Is advertising on the Web dead? Back in March there was no end to the good news about the Internet. We were all living in a bubble and loving it. The Web was touted (correctly) as the fastest-growing advertising medium in history; it had achieved penetration into more than half of American households in less than seven years -- amazing.

And the prospect of billions of Web advertising dollars fueled countless business plans: each document showing year-to-year increases that didn't stop at geometric -- they went right into exponential.

And the venture capitalists were buying it! They were putting bundles of money into companies whose main (only?) source of revenue was based on dizzyingly steep ad sales curves.

Then, just as spring flowers began to blossom, the herd mentality that is Wall Street suddenly flipped: there was no end to the bad news. The emperor has no clothes! Web stocks dropped faster than a Russian SST. The alchemy of Web investments went into reverse as gold -- AOL, Yahoo!, CMGI -- suddenly turned to lead.

But on what information was this change based? Depending on which trade source one refers to, the range of estimates of Web ad spending is breathtakingly wide: from a low of about $3 billion in 2000 to (hold on to your mouse) a high of $14 billion.

It's no surprise that people are having difficulty making rational decisions from this data.

Not to minimize the problem: estimating Web ad spending is exceptionally thorny. It's no wonder that, according to one industry source, current Web ad spending estimates are off by "30 to 100% -- and not in any consistent direction." Which makes them useless, practically speaking.

Why? Consider just these few complications when trying to estimate Web ad spending:

* In a buyer's market, rate cards mean little. Nearly every seller offers deals.

* Sponsorships have no rate cards at all. Every deal is negotiated independently.

* Bonus advertising weight runs all over the place.

* Many sites do non-cash trade deals with other sites, creating the appearance of activity.

* Many print publishers offer 'value added' packages that provide advertisers with free Web ads as part of magazine buys.

Of course, I wouldn't have set up this straw horse without offering the hope of a useful insight. Let's look at something unusual for the Web: incontrovertible, hard data. Weekly information on how many advertisers have appeared on the Web for the first time over the past year and a half.

One could, and should, make the assumption that a medium's health is proportional to how effectively it attracts new advertisers. If the medium works, new advertisers will follow quickly. And if it doesn't work, they'll run away like kindergarteners from a broccoli lunch.

So here's the good news: the trend of new brands coming to the Web has been slowly, inexorably, relentlessly, incontrovertibly positive. An average of about 200 new brands per we

This news story is important. I recommend that everyone read it.