The trouble is, PageRank only worked within a small dataset of peer reviewed academic journals. To extrapolate this into a way of life, as Google’s dreamy maths-obsessed boy wonders tried to do, was an essentially utopian gesture, which supposed that no one would try and game the system to their own nefarious ends. Only the inevitable happened, and as Google got more popular, and as the value of appearing in those top spots increased, Google gradually lost control of the algorithm which was once its muse. At the time, we remember, we gained very few plaudits for documenting this weary process - as Google was gradually gamed by desperate trinket salesmen, who built link farms to tout their wares - and by technology evangelists, who mistook overnight popularity for a validation of a lifetimes’s achievement. All were to fall to earth eventually, as technology offers no short cuts or backdoors when the calculations are finally made.

So here we are. Today, Microsoft doesn’t understand its source code. But nor can Google identify how much of its search index is comprised of robot-generated junk, designed to trick its PageRank™ algorithm. With billions of pages of "content" - pages of junk can be created on demand to populate cheap, disposable "web presences" - it’s beyond the wit of any algorithm to determine what’s real and what’s simulacra.

Like Microsoft, Google has simply been outsmarted. To read the popular press and discover that they’re arming for a billion dollar fight is like watching two dunken prize fighters hoping they’ll land a punch.

But thanks to blogger Mark McGuire for providing another dimension, one we and everyone else missed.

Just as Microsoft doesn’t have to care about the quality of its software, nor does Google (or Yahoo!, or any other want to be web destination) have to care about the quality of its product. Up to a point.

Noting the Big Daddy fiasco - Google’s attempt to weed out the spam from its search index - McGuire notes that Google profits from the irrelevance. Google makes next to no money from "search", but makes all of its money from selling advertising.

Mark notes, as we do, a webmaster’s comment that the deterioration is gradually turning the SERPS [Search Results] back into a primordial soup:

“At this rate, in a year the SERPS will be nothing but Amazon affiliates, ebay auctions, and Wiki clones. Those sites don’t seem to be affected one bit by the supplemental hell, 301’s, and now deindexing.”

Mark observes:

"Google may take some action here and there, but I believe that they actually like a little mud in the main organic results for commercial terms. Why? Because less than stellar organic results (from practices like web spam) mean higher CTR’s on their paid links and more juice for their quarterly earnings.

"A little irrelevance is good for paid links and paid links is how Google makes money."

If you want to be seen in Google, you have to pay to play - and create an Adwords account. How else are you going to be "seen"? In other words, it’s pay to play - the old economy reasserting itself with a vengeance.

This is a fascinating subject - how far can you con the public without being rumbled? Google executives may well look at Microsoft’s history and conclude they can ride the goodwill train for many years. A cynic may say the public doesn’t really care about the mechancisms, so long as they’re being delivered real results.

But the public is increasingly sophisticated, and as the web spammers have proved (the "Big Daddy" fiasco being the primary evidence) more than capable of outwitting Google.