FB users have high switching costs in terms of opportunity costs (by leaving FB they forego value) and those opportunities costs are large mostly because of network effects.

In the case of Microsoft, you say "network effects of their enterprise sales channel," when you really mean Microsoft had a large customer base to which they could quickly and easily deploy new products. The fact that one business had a sales agreement with Microsoft didn't necessarily increase the value of a sales agreement with Microsoft for a different business. (Maybe there are some network effects there, but they aren't obvious and I'm fairly sure you aren't talking about them.)

I think Microsoft did (and still does) benefit from network effects in a different way, related to collaboration: the more people who can open and use an excel spreadsheet, the more valuable it is to learn and use excel.

- Positive vs. Negative

- Same-side vs. cross-side

Network effects can be negative: if more people dine at your favourite restaurant you have to wait longer for a table. That's a negative same-side network effect. The reverse is true for facebook: more people signing up means more people to socialise with.

When you add in app developers, then cross-side effects come in to play. More people using facebook is a positive cross-side network effect, because more users == bigger market for developer apps. It's mutually positive too: more developers means more apps, more choice etc. For advertisers, the cross-side network externality is only positive from their perspective: they like having a larger audience to advertise to, whereas users generally don't like ads.

All of these combine to make facebook one hell of a sticky, and lucrative platform.

For example, they downgrade YouTube videos in people's feeds. There's no reason for that if they have a better product.

Which is the case with windows. Because the product has more users it becomes more valuable to developers/business, which then becomes more valuable to users.

95% of the office workers in the developed world are still sitting in front of a Windows PC, logging into an AD, doing stuff with Office, etc. Every significant business has an EA or Select Agreement.

Big tech companies will only acquire companies that bring something to the table that they can't get by locking 4 of their own engineers in a closet for a year. That means a brand or patents. Any software-based tech can be copied for around a million bucks without much issue.

The other thing here is that in cyberspace, the effects of the network are enforced only by artificial legal barriers like the CFAA that make it a felony for users to contact web sites in a manner that the operator dislikes. The CFAA essentially takes the ToS (and sometimes even less; just an operator's expression that he no longer wants you to access the site can suffice the CFAA's standard of "exceeding authorized access") and evals it as part of the criminal code. This is the law under which Aaron Swartz was prosecuted.

Facebook has invoked this power repeatedly, most notably against Power Ventures, which was destroyed in court for allowing users to export their own data only, the data in which they have an exclusive copyright interest. Power infringed Facebook's copyright because it downloaded a page into RAM (RAM copies that exist for milliseconds are eligible for copyright protection per MAI v. Peak) that contained Facebook's IP in order to extract the user's information.

It's why you never see third-party exporters and have to rely on the crappy ones that Facebook/Google offer and bury so they can avoid flak.

Things would look much different if our law wasn't so thoroughly committed to entrenching the extant digital players. There's no reason that another application should be legally restrained from multiplexing/multicasting one's Facebook feed.

IANAL.

However, they're ALSO preventing a bunch of old farts like myself from signing up for Snapchat accounts so I can put mustaches on every car that drives by, and preserving it as a "cool" haven for "kids these days," thereby increasing Snapchat's value to its existing userbase.

They're really concerned about the amount of information these big data stores have on them.

They're worried even more that there is a government snoop checking in on them or another unknown.

Personally, I have gotten more and more leery of putting data online. I deleted my Facebook 7 years ago. Twitter is strictly read-only for me. I share kid pics directly with family via private storage I control.

Really needs to be a peer-2-peer suite of "social" tools that cut out middlemen as much as possible.

Had she not been using it I would have written it off as confusing UI and "what's the point."

I avoid Facebook for many reasons. You have to be careful about who your posts are seen by. It's public by default. I have colleagues, family, and people I've met just once or twice on there. Any posts I make need to be inoffensive and anodyne. Scrolling down a Facebook feed fills me with FOMO, jealousy, rage, or other negative emotions. It's obvious they don't actually want to improve the core product. I feel used.

First of all, on human timescales (i.e. lifetimes) monopolies can be very long lived. The Standard Oil Trust was very destructive for decades, and though it might have collapsed due to natural causes, that destruction could have lasted for decades more. Which leads to the second problem: those long periods of stagnation are multiply destructive: not only do the victims pay a premium (which they could have spent on other goods), the monopoly's existence retards innovation.

The final irony in the case of Standard Oil is that is was worth more to John D Rockefeller after being broken up than it was when integrated. The government did everyone a service by breaking it up.

"In 1904, Standard controlled 91 percent of production and 85 percent of final sales. Most of its output was kerosene, of which 55 percent was exported around the world. After 1900 it did not try to force competitors out of business by underpricing them... Due to competition from other firms, their market share had gradually eroded to 70 percent by 1906 which was the year when the antitrust case was filed against Standard, and down to 64 percent by 1911 when Standard was ordered broken up"

sounds like if you stop trying to compete other companies will

Rockefeller paid ransom sums for decades to buy out competitors (most of which were junk operators at best) to keep them from tanking the price of oil by dumping new strikes onto the market (a very frequent occurrence). His entire goal was oriented at market share + stability. The history of oil prices back then, was truly horrific, it was a constant boom and bust cycle that tortured most companies in the oil sector. Once Standard Oil accumulated enough of the market, including distribution (not as often talked about, they overwhelmingly controlled distribution), it could heavily control market pricing without having to own all the market. Their dominance in distribution is arguably the single most important thing they accomplished after the refining consolidation. With that gatekeeper status, they were able to demand other firms they didn't own adhere to their pricing levels, or have their distribution access cut off. In that way, they figured out how to stabilize the market pricing, without having to continue to own ~90% of the refining market.

The Sherman Antitrust Act was passed in 1890 but it was a toothless tiger until Roosevelt assumed the office of president in 1901. Prior to then, the main case involving the act was successful in using it against labor unions. Following 1901, litigation broke up a number of companies. This is important.

Investigation at the time indicated that Standard Oil employed predatory pricing up until 1900. Following 1900, they resorted to a shell-game to obfuscate what was going on.

Amongst the first wave of anti-trust litigation targets were J.P. Morgan's holdings. He was cornering the railroad and steel industries successfully, the biggest coup being the formation of US Steel in 1901. The other crown jewel in the strategy was oil, namely Standard Oil, which operated synergistically with J.P. Morgan's holdings in rail and steel. During this time, we had characters like Elbert H. Gary, CEO of US Steel, who held cross-industry dinners aimed at setting uniform pricing for steel and providing management across businesses with consolidated positions against labour organization. While it had been growing at a steady clip beforehand, Standard Oil's value began to skyrocket as of 1901 with no real change in its core business practices.

The SCOTUS' exposition of the issue does a great job showing how Oil's monopoly was mediated in large part through Rail (the case also indicates that practices wrt pipelines were analogous to rail):

"the bill alleged that the combination and its members obtained large preferential rates and rebates in many and devious ways over their competitors from various railroad companies, and that by means of the advantage thus obtained many, if not virtually all, competitors were forced either to become members of the combination or were driven out of business"

Can't transport oil if you don't have rail cars. Can't build rail cars or rail if you can't make steel, and you can't run your trains or your steel plants without oil.

In 1902, the Roosevelt directed the AG to commence litigation to break-up of the J.P. Morgan headed railroad trust - Northern Securities Company. This matter was such a high priority that the case was resolved and reported through all levels of appeal in 1904. As a side note, this is mind-bendingly fast for competition litigation, even today.

With the threat of anti-trust litigation preventing outright predatory pricing and "competitive" rail back on the table steel and oil began to lose their moats. But "competitive" rail wasn't very competitive, and continued a legacy of providing obfuscated sweetheart deals to Standard. They got broken up 7 years later as a result. By the time steel's litigation hit in 1920, it had proper competition and did not get broken up.

Standard Oil was a monopoly for over 20 years. There are countless monopolies that have gone on that long or longer, only to be broken up by government eventually.

I love the free market, but between regulatory capture, legal costs, and aggressive pricing structures, monopolies can be absolutely unbeatable within the career lifetime of any person.

But it seems many aspects of capitalism lend themselves to winner take all mechanics, resulting in large companies (airliners and telecoms and entertainment companies) constantly merging and reducing competition. So without the FTC, we might have even more monopolies. Without these laws, Microsoft would be the only PC company (MS bailed apple out to keep some "competition" around) and Intel would be the only manufacturer of CPUs (likewise for Intel and AMD).

"Natural monopolies arise where the largest supplier in an industry, often the first supplier in a market, has an overwhelming cost advantage over other actual or potential competitors; this tends to be the case in industries where fixed costs predominate, creating economies of scale that are large in relation to the size of the market, as is the case in water and electricity services. The fixed cost of constructing a competing transmission network is so high, and the marginal cost of transmission for the incumbent so low, that it effectively bars potential competitors from the monopolist's market, acting as an early insurmountable barrier to entry into the market place."

There's a good argument to be made here they all software sub-markets are natural monopolies.

But we'll never know, since nobody else can reasonably enter the market.

No one except Zuckerberg, I'd think.

However, it raises an interesting question: how exactly would you break up an entity like facebook? It doesn't seem like a business that would lend itself to subdivision.

That way each competitor could offer its own interface and algorithm for filtering people's feeds, and show their own ads (or offer a competing business model). If users opted to have their posts on the competing site automatically sent to Facebook (so their Facebook-only friends could still see them) then Facebook would be free to data-mine those posts still.

The next question is, could someone code a browser extension or other tool which makes this possible without needing any changes from Facebook's side? If it would break the Facebook terms of service, would someone crowd-fund the anti-trust lawsuit to override those terms?

Then mandate interoperability with an open standard, like a phone network.

I don't think that makes much sense for breaking up something like Facebook.

Unlike a phone network, Facebook's service is not rooted in geography. Yes, it's users have homes and geographic locations, but those are just arbitrary account attributes, just like their name or if they like the Denver Broncos. You might as well have one post-breakup company start out with all the A-M accounts and the other the N-Z accounts.

If you had to divide up the users between N successor companies, I'd probably be best to do it based on friend-graph connectivity, with mandated interoperability between them. That might make the successors sticky enough to survive.

So your premise is true(ish), but the conclusion can't be drawn from it (and your comparison falls apart) because we don't need strict dependency on geography, only a (relatively strong) correlation.

Hell, the main benefit I see to instagram is that their geotagging improved immensely.

Network effects. Together everyone hates it; individually, there is nowhere else for people to go without losing the majority of your social network. Facebook has people; its software is utterly irrelevant to why people use it.

John Maynard Keynes

EDIT: And downvotes, too? I've been insulted, and I don't take kindly too it, yet I deserve to be downvoted? Thats ridiculous.

I don't have faith that the free market will solve any problem well if we just wait long enough though. Nor do I believe that it's wise to just sit around and refuse to fight injustice because the problem might solve itself out in 50 years.

In this specific case, I think the network effect of social networks (ie it only makes sense to use a given social network if the people you want to communicate with use it as well) is so strong that the unregulated free market is not a good model to solve that problem.

en.wikipedia.org/wiki/Facebook,_Inc._v._Power_Ventures,_Inc.

There was no scenario under which the electric car was going to take off until lithium batteries were radically cheaper and available at truly massive scale. Mass market consumers do not want to drive EVs that get 50 or 80 mpg equivalent and cost $50,000.

Tesla came along only when it was finally becoming possible to do a nice EV thanks to falling lithium battery prices, at a low enough price point that well off consumers could afford them. The rest is history.

Could you explain a little more? I have not heard any credible evidence to support this theory.

The current understanding of US antitrust laws is similar. The Justice Department pursues antitrust actions only if the end consumer is being hurt.

Not really. Snapchat is getting copied because its functionality enhances the strategic value of Facebook's core assets. There are plenty of other businesses that would be easy for Facebook to copy technologically, but doing so would undermine the value of Facebook's core assets. Basically Snapchat brought a knife to a gunfight, with the expected results.

Now, one could posit an argument to the effect that monopolies in areas that are highly subject to network effects are inherently anti-competitive. But AFAIK that would be a rather novel argument and it's hard to say what the remedy would even be.