Briefly, Section 230(c)(1) provides immunity from liability for providers and users of an interactive computer service who publish information provided by others:
No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.

In order to use this provision, there are three elements:

1. The defendant must be a provider or user of an interactive computer service.
2. The cause of action asserted by the plaintiff must treat the defendant as the publisher or speaker of the published information at issue.
3. The information must be provided by another information content provider, i.e., the defendant must not be the information content provider of the published information at issue.

Section 230 does not protect providers or users from all harms, for example it would not protect the provider or user from criminal copyright infringement, or from violating fair housing laws, etc. Still, the fact that it works as it does provides us with the modern internet we know today.

Why I'm thankful for Section 230 is a post from: Cyberlaw Central

Definitions

In other words, Ford is presenting New Yorkers with a choice of voting against him as a tax cheat or as the guy with the worst press operation in human history.

It's all about empowering the voters.

• Windows Phone 7: First Videos
  Closer look at what you can expect from a Windows Phone 7 experience. [Gizmodo]
• Apple to wrap digital books in FairPlay copy protection [Clarified]
  Remember FairPlay, the DRM scheme Apple mostly dropped after years of locking users into a crappy DRM scheme? Well it's staging a comeback in e-books. [LA Times]
• Malware Peddlers Look to Exploit Olympic Luge Tragedy
  Same old trick. Find what users are searching for, then fake 'em out! [Mashable]
• My 2009 Annual Report
  Math teacher Dan Meyer puts a year's worth of personal metrics into the incredible video below. [dy/dan]
• 
  
• The Smarter You Are, The Less You Click
  But then you probably already knew that, didn't you smartypants. [RWW]
• Sciencefeed
  It's like Friendfeed for Science! [via TechCrunch]
• Wi-Fi Turns Rowdy Bus Into Rolling Study Hall
  Internet on a school bus means students stop all that noisy horseplay! [NYT]
• Adobe AIR for Smartphones Wants to Be One Platform to Rule Them All
  Adobe's cross-platform desktop runtime AIR is on its way to mobile devices, starting with Android. [Gizmodo]
• Apple drops the ban hammer on iPhone hackers
  If you develop for the iPhone and App Store proper, Apple's not going to reward your spare-time hacking. [Boy Genius Report]

About twice a week I think, "Why don't I try this crazy idea and see if it works?" about some element of my career. With no label, no manager, and no inner voice of reason slow me down, I get to experiment all I want. 90% of my crazy ideas have to do with social networking -- which I spend half a lifetime doing, despite the crap I take from my family and Real Life friends. (Hey, some of us actually do bond over web comics, starship replicas, the fail whale, and photos of stuff on cats.) Mostly my nutty ideas work just a little bit. Some are epic failures. But my experimental flopping and floundering inches me closer to the day when I'll be totally financially independent as a full-time musician. Plus it's more fun than having a real job.

But every now and then a crazy idea works really really really good. Bam!

I planned to sell 20-40 of my little bootleg CD's. Silly me. I sold well over 200. My little stack of jewel cases looked so pathetic.

WholeWheatRadio.org broke every record for online listenership, CD sales, tips -- everything. The more listeners tuned in, the more tuned in, and the more money they gave, the more money they gave. The crowd online was thrilled to be breaking WWR records. I drove away from Talkeetna having earned about $4,000 in one night, with a new CD to produce in just a couple of days and an avalanche of e-mail and publicity requests to deal with. Seldom have I been so happy and so panicked.

Sponsor

As pointed out in a recent blog post by university professor Scott Shane, "ham and egging" was first coined by Columbia's professor Amar Bhide and Harvard Business School's Howard Stevenson. The term refers to the technique of convincing multiple stakeholders that others are working with you despite the fact that you're only in talks. The only problem is that most early partners only want to work with you if other reputable partners have already signed on.

Explains Bhide and Stevenson,"the ultimate ham and egging solution is for the entrepreneur to simultaneously convince each participant that everyone else is on board, or almost on board."

However, when ReadWriteWeb spoke to MobiTV CEO Paul Scanlan about forging deals between telecom and television companies, he suggested a different tact. Although Scanlan found himself caught between partners who were skittish to sign on without the initial validation of others, he decided that rather than ham and egging, he'd build contingency clauses into contracts. Scanlan's contracts stated that all partnerships were contingent on a set number of large-scale partners to launch. While this may not be the ideal method of closing deals, it seems like an ethical alternative to engaging in deals that begin with dishonesty.

Have you ever engaged in ham and egging and if so, was your deal a success?

To be fair, these two developments are really far apart in their delivery dates. The Gmail status update could come as soon as tomorrow, whereas the the speech-to-text-to-speech translation system is still a ways out. You can definitely see just how much work Google needs to do by trying to read your Google Voice voicemail transcriptions. (Voice search works better on Android 2.1 because you're talking slower and enunciating.) But both these features point in the same direction many of the company's other products have been hinting at. Here's a list of Google's major products, in case you forgot, and which sector of communication they want to dominate.

Google Voice: This is a big one, and it'll be the most natural interface for Google to slot in the voice-translation into. If you're using it the way Google wants you to use it, you're already piping all your voice calls and SMS through Google's tubes. And refining speech to text gives them a good idea of your interests and what you're talking about, allowing them to better serve up the relevant ads to you during calls.

Gmail: Having access to at least one end of everyone's email conversations, outside of business emails, gives Google the ability to be a gateway for most of your written communications. But that's not enough for Google, which is why they developed...

Google Wave: It's email, message boards, chat rooms and collaboration software all in one, except every participant needs a Google account. This closes that "openness" loophole that email has, and forces everyone into Google's biosphere. So this, and Gmail, should make sure that every medium-length communique passes through Google's maw for analysis. But what about shorter and longer forms? Update: Thanks commenters, for reminding me that Google made Wave open, so people can create their own Wave servers to talk to each other with the Wave protocol. The point still remains, that if you were going to use a service, wouldn't you rather use the service from the company that created the protocol, for performance and feature reasons?

Google Docs: For longer documents.

Google Talk: For short blasts of instant messaging, video chats and some audio chatting.

Picasa and YouTube: Communication doesn't have to be all text-based, you putting your photos and videos online count too.

Android and Chrome OS: By getting you down at the operating system level, Google can theoretically know every kind of communication you perform. It knows who you talk to, how you do it and when you do it. It can even shape the how by delivering the experience themselves.

Everything else. There's Checkout, Finance, Maps, Reader, News and other apps, which fill in the other forms of communication or expression that aren't quite covered by the major products above. One major missing piece is social networking, where Google basically failed before with its Orkut service (except for Brazil), so this new Twitter/Gmail hybrid might be their next entrance into the space.

But why do they want these things? Why would Google want to be the middleman between you and the world? To sell you ads, of course. And don't think Google is going to stop at just helping you talk over the internet or over the phone, they're going to reach into meatspace as well. How? One step is making that speech-to-speech translation portable, so you can do a sort of near-field communication with someone else with the same device while at the same time being able to look them in the face. Then, blast you two with the appropriate ads on the billboard next to you.

The move would further turn Gmail, which revolutionized online e-mail, into a comprehensive communications hub. The intent is to keep people's attention centered on Google, by making Gmail, not Facebook, people's first stop online and their default place to send and receive messages. Gmail users can already chat via Jabber or AIM, make video calls, and send SMS messages from Gmail's web interface.

As the Post reports:

Google has been trying to fashion Gmail into more than an e-mail service for years. The service currently lets users set an away message, which can be a link to a Web page, that their friends see when they instant-message them. Now, it plans to launch a new interface that will aggregate updates from more friends in a stream.

The new stream will also eventually include content that a user's connections share through its YouTube video site and Picasa photo service, according to one person familiar with the matter. But whether those features will also be announced in the coming days remains unclear.

The full extent of the new features remain unclear, but Google is inviting reporters to a launch event Tuesday on its Mountain View, California, campus promising some innovations in two of our most popular products, according to an e-mail sent to reporters.

Yahoo has included similar features in its e-mail service, letting users see what photos their contacts have uploaded to Flickr, for example.

Google could integrate updates from a user's Twitter account, since most of that is public. And it could likely make it easy for Gmail users to post to Twitter as well, due to Twitter's liberal API policy.

Facebook, however, will not likely let its rival re-publish status updates, or allow users to publish to their Facebook pages through Gmail. Facebook, much like AOL and Compuserve back in the early '90s, is a controlled and sanitized version of the larger internet, but it relies on closed protocols.

via Google to Add Social Feature to Gmail - WSJ.com.

See Also:

• April 1, 2004: Gmail Hits Webmail G-Spot
• Google Turns on Gmail Encryption to Protect Wi-Fi Users
• Gmail's New 'Undo Send' Feature Saves You From Outbox Regret
• AOL Announces Lifestream App for Twitter, Facebook, SMS
• Walled Garden No More, AOL Adds Social Networks to Homepage

The O!Play streamers are very highly-regarded here at Gizmodo, especially when considering they don't cost more than 200 notes. The addition of Wi-Fi 802.11n, along with a multi-format card reader and ethernet port, for 129.90 in Italy (around $177), is definitely worth the few extra bucks. [ASUS via Slashgear]

...it's just tough to be floored by anything less.

The Qisda (you also know them as BenQ) QCM-330, expected to debut at the upcoming CeBIT tradeshow this March before being available through Vodafone, features a 4-inch, 1280x1024 screen that could be sharper than life itself, along with HSDPA and Wi-Fi.

We don't know much else, other than that it will most likely run Android given the Home key (which makes the prospect of importing a phone for its hardware alone so much more appealing). But if you're one of those people who doesn't like to use the same phone as other people, the QCM-330 might be a decent handset to watch. [Unwired View via SlashGear]

There's the Atom N270 processor, running at 1.6Ghz, for example. And then there's the 2GB of memory and solid state 32GB drive. Lastly, the replaceable battery on this 8.9-in. multitouch tablet is clocked at a mere four hours, which doesn't seem that great (saving grace being that is replaceable).

Pricing is set at $599 when it launches in March. Impatient types can buy a non-multitouch prototype for $780 right now. [ExoPC via Engadget]

The iPad can run "virtually every one of these apps, unmodified, right out of the box." They can either run it very small, 1:1 pixel, in the center of the screen. Or they can "pixel double" it and run it full screen, in a low resolution mode.

For new apps, however, Apple is releasing a new SDK for apps with special interfaces for the iPad.

As anticipated, the operating system is best thought of as an evolution of iPhone 3.0. That means that apps are running the show, with the same tray at the bottom and the same accelerometer capabilities.

To access the screen, you slide to unlock, just like on your phone. The display is practically identical (though biggie-sized, obviously), with a Home button situated at the bottom. You call up apps the same way you do on your phone, and they automatically go to full screen. You can also swipe through pictures and pages, again just like on the iPhone.

But how does it feel in the hand? Well, it's an inch thin and weighs just 1.5 pounds, so it's definitely easily portable. And since it's intended to be a portable device, it's got a pretty crazy proposed battery life: ten hours of video playback, and one month of standby charge.

A primary concern has been how the keyboard will work. Our money was a split-screen keyboard, but it turns out they've opted instead for to copy the iPhone again here, with a keyboard taking up the bottom half of the display when called up. It's not meant for your thumbs, apparentlyyou're expected to type on it as you would a physical keyboard.

For web surfing, the page renders just like a browser, with navigation buttons on top. For email, you can bring a pull-down menu of the inbox.

So basically, it looks like the user experience is going to be just like a big ol' iPhone, for better or for worse. I'm especially curious to see how intuitive the keyboard is. But otherwise, all the multitouch features and app arrangements should feel like old hat.

The iPad can run "virtually every one of these apps, unmodified, right out of the box." They can either run it very small, 1:1 pixel, in the center of the screen. Or they can "pixel double" it and run it full screen, in a low resolution mode.

For new apps, however, Apple is releasing a new SDK for apps with special interfaces for the iPad.

At the end of a failed 15-year experiment in giving away its product, the press (newspapers and magazines) has begun to renounce free. It's slow in starting, because of the inertia of this decade and a half, but the New York Times announced recently that it would begin charging for its Website, and others are sure to follow.... But payment for Websites alone won't be enough to change newspapers' and magazines' bottom lines from red to black. Apple's tablet, however, will.

It's time for free to end. Newspapers and magazines made the mistake, in the early days of the Web, of giving away their content for free, in exchange for revenue from Web advertising.

In the past few years, tens of thousands of jobs have been lost, and newspapers and magazines are cutting back and folding all across the U.S.... Yet we need the press: the fourth estate is a necessary check for our government and business. As long as free thrives, the press can't do its job correctly. Free may be good for freeloaders, but it's bad for society. Those who want things to be free forget that there are still people doing the work they get for nothing, and those people need to be paid. As the old saw goes, there ain't no such thing as a free lunch.

What news agencies can't do is the added-value reporting, the analysis, opinion and in-depth reporting that we want to read to better understand, and that we need for society to thrive. It may be a coincidence, but in recent years, investigative journalism was severely lacking at a time when it was needed the most. Only when people pay for news can we have quality reporting.

To those who protest that "no one will pay for a newspaper on the Web", consider some very successful experiments in paid online content. The Wall Street Journal charges around $100 a year for full access to its Website, and plenty of businesspeople pay for this. This is because the Journal provides the kind of news that is not plentiful; people pay for the quality of the business news and analysis that they can't find elsewhere, as well as its timeliness.

I'm betting that Apple will get it right, as far as features, interface and usability are concerned. It will also be an excellent tool for reading the news. Newspapers and magazines will be able to package their content in multimedia bundles (either as apps or something similar to the iTunes LP) that will be designed for reading on a portable screen; this won't simply be web pages viewed on a smaller screen.

The key to hardware being successful is the software that supports it. One of the main advantages to Apple's tablet, as far as the press is concerned, is the iTunes Store. Since Apple already has this platform to sell and deliver that content, even on a subscription basis, readers will be able to easily buy their favorite newspapers and magazines and get them delivered instantly. They'll be cheaper than the print versions, and they'll be a lot greener too. And the iTunes Store will be able to provide a better selection than readers can find by going to individual Websites. Whether by subscription or by single issue, it'll be extremely simple to buy newspapers and magazines to read on the Apple tablet.

Baumgartner will join United States Air Force Captain Joe Kittinger as the only man to jump from near space altitude. Kittinger jumped on August 16, 1960, from the Excelsior III balloon, which at the time was flying at 102,800 feetthat's 19.47 miles or 31 kilometers up in the sky. Compared to Baumgartner, however, Kittinger's suit looks miserable:

In fact, his right glove failed in the descent, and his hand dilated to twice its size. Absolutely crazy.

Hopefully, Baumgartner won't have any of Kittinger's problems. He will jump sometime in 2010, after a few test jumps at lower altitudes, as part of Red Bull's Stratus mission. Kittinger will be assisting Baumgartner from the ground control, while the mission team monitors his position and body state as he plummets down to Earth, surpassing the speed of sound.

I love these nutty people.

I will do a more comprehensive year in review for Cyberlaw generally, but I thought it would be fun to take a close look at how 47 USC 230 fared in 2009. This is the first full calendar year following the Ninth Circuit's en banc Roommates.com opinion, and many of us initially feared that the case would create a huge hole in 230's otherwise solid immunity. As it turns out, those concerns have not come to pass. If anything, 2009 shows us just how strong the immunity remains.

I blogged on a total of 22 cases issued in 2009 that discussed the statute. (I blog on every case I see that substantively discusses 47 USC 230). I blogged on other cases in 2009 that were decided before 2009, such as the Woodhull v. Meinel case from October 2008 and DC v. Harvard-Westlake, a 2007 arbitrator's dismissal that came to light in 2009.

Of the 22 calendar year 2009 cases, I would classify 14 of them (63%) as easy defense wins, frequently on a 12(b)(6) motion to dismiss or state law equivalent. Even many of the remaining 8 cases contained good news for defendants. For example, in Shiamili, the defense inexplicably lost at the district court level but got an easy reversal on appeal. The Stayart court granted Yahoo an easy defense win, although co-defendant Various didn't get the 230 ruling. Similarly, the Barnes case granted the defense an easy 230 win on one theory (negligent undertaking) but denied 230 for a different one (promissory estoppel). The Certain Approval Process case said 230 did not prevent the plaintiff from amending the complaint to add a cause of action, but once added, the court instantly zapped the claim on other grounds.

This leaves four unambiguous 230 defense losses in 2009. The leading 230 defense loss was the Tenth Circuit FTC v. Accusearch case, which held a retailer liable for reselling illicit phone records. The other major 230 defense loss was the NPS v. StubHub case, which held that 230 may not apply to a lawsuit over the alleged illegal ticket scalping by StubHub's sellers. Both of these cases involve the retailing of illegal items, suggesting that 230's boundaries may not reach that far.

The other two defense losses are less consequential. The Project Playlist held that 230 does not preempt state IP law claims, a conclusion that deserves note only because the Ninth Circuit held otherwise in the 2007 ccBill case. I believe that no other courts will follow the Ninth Circuit's rule that 230 preempts state IP laws, making the Project Playlist ruling unsurprising.

In People v. Gourlay, a web host was denied a 230 defense to a criminal prosecution for child molestation- and child pornography-related claims. This case turns mostly on the web host's active role creating the child pornography (as well as the host's molestation of the child actor); with that context, this case may have little influence on other cases. Indeed, the court made clear that web hosts providing standard web hosting services could fully qualify for 230 protection against a state criminal prosecution of child pornography dissemination.

In reverse chronological order, a brief overview of the 230 cases from 2009:

Nemet Chevrolet v. ConsumerAffairs.com (4th Cir. Dec. 29, 2009). One of three federal appellate court 230(c)(1) rulings in 2009 (Barnes and Accusearch are the others). A solid defense win for a consumer review website. The plaintiff's claims that the website contributed to the reviews' development and fabricated reviews were tossed on a 12(b)(6) motion to dismiss.

Shiamili v. Real Estate Group (N.Y. App. Div. Dec. 17, 2009). In an unpublicized January 2009 decision, the trial court denied a website's 230 dismissal request for claims based on user-supplied comments. In December, this error was fixed on appeal despite allegations that the website chooses and administers the user content.

Dart v. Craigslist (N.D. Ill. Oct. 20, 2009). Craigslist got a big win in its ongoing battles with various government agencies over prostitution ads on Craigslist when the court held it wasn't liable for those ads.

Riggs v. MySpace (C.D. Cal. Sept. 17, 2009). A goofy case. The court holds that MySpace's deletion of Riggs' account was protected by 230(c)(1) on the apparent theory that Riggs (the plaintiff) was the third party supplier of the deleted content. This case would make more sense as a 230(c)(2) case.

Finkel v. Facebook (N.Y. Sup. Ct. Sept. 15, 2009). Facebook wasn't liable for the contents of a user's private group even though Facebook placed a copyright notice on the page.

Intellect Art v. Milewski (N.Y. Sup. Ct. Sept. 15, 2009). Ripoff Report wins again.

Stayart v. Yahoo (E.D. Wis. Aug. 28, 2009). An convoluted, and possibly confused, ruling that Yahoo wasn't liable for search results snippets. However, Various was denied 230 because it may have originated the content in question.

Cornelius v. DeLuca (E.D. Mo. Aug. 18, 2009). An online retailer wasn't liable for user-supplied comments despite a conspiracy allegation.

Goddard v. Google (N.D. Cal. July 30, 2009). This is a follow-on ruling to an important December 2008 ruling in this case, which dismissed the plaintiff's complaints but gave the plaintiffs another chance. The December 2008 ruling is one of the most interesting and important decisions interpreting Roommates.com. In the July ruling, the judge again found that 230 insulates Google from liability due to allegedly fraudulent ads run through its network and granted a final dismissal.

Doe II v. MySpace (Cal. App. Ct. June 30, 2009). MySpace isn't liable for users' sexual assaults on other users.

FTC v. Accusearch (10th Cir. June 29, 2009). The second of three federal appellate court rulings on 230(c)(1). The defendant was an online retailer of illegal phone records. The retailer claimed that the phone records came from third party suppliers and therefore 230 immunized the retailer from liability associated with the records. The court echoed the Ninth Circuit's Roommates.com decision, effectively extending that case to the Tenth Circuit, and said that the retailer was responsible for selling the illicit phone records despite 230.

Zango v. Kaspersky (9th Cir. June 25, 2009). This is the only 2009 ruling addressing 47 USC 230(c)(2), the overshadowed and frequently overlooked sibling of 230(c)(1). Despite the rarity of 230(c)(2) cases, this case could be fairly influential. The Ninth Circuit held that 230(c)(2) protected an anti-spyware software vendor's decision to classify software as a threat. If you missed it, you might want to take a look at my presentation slides on 230(c)(2), which distill my deep look at 230(c)(2) this summer.

Gibson v. Craigslist (S.D.N.Y. June 15, 2009). Craigslist isn't liable for physical injury caused by a gun purchased via a Craigslist ad.

Doe IX v. MySpace (E.D. Tex. May 22, 2009). MySpace isn't liable for users' sexual assaults on other users.

Barnes v. Yahoo (9th Cir. May 7, 2009; amended opinion June 22, 2009). The third of three federal appellate court opinions on 230(c)(1). The Ninth Circuit held that 230 preempted a claim against a service provider for negligently delaying the removal of user content (essentially, Zeran redux), but 230 did not preempt a promissory estoppel claim based on promises the service provider made to the person requesting takedown. The initial Ninth Circuit opinion had two other unfortunate digressions: (1) it said that 230 was an affirmative defense that did not support a 12(b)(6) motion to dismiss, and (2) the opinion had ambiguous language implying that 230 preempted only state claims, not federal claims. The amended opinion helpfully eliminated both digressions.

Atlantic Records v. Project Playlist (S.D.N.Y. March 25, 2009). 230 does not preempt a state IP claimin this case, a violation of state copyright law for pre-1972 sound recordings.

Joyner v. Lazzareschi (Cal. App. Ct. March 18, 2009). A message board operator wasn't liable for user posts.

Raggi v. Las Vegas Police (D. Nev. March 10, 2009). A union wasn't liable for messages that union members posted on the union-operated message board.

Certain Approval Programs v. Xcentric Ventures (D. Ariz. March 9, 2009). 230 did not bar amending a complaint to add a new cause of action when the plaintiff also adequately alleged that the Ripoff Report contributed to the creation and development of the content at issue.

People v. Gourlay (Mich. App. Ct. March 3, 2009). This case involves the prosecution of a pornographic web host who also molested the child actor. The web host asserted a 230 defense in trying to overturn the conviction for the charges related to pornography dissemination. Although 230 can preempt state criminal prosecutions, and web hosts are protected by 230 for their ordinary web hosting activities, this web host actively participated in the site's development and therefore lost 230's protection.

NPS v. StubHub (Mass. Super. Ct. Jan. 26, 2009). In a long-running battle between the New England Patriots and season ticketholders who want to resell their tickets via StubHub, StubHub was denied summary judgment on 230 grounds. The court cites Roommates.com in saying that StubHub may have contributed to illegal ticket scalping sufficient to potentially disqualify it for 230 protection.

GW Equity v. Xcentric Ventures (N.D. Tex. Jan. 9, 2009). Ripoff Report is protected by 230 even though it offers pull-down menus and manipulates user-submitted reports.

Nemet Chevrolet Ltd. v. ConsumerAffairs.com, Inc., No. 08-2097 (4th Cir. Dec. 29, 2009)

Introduction

Citing 47 USC 230, today the Fourth Circuit upheld a 12(b)(6) dismissal of defamation and related claims against a consumer review website. This case is noteworthy because the court rejected some common allegations that plaintiffs make to evade 230, so this case may help defendants get 12(b)(6) motions to dismiss more easily.

ConsumerAffairs.com is a consumer review website with a twist: it works in conjunction with a law firm that mines the submitted complaints for potential class action lawsuits. In June 2008, I blogged about the district court's 12(b)(6) dismissal of the case.

Development of the Reviews

Nemet tried two tactics in its complaint to draft around 230. First, it alleged that ConsumerAffairs.com partially developed 20 reviews. Nemet pled:

Upon information and belief, Defendant participated in the preparation of this complaint by soliciting the complaint, steering the complaint into a specific category designed to attract attention by consumer class action lawyers, contacting the consumer to ask questions about the complaint and to help her draft or revise her complaint, and promising the consumer that she could obtain some financial recovery by joining a class action lawsuit. Defendant is therefore responsible, in whole or in part, for developing the substance and content of the false complaint . . . about the Plaintiffs.

These allegations do not survive a 12(b)(6) motion to dismiss.

* the website "structure and design" argument fails, despite Nemet's attempt to invoke Roommates.com, because ConsumerAffairs' structure was not illegal. To me, the court's discussion reinforces that Roommates.com' real holding is If you don't encourage illegal content, or design your website to require users to input illegal content, you will be immune. Chalk this case up as yet another citation of Roommates.com for the defense.

* Asking users questions about their posts does not qualify as development.

* The unsupported assertion that ConsumerAffairs edited posts did not pass the Iqbal standard. Plus, as Zeran indicated, 230 protects editorial decisions, so the allegations needed to assert some editing beyond this protected zone.

Review Fabrication

Second, Nemet alleged that ConsumerAffairs fabricated 8 reviews. Nemet pled:

Because Plaintiffs cannot confirm that the [customer] complaint . . . was even created by a Nemet Motors Customer based on the date, model of car, and first name, Plaintiffs believe that the complaint. . . was fabricated by the Defendant for the purpose of attracting other consumer complaints. By authoring the complaint . . . the Defendant was therefore responsible for the substance and content of the complaint.

This allegation has an obvious (and IMO embarrassing) logic flaw. Even if Nemet can't use its records to validate the facts in a consumer review, ConsumerAffairs.com's fabrication of the post is only one of many possible explanations. The court notes some other possible explanations: "the post could be anonymous, falsified by the consumer, or simply missed by Nemet." (I would also add the possibility of weak recordkeeping by Nemet). To try to get around this logical deficiency, Nemet marshals up some additional allegations:

(1) that Nemet has an excellent professional reputation, (2) none of the consumer complaints at issue have been reported to or acted upon by the New York City Department of Consumer Affairs, (3) Consumeraffairs.com's sole source of income is advertising and this advertising is tied to its webpage content, and (4) some of the posts on Consumeraffairs.com's website appeared online after their listed creation date

But all of these facts are non-sequiturs; none of them show that ConsumerAffairs fabricated the posts, and post-Iqbal these allegations are not enough to state a claim. The dissent disagreed with this conclusion (about the alleged fabrication) and would have allowed those claims to proceed.

230 as an Immunity Redux

In FN 4, the court notes that the Seventh Circuit questioned if 230(c)(1) was just a definitional section. Citing Zeran, which addressed this issue explicitly, the court says "Of whatever academic interest that distinction may be, our Circuit clearly views the 230 provision as an immunity:" As a result, the court "aim[s] to resolve the question of 230 immunity at the earliest possible stage of the case because that immunity protects websites not only from 'ultimate liability,' but also from 'having to fight costly and protracted legal battles.'" It looks like there could be a brewing catfight between circuits over whether 230(c)(1) is an immunity, an affirmative defense, a definitional section or something else.

Conclusion

Given that this court was bound by the Zeran precedent, it's perhaps not surprising that the court found 230 protection for a consumer review website. Nevertheless, by rejecting another plaintiff's attempt to make hay from Roommates.com and rejecting weakly supported allegations of fabrication, this court gave defendants even more support to fend off claims that are, at their core, based on third party content.

The updated census of Roommates.com citations:

Roommates.com Cited for Defense (11 cases): GW Equity v. Xcentric, Best Western v. Furber, Goddard v. Google (and second ruling) Joyner v. Lazzareschi, Atlantic Records v. Project Playlist, Barnes v. Yahoo (note: although the case was a partial loss for the defendant, the Roommates.com discussion came in the defense-favorable part), Doe IX v. MySpace, Doe II v. MySpace, Dart v. Craigslist, Shiamili v. Real Estate Group, Nemet v. ConsumerAffairs

Roommates.com Cited for Plaintiff (2 cases): NPS v. StubHub, FTC v. Accusearch

The gadgets are flowing and they've got both publishers and subscribers in a tizzy over their options. Are they 3g? Can I put my content on it? Just wifi? What services do they deliver? Do I need to build an app? Am I locked in?

All great questions but not the one that is at the front of my mind. That question being where is the personal media hub for all of this content? Each type of media that we consume has a disparative quality of some sort that requires another gadget or format transcoder to allow usage which means, users need a hub.

I just want to know where that hub will be. I'm not sure if it belongs in the cloud or can even exist there due to limitations placed on that content by rights holders. Which is a legitimate reason not to use the cloud since publishers need to eat.

A couple reasons to use the cloud would be transfer speeds, remote accessibility and backups. With increased gadget connectivity it would make sense to do this. An example of a gadget that needs to be fed from an outside source like the cloud is the PSPgo. It relies on connectivity to fetch games, video and browse the web.

The games on PSPgo arrive from a Sony controlled hub behind a firewall. If the cloud is too limiting due to rights management the other other solution would be to offer a private hub. Another gadget, but one that resides in the dwelling of an individual. Using the Sony model for control and privacy a device like this could be the next evolution of an inclusive hub. It seems to me to be the missing link.

Media management across multiples platforms and for varying devices would require some version of a standard protocol. The protocol probably already exists and could be as simple as HTTP with SSL. The device itself a web server that connects to cars, phones, tablets, computers, televisions, etc.

A device like this could also create new opportunities for rights holders to create new models for selling content. I'm thinking in the range of micropayments for ongoing usage or payments for amount of time used. An example would be a movie that instead of a 24 hour limit would allow the consumer to view it 2 times on any device before being crippled or offered for purchase for an additional few dollars.

My personal interest would be to have a media hub that I had control over and could add content to from any device like the PSPgo, Kindle, iPhone or computer. The ability for these devices to speak a common language for file storage and retrieval would increase consumption and sales as all of a users purchases become portable, even if lockedin to a device.

There are plenty of media hubs that exist today for personal use that can be net connected, but this device would find its niche in storing and delivering content without limitation.

DISCLOSURE OF MATERIAL CONNECTION: http://cmp.ly/0

Where is the personal media hub for ebooks, music and videos? is a post from: TechStartups.com

Apple goes 3.0 with Apple TV software and all kinds of hell breaks loose. The typographer buried deep in Apple TV users reared it's ugly, but sans-serifed, head and is running amok on the interwebs complaing about the interface. A leaner, cleaner typeface has replaced the chubbier, bloated original.

This new font appears to be better designed for LCD displays. OK, who am I kidding . . . the font is the Apple standard, Helvetica. It looks better everywhere that you have to display copy for reading from more than 4 feet away from it.

So far I have been impressed with the new interface and response time to menu selections. It could be that when I blew away all the hacks it cleaned up the hard disk.

Thanks to Apple for giving some much deserved love to the ATV owners by upgrading the software. In the future, it would be welcome to actually allow us to have Boxee type content or real internet connectivity with feeds to create playlist or customs subscriptions.

For now we'll have to settle with an easier to read screen for our Scooby Doo episodes.

DISCLOSURE OF MATERIAL CONNECTION: http://cmp.ly/0

Apple TV New Font is a post from: TechStartups.com

About the author: Matt is a software developer and author living in Chicago. He is the author of five programming books, most recently Drupal 6 Javascript and jQuery and an Open Source project called QueryPath for PHP that allows developers to easily build applications from XML data with jQuery like functionality.

Matt's personal blog is at TechnoSophos.com

I saw into the future. At 10:00 AM PDT on October 25th, in a conference room seating 16 people at the Googleplex in Santa Clara, I saw into the future of the web. And it was good.

What I saw was the demise of Web 2.0, a technology grown to capacity. And it is not Web 3.0 (whatever that is) that will take its place. No, tomorrow's web is about user interfaces.

The weekend of October 24th was the annual Google Summer of Code (GSOC) Mentor Summit at Google's headquarters. Ostensibly, this is the opportunity for all of the Open Source organizations who participated in GSOC to get together and perform a collective postmortem on the summer's successes and failures. But anytime such a menagerie of geeks is assembled under one roof, much more is bound to happen.

Unsurprisingly, many of the unconference's sessions were focused on the GSOC program itself. But a healthy dose of technology centered sessions made their way onto the schedule as well, and the hallway may very well have seen more code than the conference rooms.

Beyond the physical conference space, much was happening in the virtual sphere as well. As a gesture of thanks to the GSOC participants and mentors for a summer of work, Google gave everyone Wave accounts.

Wave's utility lies in numbers. Signing in without a friend is like throwing a party but inviting nobody. Bring a friend or two into the Wave, and it feels like hosting tea in a room with too much furniture. But once the numbers start to rise, Wave's strengths surface. It is a cocktail party that comes complete with a birds-eye view of all of the chit-chat. Conversations swirl around, splintering into smaller threads of conversation only to merge back into the main discussion later. Images, maps, polls, and an API for building extensions make Wave a promising tool except for one thing.

The user interface stinks.

Yes, Wave's merits surface only when many people are in a discussion. Unfortunately, that's also where the big shortcomings surface. As one conversation forks into many smaller discussions, the wave quickly becomes visually unmanageable. The Quickest Scrollwheel in the West will still have a hard time traversing the continuously growing vertical pane that wraps the conversation. The entire advantage gained by the birds eye view of the conversation is lost to clumsy scrolling.

But this failing is indicative of something greater. Once again, Google has achieved an engineering masterpiece. And for all technical purposes, Wave is a marvel. It certainly pushes AJAX and asynchronous web interaction to its very limits, and I have no doubt that the source code for the server component would make my head swim. But the user interface, for all its visual business, simply doesn't work. Wave is an attempt to cram the internals of a Hummer into the body of a circa 1996 Honda Civic.

This is where the Mentor Summit offered a revelation.

On the second day of the summit, the PyMT team hosted an hour-long session on multi-touch input. PyMT is a set of Python application bindings for various multi-touch libraries. Linux, Windows, and OS X all support multi-touch input technologies. Last week, Apple's new Magic Mouse made its debut featuring a multi-touch surface atop a traditional laser-based mouse. Dell offers a laptop with a multi-touch screen. Wacom offers a multi-touch tablet. Microsoft's Surface technology boasts table-sized multi-touch surfaces.

Multi-touch is arriving in a big way. But what's the hubub about? What's the big feature that suddenly makes these technologies attractive? It is the extension of point and click to touch, tap, pinch, swipe, expand, drag, rotate, throw. The simple mouse model that has driven graphical interaction for decades is mid-way through an extreme makeover. And with deflection- and pressure-sensing surfaces rapidly advancing, mice and fingers are just the tip of the input iceberg.

The PyMT team took an hour-long trip to Google's hands-on room and came back with an impressive demonstration. Beginning with some Lego pieces, a flat sheet of metal, and a PlayStation 3 camera, the pair of programmers from PyMT built a gaming surface and a couple of paddles. The camera tracked the motion of the paddles on the surface, transforming the physical paddles into virtual ones in a game of Pong projected onto the wall. In an hour!

When we can build new input devices with an Open Source library and $35 worth of corner-drug-store toys, a whole new universe of possibilities appears.

And that's where Wave returns to mind. As I walked out of the room with UX Design guru Steve Fisher, he turned to me an remarked, When I watch a demo like this, it makes me wonder What is the Web going to look like in a few years. Yeah, it makes me wonder, too.

Wave is a fantastic architecture. But the architectural gems are obscured behind yesterday's user interface limitations. In that way, Wave is a milestone that marks the death of Web 2.0. And it is more than that. It's a fingerpost pointing not to the technologies touted as Web 3.0, but toward a new mode of human-computer interaction. What is going to make tomorrow's web compelling? Not metadata. Not cleaner layout. Not even native support for videos. Better user interfaces. Interfaces tuned to convey information more effectively. Reactive interfaces. That's where tomorrow's success stories are waiting.

HTML 5 and RDFa are good and all, but the real sea-change is coming from your fingertips. All ten of them.

DISCLOSURE OF MATERIAL CONNECTION: http://cmp.ly/4

The Demise of Web 2.0 is a post from: TechStartups.com

You can only neglect your users for so long before they leave you. The same goes for lovers. And I, was in love with Google Reader once.

Then one day I realized Google Reader wasn't loving me as much as I loved her. So I walked out. We see each other, maybe once a month or so, for short periods of time while we trade feeds.

Reasons

At first she didn't accept my privacy when I wanted to put authenticated feeds in the system. Then she wouldn't allow me to have my shared items back when I asked for them. I wanted all of them . . . but she only gave me 20 at a time. What about the thousands of items I had shared with her? Enough. I couldn't take it any longer. I had to do something about it.

Solution

Two years ago I began caching my Google Reader shared feed so I could access to all the items. Back then Reader hadn't added any features around sharing like search. But I had it once I was storing the items. What I had created then was a strange knowledge base that I could now query to find content that I found of value and had filtered for myself.

Value

Quickly, I added about 10 more shared feeds from friends and other people that I respected online. After about two weeks I had forgotten about Google Reader and found myself hanging out with my new friends . . . well, hanging with their knowledge bases. Here I was capturing tacit knowledge from some of the people I respected the most for their minds.

Filtering

What was now stored for my querying pleasure was content from the best publishers on the planet filtered by the sharpest people I could find. I began to build other tools around the data like grouping by publisher, sharer, keywords and gobs of new feeds.

I built tracking around it to see how robots traversed the feeds since the actual data was locked in a password protected site. Which turned out not to be that big of a deal since the title links were directed back to the publishers.

By adding new user controlled filtering mechanisms on top of pre-filtered data that was pouring into the system, it became much easier to produce pages and feeds for topics that interested me.

Here are a couple of feeds to demonstrate what I am talking about:

Keyword: http://www.filome.com/key/1/micropayments.rss
Group: http://www.filome.com/group/ksmith/1/Taminania_Brain_Science.rss
Sharer: http://www.filome.com/1/robdiana.rss
Likes: http://www.filome.com/likes/1/08100556675301148205.rss

What Reader is Doing Now

Since we broke up, she's been adding features but they are all at the feed level and not down to the individual publisher post level. The Bundles that she allows you to create are feed based. Instead of receiving 1 or 2 items of an interesting topic from a few publishers you get all the items in those feeds . . . many more than 2. It's like being in a forest and finding a twig that you want and your date cuts down 10 trees and hands them to you saying, Look, I got your twig. Happy now?

Where Reader is Going

For fear that they are going to be crushed by Facebook and Twitter, Google appears to have put some emphasis on the Reader team and either given the resources or freedom to improve the system. I would even venture to say that members of the Blogger team might be instrumental in some of these improvements. She should look to her friends for support during difficult times.

Where Reader Should Be

I'm gonna break it down nice and easy.

1. Grouping content at the individual post level
2. Feeds for everything
3. Portability of all shared/liked items from the day a user signs up
4. Content shopping cart
5. New UI 85% of web users don't read feeds. Get pretty.
6. Open up as a hub for syndication
7. Give publishers real metrics about subscribers, sharers and likers
8. Allow publishers to create community around these users (within Google)

I am still in love, but I have better things to do

My favorite part of her is still the shared feeds. Like perfume they remain long after she has left the room. They add value to the ecosystem and to the lives of those that have access to them. And since Google Reader is sitting on this massive mountain of filtered and expertly curated data, they should use it as their greatest asset in the coming walk-off with Facebook and Twitter.

DISCLOSURE OF MATERIAL CONNECTION: http://cmp.ly/0

Post from: TechStartups.com

Tweet This Post

Ownership matters, because otherwise Blizzard and other software vendors can wipe away important consumer rights with legalese contained in license agreements. For example, in Section 117 of the Copyright Act, Congress gave owners of computer software the right to use their legitimately purchased software without having to rely on permissions in license agreements. Blizzard and other software vendors are arguing that customers are not owners, but mere licensees, in an effort to eliminate our rights under Section 117.

This "owner-versus-licensee" trick is not just an end-run on Section 117, it's inconsistent with the law in other areas--the courts and Congress have long rejected efforts by copyright and patent owners to impose all kinds of post-sale use restrictions on books, patented machines, and compact discs. Why should software be different? Just as with those other copyrighted works, if you bought the disc that the software comes on outright (as opposed to leasing it, for example), you should get the privileges of an owner (i.e., the right to resell and the right to make copies and adaptations as necessary to use software).

In short, Blizzard's legal arguments here are all about using copyright law to take away consumers' rights in the software they purchased.

Photo by Joel Zimmer.

The key, according to tips compendium WikiHow, is to reuse as many common bases as possible. So "a roasted chicken can be eaten as straight chicken with side dishes (e.g., mashed potatoes and veggies) the first day, contribute to a chicken skillet, and wind up as the basis for soup." The corollary of the solo sequence technique is to properly freeze the remainders.

According to the post, if you're dining solo, you should ideally freeze food in one-person portions, meaning that while you're free to buy in bulk, you should divide the pre-cooked goodies into "half-pound or smaller pouches before you freeze it."

Browse the post for other tips on how to cook for one, and if you've got a lot of experience preparing solo meals, chime in with your own advice on how to best do so in the comments.

If you're on Twitter, definitely follow me at twitter.com/joshtpm. I keep you updated on the best offerings of the day from our news blog as well as give updates on new stuff we have coming from TPM -- new features, hires, stories we're going to start digging in on, etc.

I'd also strongly encourage you to follow the main TPMmedia twitter feed which we've just relaunched -- ditching the old extremely lame RSS-based system and placing under new bona-fide homo sapiens management.

We've all been there. You climb into bed, you turn off the nightstand lamp, and you think "Did I close the garage door?" Save yourself the midnight trip to check on the garage with this ingenious hack.

Instructables user kcj2010 hated peeling himself out of bed to double check on the garage door and hated waking up in the morning realizing he'd left it open even more. His solution was to wire a sensor to his garage door and place an indicator light hidden in the light fixture of the master bedroom.

When the garage is open, a small red LED shines on the light fixture and gives an immediate and highly visible indicator that the garage door is open. No light? The garage is closed. Seems like an idea that could be applied to any swinging or sliding door that causes you concern. Check out the Instructable for the parts list and wiring diagram. Sound off in the comments if you've undertaken a novel hack of your own to solve a problem around the house.

Database normalization is a technique for designing relational database schemas that ensures that the data is optimal for ad-hoc querying and that modifications such as deletion or insertion of data does not lead to data inconsistency. Database denormalization is the process of optimizing your database for reads by creating redundant data. A consequence of denormalization is that insertions or deletions could cause data inconsistency if not uniformly applied to all redundant copies of the data within the database.

Why Denormalize Your Database?

Today, lots of Web applications have "social" features. A consequence of this is that whenever I look at content or a user in that service, there is always additional content from other users that also needs to be pulled in to page. When you visit the typical profile on a social network like Facebook or MySpace, data for all the people that are friends with that user needs to be pulled in. Or when you visit a shared bookmark on del.icio.us you need data for all the users who have tagged and bookmarked that URL as well. Performing a query across the entire user base for "all the users who are friends with Robert Scoble" or "all the users who have bookmarked this blog link" is expensive even with caching. It is orders of magnitude faster to return the data if it is precalculated and all written to the same place.

This is optimizes your reads at the cost of incurring more writes to the system. It also means that you'll end up with redundant data because there will be multiple copies of some amount of user data as we try to ensure the locality of data.

A good example of a Web application deciding to make this trade off is the recent post on the Digg Blog entitled Looking to the Future with Cassandra which contains the following excerpt

The Problem

In both models, we're computing the intersection of two sets:

1. Users who dugg an item.
2. Users that have befriended the digger.

The Relational Model

The schema for this information in MySQL is:

CREATE TABLE `Diggs` ( `id` INT(11), `itemid` INT(11), `userid` INT(11),
`digdate` DATETIME, PRIMARY KEY (`id`), KEY `user` (`userid`), KEY `item` (`itemid`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;   CREATE TABLE `Friends` ( `id` INT(10)
AUTO_INCREMENT, `userid` INT(10), `username` VARCHAR(15), `friendid` INT(10), `friendname`
VARCHAR(15), `mutual` TINYINT(1), `date_created` DATETIME, PRIMARY KEY (`id`), UNIQUE
KEY `Friend_unique` (`userid`,`friendid`), KEY `Friend_friend` (`friendid`) ) ENGINE=InnoDB
DEFAULT CHARSET=utf8;

The Friends table contains many million rows, while Diggs holds hundreds of millions. Computing the intersection with a JOIN is much too slow in MySQL, so we have to do it in PHP. The steps are:

1. Query Friends for all my friends. With a cold cache, this takes around 1.5 seconds to complete.
2. Query Diggs for any diggs of a specific item by a user in the set of friend user IDs. This query is enormous, and looks something like:

   SELECT
   `digdate`, `id` FROM `Diggs` WHERE `userid` IN (59, 9006, 15989, 16045, 29183, 30220,
   62511, 75212, 79006) AND itemid = 13084479 ORDER BY `digdate` DESC, `id` DESC LIMIT
   4;

   The real query is actually much worse than this, since the IN clause contains every friend of the user, and this can balloon to hundreds of user IDs. A full query can actually clock in at 1.5kb, which is many times larger than the actual data we want. With a cold cache, this query can take 14 seconds to execute.

Of course, both queries are cached, but due to the user-specific nature of this data, it doesn't help much.

The solution the Digg development team went with was to denormalize the data. They also went an additional step and decided that since the data was no longer being kept in a relational manner there was no point in using a traditional relational database (i.e. MySQL) and instead they migrated to a non-RDBMS technology to solve this problem.

How Denormalization Changes Your Application

There are a number of things to keep in mind once you choose to denormalize your data including

1. Denormalization means data redundancy which translates to significantly increased storage costs. The fully denormalized data set from the Digg exampled ended up being 3 terabytes of information. It is typical for developers to underestimate the data bloat that occurs once data is denormalized.

2. Fixing data inconsistency is now the job of the application. Let's say each user has a list of the user names of all of their friends. What happens when one of these users changes their user name? In a normalized database that is a simple UPDATE query to change a single piece of data and then it will be current everywhere it is shown on the site. In a denormalized database, there now has to be a mechanism for fixing up this name in all of the dozens, hundreds or thousands of places it appears. Most services that create denormalized databases have "fixup" jobs that are constantly running on the database to fix such inconsistencies.

The No-SQL Movement vs. Abusing Relational Databases for Fun & Profit

If you're a web developer interested in building large scale applications, it doesn't take long in reading the various best practices on getting Web applications to scale such as practicing database sharding or eschewing transactions before it begins to sound like all the advice you are getting is about ignoring or abusing the key features that define a modern relational database system. Taken to its logical extreme all you really need is a key<->value or tuple store that supports some level of query functionality and has decent persistence semantics. Thus the NoSQL movement was borne.

The No-SQL movement is a used to describe the increasing usage of non-relational databases among Web developers. This approach has initially pioneered by large scale Web companies like Facebook (Cassandra), Amazon (Dynamo) & Google (BigTable) but now is finding its way down to smaller sites like Digg. Unlike relational databases, there is a yet to be a solid technical definition of what it means for a product to be a "NoSQL" database aside from the fact that it isn't a relational database. Commonalities include lack of fixed schemas {TODO}. Below is a list of some of the more popular NoSQL databases that you can try today along with a brief description of their key qualities

1. CouchDB: A document-oriented database where documents can be thought of as JSON/JavaScript objects. Creation, retrieval, update and deletion (CRUD) operations are performed via a RESTful API and support ACID properties. Rich querying is handled by creating Javascript functions called "Views" which can operate on the documents in the database via Map/Reduce style queries. Usage: Although popular among the geek set most users seem to be dabblers as opposed to large scale web companies.

2. Cassandra: A key-value store where each key-value pair comes with a timestamp and can be grouped together into a column family (i.e. a table). There is also a notion of super columns which are columns that contain whose values are a list of other key-value pairs. Cassandra is optimized to be always writable and uses eventual consistency to deal with the conflicts that inevitably occur when a distributed system aims to be always writable yet node failure is a fact of life. Querying is available via the Cassandra Thrift API and supports fairly basic data retrieval operations based on key values and column names. Usage: Originally developed and still used at Facebook today. Digg and Rackspace are the most recent big name adopters.

3. Voldemort: Very similar to Cassandra which is unsurprising since they are both inspired by Amazon's Dynamo. Voldemort is a key-value store where each key value pair comes with a timestamp and eventual consistency is used to address write anomalies. Values can contain a list of further key value pairs. Data access involves creation, retrieval and deletion of serialized objects whose format can be one of JSON, strings, binary BLOBs, serialized Java objects and Google Protocol Buffers. Rich querying is non-existent, simple get and put operations are all that exist. Usage: Originally developed and still used at LinkedIn.

There are a number of other interesting NoSQL databases such as HBase, MongoDB and Dynomite but the three above seem to be the most mature from my initial analysis. In general, most of them seem to be a clone of BigTable, Dynamo or some amalgam of ideas from both papers. The most original so far has been CouchDB.

An alternative to betting on a speculative database technologies at varying levels of maturity is to misuse an existing mature relational database product. As mentioned earlier, many large scale sites use relational databases but eschew relational features such as transactions and joins to achieve scalability. Some developers have even taken that practice to an extreme and built schema-less data models on top of traditional relational database. A great example of this How FriendFeed uses MySQL to store schema-less data which is a blog post excerpted below

Lots of projects exist designed to tackle the problem storing data with flexible schemas and building new indexes on the fly (e.g., CouchDB). However, none of them seemed widely-used enough by large sites to inspire confidence. In the tests we read about and ran ourselves, none of the projects were stable or battle-tested enough for our needs (see this somewhat outdated article on CouchDB, for example). MySQL works. It doesn't corrupt data. Replication works. We understand its limitations already. We like MySQL for storage, just not RDBMS usage patterns.

After some deliberation, we decided to implement a "schema-less" storage system on top of MySQL rather than use a completely new storage system.

Our datastore stores schema-less bags of properties (e.g., JSON objects or Python dictionaries). The only required property of stored entities is id, a 16-byte UUID. The rest of the entity is opaque as far as the datastore is concerned. We can change the "schema" simply by storing new properties.

In MySQL, our entities are stored in a table that looks like this:

CREATE TABLE entities ( added_id INT NOT NULL AUTO_INCREMENT PRIMARY
KEY, id BINARY(16) NOT NULL, updated TIMESTAMP NOT NULL, body MEDIUMBLOB, UNIQUE KEY
(id), KEY (updated) ) ENGINE=InnoDB; 

The added_id column is present because InnoDB stores data rows physically in primary key order. The AUTO_INCREMENT primary key ensures new entities are written sequentially on disk after old entities, which helps for both read and write locality (new entities tend to be read more frequently than old entities since FriendFeed pages are ordered reverse-chronologically). Entity bodies are stored as zlib-compressed, pickled Python dictionaries.

Now that the FriendFeed team works at Facebook I suspect they'll end up deciding that a NoSQL database that has solved a good story around replication and fault tolerance is more amenable to solving the problem of building a schema-less database than storing key<->value pairs in a SQL database where the value is a serialized Python object.

As a Web developer it's always a good idea to know what the current practices are in the industry even if they seem a bit too crazy to adoptyet.

Further Reading

• How FriendFeed Uses MySQL to Store Schema-less Data by Bret Taylor
• Looking to the future with Cassandra Digg Blog

Now Playing: Jay-Z - Run This Town (feat. Rihanna & Kanye West)

Take subscription services for example. Sure, I'd love a service that would allow me to download unlimited content in high bitrate MP3 format for a reasonable fee every month. Except economics and greed will never let that happen.