Skype has finally resolved the "outage" that the eBay-owned Internet telephony service experiences last week. The explanation provided: It's all Microsoft's fault. The software company has a regular schedule for downloading updates to its Windows operating system. Skype engineers claim that a large number of reboots following Microsoft's "Patch Tuesday" disrupted its network. Microsoft makes for a convenient scapegoat — especially considering the fact that it offers a competing VOIP service, Windows Live Messenger — but this excuse doesn't hold water.

After an outage that's starting on its second day, Skype, the eBay-owned Internet calling service, continues to reassure its users through its Heartbeat status blog that, although significant login problems persist, Skype's programmers are making progress and that many Asian and European users are now able to use, once again, their computers as telephones. However, the periodic updates do little to clarify the situation.

Skype, the Internet calling service, is experiencing major issues preventing users from logging in. The "outage" began yesterday evening and is likely to go unresolved for the next 12 to 24 hours. In the meantime, Skype would "like to thank everyone who has taken the time to send us their thoughts, concerns and good wishes. It means the world to those working so hard to resolve this thing." Little consolation to anyone who has become dependent on the peer-to-peer VOIP service as a telephone substitute. Skype's provided little details, calling it a "software issue" — Skype is software! That's like calling a hurricane a weather problem. Sounds like someone is covering for a drunken ... oh, never mind.

As suspected, Facebook finally forced Google to shut down Blogger-hosted Facebook Secrets, which posted snips of the network's accidentally leaked source code. I'm waiting for the t-shirts. [TechCrunch]

Microsoft wants to beam broadband Internet using dormant television airwaves, claiming it won't interfere with broadcasts. One small hitch: The sample device it sent to the FCC for testing to prove its claim was broken. [The Inquirer]

Facebook has suffered another software bug this weekend — one that displays the site's source code on users' screens. How ironic: Instead of violating its users' privacy, as it did last time when a bug let people see other users' personal data, Facebook has now violated its own. The Facebook Secrets blog has posted the code for the curious, as have others. Facebook lawyers have already started sending cease-and-desist letters asking that the code be taken down, and spokesperson Brandee Barker has requested that people not post it. Which raises the question: What's in the code that Facebook doesn't want you to see?

The iconic SiliconView electronic billboard, a distraction to drivers since it first went up in 2000, has crashed like a datacenter employee on a bender today. Additional photos after the jump.

A power failure at a Samsung factory in Seoul, Korea forced a partial shutdown of chip production for the world's largest memory supplier. The breakdown is likely to boost competitors Hynix and Toshiba and impact manufacturers of consumer products using NAND Flash memory... particularly Apple, makers of the iPod and JesusPhone, who had — until recently — been reaping the benefits of high margins. [AP]

365 Main, the troubled datacenter operator, has finished its investigation into the failure at its San Francisco facility that knocked some of the Internet's most well-known websites, from Craigslist to LiveJournal to Technorati, offline back in July. Ridiculously, the company first tried to blame PG&E for the failure, knowing full well that its clients pay it for reliable power even in a blackout. (Equally ridiculously, I ran a suspect tip that a drunk employee had wreaked havoc in the datacenter.) Now, the company has completely exonerated itself, pinning the blame on a component in its generators. Here's why you still shouldn't believe a word the company says. My analysis, and the company's press release, after the jump.

Facebook appears to be having a privacy crisis. Earlier this morning, The Register reported that some people were able to see other Facebook users' private message inboxes. Blogger and IBM employee Matt Dibb reported Facebook exposing other people's email addresses on its login page. VentureBeat speculates that a bad update to Facebook's codebase got rolled out — and today's supposed "upgrade" was actually Facebook's panicked attempt to fix the problem. Since Facebook's fine-tuned privacy controls are a big selling point, this mucked-up code episode is especially embarrassing. Mark Zuckerberg, if you can take a break from counting your fictional billions of dollars, care to take a look at your site's code? Update: Facebook says it has fixed the problem. The statement, after the jump.

Forget Facebook fatigue: The new symptom sweeping the Valley is Facebook addiction. A brief outage this morning made most Valley workers more productive — but left some, like strung-out addicts, completely unable to function. One particularly sad case expressed his relief that Facebook was back up. "I've already forgotten the cold panic, the unbearable neediness," he wrote. Also left unable to function: Facebook's PR apparatus, which promised a statement about the outage that has yet to materialize. Perhaps spokeswoman Brandee Barker was hoping to send it out through the Facebook group she normally favors, instead of boring old email, for press releases?

Facebook has been down for most of the morning, ostensibly for an upgrade. Which is nonsense, of course. No well-run company updates its servers in the middle of the day; Facebook's unexpected outage can only be, well, an unexpected outage. But we can't wait to see what happens when Silicon Valley's entire workforce has to actually stop using Facebook and get something done.

The upside of that suspect tip I got about what caused Tuesday's massive Internet outages? Tons of tipsters have written in to tell me what they saw on the scene at 365 Main, the troubled San Francisco datacenter whose generators failed during what ought to have been, for the expensively-built facility, a humdrum power outage in the city.

I'm a dunce. I was wrong. There, I said it. In running a tip on Tuesday that a drunk employee brought down 365 Main, the San Francisco datacenter which hosts servers running some of the Web's most important sites, I trusted a source I shouldn't have. Here's the story behind my 365 Main post. A warning to readers of sensitive dispositions — I'm about to take you inside the sausage factory, and it's a bloody mess.

I'm continuing to investigate the story of the outage Tuesday at 365 Main's San Francisco datacenter that brought down some of the most well-known sites on the Internet. Right now, a 365 Main executive is blaming failures at 5 out of its 10 generators. That's right: Fully half of 365 Main's generators failed right as San Francisco experienced a power outage. More to come on this soon, but for now, here's the memo from Marcy Maxwell, 365 Main's head of security.

After killing most of the websites you care about on Tuesday, 365 Main, the troubled datacenter in downtown San Francisco, is back to business. The business of making excuses, that is. Cynthia Harris, the same flack who issued an immaculately timed press release Tuesday morning crowing about how RedEnvelope moved all of its Web operations to 365 Main, only to have them taken down by the outage, is going around telling everyone who will listen that nothing untoward happened. To which any user of Craigslist, Technorati, Six Apart's LiveJournal and TypePad, and AdBrite might respond, rrrrright. Data Center Knowledge has a detailed report. Here's what else I've learned — and why 365 Main's performance remains highly suspicious.

Seen at 365 Main, the troubled San Francisco datacenter: A man being lead away by police, in handcuffs, screaming, "You have been trolled by nut rollers!" Could this have been the employee responsible for the outage? (I no longer know whether to trust the tipster who sent this in, or the tip. -Ed.)

We've now learned more about the outage at 365 Main's San Francisco datacenter that knocked some of the Web's most popular sites offline. The latest theory: An employee, reportedly drunk, hit the emergency-power-off switch in 365 Main's Colo 4 room. (Update: I no longer know whether to trust the source who sent in the tip about a drunk employee.) Other sites located in other rooms were unaffected. This isn't the first time 365 Main has suffered an EPO-induced outage; a major one still remembered by customers occurred back in April 2005, and another took place last year. After the jump, a gallery of the carnage caused, and a roundup of reactions.

There's a reason most datacenters are located in distant office parks: It's harder for angry customers to line up at your door. And that's what's happening to 365 Main, the downtown-San Francisco datacenter which is suffering a major outage, caused, a tipster says, not by local power fluctuations but by a drunken employee on a bender. (Update: I no longer know whether to trust the source who sent in the tip about a drunk employee.) An eyewitness says that in addition to the customers lining up, bicycle messengers are constantly whizzing by to drop off packages — legal notices, one presumes, informing 365 Main that it has breached customers' service-level agreements. Anyone else on the scene? Drop us a line.