[Foundation-l] Benefits of advertising (was Re: Our exponentially increasing costs)

Tim Starling t.starling at physics.unimelb.edu.au
Mon Oct 24 13:15:53 UTC 2005


Neil Harris wrote:
> Hmm. 2500 hits/sec * 86400 secs/day * $1 CPM = $216,000 / _day_, or
> $78,000,000+/year. Have you considered that the Wikimedia foundation
> board might be aware of this, and that its decision not to put up
> advertising might be a principled decision, rather than motivated by
> "fear of money"?

We don't have 2500 hits/sec, we have 2500 requests/sec, i.e. including
images, stylesheets, etc. The difference is roughly a factor of 3. The
income would thus be closer to $26M.

By these figures, we could cover our current operating costs by putting
ads on the site for two weeks per year. I'm not sure if it's a good idea
though.

Like Jimbo, I think distribution of Wikipedia around the world,
especially offline distribution to poorer countries, should be our top
priority. But Wikimedia lacks the infrastructure for such hands-on work.
It also lacks the moral high ground in the eyes of funding agencies,
because its focus to date has been to provide a service for relatively
wealthy Internet users. Our donors give money out of reciprocity, not
out of compassion for those less well-off.

There are other organisations who have been distributing information and
technology to developing countries for many years, and these
organisations are often well-funded and have a well-established HR
infrastructure. For my part, I've been working with WiderNet, which is
entirely capable of delivering Wikipedia to African universities in the
next 6 months, without funding from us. Andy Rabagliati periodically
updates installations of Wikipedia in secondary schools in South Africa.
All we have to do is provide a free encyclopedia for these people to
distribute, and they will distribute it.

It's true that they could probably distribute it to more places in less
time if we gave them millions of dollars, and we could probably raise
that sort of money by advertising. But we would have to make it *very*
clear to readers and editors that the ads are to support such activities.

It's possible to display advertising and keep the moral high ground,
thehungersite.com is a good example of that. But the connection between
the advertising and the charitable goal has to be eminently clear.
Editors who are thinking of leaving in a huff have to know that it is
the poor children in Africa who will suffer if they do. The Board would
have to have a well-documented plan for the immediate distribution of
the funds that are collected, and that plan would need to be advertised
relentlessly.

The Answers.com deal, on the other hand, is just a recipe for offence.
The benefit is unknown and will probably be small. The Board has agreed
to it in secret, apparently to repay favours such as the Wikimania
sponsorship. They've given a single organisation preferential treatment,
which brings their motives into question: are they doing this to help
Wikipedia or Answers.com?

That's why I join the chorus of editors in asking the Board to withdraw
from this deal. If we are going to have advertising, it needs to be done
with extreme sensitivity to the sensibilities of the community. Claiming
that it's not advertising most certainly doesn't help -- even if the
Board believes that line, the community certainly doesn't.

I don't believe that we need to have advertising to keep the site
running. We can still go a long way by pressing our readers to donate
more money, and by making better use of strings-free hardware donation
offers such as the Google deal. Some people have expressed reservations
against "begging", but it's generally seen as a more acceptable activity
than advertising.

If the Board wants to perform "special projects" such as offline
distribution, these projects first have to be rigorously justified to
the community. Supporting them by advertising can only be considered
after the community has been convinced of their benefit.

-- Tim Starling




More information about the foundation-l mailing list