Thoughts on Twitter #6: Why I’m re-booting my Twitter follow list

[Notes: You can view all my “Thoughts on Twitter” posts displayed chronologically here: Second, after starting to write this, I noticed ubber-blogger Robert Scoble wrote a post on the same topic. I thought of just pointing to it and saying, “what he said,” but had, by then, written too much.]


It’s rare that I post on a topic because someone requested me to — primarily, because it’s rare that I get such requests. However, I’ve been asked to explain why I’ve decided to start all over in building a list of people I follow on Twitter.

Until yesterday, I followed about 1,300 people on Twitter. Wait, let me correct that. I had about 1,300 people on my following list. Upon cursory review, I discovered that most of those people rarely posted a tweet on their account, so I wasn’t actually following them. However, I had decided to take a big tent approach to following lots of people for a couple of reasons:

1. To receive a ‘direct message’ from someone, you must follow them. (If you want to DM me, you can e-mail me at rexhammock (at) or call me at 615-852-REXH — okay, knocked off that reason.)

2. I follow lots of people who list Nashville on the location line of their Twitter bio because I’ve been (for a long time) fascinated with the role of social media in emergency situations. Twitter is, to me, a natural early radar tracking system of breaking news of an emergency nature. I discovered fairly early in my blogging experience that it’s too late to put together a meaningful list of potential eye-witnesses to a local widespread emergency event if you wait until the emergency occurs. (I now know of ways to accomplish this without having a long following list on your primary Twitter user account. Two quick solutions: Outsource such tracking to someone like Christian Grantham at WKRN’s Nashville is Talking or to an automated service like the very impressive one being developed by Chris Ennis,

That’s how I started out building my original follow list, but when I crossed about 500 followed people, I began to notice something: If you follow lots of people who tweet alot, only a fraction — sometimes only a small fraction — of those tweets begin to show up in your “tweet stream.” If Twitter algorithms were filtering what tweets I was seeing, I decided that it wasn’t my follow list they were streaming, but theirs. So I decided that I would scale back to a number of followers they could actually handle.

Then, a couple of weeks ago, I heard NYU journalism professor Jay Rosen talking with Dave Winer on their podcast called, Rebooting the News talking about his approach to “hand-building” a following list, an approach to creating such a list with a sense that you’re curating something you perceive to be a service to those who may be looking for advice on finding the best voices on Twitter in a particular field.

In part, the idea is a rejection — or, a reaction, at least — to the decision by Twitter’s management to include a “Suggested User List” (of people you should follow) in the registration process. The Twitter official SUL is a bug, in the opinion of many people, as it is heavily weighted with celebrities (and friends of the founders). Whether or not the SUL is flawed (and I tend to believe it is, in the case of Twitter), having alternatives to it (i.e., experiments to discover a better way of finding those who you may follow who will make using Twitter a better experience for tracking the types of information you personally find important) are important. We all benefit when people like Jay — and ubber-blogger Robert Scoble and others — experiment with conventions, even if the conventions are only just a few months old.

So I decided that my “August” project (one of many) would be to take control of my following list. I didn’t know it was going to be a fad, or that people would ask me to explain why or how. But, as I found out a long time ago, when your name is Rex, on the internet, people think you’re a dog — so you might as well behave like one from time to time.

[This post has officially ended. However, for hardcore Twitter geeks, I’ve included the following notes that explore some esoteric or technical aspects of re-botting my follow list. I recommend you skip this part, really.]

I’m not advocating that everyone do this. Twitter usage is a very personal thing. Everyone has a different approach. No one gets Twitter, remember. As soon as you think you do, it’s a sure sign you don’t.)

In the re-boot of my list, I have used the same method followed by Robert to start over: a $25 premium tool from the service In re-building the list, I’m using methods like the Twitter feature that allows a user to authorize Twitter to find people among ones Gmail contact list who are users of Twitter. I’ve also gone back through my list of individuals who I’ve DMd (direct message) to see who I might want to include. If my username did not generate so many false replies (I’m not complaining), I would review those who have placed an “@” in front of my username in a tweet to look for others to add to my list, however, if you do a search on the term “@r” you’ll see why that’s not very efficient (although I’m not complaining).

I have some basic criteria I’m following in rebuilding the list. The person needs to meet just one of the following:

1. The person must tweet regularly and insightfully about a topic in which I’m interested. Typically, these will include: media (especially magazines and new media), technology or humor (especially ad-libbed commentary on breaking news).

2. The person is from Nashville and seems to be the kind of individual whose knee-jerk reaction to seeing something noteworthy is to pull out their phone and tweet it. I follow these people for the “emergency radar system” reason.

3. I know the person — or know who the person is.

4. I’m intrigued by the person for reasons I can’t quite explain — as in, they’re a trainwreck waiting to happen. (I find these are the people I unfollow most quickly, also.

Other “helpful hints” for getting me — or anyone — to follow you:

1. Have an avatar. The default one is ugly for a purpose.

2. Fill in the “bio” fields with enough information to let people know where you are located and to click through to a website where they can find more information about you.

As I post this (Thursday morning, August 6, about a day and a half after re-booting), I’m back up to about 150 on the list and expect to climb higher this weekend when I’ll have more time to devote to reviewing my OPML file of those I previously followed.

Don’t worry. If you’ve read this post this far, I feel certain you’ll be on the list.

People who are power users of Twitter have fine-tuned third party applications (like Seesmic or TweetDeck) that are designed to organize and control all the spinning plates that one must keep in the air so as not to get overwhelmed by Twitter. And an even smaller sub-group of social-media devotees become power users of the service to keep up with a multi-channeled flow of their river of “social media.” However, these methods don’t serve as means to “share” ones following list on Twitter, itself. A curated following list does.

One reason I’ve chosen now to reboot the list is related to another experiment Dave Winer is running. One of the key philosophical underpinnings of the metaphor “social” in the concept of the social web is the belief that ones connections — what the Facebook people term “the social graph” — belong to an individual, and thus, an individual should have the ability to import or export such lists. In other words, when you sign onto a service that includes building a network of connections, you should have the ability to use that list in ways similar to how you’d use any list, even if its as simple as adding that information to a program that manages your personal contacts.

To this point, Twitter, to the credit of its owners, has focused on what it believes are its most challenging needs: the engineering ones necessary to scale the infrastructure (including the constant battle with the dark forces of spam and hack — the kinds — perpetrators) and grow the network of users (which the tsunami of hype is making relatively easy). One of the key components to Twitter’s success and popularity among the eco-system (or, geekosystem, in this case) of early third-party developers who use it, are its flexible APIs that allow the data generated from its system to be utilized in near endless ways.

So, for instance, if Twitter’s management think that list portability is not among the most critical feature necessary at this point, individuals like Dave can come up with creative hacks of the Twitter API (the good kind) to export following lists as an OPML file that can be imported into a newsreader and followed via RSS — again, to Twitter’s credit, for including, from day one, RSS as a feature. So, when Dave posted his hack (the good kind) for such an OPML export, I decided to use it to record and export my following list and to reboot the list on my account.

Having this record of my longer following lets me know I can still following them via RSS or via Twitter again if I want to review the list.