One buck a share. That’s how close it may have been.

When I spoke on Sunday to Legg Mason’s Bill Miller (Yahoo’s second-biggest shareholder, with 6 per cent or so of the stock,) he wouldn’t be drawn on whether he personally would have taken the $33 a share on offer if Microsoft had taken its bid direct to shareholders. But he did say this:

“Something in the $34 or $35 range would have satisfied most shareholders.”

Yahoo’s board held out for $37, while Yang himself apparently told Steve Ballmer on Saturday that he and co-founder David Filo would be happier with $38.

Miller chose to blame the Microsoft side for letting the deal get away rather than paying the extra $5bn to come up to Yang’s price. But if he is right about how little extra it would have taken to get a deal done, that is as much an indictment of Yang’s negotiating stance as it is of Microsoft’s unwillingness to pay the extra buck. It only makes the failure of these negotiations seem even more bizarre.

What should we make of the odd limbo into which Microsoft’s pursuit of Yahoo has fallen since the weekend, when it failed to follow through on a threat to go hostile? Here are the possible answers:

1. The two sides are just buying time while they stitch up a deal behind the scenes. Probability: miniscule. We’re still hearing that there haven’t been any discussions since early last week. When the bankers from both sides put their heads together at that point there was still no serious talk about price (one source suggests nothing has changed since day one, with Yahoo still holding out for $40 a share and Microsoft unwilling to budge from its opening offer, which is now worth more like $29.)

2. Microsoft is getting ready to walk away. Probability: a bit higher than miniscule, but not much. Proponents of the walk-away scenario hold that Microsoft has grown disgruntled at being rebuffed and that dissent in its own ranks, where a Yahoo deal is viewed with dread, has dulled its appetite. This sounds too trite. The move on Yahoo is a huge break with Microsoft’s past: the enormity of it points to how vital it is to control one of the Web’s leading advertising platforms (this could eventually become an important monetisation engine for much of its core software business as it moves online.) This takeover approach was not mounted on a whim. Whatever Steve Ballmer’s other character flaws, a lack of resolve is not one of them.

3. Microsoft is still trying to decide how tight it can turn the thumbscrews on Yahoo without incurring a backlash. Probability: Pretty high. The software company has shown an uncharacteristic restraint all along. There have clearly been people in the Microsoft camp who would have gone in all guns blazing from the start, but that is not how Ballmer has chosen to play it. He has probably calculated that there is a better chance of reeling Yahoo in faster by not forcing it to throw up the defences against a hostile bid. Ballmer was on the road in Europe last week, arriving back only over the weekend - after the Saturday deadline he had set three weeks before had already passed - so it makes sense to deliberate thoroughly on this before taking what could be a critical step. The next turn of the thumbscrew seems imminent.

4. (My favourite explanation.) Jerry Yang just had a baby. Maybe Microsoft just decided to take the heat off for a couple of days. Wouldn’t it be nice to believe that was true?

Has Red Hat scrapped its plans to take on Microsoft with a Linux desktop machine for the emerging world?

Unfortunately it’s not easy to answer that question. A Reuters report from India suggests the answer is yes. But my own questions to the company over the past 24 hours have brought nothing in the way of enlightenment, and Red Hat has now instead churned out this verbose and obtuse posting on its company “blog” (if companies keep using their blogs for this kind of corporate-speak we’ll even look back on good old press releases with nostalgia.)

To save you the pain of reading the whole thing, these are the key sentences:

We originally hoped to deliver RHGD [the “global desktop” product] within a few months, and indeed the technology side of the product is complete. There have, however, been a number of business issues that have conspired to delay the product for almost a year. These include hardware and market changes, startup delays with resellers, getting the design and delivery of appropriate services nailed down and, unsurprisingly, some multimedia codec licensing knotholes.

“Hardware and market changes”? That’s pretty broad. The only thing to emerge from all of this is that we will probably have to wait at least another year to find out if Red Hat is still interested.

Another Semantic Web company looking for cash: William Tunstall-Pedoe of True Knowledge says he needs $10m in venture capital to back the next stage of his Cambridge (UK)-based company, which is trying to build a sort of “universal database” on the Web.

Tunstall-Pedoe’s plan is to collect information in a structured way in an online knowledge base, making it much like MetaWeb’s Freebase (which has raised $42m.) Users could then query that using natural language questions, and an open API would make the information machine-readable.

Like Nova Spivack of Radar Networks ($20m in venture capital), Tunstall-Pedoe is a computer scientist who has been at this a long time. While on a visit to Silicon Valley this week, he told me that he has worked on the technology for ten years. Also like Spivack, Barney Pell at Powerset ($12.5m) and others in the field, he says that while it will take years more for the technology to reach its full potential, a usable consumer version could come relatively soon (in True Knowledge’s case, about a year.)

The danger for all these companies is that they stretch themselves to build a consumer service too soon, wasting cash and disappointing users in the process. The technologies of the Semantic Web may well eventually yield the next breakthroughs for organising and finding information online, but the early pioneers should be wary of getting swept up in the venture capitalists’ latest rush to find the Next Big Thing (Our own fuller analysis of the movement is here.)

The clicking habits of Google’s US users continue to cause concern.

For the third month in a row, the number of “paid clicks” (or impressions on adverts) in the US was little changed in March from a year ago. That is according to data from comScore that started to seep out late on Tuesday (the information is given first to analysts before being made public officially later.)

Overall, for the first quarter, Google’s paid clicks rose only 2 per cent, despite a big jump in the number of searches in the same period.

How much of this is due to Google improving the quality of its market by weeding out low-value or unintentional clicks (which could lead to an off-setting jump in revenue per click,) and how much does it reflect a cyclical downturn? John Aiken at Majestic Research says he things it’s 75 per cent cyclical (and at least one Google executive has hinted that economic weakness is an issue.) If so, that could hit the entire internet sector. The results will be out on Thursday.

That is the new mantra for a band of internet operators which until now has lived in a twilight world where the value of domain names can rise and fall as fast as a dotcom stock price, and where mastering the latest tricks for syphoning traffic from Google can make the difference between fortune and bankruptcy.

Communicate.com (now renamed Live Current Media) is a case in point. The various businesses this company has been through, as described to me by new boss Geoff Hampson, read like an opportunists’ history of the internet:

Domain name trading. Communicate.com was one of those to see the value of registering popular domain names, and later made money selling off names like Rugby.com and Beijing.com.

Search engine arbitrage. This involved acquiring traffic by advertising on search engines, then trying to turn a profit by filling its sites with ads from Yahoo that would draw a higher price per click. Many of the companies that tried that tactic have been put out of business: the search engines cracked down on the practice and started to block “landing pages” that contain only adverts, rather than the real content users are looking for (this story in the Financial Post about how Canadian company Geosign blew $160m on the strategy is instructive.)

Ecommerce partnership. The next step was to lease some of its well-known “geo” names (which include Brazil.com) to Frequent Traveller, which would turn them into ecommerce travel sites. That failed, says Hampson: “They were trying to do customised travel deals at discount prices.”

Building a bona fide “media” company. Hampson, who took control last year, now says he’s out to create a real media business around the most popular domain names, like Cricket.com.

It’s no surprise that other, bigger domain name companies - like Demand Media, Marchex and Internet REIT - all say they are heading in the same direction. Unless they can fill their sites with more useful content, they risk losing traffic from the search engines (notwithstanding the dark arts of search engine optimisation, which they all use to stay as high in the “natural” search rankings as possible.)

Don’t expect too much from these new “media” sites, though. Not surprisingly, there is much talk of trying to harness user generated content and social networking, spiced up with a sprinkling of “expert” commentary. How many compelling sites will come from this ultra-low cost approach to domain name development?

Vertical search, social search and now… identity search?

The attempts to outflank Google continue. This time it’s Barry Diller’s IAC, which has already failed to make much of a dent in the three years it has owned Ask.com. Diller’s latest gamble: that a search engine aimed at a particular demographic has a better chance of success.

Hence Rushmore Drive, which launches today - a site which wants to become the Google of the US’s 50m-strong black population (African-American is too narrow a label, says Johnny Taylor, the IAC executive in charge.) Taylor says the plan is to deliver a full Web search, but tilted towards the particular interests and outlook of its target audience (the patented algorithm that does the tilting could be used to build search engines for other broad demographics, he adds.)

So is this a better way to attack the search market? Logic suggests that if you can create a differentiated product and brand and market it well, there’s no reason you shouldn’t be able to carve out a slice of a market, even against a dominant company like Google.

Distribution will be one challenge. Inertia is a powerful force. How many internet users take the trouble to reset the default search engines they are first presented with?

The acid test, though, will be whether Rushmore Drive (named after the search engine’s street address) really can build and brand a “better” product for its target audience. A former senior search executive I spoke to this week put it this way: “The basics of search aren’t such a big deal - but there are many, many tweaks.” This process of “tuning” a search engine is a difficult art. There is also the not inconsiderable challenge of delivering results in 0.3 seconds (that’s how long it took for Google to come back with results to the query for “Rushmore Drive”.)

But unlike Ask.com, which tried to compete with Google head-on, at least Rushmore Drive has a different story to tell.

Well, soon you can. The announcement last night of Google’s App Engine (currently in closed trial) is symbolically important: until now Google has been a vertically integrated company, building one of the world’s most powerful computing networks in order to deliver its own search and other services. Like Amazon before it, Google now says it will make that massive computing resource available to other businesses that want to run their software on its servers and access it as a service over the internet.

Once again, Google has beaten Microsoft to the punch (though to be fair, this is still only a very low-key trial, and what’s being played here is a very long-term game.) Microsoft has promised to reveal more about its own “cloud computing” platform this year, with much expected at an October developers’ conference in Los Angeles.

This is shaping up to be yet another fascinating face-off between the leading internet and software companies. Microsoft is a platform company in its DNA and has a massive following among developers, but it has less experience of delivering high-volume services over the internet. Google has unparalleled brains and brawn in running internet services, but it is still at the early stages in winning over developers (a sign of its growing influence will come at a big developer conference in San Francisco next month.)

Ultimately, much will come down to trust. The mantra of IT professionals used to be, noone ever got fired for buying from IBM. Maybe, one day, Google or Microsoft will step into those shoes.

Microsoft and Yahoo continue to circle each other like wrestlers, each waiting for the other to show its weakness before getting into a clinch. At a time like this, just about anything either side says is pure rhetoric.

That is the best interpretation of today’s report that Microsoft is ”evaluating” its bid in the light of deteriorating market conditions. This is a thinly-disguised version of the Larry Ellison treatment: soften up the target by cutting the value of a takeover offer (something Ellison did while Oracle stalked PeopleSoft) or at least threatening to (his treatment of BEA Systems.)

Like Reuters, I also talked to “people close to Microsoft” today, but I came away with the distinct feeling that not much has changed. No matter. With the arbs starting to sweat (most were probably thinking they’d have banked a nice profit on this deal by now) Yahoo’s stock has slipped 3 per cent.

Only two things are certain. One is that, whatever the rhetoric, Larry Ellison eventually ended up raising his offers to win PeopleSoft and BEA. The other is that the real negotiation between Microsoft and Yahoo will be conducted behind closed doors, not through these public posturings.

Not got around to upgrading to Windows Vista yet? In that case you might want to delay a bit longer: it looks like the next version, Windows 7, is coming sooner than we thought.

That reaction is exactly what Microsoft is afraid of - and why it has been scrambling today to stamp on the suggestion that unlike Vista, which arrived years late, 7 (codenamed Blackcomb, after a Canadian ski resort) may actually arrive early.

Unforturnately the speculation will not be easy to kill. Bill Gates himself let the cat out of the bag by declaring he was “super-excited” about the next Windows in “the next year or so”. Yesterday, Microsoft said it planned to offer a version of Windows XP until June 2010 or a year after the launch of Windows 7, “whichever date is later” - an apparent indication that a date of June next year had been pencilled in for the launch of the new operating system.

A Microsoft spokesperson says nothing has changed, Windows 7 is still expected to take around three years from the consumer launch of Vista (January 2007) and maybe Gates was referring to the first beta version of the software.

Regardless of this attempt at obfuscation, what seems to be emerging is this: rather than slipping later into 2010, as some had expected, the next Vista is well on track and at least an advanced version of it will be on display in the first half of next year. That won’t help Vista, which is still recovering from all the bad publicity of its first year.