The Social Operating System: A Reader

For the sake of my pal Valdis Krebs, I am collating a list of posts I’ve made in recent years on the idea of a social operating system. The basic notion:

Stowe Boyd, Rockmelt: Why The Social Browser Won’t Matter

The next generation of operating systems will be social at the core.We won’t be fooling with files and folders. We will be connecting with others, reading streams from our friends, and tossing observations and hopes and insights into the wake we leave behind, spreading out to all that think we matter.

So here’s some links to pieces I’ve written mentioning the idea:

Please send along any references to other people writing on the subject.

Apple, Twitter, And The Social OS

Mathew Ingram wonders — apparently based on some thoughts by Barry Ritholtz — whether Apple should spend $10B and buy Twitter:

Mathew Ingram, Should Apple buy Twitter?

Apple’s best effort by far at adding those kinds of social elements came when the company integrated Twitter at a deep — and for Apple, a fairly radical — level into the operating system on the iPhone and iPad (and even into its new desktop OS, OS-X Mountain Lion). Never before had Apple built support for a third-party service into its devices and software in such a fundamental way. This helped fuel rumors about an Apple acquisition, just as Ritholtz and others have used it to justify such a deal: if Apple wants to integrate Twitter so deeply, why not just acquire it so that it has full control?

The fact that Apple likes to control things from end-to-end is well known, which is just one of the reasons why the deep Twitter integration was a bit of a surprise. But does it really need to own Twitter in order to get the benefits of that integration? I don’t think so. It can get all the positive aspects of Twitter support without having to own the company — and it doesn’t have to worry about the hassle of maintaining a third-party service that is used for a wide variety of different purposes that Apple has no real interest in.

Not only that, but buying Twitter could actually harm Apple’s attempts to integrate more social aspects into its devices, because it would make it even less likely that the company would ever strike a similar deal with Facebook — something it has tried to do a number of times. It could be that Facebook has no intention of ever partnering with Apple, and the two may wind up becoming adversaries as their interests converge, but acquiring Twitter would likely remove any chance of the two ever working together in even a small way.

So, Mathew comes down pretty strongly on the negative side of a possible acquisition, but omits the long-range view: the next generation of operating systems will be social at the core.

Most of today’s operating systems are still based on 1990 thinking. They are based on WIMP (windows, icons, menus, pointer). They don’t know about the Web, so users have to move back and forth from their local store of docs and files to the cloud, a thousand times a day. And the biggest surprise of the Web has been the rise of social, which is supported on our computers through apps.

All of these limitations will be attacked in new operating systems, which will be web-aware, post-WIMP, and inherently social.

Apple is headed into a battle with Google, Facebook, and maybe Microsoft (Windows 8 looks pretty good), and one of the primary areas of contention will be building social primitives into the operating environment.

Google will build its social architecture in Android. Facebook will become more than just an app platform: it will become a mobile OS. Windows 9 or some future version will incorporate some approach to social. And iOS and Mac OS X have started to move this way by including Twitter in the mix, as a fundamental social protocol.

Apple should pay the $10B for Twitter, and make it into the social layer of its OSs, and as the social framework of its apps. For example, Ping in iTunes could be rewired to rely on Twitter, fixing its design as Barry Ritholz points out, and future social TV and second screen apps could be based on Twitter, as well, which makes sense because Twitter is the leading second screen app today. The coming battle for social TV will be hugely important, and Twitter really positions Apple in that space.

So, Mathew is being too conservative, because he thinks Apple may want to ‘work with’ Facebook in the future. But that can’t be where Apple is headed.

MyZone: A Next-Generation Online Social Network

Alireza Mahdian, John Black, Richard Han, Shivakant Mishra (Submitted on 24 Oct 2011)

This technical report considers the design of a social network that would address the shortcomings of the current ones, and identifies user privacy, security, and service availability as strong motivations that push the architecture of the proposed design to be distributed. We describe our design in detail and identify the property of resiliency as a key objective for the overall design philosophy.

We define the system goals, threat model, and trust model as part of the system model, and discuss the challenges in adapting such distributed frameworks to become highly available and highly resilient in potentially hostile environments. We propose a distributed solution to address these challenges based on a trust-based friendship model for replicating user profiles and disseminating messages, and examine how this approach builds upon prior work in distributed Peer-to-Peer (P2P) networks.

This is exactly the form that social operating systems will take: a distributed model (mediated by server-side replication) for peer-to-peer social relationships and messaging.

Facebook's iPad App Works Kind of Like an Operating System - Technology - The Atlantic Wire

Adam Clark Estes

It took months of waiting, a couple of false starts and a whole lot of speculation, but Facebook has finally launched an iPad app. However, the upgrade is much more than a tablet-friendly version of the website. Facebook is also carrying over its developer platform to mobile. This means that all of the slick new class of Facebook apps that Mark Zuckerberg announced a couple of weeks ago at the f8 developers conference will be more integrated into the mobile experience. The specific details are a little bit confusing at first as Facebook is spoon-feeding the functionality to users, but we can already tell: Facebook is starting to function like an independent mobile operating system.

[…]

As [Facebook’s Luke] Shepard explains [here], Facebook apps will now be fully integrated into the mobile experience as their own apps within the Facebook app. If you receive a notification or request from a friend in a compatible app, tapping the update will switch to the mobile app if you have it on your phone or take you to Apple’s app store to download it. Shepard uses Words with Friends as an example. Let’s say your pal plays the word “quixotic”—a high scorer, by the way—you’ll receive a notification and tapping it will take your straight to the app to make your move. Facebook is also extending Credits to the mobile apps so you can also buy things within their framework. Again, it’s like its own little operating system within Apple’s iOS.

Facebook is headed toward a direct confrontation with Apple and Google (and Amazon and Microsoft) for the future of computing: the social operating system.

Apple certainly shouldn’t let Facebook create an independent app store.

When is Facebook going to develop its own tablet?

Daring Fireball: All His Life Has He Looked Away, to the Future, to the Horizon. Never His Mind on Where He Was. What He Was Doing.

Gruber responds to a Windows 8 fan boy, Paul Thurrott, who twittered ‘Hello, Windows 8? This is iPad. You win.’ after seeing a demo on a Samsung-produced tablet:

John Gruber via Daring Fireball

What did Microsoft show, though? They showed a Metro-style touchscreen tablet user interface that is, without argument, original. No accusations of ripping off the iPad here. Microsoft is admirably blazing its own trail.

But the OS reportedly isn’t coming out for at least a year. The demo tablet hardware from Samsung they’re showing it on (and giving to Build attendees) is a Core i5 Intel-based PC replete with a fan. Spec-wise these units are much more like MacBook Airs than iPads. Presumably actual shipping iPad-competing Windows 8 tablets will use low-power mobile CPUs — be they ARM, Atom, whatever, just so long as they get iPad-caliber long battery life and low temperatures.

How will Windows 8 run on such hardware? When will they actually ship? How many as-yet-unannounced iPad 3s will Apple have sold by the time the first Windows 8 tablet hits stores? (Not to mention the many tens of millions of iPad 2s Apple will sell in just the next quarter alone.)

It’s all in the future. All potential, nothing actual. Think about how different Apple’s and Microsoft’s approaches are. Apple unveiled the iPad to the public only when it was a completely finished product, two months before it hit stores. The demo units we in the press had access to that day were exactly like the mass-produced iPads that shipped to customers two months later. Can you imagine Apple doing with the iPad what Microsoft is doing with Windows 8? Say, showing a prototype iPad at WWDC in June 2009, running on MacBook Pro-caliber Intel hardware? Letting the public and the press play with the OS in half-finished alpha state on prototype hardware? Impossible even to imagine. (There were no hands-on demos, let alone take-home prototypes or developer downloads, when Apple showed a “sneak preview” of Mac OS X Lion at last year’s “Back to the Mac” event.)

I’m not passing judgment here — at least not yet — regarding which strategy is superior. I simply wish to direct your attention at how utterly different the two companies are.

Gruber seems to be making a corporate culture argument, but I think this is also a case of Microsoft playing catch-up, from way back in the pack.They hope to stall some buyers from buying an iPad, thinking that some razzamatazz might hypnotize them.

But iPad is clearly dominating the tablet space, and I don’t think Microsoft has a chance.

I am still betting on a Microsoft/Facebook social OS alliance. I wonder how closely integrated Facebook will be in Windows 8, when it launches.

Learning From The Google+ Experiment: Operating System, Platform, Apps

As part of the chorus singing about Google+ (see Armano’s insightful The Social Layer: Six Thoughts On Where Google Plus Is Going as just the most recent example), let me make a few observations:

It’s very hard to separate foundational concepts of Google+ from what might considered features or apps. Foundational elements would include identity, following, streams, and sparks. But Circles, Hangouts, and Huddle are best considered apps, in the broadest sense. So apps are a foundational element of the Google+ architecture, and they can closely integrate into the user experience of Google+, like Circles does.

But we are moving toward a world where most of the foundational elements of Google+ will be part of a next generation version of Android, and the things that feel like apps on Google+ will be, in fact, apps running on that future social OS.

This means that I could drop Circles, and use some other app as a mechanism for organizing my sociality. Imagine an imaginary app, called Groupings, that works very differently than Circles, but does build on the foundational elements of identity, following, streams and sparks.

But I would want to follow people not just on the Google+ enhanced version of Android, but the Twitter-enhanced, social versions of iOS and OS X, as well. So long as these two operating systems provide similar social foundations, Groupings could run on my OS X laptop and on my pal’s Android smartphone.

In this model, the operating systems become the platform, and apps like Circles or Groupings could run on either, or on a future, social Windows 9 (once Facebook acquires the phone parts of Microsoft). 

I could opt to follow someone, with a globally unique identity provided by the operating system of choice: in my case, let’s say by OS X, and the person I want to follow, David Armano, by Android. We would also be able to use those identities on any device.

Once I opt to follow, the basics are provided: I will get what he drops in his public stream, and it will appear in my ‘upstream’ — the unfiltered collation of all those I follow. What I post or repost falls into my ‘downstream’ which would be directed to everyone who is following me.

Obviously, the various operating systems have to support the fundamental protocols for this social messaging to work, and we will see this in due course, although it’s likely that we will see several contending models that don’t interoperate, and closed worlds built by the various operating systems providers.

We need the social operating system equivalent of http and email protocols to arise, so that an open social web can emerge.

We need the social operating system equivalent of http and email protocols to arise, so that an open social web can emerge.

So one thing we can learn from the Google+ experiment is this: I shouldn’t have to login to Google+, and use Circles, to follow David Armano’s writing over there. The works of those I follow should find me no matter what applications or operating systems I use. I don’t have to have Outlook running to read Armano’s email, and I don’t have to browse his website with Chrome, just because those are the tools he uses. 

And the developers of these applications, platforms, and operating systems need to be pushing aggressively in that direction, because in the meantime we are dividing the space for social discourse online into a maze of contending, non-interoperable models that don’t harmonize yet.

As Deal With Twitter Expires, Google Realtime Search Goes Offline

Danny Sullivan via

Yesterday, we reported that Google Realtime Search had mysteriously disappeared. Today comes the reason why: Google’s agreement with Twitter to carry its results has expired, taking with it much of the content that was in the service with it.

The details are interesting, but the lines are being drawn: Google+ is a direct competitor to Twitter, and so the orientation of Google as a whole to the streaming service will change.

It is starting here, in Google’s real-time search offering, but that’s just an initial foray, with Google decreasing the central role that Twitter plays in the real-time communication space, and trying to elbow Google+ into parity.

More important in the long run will be the nature of Twitter’s relationship with Apple, because the long-term battle is the social operating system war between Apple’s iOS/OS X and Google’s Android, with very different and potentially incompatible social worlds built in.

Fooling With plus.google (Google+): What Does ‘Share’ Mean?

I got invited to the plus.google beta (I begged my way in and Bradley Horowitz caved).

I am calling it ‘plus.google’ because that’s the URL, and avoids the problems with searching for a name with a plus sign in it.

It’s too early for more definitive thoughts, but here’s something I posted there:

‘Share’ in plus.google doesn’t have the same semantics as in the outside world, where it leads to a list of services. I guess there is no integration with Instapaper, Tumblr, Twitter? We are confronted with a gigantic plus.google system, where things can be shared among its parts, but not externally? (Ditto on pushing stuff into plus.google).

Someone may tell me that plus.google is in test phase, that those features will come later, etc.

But operating in the context of my existing flows will be the only true test for me.

This sort of experiment might be good for people to compare plus.google to another world-straddling social network — Facebook — that has attempted to do everything, but it isn’t for me, because I don’t rely on Facebook.

So I see the ‘share’ shift of meaning as indicative. plus.google is intended as a Facebook killer, but I have already defected from Facebook and I don’t believe in a single, monolithic, all-encompassing social world in the hands of one bunch of overlords, however benevolent.

I also think that the real angle for Google isn’t his giant social Disneyland they have constructed, but the primitives that underlie it, and the way that those will be built into Android, so that other app developers can take advantage of them.

Imagine how much more interesting this would have been if five partners had built social apps that were accessible on plus.google right now. Imagine if Instagram or a With were integrated?

And the rift between iOS and Android pops up in here even at this early date: I can’t upload photos from my iPhone because Google decided not to put that functionality out, yet, or ever.

So the apparent competition with Facebook may turn out to be the big florid opening act, while the long term war is with Apple over the social operating system of the near future.

So, instead of clicking the ‘share’ button and posting this to Tumblr, I will cut and paste from the closed world — at least currently — of plus.google.

More to follow.

Twitter: The Social Kernel For iOS 5 

[Update: I pulled a section of this out, as I was corrected about iMessage being the next version of Messages:

Project via comments

Actually, iMessage and the SMS app in iOS5 are one and the same thing. If the person you are sending a message to doesn’t have an iOS device, it sends it as an SMS.  If the person does, it recognises it in the To field and sends it as an iMessage.  In effect, Apple are embracing and extending the SMS protocol with this service. It is a huge move.

I stand corrected.]

Yesterday’s announcements from Apple included the new iMessage: an iOS-only messaging system, which is apparently intended to remove the last rationale that BlackBerry users might have to not adopt iPhones:

Darrell Etherington, iMessage: Biting RIM’s style and sticking it to network operators

BBM is one of the few remaining advantages RIM’s aging platform has over its younger competition in the smartphone market. (Check out this tweet representative of reaction toiMessage’s announcement if you don’t believe me.) People appreciated the way it integrates tightly to your device, and its delivery and read receipts let you know your messages aren’t getting lost in the ether. It’s been a life raft for RIM in the violent sea of the ongoing mobile battle BlackBerry faces with iOS and Android.

However, iMessage brings a lot of what’s good about BBM not only to the iPhone, which just passed RIM in terms of U.S. smartphone ownership trentds, but also to all iOS devices. With iPad and iPod touch users factored in, the potential audience for iMessage is huge, and it should cause at least some BBM-faithful to flee RIM’s platform for greener pastures.

Apple’s move also pushes the mobile carriers down in the stack, allowing iOS users to bypass SMS or proprietary messaging solutions. This is a painful but inevitable evolution.

iMessage is just a tactical play targeting BBM: a old-school pre-social, buddylist-style proprietary messaging system. It’s not strategic, really.

In the long run, it looks like Apple is planning to use Twitter as the platform for social communication, building Twitter into iOS instead of building protocols on which Twitter and other networks could run.

As reported by Marshall Kirkpatrick, Twitter on iOS 5 will be a platform for social apps:

My summary, in a sentence: iOS apps will look like, feel like, read from and publish to Twitter like never before. And they’ll do that in many cases instead of using Facebook.

[Jason] Costa [the newly hired Twitter Developer Relations leader] summarizes thusly.

“There is single sign-on, which allows you to retrieve a user’s identity, avatar, and other profile data.” That sounds like Facebook Connect, but I’m going to guess that Twitter will not prohibit developers from caching that data for time-shifted, aggregate, offline or other interesting types of analysis. Letting users skip having to create an account with every new app they download and instead click to log-in with their Twitter accounts is going to make many users very happy and encourage every iOS owner to get a Twitter account if they don’t have one already. App developers will get more and better populated user accounts, faster.

“There’s also a frictionless core signing service, allowing you to make and sign any call to the Twitter API.” To be honest, I’m not really sure what this means. Perhaps it means that parts of the Twitter API that require user authentication will be accessible via the same single sign-on feature discussed above.

“There is follow graph synchronization, which enables you to bootstrap a user’s social graph for your app.” In other words, apps will be able to offer users to find their Twitter friends who are also using a new app they’ve installed, and connect with them there too. That’s the kind of solution to the user-level “cold start problem” that Facebook Connect has been so helpful with for web apps.

“Furthermore, there is the tweet sheet feature, giving your app distribution and reach across Twitter.” Again, like Facebook Connect, this is a feature that appears to make it easy for apps to publish user activity and promotional messages out into the Twitter streams of a user’s friends. Facebook has a complicated algorithm that determines how often an app is allowed to publish messages out into the Newsfeed of a user’s friends, based on how much interactions messages from that app have received in the past. That’s a spam control mechanism that I’m going to guess Twitter will not replicate, at least at first.

It looks like Apple is going to give Twitter this deep and central role in its social OS plans, and allow the smaller more agile company to manage the building of an ecology of social apps on top of the paired architecture.

If even remotely successful, Apple will want to acquire Twitter, and Twitter will want to be acquired. These two will become as inseparable as NeXT was to Apple, when they regrooved Mac OS to be built upon the Mach Unix kernel. Here though, Apple will be making Twitter — and the open follower model Twitter resides on — the social kernel for iOS going forward.

This is a grand land grab by Apple and Twitter, an effort to block a Google/Facebook coalition on Android, or a Microsoft/Facebook partnership on Windows 8.

What about the competition? I predict Facebook will be too reluctant to partner with anyone, and may be at work on plans to launch its own hardware. Google is too slow on the social network side (the most expensive error of all time?), so they are stuck in the water. Microsoft is making credible efforts with Windows 7 and 8, but have no social network story. Microsoft is far enough behind the curve to possibly cede the social sphere to Facebook, too. RIM is falling like a stone, and would probably like to be bought, and either Google or Microsoft might bite, but that’s just tactics. None of these players has a strategic answer to the Apple move with Twitter.

What I don’t understand, though, is why iMessage isn’t written as a social app on top of Twitter. That would be the right path, and would simplify the Venn diagram tremendously. But Apple is opting to run both worlds — the pre-social and the social — in parallel, at least for a time, instead of doubling down on its social push with Twitter.