On Tue, Sep 2, 2008 at 7:52 PM, Mr O <[EMAIL PROTECTED]> wrote:
> We as a country and even state of Oregon SUCK. There are only 465 registered 
> users in Oregon and 33 in Eugene. If you're not registered and you do use 
> linux regularly then step up to the plate! The United States is one user away 
> from losing 50th place to UK. Do it for EUGLUG, do it for Oregon, do it for 
> the USA! (insert patriotic music here)
>
> http://counter.li.org/reports/place.php?place=USe

Self-selecting web polls don't make real world statistics, Mr. O.
135,395 total Linux users globally? The number of global Linux users
peaked in 2002, dropped radically the next year, and isn't yet even
close to being back to 2002 levels?

That's how Joe Math-Dropout will read that graph; that's what it says
the statistics are on that page: "Linux users in The World" right on
top of that graph.  You have to switch to the Project page to learn
that it really isn't statistics about how many Linux users are in each
country and in sum, that it's statistics on the annual number of folks
who take the time to fill out a registration form on that site.  Even
on that page, they don't caveat what was said on the page you linked.

How many people will go looking on the site for what the data
represents? Assume Joe Sixpack has read that there are 450 million
Windows users in the world but in curiosity found that site looking
for how many Linux users there are globally; what does that page tell
Joe?

IMHO, that's a web page that needs to be taken down or be rewritten,
not publicized. There are other pages on the site that need to be more
frank about what the statistics are derived from. As is, the site
helps Microsoft more than Linux because more people are ignorant of
more relevant statistics and the weakness of the data. Not everyone
understands enough about statistics to know that self-selecting polls
rarely have much to do with reality.

Graphs of web poll results with misleading titles suggesting to the
uniformed that they reflect hard real-world and relevant data? Not
something I care to contribute to. Page should be retitled, "How many
people who claim to be Linux users and have registered on this web
site." Or, "International Linux user registration contest," "Find
other registered Linux users in your nation and globally," something
that paints a more accurate picture of what the statistics represent.

As a nerd myself, I know we all have the problem of often assuming
things are obvious to everyone that are obvious to us. It's human
nature. Being a nerd of the legal variety whose clients won or lost in
large part because of my success or failure in explaining the
technical to the uniformed, perhaps I'm more sensitized to that human
failing than you?

I'm not suggesting in the least that you caught that the page you
linked to is misleading for the uninformed. There's no doubt in my
mind on where you stand on Linux in relevant regard. You're among the
refreshingly frank and blunt people of the world. Haven't met you
face-to-face but I already like you from what you've posted. No
hostility toward you here. Just questioning whether or not that link
should be pushed before changes are made on the linked site.

In fellowship,

Paul

-- 
Universal Interoperability Council
<http:www.universal-interop-council.org>
_______________________________________________
EUGLUG mailing list
[email protected]
http://www.euglug.org/mailman/listinfo/euglug

Reply via email to