them where that is - what precinct its inside (draws those lines as overlay on the map using existing mapping API) - and directions to the polling location address!

Thanks, DW

-------- Original Message --------
Subject: Re: [election-services] Fw: VIP/EML CONVERGENCE?
From: John Borras <johnaborras@yahoo.co.uk>
Date: Wed, July 30, 2008 5:01 am
To: "David RR Webber (XML)" <david@drrw.info>
Cc: EML TC <election-services@lists.oasis-open.org>

David

As I see it there are two distinct voter questions that VIP is trying to handle. First, which is my polling station, and second how do I get there.

The second one is clearly a Google maps type of scenario and we shouldn't try and compete with that, just provide hooks to jump out to those type of applications.

The first question is very different and is not in my opinion a Google maps type of solution. The "which is" bit depends on a number of factors and will/can be different for different elections. This is where the business rules come into play. So determining the voter's polling station depends on which precinct/ward they are in for a given election and this is/can be affected by population headcounts, local authority boundaries, electoral jurisdictions etc. I don't see mapping software handling all that complexity. All we need to be able to do is handle a simple relationship table that links addresses to precincts to poling stations for a given election. Let the States admin folks use the business rules to work out the boundaries etc and enter them into their EMS system, from which VIP can then take an extract.

The EDXL solution seems to be fine to answer the question of where is the nearest polling station to me but that may not be the one I should use.

So as I said yesterday let's not get sucked too deeply into geospatials etc.

Regards
John

M. +44 (0)7976 157745
Skype: gov3john

----- Original Message ----
From: David RR Webber (XML) <david@drrw.info>
To: John Borras <johnaborras@yahoo.co.uk>
Cc: EML TC <election-services@lists.oasis-open.org>
Sent: Tuesday, 29 July, 2008 8:23:10 PM
Subject: RE: [election-services] Fw: VIP/EML CONVERGENCE?

John,

Yes - I agree a new EML 115 can capture the geographic details of the area associated with the 110 polling place.

I'm seeing there is potentially an "easy" solution - use almost the exact same method as OASIS EDXL is already. We can "borrow" their schema imports of the CIQ and Geospatial XSDs. They already have this. It would be a nice primer for us to move to CIQ v3 next in EML V6.

Think we need to collect together the conceptual model first on how this all would work (Paul - novel concept - do the information model first!) - using street / GPS / map references to markout the physical boundaries - store that in XML - but I'm seeing the means to do this is probably already in the EDXL stuff.

Then ensure we have the right pieces from the VIP side too - and the linkage to the 110.

Googles mapping engine should be able to place people's physical address inside the boundary then - drawing it on the map and pinpointing their house.

This is going to be a very cool application. Think the states will have huge incentive to provide those EML 115 and EML 110's to drive this with - and it will make their lives much easier - instead of having to figure out all kinds of ugly addressing based rules - let the mapping software do the work...

Thanks, DW

-------- Original Message --------
Subject: Re: [election-services] Fw: VIP/EML CONVERGENCE?
From: John Borras <johnaborras@yahoo.co.uk>
Date: Tue, July 29, 2008 4:17 am
To: "David RR Webber (XML)" <david@drrw.info>
Cc: EML TC <election-services@lists.oasis-open.org>

David

I'm not convinced we should handle geography specifically within v6. I think it's out of scope for us. We can leave hooks for users to insert their own preferred mapping tools, eg Google maps, but let's not get sucked too deeply into this aspect.

I agree we have within 110 most of what is needed for VIP but the one bit we don't have is the link between a voter's home address to their precinct and thus polling station.  The business rules that determine this are usually within the election authorities' systems and all we've done is handle the output from that which tells the voter on their polling notification where to go to cast their vote. What VIP want to do is handle the business rules as well so the third parties can provide the searching facility direct to the voter. So that's where I think we need the new message, perhaps a 115.

Regards
John

M. +44 (0)7976 157745
Skype: gov3john

----- Original Message ----
From: David RR Webber (XML) <david@drrw.info>
To: John Borras <johnaborras@yahoo.co.uk>
Cc: EML TC <election-services@lists.oasis-open.org>
Sent: Monday, 28 July, 2008 2:31:30 PM
Subject: RE: [election-services] Fw: VIP/EML CONVERGENCE?

John,

Good thoughts. I'm seeing for EML v6 we need to handle geography! Mashups and Web 2.0 are upon us...

KML is Googles markup for maps, GPS and more. Most GPS devices support KML.

You can do very cool things with it - such as this -

http://www.drrw.net/inside/travel/cycling-mashup.html

which obviously could be used to provide recommended route to and from a polling station from nearby highway - etc.

Back to polling stations. Actually if we look at the EML-110 election event - we have a chunk of this already.

Should give us a good model to follow to add in the additional parts. I actually think this is closer than we realize. Not sure how much new stuff we need - may only be one new transaction that provides voter polling place linkages.

I worked on alot of things this weekend (including the EML 110!) - but still have some more testing and crosschecking to do before everything is ready to share the full analysis with the group.

Thanks, DW

-------- Original Message --------
Subject: Re: [election-services] Fw: VIP/EML CONVERGENCE?
From: John Borras <johnaborras@yahoo.co.uk>
Date: Mon, July 28, 2008 5:43 am
To: EML TC <election-services@lists.oasis-open.org>

Thanks for all the very useful contributions. Looks like we will need to have a very flexible new suite of messages to accommodate the address - precinct- polling station, including opening days/hours, set of relationships, and their linkages to contests.   Not sure I've seen this complexity in the VIP stuff so far but maybe as they roll it out they will come across the need for it.

David - two questions on your note below. First what's KML? And secondly the VIP requirement is primarily about which polling station do I have to use not how do I get there. So EDXL HAVE might not be appropriate if it's focus is just on giving directions?

So where does all this leave us and how do we take it forward?   Well in my opinion we need to review David's address handling facility first to see if that meets the VIP needs. Secondly we then need to build the new messages to handle the geography elements for polling station allocations. And finally develop any further EML schema localisations needed for the remainder of the VIP requirement. Having got that set together we can then ask them to validate it before working out a cross-over strategy with them.

Agreement or alternative views would be appreciated plus views on whether we need to have a TC telecon to discuss this further.

Regards
John

M. +44 (0)7976 157745
Skype: gov3john

----- Original Message ----
From: David RR Webber (XML) <david@drrw.info>
To: Richard J Cardone <richcar@us.ibm.com>
Cc: EML TC <election-services@lists.oasis-open.org>
Sent: Friday, 25 July, 2008 9:20:28 PM
Subject: RE: [election-services] Fw: VIP/EML CONVERGENCE?

Richard,

OK - thanks for reminding me one of the core values here for VIP - pointing people at the right polling place!

So -thinking about this - its not a pollbook type of function - its a public information one.

What EML is setup to carry is the ballots, candidates and polling place information.

So this is an additional function associated with polling place - which relates to address boundaries.

And then now we're seeing another - which is early balloting and those mechanisms. I'd add to that absentee ballots.

So there's potentially two new types of EML supplemental transactions. For the address boundaries I'd add geospatial and support for KML. Fortunately this has already been done for us using CIQ!!!

The new OASIS EDXL HAVE schema is doing exactly this for hospitals and emergency management. In that case its "how do I get to the right hospital?". Substitute polling station for hospital and we can borrow all that part of their XML!

Then for the "means to vote" - I don;t think we have anything specific right now. However in both instances - we can re-use the eml-common lexicon of xml parts - to assemble these pieces from is my sense - and hence tie them neatly into the "EML family".

Thanks, DW
-------------------------------------------------------------------------------------------------------

Rich -

I would agree that there are locales where that happens -- the polling locations have set hours/days but an individual would still be in a precinct (the precinct determines what contests are contained on the individual's ballot).

In Texas there are places where early voting is done at regional locations and anyone can go to any location. They are still assigned to a precinct, it is just that all precincts can go to all early voting locations. On election day the polling locations become localized and individuals are routed to their local polling place that handles one precinct (or a small number of precincts).

Polling locations actually change frequently (a church or social hall is under repair and they move to a different facility for the current election). Precincts usually change only when redistricting or such takes place.

- Peter

-------- Original Message --------
Subject: RE: [election-services] Fw: VIP/EML CONVERGENCE?
From: Richard J Cardone <richcar@us.ibm.com>
Date: Fri, July 25, 2008 2:30 pm
To: "EML TC" <election-services@lists.oasis-open.org>

Peter and John,

There is the case where polling places can change over time in an election. For example, voters may be directed to centralized polling places during early voting. At some point, early voting ends and election day--the final day of voting--arrives. On election day, voters may be directed to their neighborhood polling places, which are different than the centralized polling places.

Rich
--------------------------------------------------------------------- To unsubscribe from this mail list, you must leave the OASIS TC that generates this mail. Follow this link to all your TCs in OASIS at: https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php

Not happy with your email address?
Get the one you really want - millions of new email addresses available now at Yahoo!

--------------------------------------------------------------------- To unsubscribe from this mail list, you must leave the OASIS TC that generates this mail. Follow this link to all your TCs in OASIS at: https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php

Not happy with your email address?
Get the one you really want - millions of new email addresses available now at Yahoo!

election-services message