Currently, Digg is the #2 news site on the internet behind cnn.com. But Digg is different than other major online news sites, Digg is a social news site and should be evaluated not just as a news outlet, but also for its potential as a major social network.
According to compete.com, Digg had 23.5 million uniques in May, and Facebook had nearly 32 million. Although they had similar uniques, Facebook had over 360 million visits while Digg had only 69 million. How can Digg increase their number of visits and subsequently raise the advertising inventory capable of validating their $225 million valuation?
Sociology.
While looking for key differences between social interactions on both Facebook and Myspace, I noticed a lot of similarities. Both sites have profiles, activity feeds, friends lists, and a wall (similar to shouts on Digg). There was also one glowing difference-the friends I have on Facebook are, for the most part, my real friends.
While I share real connections with people on Facebook, nearly all of my friends on Digg are total strangers that share a common denominator which is the mutual appreciation for a specific story or topic. So how does Facebook solve the problem of creating community around specific topics? Groups.
Wikipedia describes a group as, “a collection of humans who share certain characteristics, interact with one another, accept expectations and obligations as members of the group, and share a common identity”.
When people come together they communicate and work towards a common purpose. This creates a more engaged community and one that increases visits. There are plenty examples of extremely popular Facebook groups that created quite a bit of traffic, no doubt.
The hardest challenge Digg would face in creating a Groups functionality would be the ability to game posts. A simple algorithm change to devalue diggs from fellow group members could be a good starting point to solving the missing link for Digg..
There’s a lot of people who are unhappy about the current situation at Twitter. Twitter has said that the reason for their downtime is that their super users with thousands of followers are the ones to blame because of the load they cause of Twitter servers every time they post.
So who’s to blame? Twitter for not properly setting up the site’s architecture or the super users who are using Twitter to promote themselves and their brands — for free.
The Two Sided Coin
Though Twitter never intended for their site to be used as a over hyped marketing machine, they should have at least setup limitations in what users can do (limit number of followers, messages per day, services used, etc.). On that note, people who are (ab)using a free service to ultimately make money should be kicking back some of that money back to Twitter.
Example: Robert Scoble relies heavily on his name (his brand) to market his thoughts, and to promote FastCompany, and his blog (which runs ads). He has over 20,000 followers on Twitter, and posts constantly on Twitter — effectively “clogging the tubes.”
Yet Another Example: Every single blog that submits to Twitter every time they post a new post. There have to be hundreds of blogs, with hundreds of thousands of followers on Twitter. The blogs are using Twitter as a very heavy lifting marketing tool. How much would the same tool cost 3 years ago?
I’m not saying I have the final answer, but do think a premium service is needed for Twitter. This will help Twitter bring in some well needed revenue in order to hire more rockstar engineers to fix the scale issues that the big players are causing.
Follow me on Twitter: @ryanmerket (but I promise if I get too many followers, I will start paying for it)
It seems since Twitter architect Blaine Cook left the micro-blogging site has been going down nearly everyday for hours at a time. With the emergence of FriendFeed, and growth of Pownce, can Twitter stay at the top of the micro-blogging sites?
It seems it’s only a matter of time now before Twitter gets replaced. It might have to take next year’s SXSW to claim a new king, but at this point, I’m ready.
With Twitter saying they will be recoding parts of their site in the next few months, how much more downtime are we talking? How long can the users go before jumping ship?
Michael said that data portability should not be left in the hands of the big companies…
I think it can be left in a big companies hands as long as the big company keeps user privacy as priority numero uno.
If it wasn’t for Facebook recognizing the fact that there is more to social networking then just data mining, we would never be at this point of, “what is the right way to export the social graph” outside of the “walled gardens.” Facebook has control over the most real social connections and the largest collection of user photos on the internet — all in all they have the most accurate representation of a the social graph, as we know it.
Facebook did NOT get this way by having lax rules on privacy. We all know they have VERY strong privacy regulations. That is NOT the issue, we all feel relatively safe giving Facebook our data…
So why are we so scared to let Facebook rule the Data Portability space? As long as they use open standards, I really don’t see the issue with letting them control how we get our data out…
Someone has to hold the keys to the gate of the walled gardens. Someone has to control that data, store that data on their harddrives and only allow people with the right credentials in — if not you will have a HUGE spam issue.
Why not let the company with the most real social connections, and the company that has kept privacy a top priority?
Why should they allow someone to piggy back (GOOG) on their success and right software to swoop in and undermine what they have been trying to setup for quite awhile?
Around 5 years ago I made a post on my then LiveJournal account about a new website called: Friendster.com. Enjoy:
Just found out about a cool new website…
It’s called Friendster.com — Its really awesome.. because what happens is, when you join you need to specify me as your friend… then you have access to all my friends, and thier friends, and thier friends… and then you invite YOUR friends… and it becomes a huge network of friends… you can setup a profile, pictures, interests, etc etc… its really COOL.. and i promise it will be the next big thing…
So SIGN UP and get on my friends list… then invite your friends who dont have it already…
Today I was reading through some startup blogs, and found a link to Fanista.com. See design below:
The first thing that catches my eye about this design are the Featured entertainment section on the right side. After I play with that area for a few seconds I try to figure out what the site is about — but the diagonal text hurts my eyes to try to read.
Not only does the diagonal text hard to read, but the kerning of the text gets distorted, making the text even harder to read. If you are trying to pull users in by informing them what your site is all about, then you need to be sure that the copy is can be easily scanned and not use descriptives such as “indulge your passion”, and headline it with what makes sense, “What is Fanista?”.
The next issue I have with this design is the fact that the items on the right have no names next to them. Users are forced to hover over the items to figure out what they are, and when they perform said action there is nothing for them to do except “Send to a friend”. Once you ask your user, especially new users to perform an action, you better have some “gold at the end of the rainbow”.
So what is the user supposed to do? Drag the icons to the little dock below the items? I did that, now what? There is nothing, as it seems, that I can do with the items once I drag them down to the cork board. Why even have it vying for my attention?
After looking over the page, it became very apparent that the Fanista designer is trying to get the user to do 2 main actions on the page. Drag on the items on the cork board, which ultimately does nothing, or be forced to read diagonal text, that hopefully leads the user down to a little button:
This leads us to the point of the page… If the point of the page is to try to 1st: Educate the user, and then 2nd: Acquire the user. Then don’t waste precious time on site with a drag and drop gimmick. Spell out exactly what the site is in clear readable copy, and then make the next action clear as day.
Ice cream giant Baskin-Robins created a Facebook event to help promote their 31 cent scoop night. The lucky night is on April 30th, mark those hCal enabled calendars.
I think what is really interesting here is that Baskin-Robbins is utilizing the already existing Facebook functionality to promote their campaign. They didn’t need to go to any advertising networks, or agencies to create a viral channel. Because their campaign is event driven, it just makes sense for them to create a Facebook event. At current count, there is over 100,000 users who have confirmed that they will be ‘attending’ the event, with over 80,000 users who have been invited — and the campaign started yesterday!
I think companies really need to look at doing more natural, immersive social campaigns like this. If you have a special incentive that you think your audience will be excited about, instead of sending a direct mail piece, try to put your brand in the paths of the communication model — such as social media.
OpenSocial API has been proven to be more unpredictable, and actually more decentralized. Though there is one name, there is many different flavors for each container (social network).
For instance, there is custom functions that apps call for each container. How do you possibly port your code from one container to the next if there are completely different functions for each container?
Also, OpenSocial requires the application developers to code advanced JavaScript. Facebook required developers to learn a variance of HTML, called FBML, which is VERY easy to learn.
OpenSocial is also VERY buggy. It is still very young, and is nearly a year behind the Facebook platform.
Why should Facebook change to a platform that is harder for developers to learn, not truly portable and buggy?
Bebo is running a licensed version of the Facebook platform, and it is doing great!
If anything — everyone should license the Facebook platform.
Facebook announced yesterday a new site of APIs they are referring to as “Permissions APIs”. These new APIs will finally give sites such as adonomics.com and developeranalytics.com the needed data to actually have accurate data. Application developers will be able to grant these companies access to their metrics by allowing their applications to call admin.getDailyMetrics.
With the permissions API, an application can authorize another application to call certain API methods on its behalf. The application naturally can revoke this access whenever necessary. At this time, the permissions API allows applications to grant others access to three methods under the admin namespace – admin.getAppProperties, admin.getAllocation, admin.getDailyMetrics More methods and namespaces will be added to the list going forward as needed.
Now, Adonomics will actually have real data to base their valuations on. In the past, companies had to take an applications daily active users number and multiply it by the percentage Facebook gave (ie: Scrabulous has 694,511 daily active users, which was %21 of their total installs — if you do the math that is somewhere around 3,300,000 — not quite accurate.
On top of detailed daily active users, the metrics companies will have a plethora of data to store and analyze. According to the Facebook developers wiki, here is the data that they can now pull:
daily_active_users — (The number of daily active users.)
unique_adds — (The number of users who added your application.)
unique_removes — (The number of users who removed your application.)
unique_blocks — (The number of users who blocked your application.)
unique_unblocks — (The number of users who unblocked your application.)
api_calls — (The number of API calls made by your application.)
unique_api_calls — (The number of users on whose behalf your application made API calls.)
canvas_page_views — (The number of canvas page views.)
unique_canvas_page_views — (The number of users who viewed your application’s canvas page.)
canvas_http_request_time_avg — (The average time to fulfill an HTTP request to your application’s canvas page.)
canvas_fbml_render_time_avg — (The average time to render FBML on your application’s canvas page.)
