Meanwhile, Facebook gets the capital to keep buying talent and startups, and to fuel its expansion in all kinds of other ways — and it gets to sell stock in what amounts to a shadow stock market that’s growing faster than regulators seem willing or able to understand, much less deal with.

This looks like a better deal for Facebook than its investor, putting Facebook’s value at $50 billion, which makes sense in today’s increasingly bubble-like market. Silicon Valley is going a bit wild again– not as crazy as the late 1990s, mind you, but there’s a froth element to the local economy.

An interesting question now is whether Facebook will do a a real public offering anytime soon. Federal rules require significant data disclosures when a company has 500 or more shareholders, and surely Facebook is at that point or nearing it. The Goldman deal may be an end-run around the rule, with Goldman not selling Facebook shares to its clients, but rather selling shares in something it (Goldman) owns. If this is the game, and if the SEC lets it happen, the 500-shareholder rule has become meaningless — and markets are all the more opaque at a time when transparency is more needed than ever.

Opacity is a growing issue. A thriving shadow marketplace has emerged for big startups that haven’t done IPOs, so big that the Securities and Exchange Commission is, at least in that space, looking into the wheeling and dealing. For good reason: Many if not most of the investors in these markets have no idea what the true financial picture may be of the shares they’re buying.

Facebook seems like a no-brainer right now. It reportedly has passed Google as the most visited website, and it’s growing in power and people. And Goldman, for all its sleazy ways, has smart people making investment decisions.

But Goldman was also a big investor in the financial bubble that nearly toppled the global economy. It escaped ruin only because we, the taxpayers (actually our children and grandchildren) rescued it and the rest of the banking industry. That was and remains Goldman’s real genius: making giant bets with other people’s money.

But the most important result may ultimately be what this move, among others. does for public understanding of the role of Internet service providers of all kinds. As Amazon.com’s recent takedown of the Wikileaks site it was hosting demonstrates, we are at the whims of the companies that provide the services, and they are increasingly demonstrating that we should be highly skeptical about their commitment to our data’s longevity.

We put our data — our websites, photos, bookmarks, email and more — on their sites. But they can, and do, change their terms of service at will, doing what they please with what we’ve put on their servers. And sometimes they just shut down the services they’ve been providing. They may do it for good reasons, or absurd ones. It doesn’t matter. The point is, they can.

As noted here some months ago, we all need a Plan B for just about everything we do online these days. If we give others a choke point over our communications, we are inviting them to throttle us.

Note: The original version of this piece said Yahoo was closing Delicious. That was based on a variety of credible — and, as noted, unrefuted — news stories that started appearing more than 24 hours ago. They were based, initially, on a Twitter posting that linked to a screenshot taken at an internal Yahoo meeting. The screenshot, which has now been taken down, had Delicious among a group of Yahoo services that were being “sunsetted,” which is corporatese for end of life.

Whatever Yahoo’s intentions with Delicious, my points here stand. Even if the service is sold, a new owner might radically change the terms of service (as Yahoo itself could do at any time). The users’ insecurity remains, whatever the ownership may be.

These changes appear to make plenty of sense for Netflix, because the company will avoid the cost of buying and then mailing the millions of DVDs customers like me have been receiving. And, indeed, on Monday Netflix announced it was going to offer customers an all-online streaming plan for $8 a month.

I suspect there’s been a misstep, however, if I’m any example of the Netflix customer base. I’d been paying $17 per month for a plan that allowed us to have three DVDs out at a time, plus being able to view streaming content anytime. But Neflix has raised our rate by $3 a month, or about 18 percent.

There are way too many problems with the streaming-only plan to even consider it at this point. At the top of the list is the fact that the Netflix catalog of DVDs is vastly, vastly greater than what you can watch online. If the company really wants to be a streaming-only outfit, it needs to persuade the robber barons of Hollywood to digitize everything sooner rather than later.

And while the quality of the streaming is generally OK if you have a fast enough broadband connection — though it doesn’t look as good on my computer as a DVD — network congestion (in my experience) can cause the video to degrade in quality or, in some circumstances, pause altogether. I tried it on a hotel Wi-Fi recently, and finally gave up as the film kept stopping while the stream caught up.

So when the e-mail arrived announcing the price hike, my reaction was: Sorry, no sale. We’ve moved to a lower-cost plan that allows one DVD out at a time, for $10 (also more expensive than that plan used to be), plus streaming. The various plans Netflix offers now range up to $56 a month, and slightly more if you’re renting Blu-ray discs.

Netflix has leveraged the broadband Internet structure like no other company. It now accounts for a significant amount of evening data traffic, by all accounts. I’m guessing that heavy Netflix users are going to pay for the money they save in other ways when they start running into data caps that some carriers have put on their basic Internet service.

Wall Street was thrilled with the latest Netflix maneuver, pushing the stock price way up on Monday (though it eased off slightly this morning). The share price has roughly quadrupled in the past year — evidence of investors’ love for the company, an infatuation I believe has been mostly justified.

But I’m convinced that this move by Netflix is too little, too soon. And I’m betting I’m not the only one who feels that way.

The complaints grew loud enough to persuade Google to revise the Gmail settings so users can turn off conversation view and unravel their messages.

“We really hoped everyone would learn to love conversation view, but we came to realize that it’s just not right for some people,” Google software engineer Doug Chen wrote in a Wednesday blog post.

The aversion to conversation view doesn’t seem to be widespread. Gmail ended July with nearly 186 million worldwide users, a 22 percent increase from the same time a year ago, according to the research firm comScore Inc. Both Microsoft’s Windows Live Hotmail (nearly 346 million users) and Yahoo’s e-mail (303 million users) are larger, but aren’t growing nearly as rapidly as Gmail.

As speakers like Google’s Eric Schmidt held forth on the conference stage, buzz in the conference center was about the confirmation of rumors that Arrington has sold the operation, which includes what appears to be a highly profitable conference arm, to AOL. The terms weren’t disclosed.

AOL has been bulking up on content for some time now. A few years ago it bought a collection of blogs, including the popular tech blog Engadget, and it’s investing what looks like serious money in Patch, one of many competitors in the local-news site category that’s growing like topsy.

Arrington started out in Silicon Valley as a lawyer. With TechCrunch he created a record of pugnacious journalism and advocacy. Rumors were as much the rule as nailed-down facts, but he got enough that was right often enough and early enough to be a major force in the genre.

TechCrunch sites employ some excellent journalists. But I wonder how the overall operation will work if Arrington leaves (I assume he has a lock-up period where he has to stay before he can fully get his money out of the deal). Some media companies can do just fine without their founders; I suspect this isn’t one of them.