Monthly Archives: April 2011


Automated curation and why I created @Lblogs.

A couple people asked who was behind @Lblogs, so I thought I would use this post to lay out my thoughts behind why I created the account, and since I can’t keep anything short, a couple thoughts on curation.

For those who aren’t familiar with this Twitter account, Lblogs uses RSS feeds to post to Twitter the title of the blog post, a link to it and an @ reference acknowledging the author.  All the blogs included are published in Lawrence.

Example tweet from LBlogs. Click on image to check out the hilarious Sandbar blog post referenced here

Twitter as RSS reader

Lblogs came out of my experiments with posting RSS/Atom feeds to Twitter.  I have long looked for an RSS reader that would act like Twitter and show the most recent titles with links across all my RSS feeds as they are posted.  I never found software that would accomplish this. But in the process I realized I could actually do better than what I originally wanted.  I could feed all of the blogs I follow into Twitter.

I created a separate Twitter account, made it private and my main account @lawrenckslive was the only account that followed it.  I then used external software to route all the RSS feeds I follow into this new account.  In the two months I have been using this system, this RSS system has published over 7000 tweets, but I am the only person who views them.

RSS > > Tweeted by Private Account > Viewable by @lawrencekslive

In the process of using this RSS system, I realized that curation can take many forms.  The best is when someone is looking at information, chooses the the highest quality and shares it.  But curation can be done on a less effective, but still useful level. That level is choosing blogs that have related or high quality content and them grouping them together on one account and publishing it.

Lblogs focuses on blog posts originating from Lawrence.

Because the Lawrence community has diverse interests, Lblogs isn’t really about what is going on in Lawrence, but rather what bloggers in Lawrence are saying.  After a couple days, the content of the feed includes a variety of subject matter.  Sometimes the post has nothing to do with Lawrence, sometimes it is focused on Lawrence, but the majority of posts naturally relate to a Lawrence audience just by the nature of where the author lives.

At this point if I wasn’t running the Lblogs account, I would still be following it because I like to see longer pieces that are being written by people in the community.  Since I am putting almost no restrictions on the type of blog accounts that are included, beyond that they must be published in Lawrence and shouldn’t be too commercial, I imagine over time that the number of tweets could be significant.  So if the number of tweets becomes too much, feel free to unfollow the account.  Having your blog included on the feed does not come with any obligations to follow the account.  Even if people just occasionally look to see what is new being published in Lawrence, I think it will be helpful. And since it is automated, the time required to keep it going is minimal.

Since the account is publishing links to other people’s material, anyone can opt out at any time by @ messaging the account or letting me know by any other method. I have purposely not included content from media companies to avoid using RSS feeds that have restrictions against doing so.  No one has yet to opt out, but understand I will not take it personally if anyone wants to do so.

If you know any blogs in Lawrence that you think should be included, feel free to add them in the comments or message me.  Also, if there is any interest, I will happily do a blog post explaining the step by step process of setting up your own personal RSS to Twitter account.  Just let me know.


Feel free to view any one of my tweets. Just don’t look at all of them.

Most of us know Moore’s Law:  the number of transistors on circuit board will roughly double every 18 to 24 months.  While this number has not held absolutely firm, it is the fulfillment of this prediction that has led to the exponential increase in processing power and storage of our computers.  These computers have led to an increase in the amount of data that can be processed in any set period of time, but perhaps as importantly, it has led to the ability to create a small enough computer that it is always with you . . .  the phone.

Online communications isn’t new, but the ability to never be more than a few seconds from the network has made it so we can communicate and express ourselves at any time. 

If you are active on Twitter, you are likely to create in the range of 1000 “what’s happening” tweets or more appropriately “what’s happening, interesting or valuable” data points a year.  Those 1000 tweets result in over 10,000 words. Those words can be linked to the time posted.  If you include where you are tweeting from or use a social check in service like Four Square, your location is thrown into the equation. Even if you don’t share this information, there are ways to track locations in the photos you share. Follows, followers, retweets and @ messages then produce a social graph of who you are connected to.

With just those data points, the who, what, where, how and sometimes why, can all be determined. Every tweet might not provide that information, but with a click of a button and less than a second, all your Tweets can be examined with modern computers.

The value of all our 140 character tweets has not gone unrecognized.

You might not think Klout is very advanced yet, but they are just an example of what a publicly facing startup can do with a couple million dollars. They now have full access to all tweets, they are updating almost every active account on a daily basis and they are providing a meaningful metric.  Influence is just one of hundreds of metrics that will eventually be based on our social media usage.  If Klout can figure out how to use our tweets to determine influence, other metrics won’t be far behind.

While I don’t care if any one person or even a thousand look at my tweets one by one, I am uncomfortable with the idea that I am providing the entirety of my tweets to any organization who can pay a couple hundred thousand dollars a year to Gnip (one of Twitter’s preferred data resellers), and come up with algorithm to turn my tweets and everyone else’s into a detailed personal profile.

A consumer product company marketing to me based on my interests is the least of my concerns.  These profiles could be used to target vulnerable individuals for scams, exclude people from participating in juries, decide how credit worthy a person is, decide if someone should be hired or fired and used to profile individuals based on their political beliefs.  And unlike credit bureaus that have some minimal level of accountability, many of these metrics will be developed outside of the view of the public, making it impossible to monitor or opt out of them.

The ideal solution to all of this isn’t clear. Private Twitter accounts aren’t trackable, but if we all go private, the Twitter experience will be lessened.  Twitter has yet to find a way to make money, and not surprisingly data is their most desired product at the moment.  At this point, I would much rather have multiple ads, then have all of my data sold to preferred Twitter partners, who can then sell that data to whomever they choose.

Working with corporations, I understand that they are often slow to act. This inertia will keep them from acting quickly on our individual data, but well-funded, focused independent organizations and even governments are already understanding the value of this information and taking steps to use it.

It is possible that this issue will gradually come into the public conscious, and if enough outrage exists, there might be more clear regulations on social media data collection.

While I wait, I will continue to use Twitter.  It is too valuable a communications platform to give up on. But I am going to give serious consideration to making my account private.

Do you have any thoughts on ways to maintain data privacy while not hindering the power of social media?  I continue to struggle with solutions and am very interested in your thoughts.