[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: FWD: Re: [atlarge-discuss] Election SOddities and DanIrregularities (was something...

OK. I will still waste 10 minutes on that because outsiders want to know all and read nothing.

At 00:01 07/06/03, Micheal Sherrill wrote:
When an email message gets sent (such as the returned email ballots), each
email contains a message-ID --
No. The mail mai containt
- a Message-Id
- a Message-ID
- a Message-id

Since - by mistake of mine no one noticed - and of no real importance I just looked at one of them, only a small part of them where recorded. So I gave the whole list so Danny could find at least by himslef that this kind of bug are not easy to think of. He had the whole problem nicely listed line after line in front of him and nother him nor Michael saw it.

It also happen that there are more Message-ID that I did not looked. ... Anyway the entire list was avalaible in the proper order....

The message-ID in your last email to me was:
<200306061541.AA21496100@mail.beethoven.com> Oftentimes it is coded to indicate
the date that the message was sent. In your case it was 2003-06-06.

If the ballot information was properly recorded, then every ballot would have
a message ID in addition to the ballot code generated by Jefsey.
Yes.
Far too many ballots that contained votes did not have a message-ID (only
19% contained this identifier). This is rather unusual when balanced against
the fact that 96% of the blank returns contained a message-ID (most all of them
in fact had two message IDs).
RIght observation. Wrong conclusion. Not a developper. Thinks the code is
correct abd the intent is bad before checking if the code is correct. A user
think the code is corect, a developpers knows that the code is not :-)


For an example, consider ballot #0034  -- it had two message-IDs:
<20030530062844.A51C7E3B7
<20030525055620.8317AE205

No. As explained several times - but who reads before explaining - all the
mails received are recorded. It only means that the first mail is a non
ballot mail. May be Danny sending a comment to atlarge@excelub.org.


This tells us that the ballot was sent twice --
once on 2003-05-30, and earlier on 2003-05-25
That it should be sent twice with no recorded vote on both occasions is a
signal that something is wrong somewhere.  This is not an isolated example.

Yes something is wrong: you did not read the explanation and never
figured out that a non ballot mail cannot produce ballot lines :-)


That we should have 53 consecutive ballots all with no vote recorded (ballots
6 through 58) is a statistical improbability, especially when the coding on
several of these indicates that multiple ballots were sent on different days
(for reference see ballots 31-35 and 45-49).

Do not understand what this may mean. The figures only shows that the
first sending (before I started removing the bounces form the table generated
by watch.c were more numerous than further on. An empty ballot is a bounced
mail. It is resent to the sender and all its voting lines are empty.
When one person only has access to the raw results, is in a position to strip
the ballots of identifiers, and then forwards the stripped and possibly
altered votes to a set of "watchdogs", one does not have an integrity of process.
Again never even read a single line of the description of the vote, yet
doctorates about it.
1. the three panelist received the same mails at the same time
2. I only carried a help that any other two could have carried
3. they checked all my results agains the raw data daily and
    at the end.
4. the raw data are une sealed.zip file with each of us and at the
    disposal of the Election Committee by the new Panel.

So, basically every thing you said made me to waste again 10
minutes of my time for which return? Just because you did not
read and understood basic things. Since I suppose you will
continue not to read because you are mainly interested in showing
that a solution - all the pros consider as extremely elegant - is
biased, dumb, a mistake etc....

So you will excusie me, but for the time being I will not respond
anymore to that type of outdated questions. So please enjoy
wandering if....

all the best
jfc


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