This is the first of probably two columns about the crisis in advertising. Maybe you didn't know there is an advertising crisis, but there is. We are suffering right now not just a recession, but an advertising DEPRESSION, and that depression is based solely on the growing realization of major advertisers that what used to work isn't working any more. And while this may not matter to you or me, it eventually will come to matter as our activities that are financed by advertising are forced to change or to go away completely. Magazines and newspapers, television, and the Internet are being greatly affected, with the result that each is likely to get more expensive IF they survive at all. Today's column is just about the Internet part. Next week, I'll consider television.

Looking in the I, Cringely 489,000-word archive of grotty old columns (under the "Old Hat" button on this page), I see that I have written twice before about Internet advertising. In 1997, I wrote that media buyers were keeping Internet advertising from becoming a big success because it scared them and made their lives too complicated. In 1999, I wrote that Internet advertising didn't work at all, at least not in any traditional advertising sense, and if advertisers thought it did, well, they were going to be disappointed. Both times I was right, but what we are facing today is something different, a realization not just that banner ads don't work, but that what does work might kill us.

Internet advertising has changed dramatically over the time I have been writing this column. One obvious change that longtime readers will notice is that this column used to have a banner ad at the top for something other than a PBS show. We used to sell ads, but now as far as I know, we just give them away. It was costing us as much to sell the ads as we were making from them, so why bother? Frankly, this pretty much encompasses the banner ad experience for advertisers, which is why they are so much cheaper to buy today than they used to be, and why they have been supplanted by so many other methods of Internet advertising.

In terms of display ads, the banner has been supplanted to a certain extent by pop-over and then pop-under ads with the idea that bigger is better and maybe the element of surprise will help. It hasn't. The idea of using the Internet for a billboard is fading fast.

Most advertising revenue on the Internet today comes in three areas. First are banner ads as described above, then come the search engines, which primarily make revenue through paid listings. There is a lot of bang for the buck in these because the ads respond to surfers who are actively looking for something specific that can be easily identified from a search query, and just might be close to what an advertiser might actually have for sale. Searchers are generally not put off by such paid listings if they are labeled as such, rather than being sneaked onto the page in the guise of listings that deserve to be there. Google was a pioneer in this novel technique of actually identifying paid listings as paid listings. But there are very real revenue limits for such ads simply because their volume is relatively low. Google has 150 million page views per day, which is a lot, but certainly doesn't translate into the kind of revenue that can support anything more than a search engine. That's why Google is Google and not Yahoo. What Yahoo is trying to do, which is to supplement searching with original content, is much harder to do profitably. That's why Yahoo suddenly wants us to pay — because they have figured out that advertisers won't pay, at least not enough to support the kind of earnings growth demanded by Wall Street.

The third form of Major Internet advertising is e-mail. This encompasses both spam — unsolicited e-mail offers for second mortgages and Viagra — and more traditional e-campaigns for charities, political causes, and offers from people we've bought from before for things we actually might want to purchase. It's not that we don't want to buy Viagra and second mortgages, too, but I certainly don't want to buy those things 15 or more times per day. Do you?

E-mail advertising is incredibly successful because it is cheap and effective. It is incredibly effective for outfits that use the Internet to promote social change, which isn't what you'd normally think Internet advertising would be used for. Organizations like MoveOn.org have e-mail petitions and e-mail fundraising campaigns that are based on lists generated by friends telling friends to sign this or support that. Because such campaigns are based on what respondents and potential respondents know about each other, they are pointed and very effective. MoveOn can get 750,000 signatures on an electronic anti-war petition within hours, and raise hundreds of thousands of dollars within days to pay for non-Internet advertising.

This kind of advertising really shows the potential of the Internet more than any other. Think of the impact it has in states like California, where I live, that are great proponents of referendum elections. The way citizens get a referendum on the ballot is by circulating petitions and getting thousands or tens of thousands of signatures showing voter support. This has always been a high barrier, but not in the new world of e-mail advertising, where people opposed to excessively shedding cats could get enough electronic signatures to launch an anti-feline referendum in less than a day. That's political power.

Spam, however, has a different kind of power — the power to annoy. Spam is unintelligent e-mail advertising that goes to 1,000 people in hopes that one will respond. While automation makes it cost-effective to do this, spam is neither culturally nor socially effective. Spam has become so pervasive because it works. If it didn't work, people wouldn't do it. If other forms of Internet advertising were equally effective, spam wouldn't be so popular. So spam proves by its own success that most other forms of Internet advertising are ineffective.

But because of its success combined with its repugnance, spam is changing the very culture of the Internet with sorry results. Spam's success has led to anti-spam software and services. These simply aren't effective. In fact, spam is worse than ineffective, it is destructive. Those of you who subscribe to the woefully inefficient "Tell me when" notification service for this column may have noticed that some anti-spam software rejects these notices that you have asked to receive. This is apparently because of the line of asterisks used in the body of the message, which are apparently a common visual characteristic of spam. Along these same lines, entire companies have found that their employee e-mail is bounced back by thousands, even millions, of recipients as being spam, whether it actually is or not.

At the same time that the sheer volume of spam is clogging networks and demanding server upgrades to handle all the sludge, crude anti-spam techniques are making our Internet communications less effective and less efficient. We are both drowning in spam and trying to stop it by hitting ourselves in the head. But wait, there's more! Politicians, who think their job is to make laws that tell us when to breathe, now want to make spam illegal, which might be nice if it didn't at the same time trample on the U.S. Constitution. One minute spam will be illegal and the next minute it will be against the law to say vice president Dick Cheney really ought to lose a little weight. Killing spam isn't worth the loss of free speech, at least not to me.

E-mail is the Internet killer app. Don't let anyone tell you any different. Web surfing is nice and downloading music and video files is great, but e-mail has more effect on the world than those other things will ever have. But spam is killing e-mail. It is killing our killer app. And through knee-jerk anti-spam technologies, we are helping to kill e-mail. This is not good.

If spam is the future of Internet advertising — and I think it is — and if spam and anti-spam alike are killing the Internet, then there is only one logical response. It's not what you'd expect and certainly not what I am hearing suggested anywhere else. Spam can't be stopped without crippling the Internet or destroying the Constitution, so we just have to make it better. We have to find a way of making spam an even more effective sales tool.

Huh?

It is a simple idea. Right now spammers waste 99.9 percent of their effort contacting the wrong people. They don't particularly want to do that, but at this point they simply don't care since the cost of such inefficiency is so low. But they also want to remain in business. Here is the key: If they were contacting only the right people, it wouldn't be called spam. So we — spammers and the spammed alike — have a common interest in finding a way to efficiently target only the people who really want to buy Viagra. I believe it can be done.

Any ideas?