Re: Interesting Twist to BANNER ADVERTISING

From: Matt Magri <matt_at_netmeg.net>
Date: Wed 29 Dec 1999 15:12:12 -0500

FRED WROTE:
> development we've come upon another interesting bit of
> coding genious that we just had to jump on right away.
> I'm posting it here to see if anyone on this list
> thinks it's interesting or worth the trouble to code.
 
TO WHICH WILLIAM HENNING REPLIED:
> > I think it would cause you far more difficulty than
> > any possible gain. Why? Read below...

The problem is, in reality it won't cause him the least
bit of difficultly. If implemented properly, an
ad-replacing proxy would be indistinguishable from
normal, pass-through web proxy (like a webcache). Your
only chance would be someone reporting them (or maybe
seeing something on their website that indicates they
are doing it). Since it will be hard to detect and easy
to implement, arguing against people adopting it is not
a real winning strategy, if history is any gauge. Lots
of folks won't even see your arguments, for one thing.
Of those that do, a percentage will only see dollar
signs. You would be better off adjusting to either make
it harder to do or to take advantage of the technology.
Or both.

> For the past three years I've been saying the whole
> advertising landscape on the web would change
> dramatically. At some point people will become more
> interested in local ads than remote ones.
 
> > I happen to disagree with you; the ideal way of
> > attracting local advertising dollars is to build a
> > local content site people will want to visit.

That's certainly the current way. Is it really the
"ideal", however? Wouldn't it be 'more ideal' to have a
model like the cable channels use to let cablecos sell
local ads? The thing is, there would have to be a way
for the local provider to pass some of that ad revenue
back to the site or ad network that is allowing the
exchange. It may not be possible to come up with such a
model, but now is the time to make sure that's the
case. If you can come up with a win-win alternative to
ad replacement that will be a more compelling argument
to providers who would be tempted by ad replacement.
Actually, last month would have been a better time to
have had it in place...

> Local advertisers can't buy ads on the big web sites,
 
> > Not true; geo-targeting is available;

Based on what? The domain name they come from? That's
not targeting (assuming targeting implies some
accuracy, anyway).

> > and if they can pay for expensive TV and print ads
> > they should not balk at paying for geo-targeted
> > advertising from the reputable agencies.

He's going to be able to sell to folks who would balk
at paying those rates, whether they "should" in your
opinion or not. And, since he isn't paying you squat
his overhead is ridiculously low.

> > You could always do what AllAdvantage does and give
> > your dialup users a special browser that displays an
> > additional ad outside of the browser area; as long as
> > no attempt is made to make it look like part of the
> > site being displayed you would probably be OK.

"Probably"???

> So I said: "How about we come up with code that TRAPS
> the banner ads on ANY web page being surfed by anyone
> on our private (dial-up subscriber) WAN, and replace
> it with an ad for a local business?" They liked the
> idea, so they built the code.
 
> > Well, there are a number of things wrong with this
> > idea; allow me to list them:
> >
> > 1) Copyright violation
> >
> > The ad code embedded in the web pages is part of the
> > copyrighted work, by replacing it you are violating the
> > copyright of the web site owner.

This is interesting. I don't think it's going to be as
cut and dried as you make it out to be, but it's a
reasonable approach to try in blocking sites from doing
it.

> > 2) Theft from websites
> >
> > Such a scheme would be usurping revenue from ad
> > supported web sites; replacing ads will result in a
> > drop in CTR rates for the web sites; which will cause
> > them to lose advertisers; therefore lose revenue.

Very true. The ad replacing provider is pocketing money
for views of content they didn't produce. If they don't
kick some reasonable recompense back to the content
provider it turns the current incentive system on it's
head. What a mess.

> > 3) Immoral & Unethical
> >
> > See (1) and (2). This would piggyback on top of the
> > hard work of other people to make a quick buck.

Yep.

> > 4) Self defeating
> >
> > Enough people do this, most content sites will shut
> > down as they won't have the revenue stream they need to
> > build content. Their ad space could no longer be
> > stolen, and the ad space abductors will no longer make
> > money. Fewer interesting web sites; people will surf
> > less. ISP's lose out. People lose out. Everyone loses.

True, but this gets into the
externalizing/internalizing aspect of economics. An
individual provider is unlikely to conclude that their
particular proxy will make that much difference.
[Repeat that calculation for enough providers to make a
difference.]

> For instance: say any of the 60,000 dial up customers
> log on and surf to any site with BANNER EXCHANGE, VALUE
> CLICK or other banners purveyors on the target page...
> rather than seeing some ad for some web site they're
> not interested in, they'll be greeted ads for business
> right in their neighborhood. Local business they'd
> rather support. (Think about how many you'd rack up
> with Yahoo alone!)
 
> > Bad idea, illegal, unethical, anyone doing this would
> > get sued by the ad networks, and a class action suit on
> > behalf of web sites would also be a strong possibility.

That's just not realistic. For one thing, they'll never
be able to detect everyone who will do it. For another,
the provider would have to be huge to make it worth the
effort of going after them legally. A better option
would be for ad networks to maintain a list of
providers that participants would have to block www
access from. That would provide them with some leverage
that would make removing the proxy worth the
dollar-sign-blinded provider's while. If they could
offer the carrot of an ad-network-blessed proxy that
the local provider could use to sell ads (and
recompense the ad network), that would be even better.
You would want to have one unified proxy, however,
rather than one for each ad network.

Anyway you'd still be faced with the problem of
detecting the sites that were doing it.

> So I told my head tech guy, "WHOA... what if the Post
> Office captures every incoming issue of NEWSWEEK and
> replaces all the ads with local ads before delivery to
> the subscribers' mailboxes!"

> > Newsweek would go after the post office. There are
> > actually federal laws dealing with tampering with mail
> > that may also be applicable with such web site
> > tampering.

What if it's done with the consent of the
end-recipient? I can also say (based on my experience
as a juror... really! ;-) ) that the USPS does tamper
with mail that they decide has illegal substances in
it. Anyway, laws tampering with USPS mail don't apply
to web proxies. In case that isn't obvious, just look
at all of the web caches and ad-removing web proxies
out there that are tampering already. There are still
folks who argue that they should be able to go after
web caches for it, but I don't see any class action
suits against AOL materializing yet.

> > In case you haven't guessed by now, I run a fairly
> > large content site. The site pays my mortage payments
> > and puts food on my table - via ad revenues. If I
> > found someone violating the copyright of my content
> > (by copying it OR modifying my ad codes to divert ad
> > revenue from my site) I would have to explore all
> > of my options in seeking redress.

The problem is that you aren't going to find it. You
can try to work around it by finding ways to break
their proxy's detection mechanism, but that may make
your site crufty for everyone else. Ad networks and big
content providers should be thinking about organizing
to produce the stick of a unified blocklist (ala the
MAPS RBL, but for port 80 instead of port 25), and the
carrot of a unified system which allows a local
provider to sell local ads ala the cableco model. If
such a carrot could be devised, you'd be adding the
ISP's sales folks to the list of people who are selling
ads for you (and, one hopes, converting a disaster into
an opportunity).

Matt
--
Matt Magri
Netmeg Internet





Received on Wed Dec 29 1999 - 14:12:12 CST