Like Apple, RIM offers not just the hardware but the software and services that run on top of it; RIM does Apple one better by also selling back-end servers that companies install to manage their workers' email. Microsoft is in that same business, but it's not as good as tying everything together as RIM is. The speculation is that RIM shares would have to drop to $40 or so, at which point Microsoft might bid $50 a share, or $28 billion for the company. This much is not speculation: RIM would be a better buy than Yahoo.

As it should be. It's always at the height of a bubble that some fad — bandwidth delivery! online ad-networks! social media marketing management! — starts looking like a sure thing. But most startups fail, tech startups included. Few make any money for their investors; VCs, like Hollywood studios, live off the hits.

What to do if you work at the 17th Ajax-enabled online calendar to hit the market, the 217th YouTube clone, the 397th online-ad network ? Give up, move on. Despite the headlines, people are still hiring — even, sometimes, as they shed jobs. Wall Street may have failed in its job at allocating capital. But Silicon Valley's particular genius is in matching talent to markets.

So embrace it! The world has countless real problems to solve. You're not going to make a fortune copying someone else's idea. Come up with your own genuinely good one, and you'll never have to read about your company's fire sale in Valleywag.

First, there are the conditions of his grant from the Knight Foundation, which is currently funding EveryBlock. The Knight Foundation used to have strong ties with the now-defunct Knight-Ridder newspaper chain; some of those ties now reach beyond the corporate grave, a tipster explains:

Part of the deal with the Knight grant was that Adrian will turn over the sites to the local papers after the year (or however long the money lasts) is over. That's why the sites include second-tier cities like Philly and Charlotte, former Knight Ridder papers. Of course, that may just be true in those cities, so Google could still buy New York, etc.

To be clear, this is just newsroom scuttlebutt, but it's a compelling theory.

The other hitch? Google's reputation for killing startups by neglect. A tipster tells us Holovaty is making a series of "asks" to make sure that doesn't happen to EveryBlock — and that he's reluctant to admit to the talks, lest his unusual demands be misinterpreted. In that light, the secrecy makes sense. It's certainly how an entrepreneur would act. But perhaps not how a journalist would.

Google's such a natural home for Holovaty, it's hard not to see the deal going through. Besides his journalism work, Holovaty's also the creator of Django, a set of tools for coding in Python, a programming language that's strongly preferred at the Googleplex. (Guido Van Rossum, the creator of Python, already works at Google.)

If Holovaty does land at Google, expect him to transform Google News into a site that's more of a database of information than a news archive. He's long been critical of the newspaper industry's focus on stories, rather than information. A police-blotter news report, for example, is not as useful as a website which displays crimes on a map by type and date. If Holovaty's going to save journalism, he may have to do it at a search engine that many believe is killing the newspaper business. They can't say he didn't warn them.

Jabber isn't just another AIM wannabe. It uses XML trickery to connect to every popular instant message service — AIM, ICQ, Windows Live Messenger, and Yahoo — and to let programmers connect it to other services, be they for man or machine. It's already widely adopted by the IT workers whose managers sign the purchase orders for Cisco networking hardware.

By building Jabber support into its switchers and routers, Cisco can make it easy for admins to get alerts from their hardware in the same IM window as their buddies. Cisco can also sell companywide IM setups that are closely tied to Cisco network gear for security and monitoring.

Cisco recently picked up PostPath, which makes Linux-based email, calendar and collaboration software. I'm sure someone at Cisco plans to bundle Jabber's instant messaging with PostPath's Outlook-like features and dub it a "platform" to compete with Microsoft.

But Jabber's main competition isn't Redmond, it's Dulles. Cisco can now offer managers a way to ban AIM from the workplace, or at least to manage it locally with Cisco equipment rather than routing employees' conversations straight to AOL.

Case's quest to transform the healthcare industry has tragically earnest roots; he started the effort after his brother, Dan, died of brain cancer. Hospital bureaucracy always frustrates the patient's kin — but suffering does not always lead to wisdom. Dan, a tech investment banker, might have cautioned Case against this plan.

Revolution Health has ended up as a mere information middleman, buying ads on Google and then selling the users who click on them to other advertisers. That's why it inevitably needs to sell; online advertising is a game of volume. With Google eager to bulk up its own health efforts, it's not clear how much longer there will be room for third parties to play. And arbitrage is hardly a revolutionary exercise.

Someone needs to fix the healthcare system. Case's story is suitably tragic; his motivation, noble. But he's not the man for the job.

Wrong. Roxio, a CD-burning software company, snapped up the Napster name and the technical assets of Shawn Fanning's file-sharing startup on the cheap. But sometimes you get what you pay for. Roxio shed its software business and took the Napster name, but never figured out how to profit from it. In the last year, it lost $16.5 million.

And yet Napster managed to live on. If anyone can lay it in the ground once and for all, we're betting it's Best Buy. The retailer has stumbled from one unsuccessful online-music strategy to another, most recently through a partnership with RealNetworks' also-ran music site, Rhapsody. Why doesn't Best Buy just ask Steve Jobs for more iPods to sell? That seems easier.

Korean web users mostly use Yahoo-like "portal" services and never really venture out. Part of the reason for that is, Korean portals are so good. But portals have built too thick of a comfort zone for Korean web users, leaving little room for startup innovations.

The first, most famously, was AOL, which he offloaded on Time Warner's shareholders. The second, less well known, was Flexcar, a car-sharing startup which he ended up combining with Zipcar. Dumping Revolution Health would just be the latest face-saving exit for Case.

That the rationale for the deal with Everyday Health is consolidating Web traffic speaks to Case's diminished ambitions. Instead of transforming healthcare, as he loudly said he hoped to do, Case has ended up optimizing websites for search engines. Case's holding company would retain some other health-related startups after the proposed deal, including clinic chain Redi-Clinic and insurance broker Extend Health. None show transformative promise, but unlike Revolution Health, they at least sound like sensible businesses.

Newspapers have stuffed their Sunday editions full of coupons for as long as we can remember, so there must be some money in being the middleman who passes on word of discounts and deals. The difference is that coupons are simple. They are delivered to you and then you just cut them out and hand them to the cashier when you buy your milk. Microsoft's cashback plan has a higher barrier to entry. We're skeptical as to how many searchers will take the time to give Microsoft access to their bank accounts in order to earn a couple pennies back on the book they just bought from Amazon.com.

Shelfari CEO Josh Hug — how can you dislike a man whose last name is "Hug"? —blamed the whole mess on "an unexperienced but well-meaning intern." Ironically, the company that exposed Shelfari's tactics was rival books social network LibraryThing. AbeBooks, a recently acquired Amazon.com subsidiary, owns 40 percent of LibraryThing. Who will be motivated to report on Shelfari's user-gathering tactics now? Oh, right — that would be you, gentle reader!

It makes sense: Why pay to develop your own social software when you can buy up existing technology and communities of potential customers as well? Never mind the apparent conflict of interest, or Shelfari's spammy history. Those concerns are for companies who aren't trying to corner nascent markets through vertical integration. (Photo by Tim Mansfield)

InStranet went through its third round of funding way back in 2001, when it raised $14.7 million at a $50 million valuation in a round joined by Benchmark Capital. So it's not a huge exit. Then again, at least it's an exit. Salesforce.com will also report its second quarter earnings today. Analysts expect $260.56 million in revenues for a 47.6 percent year over year growth.

Acquired in October 2007, Twitter rival Jaiku still doesn't accept new users. Its current ones complain of system slowdowns and malfunctions. On May 30, 2008, founder Jyri Engeström wrote:

Contrary to some voices out there, we DO have plans for future development and we will involve our developer community as much as we can. Just to reiterate, we are working very hard to ensure you have a useful and usable service. We feel the short term pain, too.

Acquired in October 2006, JotSpot is Google Sites now, and according to longtime users, it's not what it used to be.

Purchased in 2005, it took Google six months to assign any new engineers to the project. The founders quit in 2007, and one, Dennis Crowley, will tell any entreprenuer who will listen to reject Google's siren song.

Google acquired GrandCentral, which provides a suite of telephony services, in July 2007, immediately closed it to new users and hasn't opened it since.

Google acquired Measure Map in 2006, hoping to incorporate its features into Google Analytics. "And we did that," reports Google VP David Lawee. Too bad for bloggers who missed Measure Map's blog-specific features and don't use Google's Blogger.

But Microsoft and Yahoo, both looking ofr more heft, are still in talks to buy AOL. Once the separation — mostly "a bookkeeping exercise," reports the Wall Street Journal — is complete, selling off the advertising business should prove easier. Discussions with Yahoo are "the more advanced of the two," says a source who describes the deal as one that would combine AOL and Yahoo and give Time Warner a $10 billion stake in the company. It's a proposal AOL and Yahoo began discussing in April. Whether or not a deal is consummated, the fact that Yahoo CEO Jerry Yang is still considering it just shows how much pressure he's under from shareholders, even after last week's subdued shareholder meeting.

And who can blame them? Google coddles engineers, but it also suffocates them. With the free food, massages, and laundry come a quirky set of in-house technologies and an increasingly bureaucratic, insider-driven culture. A favored clique of Google-IPO lottery winners rule over what's supposed to be a meritocracy. Marissa Mayer, Larry Page's ex-girlfriend, rules with an iron fist over what features see the light of day in Google's all-important search engine.

Google used to pitch startups on the notion of selling to them rather than give a stake to VCs; Chris Sacca, a former Googler expert in peddling empty promises, led this effort. Not surprising that it didn't work out. Google is now getting into the VC business itself — a tacit acknowledgement that it is no longer an attractive destination for startup founders. As an investor, Google gets a look at new technologies and talents. Entrepreneurs get to keep their freedom. Funny, freedom is exactly what Google used to promise the companies it acquired — and what it no longer has to offer.

(Cartoon by Pepper & Salt/WSJ)

1998: Businesses will maintain separate networks for voice, video and data.

• ''What we try to do is demystify hype,'' Forrester's Maribel Lopez told the New York Times. ''The buzz, a lot of it has to do with data guys looking to sell the next router upgrade.''
• Jupiter's Abhi Chaki disagreed, correctly calling the convergence of phone and video networks onto the Internet "an inescapable reality."

2008: Businesses will not support the iPhone for a long time.

• "The features that make it a consumer success don’t necessarily translate to the enterprise," wrote Forrester analysts Benjamin Gray and Robert Whiteley. "IT can’t be expected to support each and every operating system their employees have brought into the company."
• Jupiter's Gartenberg spotted the Achille's heel in Forrester's argument: If the CEO, rather than the IT guy, brings one to work, "it becomes a de facto enterprise business tool."

(Photo by Michael Neel)

Basic tools to help creators and other uploaders — like an upload status bar or a timer to let you know when an uploaded clip has been transcoded — are missing. For large clips, this can be maddening. Make a mistake uploading a clip? Good luck trying to replace the clip you've already uploaded with another. And if you accidentally upload the same clip twice, that's just time lost, since even with new descriptions set it'll be flagged as a "duplicate" and deleted. If your audio suddenly sounds terrible, its because YouTube forces it through a blunt compression filter. But hey, you can add funny captions to your videos!

When it comes to user experience for content creators, Vimeo and Blip.tv beat it soundly. But then why should YouTube care? If you want access to viewers who will inevitably slag you and your work in the comments, you'll have to put up with it to a degree. Better to post your content to YouTube via third-party tools like TubeMogul, which will also cross-post your video to multiple sites — making it the one-stop shop for content creators looking to publish that YouTube might have been.

When Apple bought PA Semi in the spring, we thought Steve Jobs was looking for leverage in his negotiations with Intel for more custom designs. If Apple really is going to go back to designing its own motherboards — as it did before the Intel switch — then Intel may have called Jobs's bluff. Not that that's a bad thing for PA Semi's designers — working on Mac circuitry seems more appealing than tending to Pentagon contracts, as they were doing before.

(Intel Outside image via Loren Petrich)

(Photos by Stewart Butterfield, maguisso, doctorow, oreilly, dsifry, heather)

Shortly thereafter, reearsh firm Hitwise ran numbers which showed that Digg would be inconsequential for Google's traffic, only the 13th largest Web property, well behind Google News. Coincidence? Perhaps, but they can't have been helpful for Digg's negotiations.

One other sign that the deal has been going nowhere: Digg has been interviewing for a head of PR. That's a position they wouldn't fill if they were close to a sale. That said, we hear Digg board member Brett Bullington, who helped sell JotSpot to Google in 2006, has been pushing to keep negotiations alive.

So are things on? Are they off? Never say never in deals. But even Digg CEO Jay Adelson acknowledged this week, at a meetup with Digg users in Chicago, that his company has been too prone to leaks during negotiations. Could he be getting a taste of the same from the Google side? That's a theory I dig.

(Photo of Rose by Brian Solis/Bub.blicio.us)

Adelson said the move will open Digg up to new verticals and make it possible for stories that wouldn’t make the cut on the main Digg site to find an audience. Users will be able to control the threshold for submitted articles being promoted to the front page of these.