Chris Saad – Paying Attention

Twitter Lists and Tags

// October 30th, 2009 // 0 Comments // Analysis, Technology

In my previous post (written 5 minutes ago) I talk about Twitter Lists in relation to shared namespaces (Hint: They are not in a shared namespace).

Another under-reported fact, however, is that lists are also Tags. They are a great way for Twitter to learn how Twitter users are perceived and grouped (As a side note, they are also great for people to see how other people perceive them – one of my favorite lists in which I am listed: @chadcat/unreasonably-talented haha).

One could easily see an algorithm that can determine accurate APML data about each user not just by looking at their Tweet history, but by also checking their Bios and the Tweet History/Bios of the people they are listed with. The list name itself, in fact, is a very concentrated form of topic/tag data.

Do lists double as Twitter’s user tagging feature?

Who will be the first to ship an automated user discovery directory based on analyzing the relationship between users who are on the same lists?

I hope MrTweet is already working on this!

Twitter Lists and Namespaces

// October 30th, 2009 // 0 Comments // Analysis, Dataportability, Technology

A very important fact that seems to be getting little to no coverage at the moment about Twitter Lists is the issue of namespaces.

Twitter’s number one asset is its control and allocation of namespaces. Those little things we call ‘Usernames’. @chrissaad is not just my Twitter Name, it is a short form addressable identity that concretely links to my Twitter inbox any time someone uses it in a Tweet.

Addressable, convenient namespaces that can be used in a sentence like this are so interesting and important that facebook went to great lengths to copy them. Nothing on the open web has yet come close to this simplicity and effectiveness. Which is not to say there won’t be an alternative soon.

The important fact with Twitter usernames, though, is that they are unique. There is a finite and shared ’space’ in which ‘names’ can be allocated.

The result is that early adopters end up with all the best names and squatters rush to lock up all the best phrases. Late comers to the system end up with names like chris2423.

Twitter Lists, however, are different. They include the list creator’s username. For example my JS-Kit list is “@ChrisSaad/jskit“.

As you can see, the list ‘JSKIT’ is attached to my username. This means means that each user has their own namespace.

This result: There can’t be a landrush for List names because the list naming convention sits on top of the username. It also means that no one can own a definitive list on a subject because each list is subjective.

This is an important design decision for Twitter. One that has both pros and cons for the community. Overall, however, I think the decision was a correct one. Lists can rise and fall organically (or at least based on the influence and popularity of their creators) without the pain and pressure (for Twitter) of maintaining yet another shared namespace.

Twitter’s username namespace, however, is just rife with and waiting for all sorts of headaches. I don’t envy their position and I can’t wait for an open alternative.