About This Case


28 Dec 2007, 11:59PM PT

Bonus Detail

  • Top 3 Qualifying Insights Earn $300 Bonus


4 Dec 2007, 12:00AM PT


  • Advertising / Marketing / Sales
  • Consumer Services / Retail Industry
  • Enterprise Software & Services
  • Internet / Online Services / Consumer Software
  • Media / Entertainment
  • Start-Ups / Small Businesses / Franchises
  • Telecom / Broadband / Wireless

What's The Secret To Mobile Social Networking?


Closed: 28 Dec 2007, 11:59PM PT

Earn up to $300 for Insights on this case.

As a mobile application development team, we're trying to understand how to bring social networking to mobile in a useful manner. To date, there have either been weak mobile versions of existing social networks, or simple tools like Dodgeball. What we're looking to do is build something that is more comprehensive and useful, that takes advantage of the fact that a person is mobile (while also understanding the limitation of mobile platforms). We're not against building this on top of an existing network (via Open Social or Facebook's platform), but don't want to be locked in either. We have our own ideas, but are looking for advice from the community on which areas to prioritize of focus on.

13 Insights


The biggest restrictions in getting involved with social networking on a mobile are (in order, and looking at what you can do):

1) text input speed - you can't do much about this as a development team, apart from to ensure that cookies are handled properly so that the user never has to enter the same login form details on any given device twice (unless they manually log out).

2) screen size - design for QVGA (anything smaller is an anachronism and anything larger is out of the ordinary) and be very aware of download and image rendering speeds - see also 3) below.

3) bandwidth and cost

  i) a surprising number of 'mobile' sites still use images and javascript that are fairly heavy duty. Outside of the USA, data transfers usually cost quite a bit and there's a big bandwidth difference at the end of each day/week/month between mobile pages that total (text, code and images) 10k each and those that total 30k each. i.e. 30k doesn't seem a lot to you, the developer, but multiply it up per page viewed per day and it soon gets unworkable on many mobile tariffs.

  ii) on the upload side, for similar reasons, do what you can to speed up user image and video uploloads. For example, resample all images to VGA or smaller, as far too many new users start out into Web 2.0 and get frustrated when their very first image upload from their Nokia N95 takes forever. If you do resample images, videos or sound samples, then make this a setting that advanced users can override, if necessary!

The single biggest obstacle to any new network is its own size - if it's too small then it won't gain traction in any reasonable timeframe. Like IM chat clients which have evolved to support multiple standards, you'll have infinitely more success if you build a useable and sensibly-implemented mobile client around multiple APIs into existing networks. The likes of Facebook, Bebo, etc. are too huge to ignore. Best to work with them, do the 'mobile bit' better than they can and then get the word out!


Just make an all-inclusive rss multimedia blog tool that works easily from a mobile platform.  Facebook, etc. can import rss feeds, so a moblog tool that really is super easy would make it easy to contribute to one's social networking site.
Devin Moore
Wed Dec 19 5:31am
I thought of some more features that would be cool from a mobile standpoint.

1. Optional GPS stamp: The ability to stamp my blog post with where I'm blogging from would be cool.

2. Cached blog entry: if I can type in blog entries and upload them at my convenience, I may be able to blog from places where I don't get WIFI, i.e. the woods, and then upload them for free at my local coffee shop instead of getting hammered with online access charges.

3. Very obvious controls, manuals, and help: I tried to use a mobile browser, and it was very unobvious to me how to navigate and use the features. I couldn't tell what features it really had either. You need to make the features accessible to everyone, and make them obvious in a manual that shows with screenshots how to operate all of the features. If I as a computer expert cannot work something just from looking at it, then the vast majority of people will not sit and tinker with it, especially if the tinkering costs them money if they accidentally activate some data feature.
Twitter has already won this fight.  The way Twitter can be improved is by allowing pull-down choices for common entries.  This makes it a calendar diary (e.g. attending meeting...) as well as a tweet-like box.  I wouldn't want to go head to head with Twitter.  Loyalty is high, and it is well-integrated with Facebook.  

Adoption of any social network stems from one of two things: personal usefulness or personal expression.  Adoption of any new technology also stems from two things:  ease of use, low barrier to growing the network.  Taking these four issues into account can help you design the right system.   Here are the questions you need to ask...

First let us look at the personal expression model. Myspace and Facebook have been successful because they have allowed users to personalize their "space."  How can a mobile social network either allow people to personalize their phone or personalize their physical space?  The geotargeting capabilities mean that users could interact with their environment in unique ways, through their phone.  For instance, allowing users to tag places or leave messages for themselves or their friends, similar to what www.socialight.com does, could be useful.  Using the service to express things about themselves to others nearby will also be attractive.  Users can show their likes, interests, relationship status, etc. and other users can connect automatically if there is an interesting match.  

The other model to consider is the personal usefulness model of del.icio.us.  The site has value to me as a personal bookmarking site even if no one else ever uses it.  If others do use it and share their bookmarks, the value of the network of users increases.  A mobile social network must provide value even if there is only one user, and then provide more value if there are more users.  If all the value is only present once the network reaches some critical mass, then you may struggle to reach that mass.

 With respect to the ease of adoption of the mobile social network, it has to be intuitive.  If people are required to go online and download software, it won't work.  It has to be preprogrammed on their phones, or available via download over the cellphone networks.  It must require minimum user input to customize, and should probably function much like a text messaging system, since users are familiar with that.  Will users log on to a website to customize it?  Perhaps, as that is a model they are familiar with too.  But that could break the immediacy connection that could encourage them to use the service because they need it now.

 Now to the biggest questions you have to answer.  Why will the network grow?  Why will I invite my friends?  What is in it for me?  How do I benefit in a tangible way?  I don't want points or something cheesy like that.  I want to invite my friends because it makes my use of the service better.  How difficult is it to invite friends?  Can I just enter their phone numbers?  If I join the social network, can I interact in some capacity with people who aren't in the network?  Providing an open level of service plus an advanced level of service will encourage people to sign up as members to embrace the more advanced functionality.

 The primary use I see in a mobile social network is the serendipitous value of discovering either:

a) new places that my friends have been - for instance, I drive by a coffee shop and am told that Bob recommends it


b) new people that share my interests - for instance, I'm at a conference, the mall, a basketball game, etc and my phone tells me that someone else there is looking to hire an accountant, while I am looking for an accounting job.  Do we want to connect?  Or alternatively, someone is single and interested in Yoga, and I am in the same situation.  

 To summarize, if I were designing a mobile social network, I would prioritize things like this:  serendipitous discovery of non-network dependent stuff (things that don't depend on the size of the network - this is the personal value) , ease of use, discovery of network dependent stuff,  ease of spreading/growing/inviting others to the network, personal expression.

Using a similar framework to O'Reilly's Web 2.0 definition, let's first define what Mobile 2.0 is all about.  Once we define what is is, we can discuss applications for social network over moble.

Mobile 1.0              ;      vs.      Mobile 2.0
rigntones, ringbacks    <--->      Vringo (video caller ID)
reading           &nbs p;           <--->      publishing/participating
pull        &nbs p;                     <-->       push
connected to PC enivonment <-->  connected to my web environment
411                         &n bsp;   <-->        Google411
GPS           &n bsp;            &nbs p; <-->        Google Maps
p2p            & nbsp;            &nb sp;  <-->        p2community
Dodgeball             ;       <-->        facebook, android
casual gaming            <-->        networked gaming

Social Networking's Dirty Secret: Socially-accepted voyeurism
As my friend and Benchmark Capital Partner, Michael Eisenberg, is fond of saying: the secret to successful social networks is how well they leverage voyeurism.  People have an innate need to see what others (especially people whom they know) are doing. 

Facebook's overwhelming success points to this.  From a business user's perspective, Facebook came out of nowhere and trounced LinkedIn .  I have stopped paying for a LinkedIn subscription and migrated my network over to Facebook.  I attribute this to one thing and one thing only: Facebook's News Feed.  There, I can see all the activity in my network and decide daily whether to participate in what's going on around me.  This has major privacy implications as I've written in another insight, but these will abate with time, much like In-Text Advertising has become (almost) entirely acceptable.  Google's Open Social platform will also publish such activity streams.  As Dave McClure put it succinctly, "people who are going bananas over Beacon should understand that most people on Facebook are used to the default being opt-out (ie, lifestyle transparency), not opt-in (selective sharing)." This is exactly why we've signed up for Facebook.

Covestor and it's like allow me to subscribe to others' trading portfolios, allowing me a clear view into their decision making process and their performance.  This works extremely well because there are large financial incentives involved.

Spock allows people to track what friends and colleagues are doing online.  It's what we care about.  Effective social networking tools are not just about providing cool javascript tools, it's about lifestyle transparency.  An effective mobile version will need a way to address this.

Virality: Doing the heavy lifting for users
Marketers focus on how viral applications are the holy grail for their job function.  The truth is for a social network to be accessible to most users, it needs to make my job of recruiting/connecting with my connections easy and palatable.  Facebook is a no-brainer.   There is critical mass to the user base and now, users outside the network are beginning to feel left out, compelling them to join.  But what about smaller, more niche-oriented networks?

Meetup is an amazing network.  I recently ran an investment seminar on hedging exposure to the US dollar for American expatriates living abroad.  Within minutes of setting up on Meetup, I had over 15 people sign up for my group and RSVP!  Meetup has essentially created a network of affinity groups that share nothing other than certain predilections and the desire to meet in person.  Once I've joined a group, I'm introduced gently to 1) other groups in my geographic area that 2) share my interests.  I didn't need to 'spam' my address book (although that functionality exists).  Meetup worked for me because it allowed me to connect with people with whom I share interests immediately, without having to work at it.  I'm hooked now.

Collective Wisdom: Harnessing the Wisdom of the Crowds
Social networks allows a user to tap into he collective brain of his/her network.  Or as Tim Oreilly puts it, "If an essential part of Web 2.0 is harnessing collective intelligence, turning the web into a kind of global brain, the blogosphere is the equivalent of constant mental chatter in the forebrain, the voice we hear in all of our heads. It may not reflect the deep structure of the brain, which is often unconscious, but is instead the equivalent of conscious thought. And as a reflection of conscious thought and attention, the blogosphere has begun to have a powerful effect."

Techdirt's Insight Community is a great example of one of the ways for networks to tap into the collective experience/knowledge of its members.  This is a pretty explicit example but social networks, if used and built effectively, make every node that much more intelligent and knowledgable.  The collective intelligence acts as a type of filter and this overarching knowledge helps to make sense of all the noise in the world.

So, let's look at ways of translating/porting these salient features of a web-based social network to the hardware/usage limitations of the mobile device.

The current generation of mobile devices, including Blackberry-type PDAs, are still used primarily for reading, not writing.  So whatever level of voyeurism exists on mobile networks, it's going to focus on phone activity, not publishing activity.

Call log activity: sharing my call list, when, how long I spoke to that person would be very interesting for my connections.  Also, seeing a brief profile of the person on the other end of the call could be pretty engaging.  SMS/email activity can be added to a sort of Facebook-esque News Feed.  And clearly, any browsing or web activity can be shared out as well.

Geotracking: Wouldn't it be interesting to know where my friends are at a certain time.  If they're using their phones, it could be possible to design a system to provide geographic fingerprinting to be shared with the network.

Convergent device usage: If I'm listening to 50 Cent's new album, my network wants to know and I want them to know.  Rudimentary tatings systems can be implemented and if licensing issues can be addressed (good luck), you could find a way to share music with the network.

Conferencing my people in with your people: A technological challenge, but why not work to find a way to conference people in (maybe by calling in to a specific #) so that different network members all hop on a shared line to chat spontaneously or with advanced planning.  So, instead of learning about what my friends are up to via a news feed, I can speak to them directly.

Creating soft sales: Spamming my friends doesn't work.  I've avoided joining new networks that will require me to send out (yet another) email inviting my whole address book in.  I don't want that and it's getting harder and harder to convince my friends to join.  Instead, on every SMS I send to my friends, a simple footer would suffice with some compelling reason for them to join my network: data could lure them in or perhaps, a part of my news feed.

Spontaneous get-togethers: Meetup (as mentioned above) has a great system for connecting people with the same affinities and giving them the tools to meet up.  Functionality to get my friends to meet me at the local Burger King via a quick conference call or via an SMS round would be compelling.  What about other people in my locality?  Would they like to join?  Do I want them to know about it?

Media can make it jiggy: A song or SMS that I've sent someone can certainly take flight and become viral.  Music could be the key element in creating a compelling mobile social network.  This is rife with legal issues but if solved, music is one of the most cohesive factors in putting together affinity groups.  People who like the same music generally have other shared interests, tastes, backgrounds.

Collective Knowledge:
Share stats out to the network: Making me more informed about what my friends are up to helps me filter out the noise.  Call logs play into this while so does music activity.  Help me discover new things and new people.

Hierarchy settings: Help users figure out a way to define who are the connectors in my network, who are the idea generators, who are the leaders and who are the followers.  Given where people fit in, their knowledge should influence friends differently.  This should be accounted for in the network -- it is in real life.

I haven't focused on the concept of mobile network as extension of a web-based network.  I think the device used in a mobile network is still so different from a PC that the interaction on a mobile network will require structural differences.  Facebook will make a better mobile version but I think there is an opportunity to create a stand-alone mobile social network.

Please find below, in extreme brevity, a few points I would like to emphasize.

  • Easy logons: since recalling and typing passwords can be a chore when one is using a mobile device, it must be easy to log in, and to stay logged in.
  • Creatively capitalize on any and all status (location, presence, etc.) information the mobile device can provide. Not every data source is necessarily useful by itself, but mining through the possible combinations will likely uncover gems.
  • Provide information and services the user needs when mobile. Automate in order to avoid bothersome user interaction. As an example, one of my favourite applications is one that makes automatic directory service lookups that show within seconds, based on the caller ID, who is calling.
  • When you cannot automate, nevertheless require as little work as possible. Even disengaging the keyboard lock, launching a web browser, and opening an application can take a lot of clicks. For example, a voice command might be much more attractive (where technically possible).
  • Piggyback on existing applications and their data content, so that the user never has to enter the same data twice. Try to unify messaging channels wherever possible. Consider alliances with other actors.
  • Be universally compatible; try not to require specific hardware or software, not to mention location, or mobile operator.
  • A web site user should easily be able to use the mobile version as well, and vice versa.
  • Make sure that everything scales, also on mobile devices, even if there are e.g. 10000 contacts.

I hope this will help your endeavours.

You have to start in the right corner, and ask yourself: What is particularly social about being mobile, or mobile about being social?

First, the second one (which you were already looking at): Meeting people while mobile. This can either be people which you already know (who are in your network), people you do not know, but who is in the extended network (your friends networks), or people who are not in your extended network but who you should know, because they have a similar set of preferences to you.
In the first case, there (presumably) has to be some way to register that these people are OK, and this whitelist has to be communicated to your phone, which has to have some way of discovering the position of the people on the whitelist. This can be done in several ways, the simplest probably being to use pres ence, and combine the address book in the&nbs p;profile with the whitelist derived from the  ;application. Actually, this could be your system& nbsp;- adding the logic between presence and  the social networking site. 
There are also standardized APIs for building positioned applications, both with and without GPS (the mobile positioning would be quite sufficient in some countries, but this depends on where you are attempting to deploy the system - in Japan, the cells are approximatley 400 m in diameter or less, but in other places in other countries they may be as large as 75 km).
Another version of the same thing uses Bluetooth, but then you have to provision the Bluetooth client in the phone with the whitelist, so it reacts properly when a friend is close (discovered by Bluetooth). You would probably have to build the application and the provisioning system, if you could not use the standard provisioning system of the telecom operator (which I do not know off the top of my my head, it would have to be investigated). Your application would be the client and the server.
The first case does not exclude the second, they would be able to work toghether.

Then, we have the cases with the people you do not know.
Here, you would have to deploy a recommender engine of some kind. It would have to derive the extended graphs and create a list of users, with their mobile numbers, so they could be positioned; the system would work in the same way as in the first two cases.

Then, the first of the major questions: What is particularly social about being mobile?
In the same way as your extended graph can be enhanced with the position of people who are nearby and who you either know or should know, some positions create an instant community. This is, for instance, true of people who are attending an event, such as a rock concert. They have a common node in their graphs, and they should be able to add each other without knowing each other (at least to some dimension). Combined with the recommender system, it can also be used to derive trusted members of the extended graph.

There is an additional concern: How to address user privacy. If successful, your system would collect very intimate information about people, and connect it with very intimate information about other people. How do you approach this? You need very strong mechanisms to assure that users do not feel their personal integrity is infringed, and that they trust these mechanisms. This may be solvable by making appropriate contracts, but the users have to understand them; this is not easy and will be a showstopper in some places (notably Europe).

You probably know how to approach Facebook an d the other social platform providers (but ho w do you handle people who are not part& nbsp;of Facebook, and refuse to be for privac y reasons?)
Depending on whether you are looking&nb sp;at proximity (using Bluetooth) or positioning ( using the mobile network or GPS) you have&nbs p;to approach different partners. If you are  looking at the proximity case, the mobile pho ne companies makes most sense; here, Nokia pr obably has the best developer program. 
If you  are looking at positioning, you have a  tougher job in front of you, as the mobi le operators are notably not very good at&nbs p;this kind of thing. People who actually bui ld the systems (like Ericsson) has developer  programs for certain aspects of the system, a nd if you are able to use IMS presence,& nbsp;they will almost fall in your arms -&nbs p;if you can find the right contact in t he company. This can be tougher than it  sounds; it is probably best to sell the  idea to a mobile operator, and then have  ;them make their suppliers help you. Which me ans proving the value of the product, somethi ng which is not easy on its own.

Well, th ose were some initial thoughts. Hope this hel ps. 


Focus on FreezeCrowd! That's the secret.  If you're interested in contacting me, please let me know. We'll be known for connecting people in crowds, any sort of crowd, and mobile is where we're headed.  Imagine getting a crowd of people together and connecting in the crowd by simply putting a FreezeCrowd button on your mobile, as soon as you press it, everyone in the crowd gets identified, you "freeze" everyone in the crowd at that moment, and you reconnect with the people you met in a real life crowd online.  Why spend time building on top of an existing network, when by the time you build on top of these networks, the next best site will be launched and people will be connecting in crowds?  Please email me, contact me, and we can save you a lot of time and headaches.  Plus, we have protected ourselves on this concept.  We have working prototypes, and are open hearing more. 

Having recently developed a mobile social networking service (www.pitch.mobi/my.pitch.mobi), I have some opinions on what I think makes a good mobile social network, or rather the approach to creating a good mobile social network. The one key factor I think is that the whole underlying design - from database design to infrastructure to site structure to per-page design - needs to be approached with the mobile device in mind, and should not be driven by what features or technologies can be developed for use on the web, or using a particular application framework.

Whatever services you put on top of your mobile social network should also be approached from the mobile device perspective. This is perhaps the hardest thing I've come across when describing mobile social networking to other people - it is not about converting what you can do on the web and making it available on mobile, it is about starting and ending with the mobile device as your target platform. The guideline I used was this: If I can't do it on my mobile, then I don't do it.

Also, since mobile networks are typically very slow at data delivery, this awareness of designing for the mobile platform should underly everything that you do. With all you can eat data packages still being the exception in most of the world, anything that is image-heavy (i.e. data-heavy) will cost potential users (in raw monetary terms) more than something that is data-light (such as a text-only site), so an image-heavy site runs the risk of alienating potential customers purely from a cost perspective.

The service I designed and built (note: I stopped working for them about 9 months ago) has pretty much everything you could ask for in a social network, except anything location-based (for economic reasons - location based services in the UK are frighteningly expensive, although we did include searches of the user population that took into account postcodes, so we had semi-location-based search.) It has the usual suspects: generic text chat (works out 6 times cheaper than SMS for the same number of messages), group chat, private messaging (to 1 person or all your friends), image upload (including animated GIFs), image galleries, image comments, mobblogging, 'text walls', friends, groups, personal avatars, personalisation (WAP site personalisation, including "skins"), video upload/download (mobile to web and web to mobile, all formats except WMV - licensing issues and all that) and works across any GPRS connection and on any WAP1.0 compatible device.

Which leads on to another point: I am entirely against java-application-only-based social networking services. As soon as you start basing your service design on what you can do with a java application you start to lose potential customers (thos with handsets you don't support), and you make the user acquisition process more difficult and the initial user experience much worse, as the first thing you have to do is convince someone to download your application, with all the concommitant problems that brings.

(In my view, once you base your design on a java application, you start running into issues with data transfer speeds, data charges, security settings on the phone and everything else that comes with a java application. To qualify my earlier anti statement a little, I don't see a reason not to have a java access application for your mobile social network, I think the problem is if this is the *only* way of accessing the service. At that point, it's not really a mobile social network, it's a java-application based social network. And on that note, once you have the underlying structures of your mobile social network in place, how customers access it becomes irrelevant - in fact, you want them to access it in asmany ways as possible, and have an appropriate experience depending on the device they use - mobile, PDA, Web browser, whatever)

So I think the secret to building a mobile social network is to treat the mobile device for what it is - a mobile device. Design your service and the functions within it with the mobile device and it's limitations in mind. Think about how people use their phones, and think how you can deliver a mobile experience that is very simple and accessible for the majority of mobile users, and then add bells and whistles later.

What makes mobile social networking different from static, PC based social networking?

One aspect is the typical screen size for devices used for access, this is the same challenge faced by WAP browsers so can be overcome. Using the new service by Andrew Grey that translates standard URL into a numeric URL would be an interesting feature to have as an option to make browsing more efficient.

The key difference is the mobility of the user which opens up new dimensions to the user experience.

The opportunity to spin the view from the network based on your current location opposed to mobile access to your normal profile would be a great differentiator.

Some examples:

Users in your groups in the vicinity

You already belong to some large groups (interest groups, alumni etc) based on your current location the platform should highlight those known members nearby (to a user specified sensitivity). The physical location does not need to be too accurate, within the current cell should suffice (we don't want to encourage stalking).

Local information generated by the community

Reviews, tips, places to visit that the community recommends within the locale. Along the lines of the content available in niche social networking platform Tipped.


There are several keys points to consider moving forward:

1. which platform do you want to be married to? Android/Open Social seems to be the de facto winner of the overall future of where mobile apps will go. However, this does not preclude others in the market. At this point, it is probably too early to predict where to go platform-wise, but this could also be a viable option- something that pre-exists but crosses over.

Obviously, a pre-existing network like Facebook might be optimal. This moves to the next point to consider...


2. even with a platform, what do people want to do with a mobile social network?

I'd concur with Gary Hall(http://www.bloggingstocks.com/2007/12/28/social-networking-in-2008-friend-o r-failure/)- user generated content is probably necessary to constantly draw people back. Also, large-scale networks are destined to fail. As much as it might be cool to have the top number of friends or contacts, most people find little relevance for it. Things that consumer driven by the local market seem to be the guide that Hall might recommend. Identifying the local niche groups that will afford the most for advertising seems the logical choice. While the youth market is seemingly built-in, the boomer population and aging factor can not be ignored. (http://money.cnn.com/news/newsfeeds/articles/newstex/IBD-0001-21902986.htm) "Research firm eMarketer says 37% of U.S. adults and 70% of teens use social networking sites. It projects that will rise to 50% of adults by 2011. "

It seems with a rise with more and more social networks, and each prepared to go mobile, it begs the question how can something "comprehensive" address the local?

It might make sense to develop more "portal" mechanisms that make it easier for "self-publication" of localized niche networks. Most are centered around interests, business, or locality. If something was generated with pre-existing formats (ie all facebook or myspace users from an area, etc), it would probably have appeal.

Once you get past the the who, the what is obviously important. Many apps already exist. What niche has yet to be filled?

More and more, primary internet accessibity is going to be the key notion. Probably exlplains the popularity of any Google-based solutions. How those functionalities can best be used, why use the phone instead of the computer is the question to be asked. Not just because it is mobile, but because it allows for the easier access to needed on-the-go information.

"Best place to X," "Where to find X," or doing the standard internet activities (blogging etc) though most of those seemed to have been addressed.

The scope of where to take this question is huge. Perhaps with some further consideration, you can find what has been missed.


One of the many challenges that social applications face is that simply building a great application won't get people to use it.   Dodgeball is arguably amuch better mobile application than Twitter, yet as this simple alexa comparison graph indicates Twitter is very handily winning the battle for the hearts and minds of social networkers, even after Dodgeball's aquisition by Google:  (Alexa is not a good stats metric but it will do for this comparison):  http://www.flickr.com/photos/joeduck/2145170679/

Given the popularity of Twitter I think you may want to shoot for improvements and innovations on the basic and simple Twitter model, realizing that users appear to prefer usability, simplicity, and an existing heavily used network to more robust platforms.

I would suggest that the key mobile social challenge is maximizing interaction while minimizing typing.  

Luckily a focus point that does both of these things is creating photo-based community via picture sharing.   Typing remains a major downside to mobile socializing due to the difficulty of typing on phones.   This challenge is likely to continue for some time even with new smartphone innovations, so it may be beneficial for you to build the socializing around photo sharing applications, making it easy to caption, send, and share pictures with friends.     Pictures with only a few words of caption convey a lot of information very effectively, and there are currently no large photo sharing communities that are rooted in mobile communication, though Flickr's community features and mobile uploading are exceptional and may evolve into this.   Working with Flickr and Flickr APIs to make a mobile photo sharing community might be the fastest path to this approach, though as you've suggested in the initial question you'll want to be compatible with Facebook and the new Open Social.   In fact this cross platform compatibility would be a good "pitch" as you enter the market.    Be the "one stop" mobile application that allows users to communicate across platforms as seamlessly as possible.   Twitterfeed has become popular in this fashion, by allowing Twitter users to automatically have new blog posts appear as Tweets via an RSS feed. This type of enabling application is *very appealing* to users, especially the early adopters you'll need on your side to gain quick traction for your mobile application.

Also consider partnership with any mobile providers.    Many major players remain "behind the curve" on social networking, so tailoring your application to a specific device or wireless carrier might get initial traction, though the politics may make this prohibitive.   Also consider conference co-promotion of your application.    Twitter's rapid rise appears to have originated mostly from heavy use during the big Southwest blogging conference last year, and some would argue this made the company.

Super early adopters like Robert Scoble and Chris Pirillo have been working a lot with video. Scoble recently with mobile video.   I do NOT think mobile video applications like Seesmic or mobile YouTube applications are a good focus in social networking at this time although they may eventually become mainstream.    For mass adoption I believe photo sharing is the strong social networking tool that is currently underutilized but will become very popular.   

Encouraging mobile virality of your application is a key challenge.  This may require very active participation by your company in existing social networks, encouraging and showing social onliners the "path" to your superior mobile application.   Enlist early adopters in beta testing and be willing to accept harsh feedback, noting that early adopters may demand more features and UI pizazz than normal people.

Also along these lines you must make it *very easy* for new users to sign up and utilize services.   Best is a simple or no login for the first tries so the user can focus on the application and not the drudgery of a sign up.   If they like the application they'll sign up eventually, so at most ask for email address/phone during initial signup.   Simple or non-signups may prove to be the holy grail of those who benefit the most from Open Social applications.  

Twitter v. Dodgeball:
http://www.techcrunch.com/2007/03/19/sxsw-showdown-dodgeball-vs-twitte r/

If the internet has taught us anything, it's that people love connecting with others (whether it's in AOL chat rooms, on personal blogs, or answering online personal ads.) People of all ages need other people; they're "hard-wired" for socializing. And a mobile application that delivers true socializing can enjoy exponential growth, as friends tell friends, who tell their friends... This is not hypothetical; it's the reason for the massive, break-out success of sites like MySpace, Twitter, and Facebook.

Even using existing capabilities can be extremely effective if they're applied in a way that's genuinely meaningful to users — and adding unique mobile capabilities could create a break-through application.

Here's some ideas.
  • In the 90s anyone with a Palm would insist on beaming over their business card. They were proud of their cards, and they wanted to share them. How about a "Mobile ID profile" — name, contact info, favorite picture (or pictures) and maybe even a pre-recorded voicemail greeting.

  • Another idea would be a "mobile portfolio." A previous Insight discussed how difficult it is to share photos among mobile phones. A proud dad would love to share all his photos of the new baby, and other portfolios could include sets of photos about vacations, weddings, or a visit to relatives.

But your "mobile" advantage could come from the application's presence at outdoor locations (preferably one crowded with people). And that's when it gets really exciting...

    Evite's guest list-style application is incredibly useful after a party, offering first and last names, contact information, and the clever commments left when accepting the invitation. But a similar application would be even handier to have at the party. Imagine combining that with Twitter-style updates of opinionated real-time comments. ("Ask me to dance!" "I can't believe Joe's playing Funkytown.")
  • Imagine using this at a class reunion - digitizing the high school yearbook photos along with the "now" photos (and allowing other members of the mobile application's group to leave comments below the photos.)

    I can see an application like this being popular at big college parties or Spring Break-style events.

  • Mobile video recording hasn't really caught on — but a mobile video social network could unlock its full potential. People do a lot of things online only when they know lots of other people will see them do it. Imagine everyone at a college's big football game sharing their cellphone footage of that last play with the network (with members of the network also able to vote up the best mobile video - or enter a comment). Friends at a nightclub could share their videos of the crowd, and so could spectators at parades, concerts, political rallies or other large-crowd events.

    Arrangements could even be made to publish videos from a mobile social network on a web page. Ultimately this could even lead to a new genre — the viral mobile video. Imagine how intriguing New Yorkers would be by a social network offering the best New York City cellphone videos — this week, this year, all-time...

This doesn't even begin to address possibilities that would become available with position-based information. ("Show cute males at this party, ranked by their current distance from me....") But whatever you do, make it easy for users. (Dodgeball was never convenient, requiring users initiate the contact, and even supply their own location information...) The real drawback for mobile applications is the small screen size and the general clunkiness of navigating at mobile interface. If users can set up their mobile profiles using browsers on their desktop PC, that will speed up adoption quite a bit.

Think big. It's a new sector, which means there's room for a new player, and the "killer app" hasn't been developed. Yet....