Even as Carl Icahn rallies angry shareholders to try to force Yahoo back to the bargaining table with Microsoft, one of the “strategic alternatives” Yahoo may still be trying to work out in the background is a search advertising deal with Google. There is a 60 to 70 percent gap between what Google collects for search ads and what Yahoo collects, so simply handing over a portion of its search advertising inventory to Google would boost its cash flow and profits considerably—perhaps adding as much as $1 billion or more in cash flow.

But how could such a deal pass muster with antitrust authorities, who are already investigating the test run Google and Yahoo did last month with only 3 percent of Yahoo’s search ads?

It would all depend on how a deal is structured.

One line of thinking is that Yahoo and Google could get away with a deal that only hands over 10 to 20 percent of Yahoo’s search advertising inventory. This would need to be on a non-exclusive basis, meaning that if somebody else could come in and beat Google’s revenue-per-search-query Yahoo would be free to hand them the ad inventory instead. The assumption was that this would be the 10 to 20 percent of keywords that bring in the highest revenues for Yahoo. (We discussed this point in our interview with Citi analyst Mark Mahaney last month, for instance).

But there is another way Yahoo could get a lot more bang for its buck in a deal with Google. Instead of handing over the most valuable search terms, it would be better off handing over the ones with the biggest delta in profitability (the difference between what Google makes on those terms and what Yahoo makes). Yahoo does not have any trouble getting decent ad rates for the most desirable search terms. Call those the head keywords that bring in the most revenues. What it has trouble making money on are the keywords in the long tail and torso of its advertising inventory. And that’s exactly where Google excels at squeezing out relevant matches and clickthroughs.

If Yahoo can identify which basket of search terms represents the biggest profitability gap compared to what Google makes, it can maximize what part of its ad inventory to outsource to Google. These terms will likely turn out to be the ones that are currently the least valuable ones to Yahoo. Picking the 10 to 20 percent of keywords where the delta is the greatest between what Yahoo and Google are able to charge would effectively multiply the impact of the deal. After all, there is no point in handing over high-revenue search terms that Yahoo is already matching Google on in terms of profitability.

If the numbers work out and antitrust can be avoided, such a deal would certainly be a way to appease (or at least answer) Yahoo’s increasingly irate shareholders. But if Yahoo is serious about striking a deal with Google, it should do so before the proxy battle with Icahn comes to a head.

(Photo credit:Jack Versloot).

Crunch Network: CrunchGear drool over the sexiest new gadgets and hardware.

Celebrities are starting to take notice of Seesmic, a “Twitter for video” service that lets people have asynchronous video conversations on the fly (see my disclosure, I am an investor).

First was Deepak Chopra, who made a whole series of videos for this site. And yesterday things got even more exciting, when Steven Spielberg, Harisson Ford, George Lucas, Shia Laboeuf, Karen Allen and Cate Blanchett came on the site and had discussions with other users. Here’s one of the exchanges, between Jemima Kiss and Steven Spielberg. Here’s a Harisson Ford video. etc.

So that’s all really great, and I’m happy as an investor. But Seesmic made some terrible judgment calls yesterday around this promotion that has resulted in us removing it from our sites (we installed Seesmic video comments on all TechCrunch Network blogs last month).

First, we didn’t know about the promotion until reading about it this morning along with every one else. All we knew is that our sites all simultaneously went down three times yesterday. After the first time we identified the likely problem as Seesmic and contacted the company. They assured us there was no way the plugin could take the site down. When it happened a second time we disabled the Seesmic plugin and the sites went back up. We identified the problem - the plugin was loading an external Javascript file, and when Seesmic’s servers were down, we just sat and waited for it for up to two minutes before timing out.

Seesmic said they’d patch the problem in the next version (which will pull the Javascript call into the footer instead of the header, so TechCrunch can mostly load even if they are down), and said they shouldn’t be going down again in the meantime. We re-enabled the plugin.

Then we went down a third time late last night, and we disabled the plugin for good (until the new version is available).

This morning we heard from Seesmic that the reason for the downtime yesterday was due to multiple server reboots around the Spielberg promotion.

What They Should Have Done

A simple email to us telling us that they would need to be rebooting their servers periodically over the day would have let us prepare for this and disable the plugin as it was happening. That way, Seesmic video comments would have disappeared from the site for periods of time, but TechCrunch would not have gone down. Of course, as Seesmic grows, having properly architected plugins and server redundancy will also help ensure that this problem doesn’t occur again.

I understand that young startups need a little wiggle room to get things right, and I don’t mind testing that raw software on TechCrunch. Even if that means we go down occasionally during their growing pains.

But never withhold information from your partners and tell them that you have no idea what is causing downtime when you know exactly what the problem is. As exciting as getting Steven Spielberg on your site to talk to your users is, it is not worth being dishonest to partners.

I understand that Seesmic may have been hesitant to tell us about the promotion because they wanted to keep it quiet. But all they had to do was tell us before the downtime that it was going to occur, and we would have been happy. And Seesmic would still be an active plugin on TechCrunch.

Some of you may wonder why I’m calling out a company that I’ve invested in so harshly. The reason: I’m calling them out because they deserve it, and the fact that I invested in them means I need to be careful before giving them any kind of break.

Crunch Network: MobileCrunch Mobile Gadgets and Applications, Delivered Daily.

Facebook finally has a real problem to deal with - an exceptionally rational and well-thought-out strategy by Google that puts the leading social media cloud in the path of a wave of angry users. The only thing Facebook has going for it is that said users don’t yet know they’re angry.

With its denial of service attack on Google’s Friend Connect, Facebook is serving notice that it feels threatened. By what? Users leveraging their Friend data to form communities outside of the Facebook moat? Forget for a moment that we tell Facebook who our friends are, and those gestures are created and owned by us. If Facebook insists on freezing our data as a condition of using their service, the company is essentially recommending we go elsewhere.

Google is smart enough to realize it doesn’t need to win here to help Facebook lose. Friend Connect does more to incentivize OpenId usage than to sell Google services; OpenId proliferation amortizes the complexity of that solution across multiple cooperating Web sites, particularly those that can make money on harvesting social synergies in conjunction with Adsense. It’s a Pay-Me-Now or Pay-Me-Later offer to Facebook: Play along and leverage your social equity or raise your hand and let your customers know how clueless you are.

Facebook insists it is preserving user privacy by neutering their API for its only stated purpose: “[E]nabling users to share their information with the third party websites and applications they choose.” Instead, in a Casablanca-like statement that gambling is going on (Your winnings, sir) one Charlie Cheever notes Friend Connect “redistributes user information from Facebook to other developers without users’ knowledge, which doesn’t respect the privacy standards our users have come to expect and is a violation of our Terms of Service.”

I love many parts of this, but none more than the part about privacy standards our users have come to expect. The API enables users to share their data with site and apps they choose but somehow Friend Connect does its dirty work without users’ knowledge. If the API enables user control, then what part of its use is without the users’ knowledge? Is there an Alzheimers standard that somehow slipped in here?

Crunch Network: CrunchBoard because it’s time for you to find a new Job2.0

It seems a given that mobile social networking is going to be “the next big thing”, but squinting at tiny text is still a pain on today’s phones. To deal with this issue, Blabnote, a British startup that is currently in private beta, has created what may be the world’s first “vocal social network.”

To login to the network, you simply call Blabnote from your phone, which uses caller ID to match you to your profile. From there, you can vocally enter any number of commands. For example, if I wanted to create a group for TechCrunch fans, I might say, “Create Group called ‘Team TechCrunch’”. Members can be added by saying, “add Mike and Mark”, and you can send messages to group members in a similar fashion.

Blabnote has no shortage of obstacles to overcome, to put it mildly. For one, the entire system is going to rely on voice recognition, which isn’t exactly a perfected technology. Imagine creating a very personal voice message and sending it to an ex-girlfriend on accident - the setup is ripe for disaster. And should you get sick of talking (and listening), you’re out of luck: there is no web management interface, though Blabnote says it will provide an API for third parties.

Blabnote could be a useful organization and notification tool for established groups, like soccer teams or clubs. But if it aspires to become a large social network, this company is going to be teetering perilously close to the DeadPool.

Crunch Network: CrunchBoard because it’s time for you to find a new Job2.0

It’s coming up on a year since the iPhone was released, and the second version appears to be just around the corner. So it’s a good time to check in with our readers and see just how many of you actually use the device.

Please take a second and let us know where you stand.

Crunch Network: CrunchGear drool over the sexiest new gadgets and hardware.

Epic Gillmor Gang today. Everyone went in with guns blazing over the data portability/ownership debate that has spilled out over the Facebook/Google scuffle. DataPortability founder Chris Saad was also on the call, but failed to take a leadership position in the debate (he did, however, weigh in with a blog post on the subject before the call). Their influence may be waning.

As the podcast ended the blog posts started rolling in.

Marc Canter, who I accuse of compromising his position as a thought leader in the data portability debate simply because Facebook is suddenly telling him everything he wants to hear, says that his position hasn’t changed (nevertheless, it has). Robert Scoble simply apologized for being on the wrong side of the issue, yet again. And Dan Farber, a Gillmor Gang regular who missed the call, picked up on the analogy to the founding fathers writing the Bill of Rights and wrote about it here.

All in all, the group seems to be in alignment after the talk. Data ownership is an important issue that cannot be left in the big co.’s hands. Because if it is, they’ll serve their interests first, and that will lead to more walled gardens.

Crunch Network: MobileCrunch Mobile Gadgets and Applications, Delivered Daily.

Storyblender, which recently launched in private beta, is a new casual animation platform that will appeal to users who want customized movies in a hurry. The site was a presenter at TechCrunch 40. If you’d like to try it out, you can grab one of 500 invites here.

The site is straightforward, which is a good thing because there doesn’t seem to be a tutorial yet. New users are presented with the video wizard that offers a number of pre-created movies that can be easily modified. Each movie is broken up along the bottom of the screen into brief scenes in a manner that will be familiar to anyone that has used a video editor like iMovie.

The site has hundreds of pre-rendered characters, backgrounds, music, and effects, which can be added to a scene by simply dragging and dropping. Nearly all of these have a distinctly comical look to them - the site doesn’t seem to offer “serious” movie making at this point. Instead, many of the backdrops and characters are better suited for video greeting cards (samples include “<a href="Happy Birthday" and "Party Time!“). Users can lend their own voices to clips, and they can import media from YouTube and Flickr. Members can share their movies with friends, who can modify them further if they wish.

Storyblender is competing with other animation sites like Fuzzwich, aniBoom, and JibJab.

Here’s one of the sample videos provided by the site:

Crunch Network: CrunchGear drool over the sexiest new gadgets and hardware.

Condé Nast has acquired popular technology blog Ars Technica (ranked #5 all time on the BloggerBoard), we’ve confirmed. The site will become part of Wired Digital (which in turn is under CondéNet, run by Sarah Chubb). Wired Digital assets include Wired.com and Reddit (acquired in 2006). The acquisition price will not be disclosed, but our sources say it is in the $25 million range, which is what Condé Nast paid for Wired.com in 2006.

Effectively, Ars Technica is now part of Wired. Look for an official announcement next week.

This marks a new beginning for Ars Technica, which was originally founded in 1998 by Ken “Caesar” Fisher (based in Boston) and Jon “Hannibal” Stokes (based in Chicago). They, along with their 8 or so employees, will remain with the company as it is integrated into Wired Digital.

Comscore says Ars Technica has just 1.5 million monthly unique visitors and 4 million page views, but our understanding is that the actual number of unique visitors to the site is around 4.5 million. The audience demographic is very similar to Wired, although our sources say the overlap is relatively small.

This is also another lost customer for Federated Media Publishing, which sells advertising for Ars Technica (Digg left Federated Media last year to accept a very lucrative Microsoft deal that will pay out over $100 million over three years). CondéNet will now take over advertising sales.

Crunch Network: CrunchGear drool over the sexiest new gadgets and hardware.

Chicago’s Fox affiliate WFLD has launched AirFox Live, a mashup whose spec list reads like a technophile’s pipe dream. The site combines a helicopter, GPS, live video, and Google Maps to produce a realtime data stream of the network’s newscopter that serves as uniquely informative eye-candy.

Live video taken from the helicopter’s cameras are displayed alongside an embedded Google Map that shows its current location. It might not sound particularly riveting, but I had a hard time pulling myself away from the tiny red copter as it hopped around Chicago.

Unfortunately, the site is only active when the helicopter is in the air (technically you can watch it sit at Schaumburg Airport, but have fun with that). The only guaranteed time to see it in action is from around 5:30-8:30 AM CST during the network’s morning show, which probably isn’t going to be too exciting.

But AirFox will also be active whenever the helicopter is involved in breaking news, which is where its real potential lies. The site will add a new element of information (and excitement) to events like police chases and fires. Of course, the new technology will probably only appeal to residents of Chicago, but if the program is a success we can expect other affiliates to follow suit.

You can watch a clip of AirFox in action here.

Crunch Network: MobileCrunch Mobile Gadgets and Applications, Delivered Daily.

Just over a week ago the founders of and five engineers from VoIP services provider Jangl left for Jajah after the company failed to find a proper suitor. Following their departure, it was unclear what would happen to Jangl’s assets and remaining staff. Now we hear from multiple sources close to the deal that Live Universe has agreed to acquire both.

This appears to conclude the Jangl saga that started late last fall. Around that time, Jangl’s board began telling the founders to pursue an acquisition strategy in lieu of raising more money. The board’s decision came even when the company had closed deals (some profitable) with several partners, including Plentyoffish and Tagged.

We hear there was a disconnect between the VCs, who had a more enterprise background, and Jangl’s executives, who were set on developing a consumer-facing brand. The founders, and Michael Cerda in particular, are said to have worked diligently to carry out the board’s marching orders. But despite many companies showing interest in Jangl, it struggled to find the right company for its exit.

An acquisition deal (apparently with WhitePages.com) came close but unraveled after the terms changed and became far less acceptable. With no apparent options left, much of the company’s staff was notified that they would probably have to find new work, and it was finally announced that Jangl’s founders were indeed jumping ship.

Just what Live Universe plans to do with everything they left behind has yet to be seen. I’m sure Jangl’s partners will be interested in hearing the fate of their agreements, if they haven’t already.

Crunch Network: CrunchBoard because it’s time for you to find a new Job2.0