[atlarge-discuss] detailed visual data count under Excel

["J-F C. (Jefsey) Morfin" <jefsey@club-internet.fr>]

Dear Members,
you will please read that mail to the watchdogs. May I add that I am 
amuzed. For years our community has been mared by suspicions about Joop's 
booth and the claims to have access to programs and to data. When all the 
data are provided, when the program is given, when you are all made 
watchdogs...

- Joop proposes to check the votes in using the booth (to take back 
control? :-)
- Joop even asks if it is possible to have the list of the voters, which is 
online on two machines. And documented several times.
- NO one among ANY of you asked about the real decision which was to be 
taken about the way nearly 1/3 of you voted :-)
- yet several spent hours documenting why they disapproved not having .... 
what they do not read.

I agree this quasi fool proof voting system is a step ahead in ICANN's 
world. So it changes some of your habits, and respond to your demands (what 
is not usual). It certainly lacks a better presentation and documentation. 
But it did very well and has obviously no problem escalating, at quite no 
cost and in being made 99.5% clone proof (voters, organizers and watchogs), 
without violating privacy.

Obviously always enjoyable to spit at a Cadillac because one cannot spend 
his time as usual, feeding the horses :-)  Now, those who have an hour, 
please reproduce what I did and get your results ready. When the watchdogs 
have reviewed my Excel count - I may have made errors: they are here to 
check me - I will publish it, report the members and turn myself to the 
development of http://apolite.net and of othe projects as announced.

jfc

==================


Dear Watchdogs,
You will find enclosed my visual recount under Excel. I did not document 
that method I think the best one to visualize the problems and the 
decisions made, not to interfere with your own solutions. Every one can 
check this. I will put that file on-line after receiving your comments. I 
copy this mail to the Mailing LIst so anyone can run it separately.

The data are obtained in:

1. loading all the data of the published atlarge.wtc file in fixed format 
and creating the fields for question nr, VID, received ballot, obtained 
response, following text. Adding the first Type column for convenience

2. I sorted by obtained responses and removed all the "{ ]" responses to 
reduce the size.

3. I sorted by Question/VID and built a command to show me the duplicates. 
This resulted in 2 duplicates. I created a column OK where I marked them "NO".

4. I then sorted by Type and Received Responses and created the column 
"accepted responses"
    I accepted all the responses including "X" for votes
    I accepted all the responses including Y or N for questions
    I accepted all the responses including "X" as "Y" for questions as it 
is the obvious desire of the voter. I was surprised no one rose that issue 
watch.c has no right to address.
    I marked OK as "NO" for all the other received responses.
    I market all the accepted reponses as OK "YES".

5. I sorted the result per question, per type of response and per VID. I 
created the column "JFC Order" to stabilize the sorting for futher 
reference and gave an order to every response.

6. I created the column "Accrued" countong +1  for every line, restarting 
on Question/Accepted response change (for OKayed responses only)

7. I noted as a "result line" the line precedeing a 1 in "Accrued" column. 
I marked it as such in a "JFC Result" column.

8 I created the Result page and copied the significant elements of the 
result lines.

9. I entered in this page the results of watch.c

10. I listed the difference between the automated provisionnal count by 
wtach.c and the visual recount as explained above. This completes the 
procedure defined by the voted voting rules.

I find nothing strange. No proof of cheating. The differences with watch.c 
provisional results seem to be fully explained by my remarks of yesterday. 
The difference for the questions is often nil (but the addition of the "X" 
significantly increases the number of "Yes" to many questions.

Thank you for your comments.
jfc


---------------------------------------------------------------------
To unsubscribe, e-mail: atlarge-discuss-unsubscribe@lists.fitug.de
For additional commands, e-mail: atlarge-discuss-help@lists.fitug.de