In O’Reilly’s core design patterns for Web 2.0, one of the most important is Data is the Next Intel Inside. This pattern for experience design places a focus on control over data rather than control over the code, framework, software or hardware. There is strategic advantage to owning and controlling data, as it can provide new opportunities heretofore unseen on any computing platform. Data comes in all forms, and what can or should be done with the data depends on strategy. ZDNet suggests potential strategies, including creating difficult to recreate data, making data open or charging for access to data. Another example is by innovating around data, such as Google’s vision for using their data stores to produce more relevant results and new ways to query.
Twitter is an absolutely explosive product, with over 140 million tweets sent daily (about 1 billion per week). This certainly indicates a massive number of users generating a lot data and often white noise. This often makes it hard to find good, relevant content on Twitter (for example, as I write this, two of the topic trending in Australia are #drunkestievergot and #throwagrenade; it’s the epitome of relevance and importance). There is an overwhelming amount of data, and people are getting lost while trying to find good, relevant content (the use of Twitter in recent uprisings and during natural disasters aside). Even Xerox’s PARC research facility realises there is an issue of identifying relevant people and content on the service. This is where tools like TweetMeme come in – an application that sorts through the data to deliver useful information (specifically news) to users.
TweetMeme is a standalone product which filters through the millions of tweets sent daily for relevant news stories, stored in tweets as links to these news stories. Depending on the number of times a link appears across all tweets in a given time frame, it gets posted in the ratings, and the top links or stories are displayed Digg-style on TweetMeme as a collective news feed. As more users post, retweet or comment, the higher up the rankings it moves. Links and stories are categorised based on content, generate such Genres as Technology, Sport, World & Business plus Gaming, Comedy and Lifestyle (to name most of them).
TweetMeme can aggregate so much data from Twitter because of the very foundations of Twitter. The service is geared to display an account’s tweets publicly by default. The fact that so users enrich links with hashtags, reviews, comments and experiences provides enough extra information for TweetMeme to be able to classify the tweets into the same category and then determine the recency of the story. TweetMeme functionality is further facilitated by:
- The fact that Twitter’s data is easily searched and can be addressed and located by applications external to Twitter,
- Twitter is an open platform that is open to innovation and does not protect tweets and data from reuse,
- Twitter users are generally taciturn in their approach to reuse of their data – people enjoy novel reuse of their data,
- The barrier to entry for both Twitter and TweetMeme is low (TweetMeme actually uses Twitter’s OAuth service for login), encouraging data to be placed in the system by users,
- Data is reused – users can retweet from within TweetMeme, creating new content in Twitter and boosting the value of existing content both within Twitter and within TweetMeme,
- Data is enriched – TweetMeme allows a user to post a comment to a news story – as a new tweet. This enriches the entire data ecosystem, as TweetMeme handles more data and has clearer insight into data, and returns the data back to Twitter as a bonus,
- The service is incredibly open – third-party websites can embed the TweetMeme re-tweet functionality natively in their site or service, increasing network effects.
The short YouTube video below explains how the retweeting and platform openess works on TweetMeme.
The result is that TweetMeme can innovatively use user created data and present it in a completely different format. The only real issue for TweetMeme is that the data is so open that the barrier for entry for a competitor service is incredibly low (Digg already integrates sharing via Twitter, so they clearly have the knowledge to implement changes to their new ranking algorithms should they choose to implement similar functionality). This means that innovation is a constant pressure – TweetMeme needs to keep ahead of the game to be successful in the long term.
TweetMeme will change the way I collect and see news; what will it do for you?
I am a twitter addict (THERE I said it!) and I definitely get buried in a sea of irrelevant/boring tweets. I love the idea of this tool, and it certainly seems easy to use (thanks for the video tutorial!). With all the recent natural disasters I can certainly see how tweetMeme could help us focus on the trends we really want to read about, and identify relevant content.
Perhaps my favourite part about the whole thing is that the tool is open – and encourages other platforms to incorporate Tweetmeme into their offering. I struggle to keep track of all my social media and like using an aggregator like Postling, especially when managing multiple brands.
Come on Postling – do it! Integrate with TweetMeme!
Hi Steph,
One thing to also point out (I was attempting to keep the blog post short) is that TweetMeme retweeting can actually be integrated with WordPress blogs (along with other CMS solutions) easily. This extends the reach of Twitter and TweetMeme data way beyond just the Twittersphere. This reuse of data is unprecedented, and opening the TweetMeme platform up for reuse a fantastic way for TweetMem to cement itself as the dominant player in this part of the market. Again, ownership over the data provides a lot of value here.
I like the idea as well. You are perfectly right when pointing out that the pitfall of massive data collection and generation is the difficulty to find appropriate data both in terms of relevance and quality (and thereby being efficient). This is a challenge not only in the case of Web 2.0. For instance, it is an issue companies try to address with Business Intelligence and related techniques like Data Mining.
Your post also links to the next pattern, which is about facilitating reuse and mashing of data accessed through APIs. As you said, “Twitter is an open platform that is open to innovation”.
Hi Dave,
Excellent point there – the openness is what provides the innovation. As I replied to Steph, TweetMeme allows reuse of their retweet functionality – it’s so pervasive now that most major IT industry news sites are using it (CNET as just one example). This expansion allows more value to be added to the data TweetMeme has already collected, which is important from the perspective of the pattern concerned here – as ownership is what allows and drives the innovation you mention. Recognising this is a key strategic factor, and it would be worth TweetMeme planning for that.
Hey Dave!
Interesting article you have here. I never knew about Tweetmeme, but i always thought that there should be a way to sort the millions of tweets to be more useful for those dependent on twitter as some sort of news source, which is ever increasing in today’s world.
I’ve just penned an article on StumbleUpon on my blog at http://bit.ly/dLwCse which is similar is regards to having a platform that divides the web to be more manageable for readers.
Hi Dave,
StumbleUpon is another good example of making relevant web content available at the fore for users. It would be interesting to see a TweetMeme-StumbleUpon hybrid – finding news stories by being suggested them based on your Twitter activity. What do you think?
Hi Trent,
Very interesting article, I’m always surprised that we only have started to see more applications like this spring up relatively lately into the public domain, if anything twitter has been locking down the kinds of programs people can create.
Services like this have always interested me (I recently wrote a blog article on my website about a similar product) because as you pointed out, the amount of data posted onto even just twitter is mindboggling, and programs like this really add value to twitter.
I do not believe there is a more powerful way to quickly propogate news quickly yet out.
Hi Chris,
You actually made me realises a point I hadn’t considered before – what happens if a false story is propagated very quickly via TweetMeme? This could impact a lot of people. Imagine if a posting of a disaster was made on Twitter then subsequently promoted on TweetMeme, and the story was wrong? This would certainly cause a lot of people to become distrustful of TweetMeme and Twitter, devaluing the data. What do you make of this issue?
Hi Trent,
Very interesting article- as I am relatively new to twitter (I’ve avoided it for a while as I knew it’d eat up a lot of my time), this sounds like a great way to avoid the useless twitter overload. I’ll definitely be using this application – and to be honest the only reason I probably am is that the barriers to entry are low, I love when I can use the same log-in! I’ve already got enough passwords to remember.
Interesting that you mentioned Digg as well, my blog post this week is on Digg- I haven’t actually used it much in the past, but as I did some research for my blog I saw some of the benefits that you have mentioned in here as well for TweetMeme.
Great blog,
Mindy