DNSO Mailling lists archives


<<< Chronological Index >>>    <<< Thread Index >>>

[nc-whois] [fwd] Q 5 crunched (from: abel@able-towers.com)

I think this didn't make it to the task force mailing list, yet -  
it's something those of you who attack question 5 should definitely  
look at.

Thomas Roessler                        <roessler@does-not-exist.org>

OK, as agreed some crunched numers on Q 5, i was playing around with, i
shoved it all in one sheet, for convenience :P

Now, there is one problem; the sheet almost has to be "self-explicatory"
since i have made no comments on the sheet itself, but then it was/is a
working document.
Henceforth a little explanation here.

I first of all changed the numeric grading into "importance" with rank 4 as
the neutral rank.
I know this is somewhat bending, but imo a plausible bend :P

Then i dropped the totals in, which are in the corresponding columns.

Next i made a divide between " importance +" and "not important".

Made a "totals-per-side" column and totalled the numbers, to get an average.

>From there i took it more personal, since nothing significant had come up
(well at least reasonable usable).

So i decided to start with removing the "not stated" since on average 78+ %
did not answer in that group (we all know this is a bunch of empty forms
that should have never been put into the equasion) then i looked over the
percentages per question.
This immediately made one thing clear: these questions had differentiating
intrest at stake, and therefore combining all the answers would always end
up with "common denomitives"

For instance: all but governmental institutions, would be interested in the
"availibilty" of a domain, hence the high percentage of not important for
the government group.
Same goes for similar names and merchant verification, all other groups
however (disregarding "not stated") had a form of interest in these
questions, So i deleted the govenrment answers from the totals and made new
totals and averages.

Q D (IP enforcement) was obviously a question that has some semblance of
importance for the governmental institutions so they were brought back into
the count, however the individuals had less to no interest there, hence they
were removed and again an average was created.

In Q E (source of Spam) it was clear that the "commercials" have a vested
interest and will be less "charmed" of searching for culprits, where
registries have little to do with it and know better ways to find a source,
so i left them out this time round and redid the average.

In Q F again it was apparent that the registry has little to do with illegal
activity and law enforcement, so they were left out of the equasion.

Q G is in fact to weird for words and the "non-answered" questions takes a

Now i can draw upon some preliminary conlcusions:

hehe, i will leave yours to yourself, so not to draw upon anything that
isn't there.

As you will see, albeit it the long way round, this question gives us some
more insight, which we will be able to use as factual base for our later
thesis on free-form answers and total conclusions (i hope)

have fun



<<< Chronological Index >>>    <<< Thread Index >>>