[Date Prev][Date Next] [Chronological] [Thread] [Top]

RE: Wyle Certification Material



Tari,
    Candidate withdraw. Sounds like what your saying is something that effects only the GEMS side of the issue...Wyle wont even see GEMS, so this feature is in 1.96.1, what effect does this feature have on the Accu Vote printout...?
-----Original Message-----
From: owner-support@gesn.com [mailto:owner-support@gesn.com]On Behalf Of Tari Runyan
Sent: Wednesday, May 02, 2001 9:09 PM
To: support@gesn.com
Subject: Re: Wyle Certification Material

Charlie - to withdraw a candidate (1)  you go to the race editor then the candidate tab -  select hte candidate and check withdrawn button.
then you upload reults and the name is not on the race
----- Original Message -----
Sent: Wednesday, May 02, 2001 7:01 PM
Subject: Re: Wyle Certification Material

Global Chouteau wrote:
Any and all, attached is a list of questions relating to the feature additions that takes the release 1.94w to 1.96.1.......I would appreciate any answeres (even one) you can provide...
> 1. Name “WITHDRAWN” (recommended approach)
This can be demonstrated with any ballot by withdrawing a candidate after the ballots have been printed but before downloading.  Sorry but I don't remember how to withdraw in GEMS off the top of my head.

> 2. Handling Precincts with no ballots. (and what and why this is necessary)
It's for recounts in which some precincts have none of the races that are being recounted.

> 3. Auto Absentee feature.( here I assume new ballots will have to be printed)
Correct since they depend on separate ballot IDs for the absentee ballots.  See the Auto Absentee test databases at http://staff.gesn.com/resources/gems/databases/index.html .  From http://staff.gesn.com/lists/qatesting.w3archive/200103/msg00058.html, "The Auto Abs Test 1 is with full pct IDs and Test 2 is with sequence IDs."

> 4. How to handle “Counter Group Header Group” cards…I could use an explanation of the use of all the new Header Cards that can be used with 1.96.1
The Counter Group Header cards are designed for use with V2 ballots which are not supported by 1.96.  We do not currently supply CG header cards.  The V1 AV-OS supports 2 count groups which are predefined to be Polling (CG1) and Absentee (CG2).  Therefore 1.96 will recognize the CG header cards and will interpret the CG 2 header the same as the Absentee Count card and the CG 1 header the same as the Polling Count card.

> 5. How to handle “Counter Group Header, Poling” cards, or “No Override Header Card” after using  the Absentee Header Card..
You're mixing V1 and V2 header cards here.  The No Override Header is what I call the Count Group Separator card which is coded as CG0.  It would be used to clear a count group override in effect by using the CG header cards as CG overrides which only has meaning with V2 ballots or, to a limited extent, when Auto Absentee is enabled.

In order to implement the Auto Absentee feature, we had to implement a way that would detect the CG from the ballot ID and therefore these cards act as override cards when the AA feature is enabled.  Unfortunately using these headers with AA causes all ballots to be rejected if the CG is actually changed so there is no good way to demonstrate this mode.

> 6. How to explain the use clearing old overrides before setting new overrides.. Please explain “overrides” and how to use the ballots we now have to demo this change.
To override means to override the ID printed on the ballot.  The only override that has meaning with V1 is the precinct ID.  Ballots counted after using a precinct header card are counted to that precinct until a precinct separator card is used to clear the override.  For V2 we would also encode the CG in the ballot ID and therefore the idea of a CG override would make sense.

> 7. Same questions when using the Accu-Feed unit.
We use a different model for auto feed vs. manual feed when dealing with header cards.  For manual feed, you need to press the YES and NO buttons in order to process a header card.  Therefore there we have confirmation that you actually meant to process that card.  With auto feed we do not require holding buttons so we use the idea of a deck to ensure that you really mean to process the given header cards.  In this case, an override, pct or CG, must be cleared before another override can be used and before the ender card can be used.

> 8. Where do I get the override cards..?
You should have precinct header cards.  There is no way to demonstrate CG override with V1 but I could send you some CG header cards which would act the same as the Abs Count and Polling Count header cards.

> 9. Explain “check that the card is valid for the current count group”. I assume this is active only in pre-election mode.
The V1 AV-OS has support for 2 count groups which we call polling and absentee.  If you encode the CG on the ballot as we do indirectly for the Auto Absentee feature, then you could use an override card to override the CG in the ballot ID.  What 9 says is that we actually check that the CG is valid for the card style and reject the ballot if not.  Prior to this we would accept it anyway because the V2 AV-OS has the counters even if they're not valid.  What this really means in practice is that since for V1 we need different card style IDs for Polling and Absentee ballots when Auto Absentee is enabled, we will reject the ballot if you try to override the CG because no other CG is valid.

> 10. Need to demo the up to 48 character precinct name capability.
Then you'll need to prepare an election database with 48 character precinct names, download a VC, and print a totals report.

> 11. Is there any way to demonstrate the “download protocol version doesn’t need to be pre-selected in GEMS…?
No, not yet.  This awaits GEMS support which probably won't make sense until the pre-1.96 firmware has been largely phased out.

What prevented this in the past was that the AV waited for the host to identify the protocol first even though both ends were really supposed to declare themselves at the same time.  This meant that GEMS had to be told what protocol to use before it could download.  All we changed was to have the AV identify itself first so that at some point in the future we could use this to avoid the need to pre-identify the protocol in the GEMS AV options.
 

         Guy