Today, I discovered an exciting app. Topi is a group messaging application that also allows chatting with people in vicinity, having common interests or affiliations. With the recent wave of similar social applications that leverage geographical proximity (Sonar, Banjo, Localhero, Hear I Am...), it feels like today's mobile phones, and their users, are ready to give a second chance to PAN Dating practices that started at the beginning of this millenium. In this post, we will quickly sketch the evolution of Mobile Social Software (MoSoSo) of the last 10 years, introduce a few recent apps (including Topi), and discuss why we should all be excited about their possibilities.
1. Ten years ago, it all started with Bluetooth
When the Bluetooth technology started to appear on mobile phones, some people utilized this PAN network to broadcast information about themselves and/or what they were looking for, with the objective of meeting new people around them. At that time, this practice was called Bluedating, or Toothing. Bluechat applications like Nokia Sensor, Mobiluck, and Proxidating were used to exchange small messages through the Bluetooth network. Whereas mobile access to the Internet was still slow and expensive at that time, the Bluetooth network could be leveraged for free. Nevertheless, the population of users with Bluetooth-enabled phones was scarse and segmented across too many non-interoperable applications, and reception range of the Bluetooth network was too small (about 10 meters at most), making the odds of meeting through such applications close to zero.
2. The "check-in" era
Bought by Google in 2005, Dodgeball was a mobile social network relying on users sharing their current location by SMS, in order to receive local crushes, notifications of surrounding friends and recommended venues. After the growth of wireless access to the Internet, it had become possible for proximity-based applications and web sites to evaluate the user's approximate location automatically. Foursquare and Gowalla introduced the "check-in" practice, inviting users to pick their current venue (from a list of surrounding venues) in order to let their friends know where they are in real-time (including on major social networking sites like Twitter and Facebook), and to discover strangers who checked in at the same venue. The automatic and continuous localization enabled instant notifications of surrounding friends, and recommended venues. The actual communication between users can happen on these applications, on social networks, or by SMS. Then, other applications started to leverage these check-ins in order to recommend additional people in vicinity: Sonar, Banjo, Agora...
3. Meet locals with a goal
With the increasing amount of mobile internet users, the current trend goes beyond allowing to chat with local stangers. Now, that there are too many connected strangers in vicinity, another layer of filtering has become necessary: common ground. On Localhero, users ask questions, provide services, and post classifieds (similarly to Craigslist), with the goal of getting quick replies from locals. Posts can be tagged, so that users that are interested by the corresponding topics can be notified in priority. Hear I Am, on the other hand, grabs your list of favorite musicians from Songkick, and notifies you whenever a person with similar taste in music checks in near you, so that you can meet with a good ice-breaker.
4. Topi, and why does it matter?
Topi is like the fusion of Localhero and Hear I Am. It enables chatting with local people with any common ground: whether you are fans of the same band, were born in the same city, or studied in the same university. Each interest (or affiliation) becomes a group, a virtual room in which near-by people can chat (using text or drawing), share pictures etc...
This is exciting because users are given the possibility of promoting themselves on specific interests, and meet local people with similar interests. This feature is great for many use cases:
- meet relevant professionals in a crowded conference / networking event
- ask neighbors about current local musical events, while on the move
- find another alumni from the same university, to share good memories around a coffee, during a train ride
- and of course, for dating!
...BUT
All these opportunities have always required that a sufficient number of people use the same application. Even though our current state of technology allow much more powerful ways for rich local interaction, the number of MoSoSo applications is still high, and the target population is thus highly fragmented. In order to survive, any general Social Network (such as Twitter, Facebook) require a sufficient mass of active users; the same applies to Topi.
Topi is a promising social app, but it still has a long way to go, for its own survival! Keep up the good work, guys!
Like most people who will read this post, I'm a man, and I'm not interested in seeing woman shoes and wedding pictures, but I do love Pinterest.
Why? And why do I love Pinterest? Because it's a great social / self-expression service that I think is much more powerful than blogs like Tumblr. Let me explain this in a few points.
1. It's a tool of self-expression
People like to stay in touch with their friends (e.g. mobile phones, instant messaging, Facebook...), but also to show what makes them special. Even though most people use Facebook for expressing themselves, these two needs are completely different, and Facebook is definitely not the best place to express yourself today. Why? Because, unless your personality is a copycat of your friends', most your friends won't push the "like" button on the picture of you riding your horse, or on an extract of your favorite book. They just probably don't care about it. They might give you a "like" or a comment to cheer you up because they like you, but don't tell me that your 300 facebook friends are all interested in the same things as you!
In order to share pictures, quotes and videos that might not be relevant for their Facebook friends, many people started opening Tumblr micro-blogs. But, unless they shared content with a consistent editorial line (which is rarely the case for personal Tumblr blogs), they would keep their blog mainly for themselves, as a kind of public "I want to keep this" bin they would fill with stuff they found interesting while browsing the Internet.
Microblogs are a good first step towards self-expression, as they can give a clue of what their authors are really interested about, without constraining themselves to posting content that would be popular among their friends. Nevertheless, for most cases, Pinterest is a better bin!
2. It's visual and beautiful
Assuming that most content shared on microblogs are images, does it still make sense to display a person's blog as a chronological vertical list of posts? Of course not! Pinterest spreads images in a beautiful column-based display that stretches to your whole screen, like Google Images. It makes more sense, and you get to discover more content at once when browsing the website.
3. It gives a better visibility to your content
Sure, people can follow your Tumblr, and find it by browsing categorized directories. They would check out a few of your latests posts to figure out if they want to follow you or not. But, on Pinterest, it's not necessary for people to look for anything. If your beautiful picture shows up in the stream, they would just click on it, and may decide to follow your Pinboard, that's it! ...And of course, you can still follow and be followed.
4. It's more personalized and addictive, thanks to Pinboards
As successors of tags, Pinboards are a powerful concept. Not only they give you the possibility to group images together in a meaningful and visual way, they also emphasize the wide range of interests you want to express yourself about, and makes it quicker for you to share stuff in an organized manner.
When browsing someone's blog, the most emphasized feature is the last post, and tags may appear on the side to give a clue of what topics the whole blog covers. When people visit your Pinterest profile, they see a gallery of Pinboards containing a few thumbnails. This display makes it easier for your visitors to have a visual overview of what kind of images you posted, and, most importantly, how you relate to these groups of images. For example, your photo of a beautiful beach will have a different meaning whether you posted it in your "Places I want to go", or in your "My last trips". In both cases, you didn't have to express that when posting, you just had to choose a Pinboard from a list, which is must quicker. Additionally, the fact that you have to organize your images in Pinboards makes you want to complete these Pinboards with more images. There is no such need for completeness when posting on a blog's category. And, believe me, this is really addictive, it makes your profile look great, and thus provides a feeling of satisfaction!
5. It's dead simple and quick to use: push-button publishing
Tumblr and Facebook made it easy to re-post content from others to your own followers. But, only on Pinterest, this practice is probably the first action a new user would do. Indeed, starting with Pinterest is extremely seamless! You don't tell to yourself "Ok, I'm going to open a blog and write my first post". Instead, you just go to someone's profile, or to the home page, see a picture that you like, click on "Repin", and you're in! As soon as you've pinned your first picture into a Pinboard, you want to pin more, in order to complete that pinboard. Again, this is very addictive, and ridiculously easy to start actively using the service. Eventually, you might even start adding your own pins, by uploading images, or by adding the "pin it" bookmarklet to your web browser.
6. It makes you want to come back
Every time you pin a picture, other users will start re-pinning it, liking it, commenting on it, and even following your Pinboards. And, of course, you're notified about all that, making you come back to the website and check out the Pinboards of these people that think you're so cool. This makes you discover more stuff, follow more people and repin their images, and so on...
Conclusion
Blogging is like writing a journal, Pinning is like decorating your home and filling your living room library with books and records you love; for your visitors to see, and your own enjoyment. Pinterest deserve their exceptional user growth because they developed a beautiful, easy to start and use, and addictive solution to a growing need: self-expression.
What's next? What's yet to improve?
First, you must give it a try in order to experience this new phenomenon and make your own opinion. I invite you to check out my own profile.
Then you might realize that Pinterest is not optimal for finding and browsing videos (yet?)
After some browsing, you will also realize that most pinned content is actually kind of consensual. Therefore they only provide a shallow perception of people's personality and tastes. This is ok, as long as users still want to discover and keep beautiful images, without deep expectations on the "self-expression" side.
Finally, you might be disappointed by the irrelevance of some pins in your dashboard view. The possibility to follow specific Pinboards from a user is a good start, but the free semantics of pinboard grouping and naming makes it complicated for Pinterest to automatically recommend pins that are interesting to you.
I would love to read your first impressions, and predictions about the future of Pinterest and self-expression websites in general. More specifically, I would be very interested to read stories of people having met or had interesting conversations thanks to Pinterest. Please leave us a comment! :-)
Innovative startup Stamped is a recommendation service with a difference. You can "Stamp" anything from films to food, see your friends' Stamps and build up an archive of things you love, to refer back to later. Stamped takes the clutter and anonymity out of recommendations, while keeping things private to the wider public. Building affiliate programs for when people want to act on a Stamp is the clever business model behind this app, which we expect to hear more from in 2012. (from "The 15 Best Mobile Apps of 2011", on Mashable)
My goal here is not to review Stamped as a startup, or a service, but to identify the features that gives it potential for usability and social virality.
1) Emphasize the main function(s) of your app, in context. And give concrete examples of what the user is expected to do/enter.
2) Challenge the user, e.g. by providing a limited amount of points, and a way to earn points that benefits the virality of the app (social sharing) and the quality of contributions (number of likes).
3) It's very easy for a user to forget about your app after the first launch. Ping them regularly (e.g. email or push notifications) to make them come back to it, and don't forget to remind the goal of your app.
A few weeks ago, Loïck Müller (the designer of whyd) and I participated to the Foursquare Hackthonin Paris. We wanted to leverage Foursquare API to break the ice between like-minded music lovers, whenever they check-in to a same venue. So we built a simple web application using technologies we love and use everyday: node.js and MongoDB. Within less than 2 days of development, we were able to present a working application relying on two partner APIs: Songkick and Thecallr, and won prizes from these two partners + a "people choice award". Since then, Hear I Am has been running publicly on the internet and hundreds of users signed on without any advertising effort. Here are some details about how it works, how we made it, and the status of Hear I Am, two months after the release.
Enter your name (or nick), a short bio, your Songkick username, and your phone number
After your favorite artists are fetched from Songkick, you can close your web browser, and go out!
Right after you check-in to your venue, you receive a SMS with a list of like-minded people (when applicable) who checked-in at the same venue, displaying which artists you both like
You can contact them through their Foursquare profile, e.g. to talk about music around a beer!
What's under the hood?
Beyond the simple HTML / OAuth routine, Hear I Am subscribes to Foursquare User Push API events, so that our server is notified whenever our users check-in to a venue. When this happens, we look for other users who checked-in at the same venue and are fans of similar artists. The data model is very simple: a unique collection of users which stores the list of artist identifiers (from Songkick) as an array and the last venue identifier, for each user. If matches are found, a SMS is sent to the user, with a link to a mobile-friendly web page displaying those matches.
In order to host and run our node.js and MongoDB infrastructure, we chose Dotcloud. It's a highly flexible and reliable EC2-backed platform that offers a free account which supports HTTPS, as required by Foursquare User Push API. It was very convenient to use during the development process, thanks to the easy command-line tool provided for pushing, running and debugging the code from our GIT repository.
Today, we have more than 1200 users, all around the world!
Here are some closing notes, advice and ideas of improvements:
The hackathon was an amazing experience, and it gave a good visibility to our project (see the Techcrunch article in which Hear I Am was cited)
Node.js is a great platform for prototyping efficient services in Javascript
...but it was fortunate that Dotcloud uses supervisor to prevent occasional crashes! ^^
MongoDB is definitely a good fit for this kind of project because it's schema-less, highly efficient, and directly compatible to Node.js thanks to the JSON data interchange format.
In the near future, we intend to leverage MongoDB's geospatial indexing in order to extend matches to people in close surrounding area, instead of solely relying on check-ins to the same venue. That will increase the odds of a match.
Next project: develop a native mobile app with real push notification, instead of SMS and emails.
I just added a new feature that will help to promote Play'em, my social TV project: a button that allows users to share a video that they like with their Facebook friends.
This button will not only post this video to their Facebook stream, but also promote the use of Play'em for discovering cool videos from their friends, thanks to the "Watch Play'em TV" call-to-action link.
By clicking on "like", friends will let the user know that they enjoyed this video, but also support Play'em. :-)
In April 2010, we conducted a survey of 256 users of real–time microblogging platforms, mostly Twitter users, in order to analyze usage of those platforms and evaluate induced cognitive impact. In this article, after reporting and discussing the results of this survey, we identify opportunities to reduce information overload and frequent disruptions. We propose a novel approach for filtering status updates from real–time microblogging platforms based on contextual relevance between their authors.
How to cite: Joly, Adrien, Maret, Pierre, and Daigremont, Johann. “Between social awareness and productivity: Results of a survey about real-time microblogging” First Monday [Online], Volume 15 Number 11 (3 November 2010)
Title: A Context Management Framework based on Wisdom of Crowds for Social Awareness applications
At a time when social networking sites revolutionize the usages on the Web, it has become rich, easy, and fun to share private or professional content. Sharing personal information in real-time (such as news, moods, etc…), supports the maintenance of social ties at a high scale. However, the information overload which emerged from the growing quantity of signals exchanged on these services, often in real-time, motivates a regulation of these signals (called “mediated interactions”), in order to reduce the temporal cost for maintaining social networks, and implied interruptions, which have a negative impact on productivity on tasks that require long-lasting attention.
In the frame of this thesis, we have developed a filtering and recommendation system that relies on contextual similarity between users that produce and consume social signals, as relevance criteria. In our approach, contextual information is aggregated and interpreted on users’ terminal(s), before being submitted on-demand to a server in the form of a set of weighted tags. In this thesis, we present a broad state of the art on context- awareness, social networks and information retrieval, we propose a formalization of our filtering problem, and we implement and evaluate its application for enterprise social networking.
Keywords: information retrieval, context-awareness, recommender systems, social networks, information overload
