I've been saying this for a while now (http://news.ycombinator.com/item?id=3041689). Facebook's engagement is shifting to mobile. Their entire business model relies on web. Their app ecosystem has been crashing as a result, and display advertising is likely to follow.
Facebook is filled with smart people. They understand the trends better than we do. As you'll note, 4 of their top 5 Risk Factors relate to the web-to-mobile shift and the problems inherent with that.
From the filing:
#2 - "We generate a substantial majority of our revenue from advertising. The loss of advertisers, or reduction in spending by advertisers with Facebook, could seriously harm our business;"
#3 - "Growth in use of Facebook through our mobile products, where we do not currently display ads, as a substitute for use on personal computers may negatively affect our revenue and financial results;"
#4 - "Facebook user growth and engagement on mobile devices depend upon effective operation with mobile operating systems, networks, and standards that we do not control;"
#5 - "We may not be successful in our efforts to grow and further monetize the Facebook Platform;"
> "Most of the insane growth (in profit as well as valuation) has already been squeezed out."
I would amend that to say: most of the easy growth has been squeezed out. But there's still a huge opportunity for Facebook to grow revenue/profit if they switch to using personal data to make well-directed ads they can push to sites using Facebook Connect and/or if they can roll out a solid first-party Facebook-phone. 
Those are both far trickier than "grow Facebook", so it does make sense to capture some returns now. But I think it would be short sighted to write their future prospects off entirely.
 They could quite possibly fork Android and do to Google in the phone market what Amazon is doing to the tablet market. And they'd be crazy not to try.
I would agree with you here. Most of the low-hanging fruit has been plucked. But that doesn't mean the low-hanging fruit is the juiciest. As I mentioned in the grandparent comment, and as I agree with you, Facebook Connect is the future of Facebook. Facebook seems to understand that fact, and now it's just a matter of how keenly it does.
Facebook has (wisely) focused almost entirely on user growth and retention over the last few years. That it stumbled into being an attractive display-advertising platform was almost inevitable, and required little effort. The real money lies on the distributed mobile and web ecosystem, into which Facebook has been diligently inserting itself as a de facto credential and backend. (Think of Facebook as the "Intel Inside" of the entire content web and mobile content web, and you get the idea).
Facebook has established an impressive beachead with FB Connect, and from there, it's only a matter of time before it can monetize its FB Connect partners through contextual and interest-based advertising services. Google should be very concerned -- and, to that end, it would not surprise me in the least if the threat of FB Connect to AdSense (and, to some extent, AdWords) was a significant factor in the race to develop Google+.
It's unlikely that Facebook is trying to time their IPO at the height of their potential. I think they still have a lot of room to grow the display advertising business, but unfortunately there is a shift to mobile and the Facebook mobile UX isn't that great. I think they need to continue to worry about building products users love and improve mobile UX, advertisers will always come to where the users are.
I like that they point this out, but in theory this goes for the desktop as well. I guess the difference and reliability between the desktop OS/browser platform is seen as more established than the mobile platforms. This makes sense right now, but may well be something that grows over time into less of an issue.
I'm wondering, this might seem absolutely ridiculous, but would Facebook not benefit from having their own mobile operating system, even if its based off Android. Services are no longer apps or on the web but part of the system. Phone numbers are obsolete, you would phone facebook contacts as you would on skype. Such a system could perhaps offer better advertising capabilities and offer another competitor against Google and Apple on the mobile landscape..
Thats true, but I think there could be a way around that if we had a domain name service for telephone numbers whereby instead of having to remember a 12-10 digit number with area codes for someone, we only need to remember a word or sentence and since a string (lower alphabet only) is base 26 it might be easier. Also could we not run out of phone numbers anyway like with ipv4?
The people will still use the Desktop like before. The problem is, for messaging and status updates is mobile totally sufficient.
They talk much about Pictures, but most of the people i know don't post any. The big success of Facebook is the E-Mail IM replacement for young people, but it is very difficult to advertise on that. Who made ever money on IM and Mail? ICQ and AIM had impressive numbers, but were never a big success financially.
People in your social network may not post pictures, but for all the Facebook users I know, pictures is very critical part of their experience. [I personally don't use Facebook.] Facebook has actually solved the problem of sharing pictures with your loved ones for non technical people, IMHO.
Funny i don't use it myself, i get an email when someone sends me a message.
I know also people who share them, but max every week and probably just 5 friend really look at them. When you look at pictures, how can you concentrate on ads? They are not even in the middle like Youtube's. Pictures on mobile reduces the space for ads even further and you will always be uglier than every competitor.
I do not have any opinion about ads, I was only replying to your comment that no one shares pictures. Since Facebook, all the non technical folks I know have stopped using Flickr/Google Albums/Snapfish/Photobucket/<<Name Your favorite photo sharing Service>>. They also don't send pictures by email. They simply upload it to Facebook.
Ad serving (at least the monetizing part) isn't always about how you can make a user concentrate on an ad. How many free iOS games out there serve up unobtrusive ads that you sometimes even stop noticing?
>They talk much about Pictures, but most of the people i know don't post any.
I'm not really sure that applies universally. The majority of my stream post pics from mobile (either directly, thru instagram, etc.).
>The big success of Facebook is the E-Mail IM replacement for young people...
In mobile? I don't see that happening either, as most too used to texting/iMessage/BBM
Where I do see a lot of usage is in-stream. If they can figure out how to serve up ads within the timeline without being too intrusive then that would be the way to do it in mobile. Tho that new timeline crashes more often than it works in mobile (at least in the iOS version)
Well i am certainly not an expert, but none of your examples get's people to use Facebook on the desktop.
What is the benefit of looking at someones timeline and how often will you do it? Facebook needs incredible engagement (what it has right now on the desktop) and space do display ads. Youtube has the content, but what brings people back to FB? Don't forget all these social Startups will hurt Facebook on engagement.
A Facebook mobile ad network would be huge. Mobile publishers are already suffering with low eCPMs, so they are hungry for new solutions. Showing brands/banners is one thing, but showing mobile app ads is another.
Since a lot of mobile apps have some type of Facebook integration (ie: a Facebook app id), Facebook has a good idea of who has what apps installed and they can serve up apps that are relevant to the user both on the interest graph and the data of what other mobile apps they have installed.
Wouldn't be surprised to see them move into mobile advertising hard in 2012/2013.
* Disclaimer: This is all speculation, no insider information was divulged