MirrorManager Analysis Proposal: Measuring aggregate adoption delay for updates

[Jeff Spaleta]

On Mon, Feb 2, 2009 at 12:30 PM, seth vidal <skvidal at fedoraproject.org> wrote:
> My primary concern is how the ip-information is correlated and
> presented. Are we talking about releasing specific ip-ranges worldwide?

No.
y-axis is number of unique ips or percentage of unique ips or something similar.
x-axis is measurement of time

> More importantly what is it we get out of this data? I question the
> efficacy of this specific analysis to tell us much of anything.

That sir, is an invitation for an expansive explanation...but I will resist.
Was the originating question and reference devel thread not clear
enough? There was a question as to how quickly yum updates are picked
up.
by the userbase I can give an approximate answer to that question
based on mirrormanager UserAgent strings insofar as it applies to the
clients which connect to mirrormanager. I'm sure other caveats apply
as well which weaken this as an absolute measurement, some of which
could be accounted for in the complexity of how elapsed time is
established as an enhancement.

-jef