Introducing friendstat.us »

I've got a lot more of the wow.ly projects and business details to share, but I thought I'd take a quick break from that tonight and instead dig into explaining a few features of the latest wow.ly project to roll into production friendstat.us.

Friendstat.us is intended to help you keep up with ALL your friend's content.

To be a little more specific, friendstat.us compiles your various social network streams into one simple person-based view. It then gives you a handful of tools to consume, analyze, and interact with all that content.

Right now it's really just dealing with your Twitter stream, but all of the same features and details will pretty much apply to any new streams/services we add to friendstat.us.

Anyway, let's talk a little bit about the core features and ideas that make up friendstat.us

People-based view

A person-based view groups content generated by each person into a single stream associated to that person and then presents you with one larger stream that is a collection of those individual streams.

If you think about it, in systems like Twitter and Facebook you follow people you find interesting, but the content they generate is actually presented to you based on time. That is the newest content, regardless of who generates it, is always pushing the older content out of view. This approach awards the 'noisy' people you follow and makes it easy for you to miss out on, quite possibly high-quality, content from those you follow that publish less often.

To solve this problem Friendstat.us uses a person-based view. This makes it easy for you to see which of the people you follow have recently published content to your stream, how much they have published, and what as-a-whole they are talking about.

In-line replies

Sometimes a status update compels you to respond or ask a question. Directly within Twitter, when you click the 'reply' button, you are bounced back to the top of the page where the Tweet form is with the @username filled in...but this can sometimes be confusing, and often you'll forget the specific wording of the tweet you are responding to. Friendstat.us fixes this problem by putting the form to reply to a tweet directly below each tweet.

Directly below each tweet is a 'reply to this tweet' link. Clicking that link reveals a form in that spot, with the @username information already filled in. Just fill out what you want to say to @username and click the 'submit' button.

Old-school re-tweets

Since Twitter rolled it out, there's been an on-going debate about how re-tweeting should work. The advantage to old-school re-tweets is that it gives you an opportunity to add your own thoughts to the tweet.

Friendstat.us thinks this is the better of the two options, and so that is what we've implemented. And to make things even better, we've also included the re-tweet feature in-line. When you click the 'retweet this tweet' link below a given tweet a new form will appear in that spot with the text of the origional tweet already entered. You just make any edits you want and then click the 'submit' button. Nice.

Tweet Tagging

Sometimes a status update will contain a nugget of truth or a link that you really want to keep track of. Tagging the tweet within friendstat.us adds meta data to the tweet and saves the tweet within our system. This makes it easy for you to both search and view these 'special' tweets at any time in the future.

Once you've tagged a tweet, you'll be able to see all your tags via the 'Tags' tab. Clicking on any 'tag' word will present a full view (just like the standard person-based view) of all the tweets you've tagged with that term over any period of time.

Blog Posts

One of the things a lot of my Twitter friends tell me is that, more and more, they use Twitter as their RSS reader - that is, they find most of the blog posts they actually read via Twitter. The problem is, it's easy to miss out on links passing through your stream (I'll talk about that next), and not everyone you follow always tweets a link to their latest and greatest blog posts.

Friendstat.us fixes this problem by adding a 'blog posts' tab to every user in your stream. If friendstat.us knows about their RSS feed, we'll show you the five latest posts by that user. If we don't know where their RSS feed is yet, there's a form there for you to tell us about it so we can get it added asap!

Note: Associating an RSS feed to a Twitter user within the friendstat.us system is a slightly technical process and so for now is done manually by me. Sooner rather than later, I'll automate this process a bit more.

Links broken out

An almost endless supply of really interesting links is one of my favorite things about Twitter. But keeping up with all the links, even just the ones shared by the limited number of people I follow is almost impossible. It's even worse when they hidden deep inside the endless flow of status updates.

Friendstat.us fixes this problem by breaking all the links for your current view out into their own tab. This makes it quick and easy to see how many (and which) links there are to review. And if you choose to have your links expanded (via the settings page), we'll attempt to gather and display some additional information like page title and description for each link within this tab!

Private notes on people

When you start seeing things in a people-based view you can really start to see trends, interests, and passions that drive a person. And while tags are a great way to keep track of a specific message or thought, sometimes you want to jot down some quick notes or thoughts as it relates to the whole person.

The friendstat.us notes feature helps you do just that. For each person in your stream, there is a Notes tab. Clicking that tab will reveal any previous notes you've saved for this person as well as a form for adding more. It will also show you how the person describes themselves within Twitter itself.

Currently notes are private to the person who writes the note. So no matter what your write or associate to a person, you're the only one that has the ability to see it via the notes tab.

Soon you'll also be able to search your notes and tags for quicker and easier reference.

Tools

Another interesting side effect to a person-based view is that it becomes very easy to identify the noisy people who are cluttering up your Twitter experience. Once you've discovered one of these people, there's a really good chance you'll want to get them out of your stream, so we've added an 'un-follow' tool to make this a quick and easy process for you.

In the near future we'll be adding some additional tools to help you communicate and manage your relationship with a specific user to this area.

Statistics

Interestingly enough, friendstat.us originally developed out of the idea of providing some basic statistics about each of the people you are following. At first I just wanted to have a quick overview of who were there noisiest people in my stream, and so some of the statistics you'll find here help to answer those sorts of questions (as it relates to your current view).

As you can see, friendstat.us has evolved quite a bit from this original idea, and the stats tab is really nothing more than an occasionally interesting F.Y.I. part of the experience. Over time, I may add in more interesting statistics and ways to analyze and visualize your relationship with a given person to this area.

Mentions

One of the things that I learned with Conversationlist is that people are fascinated with who's talking to whom and who's talking about whom. But once again, it's easy to miss this sort of information in a traditional Twitter view.

The mentions tab helps make it easy to see just who's being mentioned by the people in your current stream. Mentions are sorted by the frequency that a person is mentioned and we show you just which accounts dropped the mentions, so it's dead simple to see who's being mentioned by whom.

Mentions are also a great way to discover interesting people since these are the people, the people you are interested in are talking to/about.

Lists

Twitter lists have been all the rage since they rolled out last year. And while I also think they are a great way to pay attention to a sub-set or a group of interesting people, they suffer from many of the same problems that I've mentioned above (especially in a traditional time-based view).

Friendstat.us fixes this by giving you the ability to view any public list, set up by any user, in the same person-based view as you do your own timeline. Best of all, all the other features mentioned throughout this post also apply when you're viewing list status!

Note: While the current system supports viewing lists, there is no user interface for actually picking a list to view yet. Rest assured, this feature will be rolling out in the next few days.

Other settings

The default settings for friendstat.us will pull up to 1,000 tweets from your public timeline that have been logged in the last hour and show them to you in a person-based view. The order the people are displayed in your navigation is sorted by the most recent person posting to the person who posted the longest time ago.

Currently friendstat.us also defaults to not expanding links for you (because currently expanding links requires a bit more time/processing and therefore slows down the overall user experience).

All of these things can be changed vai the settings area of friendstat.us. Any change you make to your settings will automatically be saved with your account, so any future visits to friendstat.us will adhere to those settings!

In the Queue

As you can see there's already a ton of features and options built into friendstat.us, but there's always more that can be done! In the near future, I plan to work on a few of the following things to improve your friendstat.us experience:

1. Implement list viewing features - this means making it easy for people to find and view lists associated with each person in their stream as well as the lists that the user has set up and the lists that the user has been added to.

2. Direct Message support - currently there is no direct feature for reading/reviewing direct messages that have been sent to you or that you have sent out. There is also no direct tool for sending out an direct message (you have to use the command line version which is the form 'd falicon blah blah blah' in the tweet form to accomplish sending a direct message right now).

3. Tighter integration with other wow.ly tools - this means implementing some interesting tidbits from things like conversationlist, hivemind, pu.ly, and others.

4. An API - So others can build their own friendstat.us clients (like maybe an Adobe AIR version) or just extend upon our feature set in some new and interesting ways!

Suggestions or ideas?

So that's the current state of the union for friendstat.us...hopefully it's got some nuggets that interest you enough to start using it yourself, and maybe even tell a friend or two about it.

If nothing else, I do hope that you'll check it out for a few minutes and leave some feedback here on the blog, send me an email at info at falicon.com, or ping me on Twitter @falicon.

Sometimes a project can just be about supporting other projects. »

The next project I'm about to layout details on related to wow.ly is URIData.com.

URIData is a bit unique within the wow.ly universe because it's not really a consumer focused project, but rather a developer focused one. That is, it's a web service that's only really useful to developers.

In fact the only reason it's even included under the wow.ly umbrella is because it was born out of, and supports, the needs of many of the other wow.ly projects (especially timelylinks.com). Since it's a simple, reusable service that fits in-line with the overall 'data collection' intention of wow.ly I figured it made sense to release it as it's own project as well.

With that in mind, there's not much useful information that can be filled out in the questionnaire about this project.

URIdata

What it offers:

A simple web service that returns basic data about any supplied Uniform Resource Identifier (URI).

What makes it unique:

It will expand and provide details about URI from most any URI shortening service.

Emotional tie to client:

Simple developer focused tool. Does one job, quick and simple.

Why and who should purchase:

N/A.

Where to find prospects:

N/A.

How to sell:

N/A.

Cost to reach 100X required customers:

N/A.

Expected conversion rate:

N/A.

How much it costs:

Free.

How many clients are needed to make 100k (or break even):

N/A.

Estimated time to reach 100k:

N/A.

Resources needed:

Server to host web site and run code.

Existing competition:

There are multiple API in the market that allow for expansion of certain short URI services, but none that I currently know of that are generic and all encompasing like URIData.com is.

Current Subscribers:

N/A.

Current Income:

N/A.

Current Resources:

Server to host web site and run code.

Other Notes:

This product was really built to server internal needs of other wow.ly products, but has a nice side effect of allow us to collect a large amount of data about URIs around the internet.

Next 3 Steps:

1. Better handle various response codes (ie. 404s, 200s, etc.)

2. Handle unreachable addresses (ie. things behind a proxy or paywall?)

3. Handle 'obivous' non URI values

Estimated time to complete 3 steps:

1 day.


Probably the most important thing to note about this project is that sometimes worthwhile projects aren't so much about how to treat them as businesses. Sometimes the need/value of a project comes from them supporting your other initiatives.

In this case, we needed URIData for oursevles, and it was very low cost/high potential value to also let others have it.

Diving into the business behind Hivemind »

Continuing with my theme of expanding on the wow.ly idea, today I'll give you my quick questionnaire answers related to the Hivemind project.

What's especially interesting about the Hivemind project from a business point of view is that it's the first of the wow.ly projects that has a consumer-facing revenue stream component to it. The hope being that, in addition to helping work towards the larger wow.ly goal (that I haven't laid out for you just yet), it also can be a self-sustaining, maybe even profitible, project of it's own.

We'll see how it works out in the real world, but here's my thinking on it as we just leave the starting gate:

Hivemind

What it offers:

A unique way to discover people and monitor groups of people on Twitter.

What makes it unique:

Ability to dynamically group people of a common theme or interest together. Dynamic monitoring of the groups you create.

Emotional tie to client:

Adds a sense of intelligent context to user discovery for Twitter.

Why and who should purchase:

Twitter users looking to find quality (as it relates to their own opinion of quality) people to follow and monitor.

Where to find prospects:

twitter.com, oneforty.com, facebook.com, google adsense and comments on various blogs addressing the topics of Twitter followers, figuring out who to follow, or how to keep up with who your compeition is following/paying attention to.

How to sell:

1. Get users to try the basic service.

2. Upsell premium version on timelyness and peace-of-mind that comes with monitoring a group.

Cost to reach 100X required customers:

$84,000 (assuming an insanely low $0.10 cost per-user-acquisition)

Expected conversion rate:

About 1% of free users should convert to premium version.

How much it costs:

$1 a month (6 month subscription min.?)

How many clients are needed to make 100k (annual):

About 840,000 users (with 1% conversion rate to premium product)

Estimated time to reach 100k:

About one year.

Resources needed:

Server to host site and code to run monitoring, system for accepting payments and managing subscriptions. One developer and min. freelance graphics/layout.

Existing competition:

http://tickery.net (cached system primarily built to showcase fluiddb system, no list building from results, no monitoring, no meta data added)

Current Subscribers:

332 at time of post (with 3 free monitors currently set up).

Current Income:

$0.00

Current Resources:

One developer (me) and freelance graphics/layout help (Pat)

Other Notes:

Though there is already existing competition, our model of focusing on dynamic data and monitoring is not currently being attempted or challenged.

The ability to tie into the larger wow.ly data set also gives us a unique advantage to improve performance and scale (via saving API calls for basic user information) and add unique meta data (conversationlist details, conversational social graph, etc.)

Our tracking features also add a unique component that the competition does not currently offer.

Next 3 Steps:

1. Improve the speed of the system with out compromising the dynamic nature (via short term caching and combining user data from other wow.ly data collection systems)

2. Make it simple (better explanation of what it does, how to use it, walk the user through the process better -- especially improve the upsell process and really give them a reason to try it)

3. Implement the upsell subscription bits (integrate with paypal to start)

Estimated time to complete 3 steps:

About 2 weeks.

And so there you have it. As you can see the above information showcases a few (currently) really weak points...primarily the 'how to sell' and of course all of the numbers are a complete guess/prayer at this point of the project (though I like to think fairly grounded in my experience of past systems I've been involved in that have had sim. approaches).

It's also interesting to see that this project already has a bit of serious competiation in Tickery.net...which, in it's current form, is very similar in many ways to what Hivemind does (in fact the developer behind it knows Whitney and I via Twitter but was unaware of our Hivemind application when he put Tickery together). Because of their approach to the problem, Tickery has really nice performance (when a user is already in the system) and is already starting to gain some attention (high profile Twitter user Robert Scoble even Tweeted about it).

To some that might be enough to kill the momentum or the passion for the project (given that we are so close to the starting gate and have so little invested in it already)...but to me, it's more validating than depressing and there are enough holes and weaknesses in Tickery's approach (and motivation) that I think we can co-exist quite nicely (and in fact help make each other stronger by championing the overall idea of grouping to find interesting people).

Anyway that's where I'm at with Hivemind right now. Thoughts? Feedback? Ideas? Questions?

A little insider info on Conversationlist. »

Long time readers of this blog (all two of you) might remember this post where I mentioned a simple questionnaire I like to fill out for each project I'm working on...well like anything it's evolved a little over time, but I still use a basic version of it for any personal project I'm thinking about taking seriously.

As such, I've started filling it out for each of the wow.ly projects...and once again I figured, why not share this with you. So here's the first one which covers the first wow.ly project we released - conversationlist.com

Conversationlist

What it offers:

Daily built Twitter list of the people you are talking to or about on Twitter.

What makes it unique:

Dynamic, focused, and simple (set-and-forget).

Emotional tie to client:

Focused around the user's own actions and interests (ie. people who have their attention).

Why and who should purchase:

Any Twitter user that wants to share a list of the people they are talking to and about

Where to find prospects:

Twitter.com, oneforty.com, and various comment threads where Conversationlist or it's competition is being mentioned.

How to sell:

Viral by nature (name of list is always conversationlist and description gives link to site).

Cost to reach 100X required customers:

n/a

Expected conversion rate:

n/a

How much it costs:

free.

How many clients are needed to make 100k (or break even):

n/a.

Estimated time to reach 100k:

n/a.

Resources needed:

Server to host site and code to rebuild lists.

Existing competition:

http://favstar.fm and http://twittfilter.com

Current Subscribers:

About 6,276 at the time of writing this post

Current Income:

$0.00

Current Resources:

n/a.

Other Notes:

Because of adoption of this application, we are able to build out our own social graph showing who people are really talking to and about (which may help present a revenue stream).

Next 3 Steps:

1. Start logging more complete social graph of who's really talking to each other.

2. Release API to allow develoeprs to access this new social graph.

3. Build front end report for users to see their own version of this social graph on-demand.

Estimated time to complete 3 steps:

1 week.

So as you can see, there's no direct revenue stream on this project right now...and that probably makes a lot of sense since Whitney and I really just threw it together as an interesting Twitter hack to start and weren't initially thinking about it with any business objectives in mind.

We were lucky enough to have it generate some buzz (thanks to tweets by some very influential people like @fredwilson and @ev and some press on Techcrunch) and our user number has been steadily growing ever since.

And what's probably most interesting to me right now is that, thanks to this continued growth, we now have this new social graph data/idea to play with (something that's a true side-effect of the project and hadn't ever really occurred to us as something to build towards)...again not sure if it will reveal any sort of a revenue stream or not, but it should help to continue to generate buzz and add to our overall data collection (and maybe even help get some other developers to buy-in and rely on us sticking around)...and to me, that's the kind of loss-leader worth having (especially when the 'loss' is really just some cheap server cycles!)

But what are your thoughts on all this? Would you answer any of the above questionnaire differently? See any gaps in my thinking or directions that you think are leading me down the wrong path?

Developing a plan and starting to look for funding. »

When I'm not working for reviews.com (the full time gig), finishing up writing my latest book (Beginning SimpleDB), or helping develop a new socially-minded start-up (Catchafire.org)...I'm working on (and becoming more and more addicted to) a concept I call wow.ly with Whitney McNamara.

If you follow me on Twitter or facebook you're probably already sick of me talking about wow.ly and wow.ly related projects, but this is the 'first' official mention I've had about it here on the blog, so indulge me for another quick moment and let me explain.

wow.ly is basically an umbrella project for a handful of small, focused, applications built on top of various open, social, applicatation programming interfaces (API). So little things I'm building on top of services like Twitter, Facebook, LinkedIn, Foursquare and others.

I won't go into detail about each project right now (though I do hope to do that in future blog posts very soon), but the current list of wow.ly apps that are either in-production or at least in-development is:

Conversationlist, Friendstatus, Fubnub, Fuzzypop, Hivemind, HotAroundMe, Pu.ly, Taglist, Timelylinks, Tym.ly, and URIdata.

I've also got at least two others kicking around in the back of my head that I've only scratched down a few notes about...and of course secured domain names (knowabout.it and simplejson.com).

Anyway, a few of our initial projects have already started to gain some media and user attention, and this in turn has helped to fuel my obsession with the overall concept. In fact, my obsession has now been fueled enough that I'm starting the process of trying to raise some money around wow.ly and build it into a 'real' company.

And so as I've been building out these things (with much help, feedback, and direction from Whitney) I've also been talking to a handful of industry experts that I really admire and respect.

One specific person that's been helping me flesh the overall wow.ly idea out (especially as it relates to seeking funding) is Darren Herman (also check out his awesome blog at http://www.darrenherman.com/).

Through these conversations with Darren I've compiled quite a few random thoughts that I think are worth sharing:

1. All the wow.ly stuff is basically built on top of open social APIs...what I am really trying to do is aggregate and augment that data in ways that hopefully proves useful for consumers (and eventually businesses).

2. They way I actually do each system right now is to essentially build an internal API for each service, and then build some type of front end on top of that. So the idea of letting other systems access, use, and augment our data sets is actually baked into the core and I'm very open to helping make that happen (it really just means me locking down, cleaning up, documenting, and adding a few extras to our API stuff so others could use it too).

3. From a business point of view, I'm struggling a bit to figure out where and when to monetize. I know long term, the massive and unique data collection and insight (that I'm just starting to build) will be insanely valuable, but I'm not quite sure yet how far off that point is (or just what the best initial things to do to monetize it will be).

4. For me, I think the first step is to just build out a collection of small services that people (hopefully) use...each one should be fairly focused, insanely easy to use (this is actually a lot harder than you might think to accomplish), and do something at least a little unique/interesting (ie. none have to be all that ground breaking, but since they are low barrier entry systems, people give them a shot and/or set-and-forget). I call this phase one.

So phase one is really just a masked 'data collection' phase. Though I do think there's a chance some of the apps in this data collection phase could generate subscription or ad revenue if promoted properly - it's not my initial focus and I want to be careful not to sacrifice the data collection phase (and long term revenue vision) just to make a small amount on consumer services.

Anyway, this phase one at the moment is only about the people using wow.ly stuff (and the people they are directly connected to -- ie. tagging, talking to, talking about, monitoring, etc).

So phase two is all about building out consumer and business intelligence apps.

I envision this phase two starting once we have a critical mass of interesting data (and honestly I don't know what that is just yet or how long it's going to take) but once we hit that point, I'll shift from building out more and more consumer apps that really just collect data to building out consumer and business intelligence apps that use our collected/derived data for marketing reports, branding campaigns, and/or improving consumer experiences across the internet. It's this phase where I think the 'big' (and recurring) profits can start to roll in. (from my experiences building/running supermug.com, draftwizard.com, and statsfeed.com, I'm convinced business-to-business sales beats the pants off business-to-consumer sales as a route to profitability - and so this is route I want to take here again)

With all of this in mind, I think working with other services/apps is a really intelligent way to help speed up phase one...though I'm not sure if there's any direct money path that makes sense (from either end really), I guess it just depends on the end-game for each system/service (and I think since mine is to collect as much 'interesting' data as possible, it makes sense to make my systems as open, free, and useful as possible to other systems).

And so my thinking is that raising money would really help me execute, speed up, and survive phase one. Technically I can 'survive' it just like I am, and the only thing that really suffers is speed-to-market (and again, I honestly don't know how important speed-to-market is for this stuff yet). But I really don't know enough about raising outside money, and being a developer at heart, I really just want to focus my limited time and energy resources as much as possible on building and improving the systems/apps (and I like to think that's where my strength really is anyway).

This is where Whitney and many of the other people in my extended network are coming into play (and I'm starting to initiate extensive conversations with many of them about all of this as I get myself ready to chase funding).

Anyway, getting back to the 'vision' real quick. In my mind the ultimate marketing question is "who is this person and what are they interested in?".

The data set I'm building helps identify that AND then also helps me to direct them to even more/better content that fits their personality.

I like to think of it as the Netflix movie suggestions model powered by social data/actions instead of movie rating data. When done correctly, the result is a very powerful search/recommendation/filter system for content and people in general.

I feel like I should also mention that this is all just my off-the-cuff thinking as of late with the system and direction things seem to be heading. I'm still at an early enough point that I can be easily swayed to another (better) vision if one presents itself.

Before I started doing things under the wow.ly umbrella, many of the other little systems I was playing with were at least indirectly toying with this long-term vision of 'using your social data to improve your internet experience' in some way.

A large part of the motivation behind that though is just because, done right, it's easy to see the revenue potential. But this is becoming a very crowded market space and getting more and more crowded every day.

My approach to the problem with wow.ly is instead of working a ton on one or two key products that HAVE to hit, I would build out a large number of smaller projects (until one or two DO hit and become those leading projects).

The hope is that in that process maybe I build/identify a handful of smaller, still interesting, and potentially profitable in ways I didn't expect, projects (or even companies).

At the same time, the wow.ly approach makes it easy to step back on any given project, say this is a much bigger hill than anticipated right now...let's table this one and move to something more interesting (ie. fail fast and small, win big and quick)

So that's where I'm at right now and where I'm looking to go.

What do you think?