Building Better Social Graphs
I'll say right upfront that this may be a feature that many people don't need. But I need it so I thought I'd post it anyway.
As software becomes social, the creation of the social graph on each web service becomes a chore. I do not believe that you simply want to port your social graph from Facebook and Twitter into new web services. I believe you want to curate the social graph for each and every web service. And that's how I try to do it on each new social web service I encounter.
The people I want to follow on Etsy are not the same people I want to follow on Twitter. The people I want to follow on Svpply are not my Facebook friends. I don't want to share my Foursquare checkins with everyone on Twitter and Facebook.
I am slowly but surely building social graphs on Etsy, Svpply, and Foursquare (and many other services but I thought I would focus on these three for this post). Each service has a slightly different relationship model. On Etsy, you join circles. I am in almost 100 circles on Etsy, some of which I would like to join back. Svpply uses the Twitter follow model, and close to 100 people are following me, some of which I'd like to follow back. Foursquare uses the Facebook model and there are 3815 friend requests for me there. I am sure there are a bunch of people in the 3815 that I'd like to share my checkins with.
So here is the feature I want. I would like to be able to run these people through all my social graphs on other services (not just Facebook and Twitter) and also my phone contacts and my emails to help me filter them and quickly add those people if I think they would make the social experience on the specific service useful to me.
I don't want to get emailed everytime someone follows me or friends me. I'd like each service to build up a list of relationships and let me query that list against other social graphs. I think that would be a much more efficient way to build social graphs.
I do add people myself. I use Twitter's Who To Follow service to build my Twitter social graph. I join circles in Etsy when I see a seller who has a store I like. I follow people on Svpply when I see something they added that I like.
But the more people you have in your social graphs on these services, the more valuable they become. So making it easier to curate application specific social graphs is a big part of making the services more social and richer experiences for all of us.
Comments (Archived):
The problem with this idea is context or the need for metadata. Keeping track of your network in each group is simple, but each network has a different purpose and that purpose could be different based on the person creating the network. For example, I use Facebook mainly for personal contacts, but other people use it for business. I use LinkedIn for business, and Twitter is a mix of business and personal. Each contact in each network has specific context (or metadata) that makes that connection mean something. For people, that context is implicit but for an application the context needs to be built and managed explicitly.I would love to have something like this as well, but with the number of network-enabled services, even the development becomes difficult. Also, how does this fit with the idea of a single digital identity that is shared across all of these services?
that is my point. i want to curate my social graphs for each and everysocial software i use. i just want it to be easier
If you consider that a single social graph with you at the root could have branches which are mutually exclusive (e.g. one for each application) then you have only one social graph *and* the ability to see it as a single structure that you curate like a bonsai. The added advantage is that mutual exclusivity is only one of the possible states – a single graph that you own gives you many additional options and functions that separate graphs do not.
I like the bonsai analogy. I agree with Fred, I think this would be a good feature, but perhaps not something most people need.One trend I have noticed on my Facebook is people are either 1) deleting their account and starting over, or 2) cutting down their friends to <100 (friends + family that they’d invite to a birthday party).
I feel like the opposite is happening with Facebook. I’m seeing more and more friend requests and feel like people are treating FB as the LinkedIn equivalent of non-professional life. It’s the holiday card Rolodex. The same instinct that drove all of the Plaxo requests back in the day prompts people now to look at the “You may know” box and say “yeah, it would be good to stay connected to Sally”. Not that you really want to share anything with that person today, but you may want to find them later on down the road.It’s both good and bad for Facebook. Good for obvious reasons, bad because it’s hard to fix the “I’m connected with everyone I’ve ever even kind of known” issue with Groups or Lists or some other setting or tech. It’s kind of ironic that the more successful FB is at assembling the uber-graph, the more pronounced this problem becomes.
It probably depends on who your friends are. I’d honestly put that number at 5-6% (35-40) of my +600 friends where I’ve seen someone post “cleaning out my friends list” or “don’t get offended if I delete you, trying to lock down my profile to friends and family only.” These aren’t uber tech-savvy individuals either. Sure… it’s not a huge percentage, rather just a trend I’ve noticed.
I’ve heard of people doing this too Robert. But I think for most keeping Facebook to close friends and family feels like holding back the tide. Too much work to explain/deal with why you’re saying “no” or pruning your friends list. Easier just to say “yes” to everyone and, hopefully, find some other way to share with smaller closer groups.I think Path is a really cool example of designing in constraints. I’m less likely to invite everyone I know to Path, and people are less likely to invite me, because right up front you know it’s for your inner circle and the 50-friend list creates scarcity that people don’t want to impose on. I think we’ll see more services with built in constraints that help us through the awkwardness of the friend request / reject problem.
That’s another interesting trend. MySpace had no limits on friends (that I’m aware of), while Facebook imposed a 5000 friend limit, and now services are popping up with an even lower limit. I also see this trend as a validation of services like Path.Another trend I’m seeing is people sharing URLs. Hell, 9 out of 10 times I post on Facebook, it’s sharing a news story, music video, or something else… or to see what others are sharing link-wise. Problem is, this isn’t really organized (I have to search for 5 minutes to find a music video I or someone else posted 3 days ago). Delicious 2.0, where are you?But I’ve digressed & am rambling too much. I’ll leave it at that.
A privacy-based model is needed for this. Some “contacts” are bound to overlap in various social graphs. I believe separation from a true human relationship perspective is what is needed.Who is considered “family”? – and even within family, who are close members vs distant? Coworkers? Classmates? Acquaintances? etc. If we specify the level of privacy based on the person’s actual relationship with us, then it would be easy[er] to migrate an ex-classmate, who turned into an office buddy, that later turned into a dinner with the wives and kids kinds of “friend”.I think FB started off on the wrong foot. People hoard FB friends in narcissistic way. Fixing that psychology is difficult, if not impossible. Twitter has less of that, though I’ve seen a bit of overlap in the narcissism department between the two.I believe the key is a new privacy model that lies under (over?) the current social graph. One that enables anyone to manage their entire social graph contact list from a single place. Maybe not necessarily account settings (as I doubt FB will enable a 3rd party app that manages their privacy settings) but at least a categorization of contacts, showing overlapping connections and basic hints like what they can/cannot see or do.
I’ve only signed up for Discus recently, but it seems like they’re in a good position to chase this. Most social services have a comments layer, so it’s a good place to build scale and power across multiple services. It’s also the most social layer by definition – where people talk. They can accumulate data on who talks to whom where and who adds value per the community. With that data they can build tools to help people curate their graphs in various services. Not hard to imagine the Discus Dashboard where I’d do that…As someone said above, it’s going to be tough to get big players to want to give this “graph power” up. But if Discus – or newco or whomever – could start convincing some of the services with smaller graphs to play ball, who knows. You end up with an uber-graph at Facebook, and maybe Twitter and LinkedIn don’t relent, but everyone below that size sees that the increased scale and reduced friction from a more independent “coalition graph” outweighing the downside of giving up some graph control.
my partners say the same thing to me about disqus
We’ll have to get explicit in defining our relationships. It’s a big task to define each contact I know as business, family, school friend, other friend, music lover, yankees fan, etc. But with that type of classification, you’re one step closer. Still hugely manual.
yes, but automating a piece of a manual effort does help
So basically Delicious tagging of people…probably tough to make a standalone business, but def a feature more social graphs should think about supporting (btw, since we aggregate from all these different social services [or will soon], we are working on some of these related problems/ideas with knowabout.it and tagging of people [manual and auto] is a big part of that work).Overall I like the idea of using existing social graphs as a seed for new networks…but I agree it shouldn’t be a pure import…it should be more of a pick and choose…a “here’s your Twitter friends that are also on easy…the top of this list shows you who has bought/followed similar things to you…who else on the list do think has good taste in hand made products?”
Delicious tagging of people – love it – but let’s hope whoever executes it doesn’t then sell it to Yahoo π
My delicious tags for people are going to be different than yours by a radical amount. Most of this has to do with degrees of friendship and kinship. You and Fred could have a friend in common and you would have different degrees of closeness – it would make the social graph both harder and easier to use
True – it’s all about personalization…but there would also be a large amount of overlap on how people were tagged…and those would closely relate to how the outside world ‘views’ a given person…for example, most people would tag Fred ‘VC’ but then depending on how well they know him and what they interact with him about, there would be millions of other tags that could be applied.This was also true of delicious tags on content btw…and I think that’s what the true beauty of delicious really was…it worked on an aggregate level VERY well (for social discovery), but it also had a ton of value on a personalization level (for personal reference).
Think the word, brother, mother – those tags are ubiquitous (after alleveryone here has a mom at least at one point of their lives, how else werethey born) but my mom and your mom is not the same person. That’sa large difficultly with this kind of data structuring.
And for gosh sakes, put that in there at the beginning so that when the connections or contacts get created, they get tagged right then.I’m ruthless about keeping my contacts tagged and up-to-date, but then, I’m not a “normal” when it comes to contacts. π
How are you finding Svpply, do you think it can slot in as a standalone service?They’ve a nice take on social discovery and curation. Interesting new path to purchase. Their onboarding process is very good too.We’ve begun to see them and their very similar competitors, thefancy.com and nuji.com show up in Shoply’s traffic logs. It’s always nice to see a completely new category of site start sending you traffic.(thefancy is winning the referral wars thus far)
i have not tried thefancy and nuji but i will nowi really like svpply. i’ve been posting stuff to svpply and then reblogging on tumblr latelycheck out fredwilson.vc today
Heartily agree with one personal exception to this quote:”the more people you have in your social graphs on these services, the more valuable they become”The time I can afford to sacrifice on networks limits the number of connections I can reasonably maintain. I can’t be close friends with thousands or even hundreds of people, and I certainly can’t read all the content generated by hundreds of fanatically creative folks.My attention is limited, my participation even more so.
yes. 100% agree with mark here. very important IMHO to creating value in the next phase of the web.
About 10 years ago I worked at yahoo, where everyone used yahoo messenger internally. Most employees had 100’s of contacts. There was one guy, just an ordinary developer, not high up in the org, who refused to have more than 10 contacts because he said it overwhelmed him. To get on his buddy list, a space had to open up via someone leaving the company. It soon became a status thing in the office who he decided to add and chat with, people were fighting over it.It was like the VIP room, only important because not everyone can get in.In the age of abundance, scarcity (real or artificial) creates perceived value. Best example is twitter’s arbitrary 140 char limit.
If that idea is appealing, you should check out Path. It only allows 50 friends per user, built in exclusivity.http://www.path.com/home.html
On Yahoo and AIM and most other systems other than Facebook, there are 4 states in terms of perceived status. 1): You are not my friend but sent me an IM directly, 2): You have me on your list, but I do not, 3): We are friends, 4): I have you on my list but you do not.I use Yahoo messenger quite a bit and mostly let people be at stage 2, which is similar to Twitter. You can track my online presence, but I don’t particularly care to initiate conversations with you. I curate my IM lists quite a bit. It’s easier to connect with me on LinkedIn than it is to be on one of my IM lists. :)The reason is because the people on my IM list have the ability to interrupt whatever it is I am doing.I think Quora is the most natural way to find interconnections of interests. I am sure they will have a way to quickly reference which friends are following which topic soon and display relative expertise in terms of upvotes.
there is a lot of insight in this comment ed
I hear you, Mark. The question used to be who is a close enough friend to give my cell phone number to. Now the possibilities are seemingly infinite and we as people are not.But wouldn’t this filtering mechanism — the social fredometer — help to narrow down the field — so that the social relationships would become more intentional and meaningful — and you’d be connecting based on something more substantial in terms of shared interests/experiences — and perhaps connecting with people in a more multidimensional way?Maybe I heard it differently.
i think the combo you need is a large social graph and rich filtering that actually works
Yeah but those filters will have to know what goes on in my head. It’ll have to be a virtual simulation of my tastes, and driving curiosity and needs. Plus, I’d like to, need to have the final say on those algorithm dials.I wrote up a “far out” post on virtual assistants before, there’s an issue of trust in those filters. You trust your eyes not to lie to you (besides optical illusions), what if a big ad firm built those eyes for you?The two prime motivators are personal relevance and provider profit and if they’re not perfectly aligned you’ll get low quality suggestions that are just good enough, as opposed to brilliant.
+1Filtering and curation absolutely has to improve imo. I understand the value of Twitter but can’t be bothered to follow many people because the noise quickly becomes too great. If only I could run Disqus against Twitter.
So true! There are some dating sites that we’ve worked with who have thrown around the idea of somehow incorporating user data from Facebook or Twitter and I always thought “What an awful idea!” There are too many “me too” sites that want to simply tap into the success of the existing leaders. It seems to me the only services that stand the test of time are those that create a standalone brand and serve a specific niche.My interactions on LinkedIn are dramatically different from those on Facebook, and I DEFINITELY don’t want my FourSquare check-ins broadcast on a dating site.
that’s a perfect example of what not to do
I’m busy with a startup that manages your social graph from a contacts prespective. So it is easy to have a overview of your contacts 1 places. For the moment I have finnished developing the part that will let you to pull your contacts from the different social networks.Next step is to let you curate these contacts easier and update the social networks from my application.I would be interested to discuss this in more detail so I can see if I’m heading to something that people want to use π
I don’t think people want to talk about it, they want to see it! Hurry up and code so that we can all try out what you are doing!No, you can’t sleep, or eat… Just code. π
from my perspective there seem to be a couple issues worth dwelling on here:1. downloading your social graphs — it would seem to me this requires a degree of data portability. what kind of model will enable soc nets to have an incentive to enable data portability? there are also intellectual property rights here (should you be able to download email addresses of your contacts?) and thus the ever growing issue of intellectual property rights in online platforms vs IP rights in the nation-state legal framework. 2. i think any form of merging social graphs will benefit if each graph adheres to certain technical standards.i believe a federation/union/cartel of sorts is needed to really make this happen.countdown to the governance layer has begun……
oh, one more point i want to add, which i think is important. technical standards/policies often stem from a moral/ideological viewpoint — example being google’s explanation (at least what they publicly offered) regarding dropping h264 from chrome. so, to ensure agreement upon technical standards, agreement regarding morals/ideology — values — may also be needed in this federation.
you and my partner brad are in a mind meld on the importance of values in defining web services. google was great on values for a long time, but they do seem to be ceding their leadership a bit these days. twitter’s approach to the wikileaks subpoena, which i should not talk about due to my board involvement, is another example of this
IMHO governance layer will be all about values. now is also a great time forstartups to focus on values IMHO, as wall st and govt have failed us bigtime in that department IMHO
Kid, you clearly need to start this company and call it the Social Graph Cartel.C’mon, I dare you. π
my dream was actually to go meta — to enable others to roll their own cartel, and then create a federation of competing cartels!but i may need to dream smaller and focus on creating only one cartel. maybe social graph cartel is what it is meant to be……
I’m not sure Mexico is appreciating the federation of competing cartels they have south of the border.So yeah, we’ll just use yours. π
whenever I see the word cartel I can’t help but think of Dallas (probably before your time Kid)
I think Debeers.
well diamonds are a girl’s best friend!
Yet another classic tune stuck in my head today.Actually, it was more like knowing that the CEO of DeBeers couldn’t enterthe US because of trust and trustbusting issues that have only been resolvedrecently.
…and bourbon and branch.As you can see, not before MY time.
morning Donna! and of course shoulder pads
You’re on a roll!Morning, Richard!And this really is my signal to go to bed! My “quasi” New Year’s resolution (“quasi” because I don’t really believe in NY’s resolutions) is to get more sleep and to get up earlier. As you can see, I am failing miserably. I’m getting up earlier, but not going to bed earlier!Well, have a wonderful day!
Great idea Kid, meta-social. Why not start an Internet standards body for social graphs?you can get sponsered by Sun…oops I’m mean Oracle and standardize the data model for social graphs, call it XSGML for xtensible social graph markup language
I would actually like a portable social graph.So when I land in L.A. – I can quickly match my taste graph in food with Angelinos who have similar taste and quickly amass a trove of restaurant recommendations.I’d like to filter my interaction with Trip Advisor in a similar fashionSame with a Twitter search on a topic – give me the results – then run them through a filter to match the results with: a. people who I am friends with, b. people who have a similar taste graph, and c. experts on the subject.Seems like we are only scratching the surface so far.
now we’re talking!
Right on Harry, this is the type of connective meware I’ve been looking for (or trying to build). I want to see a slice of the world according to Mark, based on key relevance points such as people, topics, philosophies layering my perceptions.
Yeah, I was thinking about the different circles that I “follow” in real life. I have high school and college friends, VC stuff, Media stuff, Y Princess Group (YMCA dad daughter camping activities – big weekend last weekend up in the wilds of Northern Connecticut), paddle tennis stuff, tennis stuff, different clubs I belong to, music – which I generally break down into classical as well as rock, Hockey stuff – not to mention different art, architecture, design, etc….That’s how I mentally organize my life – and while some people populate a lot of these buckets (and they tend to be your closer friends) not everyone is relevant to every bucket – and within every bucket there are plenty of people who are experts on the subject who I would love to “follow”, as well as people I might like to meet. I’m not sure that any of the existing social services do a great job of bucketizing these interests – pushing new information to me – exposing me to new people with similar interests or higher levels of expertise. I love Foursquare – but really in my town is it that useful? I am mayor of like 6 places – and I recognize almost no one else who uses the service in the town – so is that useful to me? Does it get me deals? Do I meet new people? Can I transport it with me to Hawaii? (actually some of the tips in places are useful). I used to use Zagat all the time when I lived in NYC – and over time I came to realize that there was a type of person who reviewed restaurants for Zagat – and I was able to mentally parse through reviews with great success – often times taking a lower rated restaurant and having a better time (I used to eat out a lot) – because I was able to filter for stuff I knew I would like as opposed to the wisdom of the crowd as represented by solely the numerical values. That’s really what we need in any given situation.
Great blog post fodder (plus paragraph breaks), I’d like to see this idea develop a little more. Social relevance > generic friends
I’ve been wanting to communicate with you Harry. You would be surprised how what you would like is achievable in a short period of time.
Harry, I’d to like to communicate with you in regards to a new start-up I’m working on:www.adhd.comYelp is the new Zagat
without the curation that the zagats always did. i totally agree that yelp is the new zagat but in the process we lost a bit of qualitiy control
I wonder if the answer is that we’ll all have to manage multiple social graphs (seems like that’s Fred’s perspective) or if we can have one graph that’s ‘smart’ enough to solve the challenges raised in Harry’s comment and those of many others.My own recent experience/struggle with this stuff: I’m working on a career transition, and everywhere I turn there’s advice to build a ‘personal brand’ online (I get the idea but dislike the terminology). I think it’s valuable to connect one’s various online identities (website, blog, twitter, google, facebook, etc.), but you quickly bump into two kinds of problems – one is that you may be broadcasting things that are of interest to only a subset of your social graph (many of my friends would not be interested in my tweets of avc posts, interesting tech news, etc., etc.) and the other is that you may be broadcasting things you don’t want certain subsets of your graph to see (do potential employers really want/need to see details of my foursquare checkins/movies I’ve seen/restaurants I’ve been to? I’m not so sure.The other aspect of this is that certain channels or topics will wax and wane in importance and it would be great to have that accounted for as well. I expect I won’t remain quite as interested long term in career-oriented blogs as I am now, but it seems I’ll have to adjust that information flow manually (though I just came across my6sense, which seems to try to automate prioritizing things that are interesting to you).Based on Fred’s post today on the independent web, I went and read John Battelle’s original piece on Identity and The Independent Web (http://battellemedia.com/ar… If you haven’t read it yet, it’s relevant to the discussion here.
We’re building this, currently in private beta, but launches to the public next week. I’ll repost the URL here when it’s live.
please let me know too
i suggested that very feature to foursquare 15 months ago
I’ve been thinking about this issue for a little while now, and my accidental (on my part, by design on Quora’s) following of my Twitter graph on Quora crystallized some of the issues enough for me to write about it: http://tumblr.absono.us/pos…In short, because I’ve been following the path of least resistance and using my Twitter group as my entry point for most new services — a practice that’s often explicitly endorsed by the services themselves — I suspect that I’m not giving new services the same attention I gave the earlier generations of “social” services. From my post:”Itβs easy: rather than having to dig in and figure out who are the interesting users on the new service, I go to my familiar list. But is that warping my perspective on that new service? Does this habit mean that Iβm not really giving the service the sort of investment that I gave to Twitter or Tumblr a few years ago?Rather than finding the new user community that makes the service interesting for me, maybe Iβm settling for finding out how my existing community views the service.”
Honestly, I knew I couldn’t do that. My facebook has school/college/other friends and a few avc’ers. twitter is mostly tech people. And I just registered for SVPPLY – I can’t follow the vast majority of you.and one of the fears for quora is it is too much of a tech centric crowd. Lucky for them, they designed a follow by community (aka type of question). So i can hang out with the exercise people there if I want
“interenting users” LOL! Please expand on that?
It just means the users that are most interesting to me in the context ofthe new service: the people that I already follow on Twitter may not be theones who are “most interesting” to me as a user of Svpply, ExFM, or Quora.That said, it’s easier — though possibly not “better” — to just use peopleI already follow on another service as the basis for my social group in anew service, rather than investing the time to figure out who’s using thatnew service in ways that really appeal to me.
Definitely an interesting point, and a common trend. Why invest the time in building a presence on a new medium when almost all the people you care to talk to are already on Facebook or Twitter? There’s a great book that talks about this in detail called The Viral Loop:http://www.amazon.com/Viral…The big social networks have hit a point of non-displacement, which means newer services have a hard time surviving because the big guys are just so well entrenched. A great read for anyone in this industry.
and that is a huge problem for innovation
Agreed. As VC’s, how does this impact your evaluation of business opportunities for your portfolio?
well we are not giving up on new networks of scale getting created.but we are cognizant of the issues you raise. we look for novelapproaches for rising above the noise. and we also are looking fartherand wider for networks that seemed to have reached a critical mass(berlin for soundcloud for example)
Thanks for the pointer — haven’t read The Viral Loop, will pick it up.I think I may be wondering about something slightly different, though: it’snot that I don’t *use* the new services, but I’ve started to wonder if myexperience of those new services is shaped (or warped) by the fact that I’mimporting my social graph from a different service.A few years ago I spent time figuring out who to follow (or not follow) toget the best experience of Twitter; now I’m more inclined to import thatTwitter circle as my starting point for a new service than building it upanew each time…I *use* Svpply, but I view it through the lens of a socialcircle that I defined on Twitter rather than one I built on Svpply.
you wrote my post way better than i didi am so glad you grokked what i was saying whitney
Good points. Each social network should have their own graph. I use Facebook with different people than Twitter. This is where I think Google failed with their Wave and Buzz. Your email/google social network isn’t usually the same network you would use in a new social network that has a different purpose.A possible solution to managing separate social graphs is designing an open standard for social graphs, so that it can be exported from networks. With this export, an app could be built to easily manage each social graph, then re-import it into a social network.
The answer here is that the internet infrastructure upon which we are building our new world needs an operating system. Your social graph is a cross-cutting concern over many applications, and there are numerous cross-cutting concerns like this – mostly identifiable as points-of-control. Traditionally, applications run on an OS, an the OS provides these horizontals across the application verticals.Somehow we managed to deliver a couple of things with the internet platform, and now re-invent the wheel for everything else, every time we build an application. You end up with every bloody application having their own social silo – great if your goal is to create a point of control, not so great if you are a user.Users need an OS, but businesses either don’t want to relinquish points of control to that OS, or want to fight to own them. When business works against the interests of customers things get messy.
I was thinking the same thing, Colin. As I read Fred’s post I was picturing a LARGE open source database project with a user friendly UI and ports to different devices such as mobile, etc. One would import the social data from each service, curate, balance, connect, then consume in pieces as needed from each device and location.The elephant in the room, as you and others point out, is the portability of the “assets” β the social graphs from each service. If each service put the user first (over business paranoia and greed) the problem is more easily solved. And, frankly, there’s probably little risk to those businesses which do take this approach because execution and respect for users is what keeps users.
Comment #2 here:So Fred, doesn’t this sort of beg the question of just how many social graphs and social sites any reasonable person is likely to use? Or do you just do everything through Facebook going forward?I’ve always been of the mind that Facebook is a great aggregator of information – but does just a few things particularly well – and that other sites do very specific things better – but don’t aggregate as well.But when you start going down this road – with every web app having a social layer – at what point do you look at the discussion on Disqus today and say enough is enough and declare social bankruptcy.Do I really need to be following or followed by people on Facebook, Twitter, Foursquare, Quora, Etsy, Svpply, Pandora and other music sites, etc… or does Facebook just become good enough for the masses – and then what happens to the data that resides on the other applications?
For most…the masses…Facebook is enough.I’ve been thinking this through from a personal branding perspective. Most people think that Facebook is creating a personal brand…when in actuality, it, like Twitter, like Tumblr are channels.For those that drive everything back to their own brand, their own URL, their own personal social brand, having a piece of middleware, a sorter like Fred suggests, is a real time saver as they deal with each of the these place or channels individually. For the remainder whose online persona ends at FB or LinkedIn or even a static web page, this is not as relevant.Not sure that this correlations plays out…but it appears to.
“For most…the masses…Facebook is enough”What a scary thought
yup and to most, that is all they think they need.The issue is that is not all that most anybody needs. If you work or are looking for work or pursuing interests or gathering a network, without something dynamic in your online string of contacts (blog, microblog, Disqus, whatever) you are working with a handicap.Seems so obvious. But isn’t to most.
Great point Arnold
as i said earlier arnold, i think this is not accurate. it may have been a year or two ago, at least in the younger crowd, but not anymore
Got it from the comment above.I love when for all the time i spend online, working with clients, building by own networks, my thinking is always iterating and changing. Nothing static in today’s world and that’s what makes it compelling.
That was said about AOL too. Imagine if we stopped there.
Funny, that comment made me think about AOL too. Facebook is the new AOL.
i hope they don’t acquire Disney and then fuck it up
Can’t you just imagine?Toy Story 7 1/2: Farmville Adventures
i totally disagree Arnold. and i am seeing it play out with my kids and their friends. they are actively cultivating distinct social networks in many places on the web. my daughter told me “facebook is like pants, you can’t go out without them, but they are not enough, you need more than pants to dress yourself”
Thanks for sharing this Fred!I can’t argue with the truth behind the clothing/fashion metaphor. I need to broaden my datapoints it seems.
Harry, I had the same thought, but didn’t comment for fear of being a “social wet blanket”. π I was thinking “good grief, how many social contact points are needed/appropriate”, and more importantly perhaps, how long will they remain meaningful to a user who has so many? At what point does one decide to pare back the list? Quality over quantity is my first interest (and I’m sure Fred’s and others here), but we obviously reach saturation at different points.
Clearly Facebook is going to get better at grouping “your social networks” into distinct pools and letting you decide how to share with each pool.You can then extend that to connecting new social products in and saying “I only want to follow my tech friends and foodie friends on Quora.”Both FB and Twitter have dipped their toes into this water with lists.But especially on FB, it’s so hard to keep people in lists and move them around. I just gave up for now because there’s no usefulness on the back end.
i don’t know anyone who wants to use facebook to do this
I agree completely. I’m just saying that I think Facebook wants to be that “master network.”
The trust isn’t there. The foundation of trust isn’t there. A large enough group of people are waiting for viable options, and they will come.
Yeah we need to go back to the social graphs of yore: Princeton, New Canaan, Stanwich….. π
This sounds like it could be just dripping with sarcasm.That seems to be the key though – those are my social networks to some degree – I am a product of my environment. The question is how to find some new and interesting people to populate those networks.
it’s not sarcasm if i neutralized it with a winky-face!here’s a full-on smile for ya: :-)hey look we’re all in the same boat (although you are particularly, covetously, blue chip)but yes i think our traditional circles get a little stale and there are parallel circles and people who could keep injecting fun into it but you may never know they exist if you go about your daily analog biz. so we are crossing and stacking circles.the discovery stage is incredibly fun. the sorting/organizing pruning and is incredibly boring and tedious.
Harry, the answer is eugenetics!You submit your “genetic / DNA profile” and the software will the connect you will “new and interesting people”
if you want to get value of of social software, i think you’ll eventually need to have at least dozens of rich social graphs. i am preparing myself to have a hundred or more over the next decade
Yes it seems inevitable but the obvious outcome of that inevitability is really only dealing with other folks who are highly active users of social media software – or you will end up friending or following a whole lot of people whom you do not actually know – but you know enough about to willingly share information.
i think everyone will be active users of social software in the next ten yearseveryone uses google now, right?
Instagram does a good initial job of this. After you take a photo and link some social networks (twitter, fb, tumblr) for sharing it, it’s very simple to curate your Instagram friends. But there’s a classic chicken and egg difficultly. When I join, I’ll only follow a few of my friends that happen to be there before me. When more of my friends join, it’s not easy to know they’ve joined unless they follow me and I happen to see that in my news stream.I would like to see Instagram do some additional cross referencing as new users join and present me – a current user – with a list of my friends new to the service, on a regular basis (daily or weekly, in-app or via email, respectively).
I want a social graph organizer for all of these social graphs
I’ve thought about this too but I don’t think most sites are open or comfortable enough to let their userbase be more easily accessible.Smaller networks benefit more from a cross-link into a larger network. How does this interplay happen between competitors then? Clearly the best executed should be able to win out, so in the end the littler competitors just help feed the bigger guys who execute better in an ecosystem like Fred suggests. But I digress, because this will happen, should happen, needs to happen. It’ll be good for the consumer.P.S. I have a feeling this post is ‘foreshadowing’ one of Fred’s latest investments; “Making the social web more useful” seems like a good tagline for something like this. But I could be wrong. π
Good guess Matt, based on the shadow of what Fred’s preoccupied with slipping into posts π
He does have a tendency to prime us with what’s needed and missing, and then magically filling in said gap. π
I feel like this is the thing I should build. I’m sick an tired oforganizing this stuff. It’s radically unhelpful. Every big socialnetworking site has different ways of organzing this information too, andall i want to do is make sure the right message goes out to the right person
There are very few individuals who’d benefit from putting the resources into this – but it would be fun to do! If I find the time and my projects go well enough to give me some ‘free time’ then I’ll see about connecting with you about this. π
I don’t agree. Think about the power of a derivative social network from the perspective of an advertiser. Being able to tap into a different network that responds/reacts differently is HUGE. Imagine if the derivative social mapping site offered virtual currency in exchange for use (e.g. Zynga, QuePasa). You could also create a system where advertisers literally pay people to pass messages to their “friends” either in real/virtual currency or in product (organically create brand ambassadors).A natural ecosystem of people who only pushed advertisements would be created, and they would in turn lose their “friends” who felt that their connection was abused…eventually, you would get people passing the right messages to the right people in exchange for the right reward. This would give advertisers a high percentage of targets reached and allow people to map their world in a way that was both fun and valuable without being overly annoying/obtrusive.
Sorry I wasn’t meaning to imply there isn’t money to be made if you created a successful ecosystem as such.
Honestly, while I think the ecosystem is important, I also think we rebate the overwhelmed point. Hence a butterfly net for catching and organizing
you’re a doll Matt
Awww *blush*I think you’re a sweetie too. π
awwww
Google and some other companies tried something like this with the “open social” initiative but it never took off. src: http://code.google.com/apis…Ostatus is having more success with open protocols and interfaces for cross network data sharing (google buzz works along the same protocols). Pubsubhubbub/Salmon/Web Finger are all pieces of the ostatus puzzle. Different identities serve as various sub network nodes, and the data is all interchangeable.Diaspora is trying their own xmpp shake on distributed social networking which would enable cross network “pollination”. Separate aspects are designed to represent different networks/groups. You could have different aspects for different groups.
I’m less worried about protocol than the necessity of getting the message out.
P.S. again what do you mean by “vaule” or “useful” in terms of a social web.I believe it is no different from the pre-Internet real work except that meeting people is easier.If your “value” is to meet people to general leads for your real estate business, then in the real world you hang out in local bars, sign up for every local chamber of commerce event, show up at every kids festival, …. and hand out your business cards.On Internet social netwoks it is the same thing. Are you looking for “friends” “sales leads” “community / political social groups” ….I believe the social web is consolidating along similar lines as the real world.* Real Friends* Business Network* Community / Political Network* Church Group* Rolodex
there’s no “oreo tell” here. i am not working on anything seriously right now that would be anywhere near this
Interesting. π
I’ve been working on a platform-agnostic graph analysis/management tool idea (TribeTrax) for the last several months that I think solves this problem Shana. Should have something to look at relatively soon.BTW, anyone out there interested in adding TribeTrax to their portfolio? π
Will there be data surfacing tools in it?
What types of data and/or tools are on your wish list?
seeing how crossovers of my networks changes influencer patterns
Also – why do people want to follow you on SVPPLY? How much do you think it has to do with the pulling in data situation? You’re a VC, not a professional furniture designer (now that would be an interesting person to follow)(not that you have bad taste, it just leads to some interesting peculiarities about this system)
pretty harsh that you just dissed fred on his own blog in front of everyone by saying he wasn’t worth following and implying his opinions are void of value. kinda brutal, shana.
Not what I meant. he’s a VC, and he has good taste in stuff. I just thingfor the vast majority of people (even people here) he’s probably lessinteresting to follow than say, Jonathan Adler(http://www.jonathanadler.co… just for reference here), who is ahuge tastemaker in decorating ones’ home. I don’t expect Fred to be atastemaker there (his wife probably, she’s a former buyer). If this wasfollow Fred along about technology choices, that would be a whole differentballgame. That would be very fascinating.It also points why there should be many social graphs. There are timeswhere I am not interested in what you all say. If I want social networkingabout beauty and makeup, you can sure as hell bet I am not here (I’m atmakeupalley or naturallycurly or refinery29 or satorialist) and I wouldwant to follow those people around (they know something) Asking Fred aboutcool lipstick and nailpolish and what clothing looks good with what is justweird. Same thing with decorating a space.Same thing with food. Fred doesn’t mention he cooks, why would I startfollowing him around on food52?Am I the only person who finds this conceptually bizarre?
I heard that Fred was a pretty mean contributor on honestlynow Shana :)and he doesn’t need to cook he has the Gotham Gal and he can’t compete there anyway!
I wouldn’t know, I still don’t have an invite to HonestlyNow. :(And you’re proving my point when it comes to the GothamGal. And also whyparts of “overlapping graphs” cam be annoying useless.I do think and there is some evidence to believe this, that people withsimilar tastes in one area should have others that overlap. However, I alsothink that its far more useful to figure out from those conversations andconnections who are the tastemakers in different fields that differentoverlapping groups look towards.
omg sorry shana i will change that NOW!BUT — you have to promise to put yourself out there and ask a question.:-)sorry i’ve been sparse on AVC but i’ve been having too much fun.yesterday i asked if i should have the mole on my face removed. (answer: NO!!!)and we learned (from Erin at RedStamp) that you CAN leave kids’ birthday thank you notes in the kids’ cubbies at school (in lieu of mailing them) AS LONG AS every child was invited to the party.and Erin learned (from Sue at thezhoosh.com) that a pricey mohair couch DOES stand up to young boys’ abuse. (this was a full-scale smackdown of the ill-advised crowd who said it wouldn’t)today i might — just might — ask you what you think of my latest bathing suit.why make a mistake if you don’t have to? it’s market research on yourself.we have MUCH to learn about what fits in our real lives.just sayin’.(PS — if anyone wants one — please send me a request at [email protected] and mention AVC in the body and i’ll take care of you.)
…and my son gets to keep his hair long, thanks to Honestly Now!But can’t convince my husband to shave his facial hair — his argument about everyone saying that he looked younger without it is that the clean shaven photo WAS 10 years younger. Oh well, I tried.(Still have my doubts about that mohair sofa.)
what is HonestlyNow?I find that phrase to be not real, when someone is speaking to you and states”honestly now” I am thinking that all the things said before that were not honest …I don’t like the phrase ‘To tell you the truth” either as I find that it just under values everything one has said before that phrase and I don’t have a desire to listen any more.Could you explain what HonestlyNow is… is it a place where you can drop the niceties and not worry about hurting anyone’s feelings etc..”Diplomacy” is when you tell a person to go to hell and they look forward to the visit. π
Attach the phrase to the question, not the answer.My inspiration to build honestly now is rooted in my experience when I lost my mother very suddenly (in fact within months due to bizarre circumstances I lost my whole family).We had all been extremely close immigrant family. I realized I asked my mom for advice on all types of decisions, multiple times a day.It could be my clothes, my body, my health, my dating + relationships, my apartment, my wedding, and ultimately my parenting skills.She was a very smart cookie and always had awesome advice. She was assured to tell me the truth because she wanted what was best for me and could see my potential. She also would not mince words when I was making a mistake. It was both for information and validation, but as much for social connection. Check-ins. Why make a mistake if you don’t have to?So I’d ask.But then one day I couldn’t.I’d ask my friends for feedback. I’d find myself asking, “honestly now?” They’d just tell me ‘great!’ ‘you’re perfect!’. They didn’t want to hurt my feelings.Since part of my background is market research, I thought, what would be the optimal daily market research tool to support personal decisionmaking? So I designed it.You can ask everybody (should I get this couch?) or privately to just your closest friends (getting chemo — which wig is best for me?). High-end pros can weigh in with specific very useful advice (when you want it).So yes. We cut through the clutter, tell you what you need to know about yourself and what you should do (which you probably can’t see yourself). But we have been painstaking in making it a positive, safe + respectful place. You may laugh (loud + hard), but there’s no snark here.Make sense? Honestly, now…Tereza
In recruiting, the company’s story is an important part of my marketing plan in generating the interest of prospective candidates — which I guess is a basic element of any type of marketing. Anyway, I’ve heard a lot of stories. That is a wonderful story.
hey wait, i don’t have an invite to honestlynow either!
email me at [email protected]. i’ll spot you one.
Leigh — If you don’t hear back right away (sometimes disqus doesn’t seem to report the comments past a certain layer) you can send your email to dwhite AT bwasearch DOT com and I’ll invite you.
Yay I got on
You’ve gotta come play with us…so fun…and actually quite helpful! You will be hooked.(Not that I contribute as much as I could, but it really is a great resource!)
Darling, you could NOT POSSIBLY take Kid seriously when he yanks your chain like that. I can hear him laughing all the way up from the Sunshine State.Kid knows damn well that Fred doesn’t know anything about fashion. Fred’s said publicly that he doesn’t know anything about fashion.Fred nonetheless manages to dress rather well.It’s this fabulous new line: Gotham Garanimals.
Have you ever seen that one jacket — some sort of motorcycle jacket/coat that he wore on one of the videos? Wow, I still remember that.
it’s a belstaff “racing man” jacketi just posted it to svpplyhttp://svpply.com/item/2442…i have an earlier year’s model without the american flag. i like it better without the flag
Thanks, nice to see that jacket again. I like it better without the flag too.
this is great. i love everything about this comment tereza.
Agreed – I still think listening to responses helps cause clarity as to what is being said.
awwww, pickin’ on Shana, Kid?you’re just acting out because you yearn to diss $teve but with his latest health issues that would be in bad taste.
i am mostly curating on svpply, not posting much. so i am taking what real designers and fashionistas are posting and picking my favorites. i think the result is pretty cool.http://svpply.com/fredwilson
I haven’t even figured out how to post! I happen to like most of the stuff you post, however I’m still a little lost about the why…
I might be wrong, but it sounds a bit like what Cafebots (part of the KPCB sFund) said they were doing – friend CRM.
Exactly!Case in point. I just learned that Andy Swan plays tennis. (I learned that on his Quora profile, not here)Harry DeMott and I ‘met’ here but we then learned we live 5 minutes from each other and in fact have used the same tennis pro.A gal who’s the mom of my daughter’s preschool sweetheart turned out was the HR person at Etsy. We became fast friends and she intro’d me to lots of people (and oh btw she grew up with my best friend in my town).A high-end photog/former model who does work supporting cancer (a cause I care about) and turns out we know some people in common…but more importantly have some very aligned professional objectives.I could go on and on with these. But it’s how my brain is wired and many are not. Artificial intelligence i’m sure can do it better.Our social graphs — the overlapping Venn diagrams in our varied, mulit-dimensional lives — and seeing those commonalities pop up is one of the great delights that exists and can and will bring a richer human element to the web.
I’m sure it is rare for Andy to lose while playing tennis, but given his focus on winning, that may be an instant YouTube classic.
http://espn.go.com/sportsce…
Natural networker, you could teach those fancy algorithms a thing or two, and you should.
here’s my secret, mark (shhh):you have to listen to them.people have an amazing ability to tell you exactly what the want and need. ask the right questions and pay heed to the ‘superfluous info’. it’s *never* superfluous. it oozes critical data.A slightly dramatized example: ‘just had surgery on my knee.’. ‘why?’ ‘tennis injury.’ = conclusion: he plays tennis! wait. my pro teaches in his town. do they know each other? omg of course!! hahaha! wait. i know a great guy you should partner with for paddle, national caliber. lemme know when you need someone off the chain so you can win your club championship + i’ll intro you. woo-hoo!.about 2 years ago i was sitting in my house alone in the deep woods between preschool pickup and dropoff and decided, “hmmm….i know tons of people but none relevant to what i want to do next, and nobody who i need to know knows who the hell i am.” so i committed to working the networks. not really sure where it was headed.chip by chip, convo by convo you learn there are people with big intersections with you, right in your backyard, or a sympatico BFF on the other side of the country.the big step function is definitely blogging because people start finding you. you’re a producer, an agenda-setter. not just a consumer. love that. but i’m preaching to the choir here.i never, ever feel like i am done meeting people. i get a nervous panicky feeling every day, ‘who do i not know yet that i’d want to?’ it’s an addiction.with the accelerated social data coming at me, i’m getting better and faster at deciding who’s in/who’s out. i use old-fashioned criteria to do it (e.g. a fully enforced no-asshole policy). i guess i could create and use an “asshole” tag within my toolkit but that would infuse all kinds of bad mojo into my circling spheres. for now, i’ll remember it in my head (a steel trap in remembering asshole-ness)what does frustrate me is lack of integration of all this with my “Family CRM” as fred has called it. my holiday card list is a damn mess. i need it to auto-refresh to reflect the year-long interactions over the soc-nets.trying to decide if i should lop off my (dead) parents’ old friends who’ve never ever called or written in years to check if i’m ok (do they think i don’t notice?). i feel guilty. but i guess it’s time.when you give give give and they don’t give back, i write them off. who has time for selfish people.out with the old + in with the new, i say.
exactly! except these are the details that never seem to pop up through a social networking application.(Also, random, is it possible to learn to play tennis later in life)
one of the most beautiful things about tennis is that you absolutely can learn it later in life, and many do. and you can play it into old age.you can also travel anywhere in the world with your racket and find someone to hit with at a hotel or nearby club.wish i had more time + money to play but it’s back-burnered for now with the startup.you learn so much about people on the court!my favorite: mixed “troubles”. i love shakin’ it up with the guys.i’m not as consistent as i need to be but have a great arm from VB and love finding holes on the other side of the net.what’s not to like?
When I have the money its going to be tennis and ballet (something about the grace I find appealing)
It is interesting because I also know many world-class talented people that try to “escape humanity” rather then spend evenings on the Internet trying to “network” with more people.Personally, I believe that there is a “quality” vs. “quantity” issue in social networks.Some I people know meet so many people in the real world that they are trying to filter out people in thier “real world” network to increase the “quality” of thier lives vs. going on Internet social network just meet even more people – there are lots of people in the world.Other people I know seem to define themselves based on how many people the “know” and blur the lines between “friends” and “business” – these are the real estate and financial advisor types that sells mutual funds to thier friends at church.You’ll find the same division “pre Interent” as current. Interesting all the same.
quantity is an issue, for sure, and in fact it’s one i’m planning to address with my business.indeed there’s a lot of benefit in quiet. i live in the woods in one of the small towns where world-class people go to hide out on weekends (i’m the trailer-trash). problem is, if i didn’t reach out, no one would talk to me! most of them alternative between over-exposed then alone. it’s how it rolls.i assume they’re not spending their evenings on the internet. they’re either recharging with the people they care about, or going out to dinner with other world-class people. because the more known you get, the more exposure you get to other people like that. it’s a fact.tricky thing about sales. i come from sales. you’re always prospecting new customers. relationships that become legitimate friendships are really the ideal. without it, you may as well sell it in an amazon shopping cart. what the human relationship brings is deeper insight, better anticipation of needs, and most importantly, when something goes south, a human being who cares takes care of the situation, making it right. indeed almost 100% of my great customers/clients have become very good friends. it is a great joy when that happens.some smarmy people abuse it.
Just gave me an idea..Putting Tereza, Shana, Mark, Charlie, Kelley, William, etc. on a ‘Quality’ twitter list – seems like maybe it could just be called AVC…
I did this already. Granted, it needs to be updated slightly (ping me shana dot carp at gmail), however, the avc.com people list still lives!
you totally get it Tereza but unfortunately getting our social software tuned to our lives is a royal pain in the butt
You’re exactly right.I’ll add one thing: you mentioned you don’t want an e-mail when people follow you. I think that depends on how you are using that social network.And the last thing that I need is yet another inbox to check. I’m constantly cutting down what services can send notifications to my e-mail inbox, but I at least want the option.So right now, I don’t get friend requests or follow notifications via e-mail. But I sure do get my Facebook messages and Twitter DMs via e-mail because I don’t want another inbox.
I want the next generation of personalization but where my social graph, reputation and associated data aren’t stored with each Web service rather they are stored with me – i choose who what where and when – i hold the knowledge over time – i control my own data.This was the premise behind my now defunct start up that was probably too ahead of it’s time – but eventually someone is going to do this and have the right timing — i for my part will be the first to join. π
the question is how do you get there? offering it at face value seems like vitamins and the gym. we all know we should do that, but we don’t. is there another way to get there that starts with something we all must have right away?
I’ve thought social search might be a great way in…..Ala – I have this ever growing network of social connections, both implicit and explicit. But as with everything in life, people have different meaning to me and if I get really utilitarian about it, they also have different value.I have a friend Dave whose social graph info may imply he’s a legal specialist. But because I personally know Dave, I know he’s the gadget guy. You want to buy a new tech toy? He’s the guy to talk to. And if I was going to “search” through a search engine about a new gadget, I wouldn’t want to search like Leigh (because I wouldn’t know the first thing about new tech gadgets, bc I just go ask Dave), I would want to search like Dave….What if I could somehow connect my search to Dave? What if I could tag to explicitly define our relationship – Dave as ‘friend’ and also tag what I think he’s an expert in – in this case, ‘IP law’ ‘technology’ ‘gadget’ etc etc.? It would allow me to potentially then use Dave’s search intelligence for my own.Well, that’s great. But what if it could then get even more interesting and we could connect Dave to all the other people who have been tagged as experts in ‘technology’ and ‘gadgets’? Wouldn’t I theoretically be able to have a socially created vertically brilliant search engine…And that knowledge could be taken with me over time and applied to all my networks – it’s not just search but more an intelligent social graph system.There you go. Easy breezy right?
Fred, is it okay if I send a little more ‘detailed’ explanation regarding that point (the brief sent in 2010)? My example regarding management was too pointed at cognitive ability.
yes
Great post. interesting to think about how these communities would need to be organized — are they symmetrical? ie – do we both follow each others activity? or can I follow you without your having to follow me?- how are privacy features enabled? but with privacy features enabled, are these social graphs actually useful anymore (since they only represent a portion of someone’s activity)- is it possible to manage “community overlaps” if someone has multiple interests/behaviors that overlap with mine, is there a way to connect that person with me?- how open is this data for other services to interpret / use?
The emerging social web is still very much in its infancy. Right now, the tools to make sense of it are pretty crude, but lots of folks are working on ways of making it easier. (Full disclosure: my startup is one of ’em.)Lately, I’ve been trying to make sense of this problem by thinking of it as two distinct pieces. First, I have a social graph that encompasses all of my offline and online relationships, regardless of service. Second, I have an interest graph, which covers everything I’m interested in. Both graphs span all the services in which I participate and both are represented differently in each service.I friend or follow some folks because they’re knowledgeable or interesting on a particular topic. For example, Robert Scoble isn’t a buddy (at least not yet), but he has interesting things to say about emerging tech. The same quality that makes him a great person for me to follow on Twitter and Quora would make him an odd choice for me on Etsy.But if Robert and I WERE buddies, I’d follow him on Etsy too. I seek out my “friend friends” in all of the services because – rather than what they talk about – THEY are what’s interesting to me.In reality, I think two things are needed. I want to be able to connect with my friend friends everywhere and I want to be able to follow the best people on a topic on any service where there’s good, relevant conversation going on.
Hi,I am working on an assignment on Social Graphs and have been asked to create Social Graphs for the audience of a local product/services search site. As i understand a local search site will have anyone and everyone logging in to look for information or searching for products and services as their audience…so in this case how can i create their social graph….can u help.Preeti
μμ°, λμ€μ»€μ€ μ΄λ°κ±° 보면 λκΈ νλ¬ν° μ λ§ μ’λ¨ μκ°μ΄ λ λ€.ν λ‘ λ¬Ένκ° λκΈμμλ νΌμ³μ§λ κ²μΈκ°
Fabulous comment, brilliant really! Thanks for sharing.
Couldn’t resist running it through translate http://translate.google.com/#“Wow, look at this online diseukeoseu think quality really admire.Discussion will unfold in the culture at”
i think it is spam and i should delete it but the reaction from JLM and you is just to good, i will let it stand
muy bien.
No, this is not a spam. I’m a Korean and can read the comment, which says “Whoa, the quality of comments in this Disqus thread is really good — looks like healthy debate culture is happening in the commenting sphere.”
You begin with, “this may be a feature that many people don’t need”…the fact is it is something everyone needs and it is a simple matter of maturing. As with Accelerating Returns, the maturing of users takes place at a more rapid pace.Admittedly, we would not assume the person who likes to post the most irrelevent things on FB would have any need to have multiple social graphs. Give it a year. That person will want to seperate contacts, the beginning point being home from work.The ‘me too’ realm of social graphs will continue to grow and what will help is a way for the consumer to take advantage without using too much labor ;D
right
Agreed. This becomes even more relevant when importing social graphs to look at friends-of-friends for example… GIGO
If you’re thinking about email as a source of a social graph, have a look at this little project made during HackMTL last November:http://nicolas.kruchten.com…
will do
I am playing on something related in my spare time, it’s more for my personal/business consumption. I think this kind of technology is for “social geeks” since probably you need to adjust some “parameters” instead of following the current recommendation/filtering systems approach based mainly in algorithms.Data Portability, APIs and feeds are what make these things possible
Fred, can’t you make this happen? Great point and idea. Seems like only a matter of time until all this is available. I think the logical question is do you want to let each of those services know everything about you that the other services know or does there need to be one central repository/company where you can manage everything. It seems like it necessarily has to be a neutral service/middleware type application who is not incented to do anything else for you buy manage your connections to the other “social sites.” Good idea and I am sure you someone will crack it – Adrian Meli
making it happen starts with getting the idea out there.
The problem I have with all these social networks is I’m starting to feel like I’ve got multiple personality disorder. Why do I have to curate multiple networks? Why do I have to do any work at all? I think the logical endpoint is that the network will (eventually) move to the edge. I’ve been thinking for a couple of years about this, but I have neither the time nor the resources, but why isn’t somebody building a peer-to-peer social network? You could probably leverage a bunch of existing open source code from Bittorrent or some other thing to deal with the whole discovery process. Then you only have to deal with one social network and what you end up nurturing or managing are your relationships, not your networks.
i don’t think you want to deal with one social network anymore than you want to deal with one cable company or one president for life
Fred, I agree that I wouldn’t want to deal with 1 social network in silo. What if, however, it were distributed using a peer-to-peer protocol like bittorrent? All of my posts/tweets/pictures would reside primarily on my local client but would also be distributed/replicated at my peers.Clearly the reason this isn’t being built is because there’s no straightforward business model, but then again the Web grew in an academic setting, not a commercial setting. Hopefully this could be built open source.Frankly, I’d much rather have just one social network with all of the people I know and/or follow. As long as you allowed for asymmetric links, you could accomodate any and all types of relationships. Then, I could focus on managing my relationships instead of managing my networks.
I’ve been thinking about this a lot lately too – seems like Hunch should be well equipped to build
I am no Luddite, and participate in my share of online social media, but let me suggest that “curating social graphs” is not something that many people will ever want to do; nor is it something that will be very good for anyone’s mental health.
I question the whole concept of “curating social graphs”. A 1% feature, expensive to implement, and probably actively harmful to those who spend their time checking and unchecking boxes on their phantom social network.
These posts are awesome… Any chance of writing an article showing an actual Term Sheet and what to look for?
you should read brad feld’s series on term sheetshttp://www.feld.com/wp/arch…
Exactly, there is a time and a place for everything, and that includes social graphs. I remember you talked about the Implicit Web, and it’s an interesting phenomenon to think about as how some information can be taken with you elsewhere, or services can crossover. For instance, the other day I saw a very unique application of Foursquare. I could set it, if I check into a venue more than once, then it would automatically be added to venues on this site.Some networking sites are telling us who we should be friends with or who we should connect with based on an algorithm. This may not hold true on the other service or site that you join. If you’re talking about social shopping versus social checkins, these are two completely different kinds of services and should be tied to different networks. Some people you may follow on Twitter you may never go to the same venue that you visit. Some relationships are completely virtual, whereas others are more real, and others are location specific. One of the challenges with social networks in general is mirroring what’s real. The better a network can mirror a real world relationship, the more likely it’s social graph will be a stronger representation of a network. As soon as a network has created an algorithm telling you who to connect with, it’s not mirroring the real world connection, yet a connection based on a virtual action.
that’s exactly right eric
I posted a longish post on how this could apply to travel – I think there are lots of opportunities for specialized social graphs as well: http://www.charleshudson.ne…. I think it will be interesting to see how new companies take advantage of the opportunity to help consumers find like-minded people outside of their Facebook social graphs.
FoodSpotting needs a social graph bad. It’s not what you had, it’s who you had it with. http://goo.gl/fb/NFnso
I agree with your fundamental statement that each social graph is unique. The friends and family graph might not care about all your professional shares but your professional graph does. Also your running graph (DailyMile) is different than the foodie graph. However being able to see which ones of my existing friends belong to the other graphs is great. I hate showing up to a new social network and have to start networking all over. However it’s always nice to see familiar faces already there.
I’m surprised no one picked up on my comment re: Cafebotshttp://www.cafebots.com/abo… “CafeBots is the first company dedicated to Friend Relationship Management. By building applications that are useful, fun and scalable CafeBots wants to let people extract more information from and make better use of their social graphs.”
i will give it a try. do you use it william?
No, I don’t. There isn’t much about it yet except a stated direction during the sFund announcement. Maybe Bing Gordon could tell you more?
It sounds to me like you’re interested in a “friend” recommendation engine plug-in for all the various web services with social graphs (Twitter, Foursquare, Facebook, etc.). I think it’s a neat and possible idea–it would just need some user input on the front-end on how you use the services, whether for news, politics, music, photography, restaurant reviews, etc.From the social services I use, I’ve had the best luck finding new people to read via Tumblr’s reblog function. It’s easy, I see the content the user generates, and I can tell right away if it’s the kind of stuff I want to read on a frequent basis. Twitter’s who to follow works somewhat well, but telling me I might like to follow Mark Suster or Chris Dixon because I follow you doesn’t seem like a real accomplishment. Similarly, Facebook telling me I may know someone isn’t that helpful or useful. It’d be much more interesting if Facebook suggested people I might want to be friends with.
twitter’s who to follow got a lot better sometime in the past weekit’s become much more random but still very accurate
Thanks for the heads up. I’d stopped paying attention to it, but now, I’ll give it another whirl.
One way to do it. What about a top 5 list generator for social applications, similar to what Joe Lazarus built for last.fm? The criteria would be different for each application, but the idea is you recommend new Twitter feeds or Tumblogs by showing the Top 5 of different users.Each social network Top 5 would rank different things, but by generating a Top 5 (for willing participants), you could use the top 5 list to discover new friends/feeds with common interests. For example, on Twitter, in addition to recommending who to follow, when I visit someone’s feed, it would give me that user’s “Top 5” list. Probably comprised of something like: which feeds did that user spend the most time on last week, which users’ tweets got the most clicks, which got the most responses, DMs, etc. On Tumblr: which blogs did you read the most (time spent), which did you reblog, where did you comment, etc.This doesn’t solve the problem entirely, but it gets you closer.
I think more people need it than you think. Thankfully, since you posted this, I’m sure there are several nerds up for the task and ready to deliver! I’ll be waiting for your follow up review when that happens π
this is completely opposite to what facebook believes in I suppose and Google is kinda talking/doing things that mimic your points. Many of their webservices latitude, hotpot, reader etc seem have to a different set of friends. Boutiques seems to be mimicking twitter but only for fashion clothes. Rather interesting.
yup, i believe in the exact opposite of what facebook believes inon purposesame with apple
but I wonder which model will win, looks like facebook is winning for now. People spending more time on facebook, facebook making lots of money etc. I also think this is a smart move by Google to differentiate itself from facebook, instead of simply copying the facebook model. I guess these are the most exciting times ever for the tech industry, so many different models at work here, so many different kinds of platforms at play. Microsoft is looking totally lost π despite their recent xbox, kinect successes
“But the more people you have in your social graphs on these services, the more valuable they become”This is a great post. What do you mean by “value”?Example, do you “value’ your social network friends and contacts by some financial measure, like potential adv spend?Is it like the those real estate agents or financial advisor types that hang out in bars or meet-up events to make friends and then give you thier business card?Because I know people that see Internet social networks as the ultimate sales opportunity, I see that as very sad.I personally stick to 2 types of social networks. Those that are are only to value people in the old fashioned hands-on frienship way that I meet in person and then use the Interent to keep in touch with them and those that are pure “business” contacts from which I’m see some future sales $$$ potential.
value to me means making my life richer emotionally, not financially
Exactly! Cross linking all these social networks in terms of people you know would make our on-line life much easier.
This was some of the initial thinking underlying Qwerly. We’ve been pulled in a slightly different direction as most early customers think that social network to social network resolution (e.g. Twitter handle to Facebook user ID) is already valuable enough to pay for. But I’d love to come back to this direction eventually.What everyone (except Facebook) should do is publicly publish friends and mark up that contact list with microformats. That would make it much, much easier for someone like me to crawl, aggregate/diff the lists and provide an API that could service what you outline.Shameless plug: http://dev.qwerly.com
max, i’d like to talk live about qwerly. albert might want to join. a skype next week?
I’m in NYC Wed-Friday. Seeing Albert Thurs.To be honest, I don’t think there’s an investment case, but I’d really appreciate your advice.
that must be next thursday, right?
Yes, next Thursday lunch.
Hi Max (and Fred, great post). I would have loved to been at the table listening to you work through the problem set and opportunity.I’ve been busy trying to make dating more social, API’s for profile/communication sharing, but our industry is not open enough to consider this stuff even though the upside is huge for a few big players that get into the meta-search and cross-site communication stuff.
I’ve used and admired an application called “personalbrain” since 1999. I think if it this company were to put some time & effort to mapping one’s social connections across different services, they could be well poised to make a name for themselves and possibly lead “social graphing” revolution.
Maybe this is for you: http://www.mynameise.com(I’m not involved in any way)
Great article Fred. Couldn’t agree more. We’re a startup out of Denver that is taking this “problem” and turning it into a great business application. http://www.SocialBios.com
that would great if someone would launch something to manage them – soon. π
It’s not so much about the “social graph” as the “interest graph”… Start with common interests and let the socialising come after that.Call me silly, but I still think that Google is best-placed to solve this situation. After all, they’ve been building groups around shared interests forever — the only problem is that members of their ‘groups’ have no way of finding out who one another are, because Google only facilitates groups so they can targeting ads at them (ie. based on common search terms).The #hashtag has already become a useful tool (and community-driven, mind you!) but I don’t think that it is quite intuitive enough…Here’s my thoughts in more detail: http://www.webe.com.au/tech…
I agree this is definitely becoming more of an issue and a solution would be great…. I wonder if the Google Me initiative is working on this. One of their employees put out a video a while back talking about the problem of different and overlapping social networks.
Take it a step further and give me a full-blown platform to manage multiple identities and inter-network connections: http://blog.vkazartsev.com/pos…I guess even a startup could pull it off with a sale to Google, Microsoft or whoever-else-is-left-in-the-webmail-game as the obvious exit. What do you say?
If only there was some programatic way to sort out reciprocity.
It sounds to me like you need DataSift hooked up to a social graphing application of some sort. It sounds like a really interesting project for someone to create on top of our service! And if a UK startup decided to make a go of it they could even get some potential funding, details here: http://blog.datasift.com/20…
Poor Middle Class Charlie, I hired Accenture to “integrate” all my social graphs for $200 per hour.
First you brought in McKinnsey to give you advice on how to go about doing things @ $400 an hour, they gave you the strategery and you then hired Accenture @ $200 an hour who then brought in some folks from India ( paid them $25 an hour) and you were told there was scope creep. Now you have a half finished job and you hire someone else to tell you who screwed you over, and still you don’t have the strategery ( yes thats how G.W.Bush said it) or the product you wanted…