Tag Archives: twitter

What do IRC, Twitter and a sense of bewilderment have in common?


My first run in with IRC happened about 10 years ago. I was taking CS3210 class (now defunct) at Georgia Tech where we had to learn about OS internals by finding our way around the Linux kernel. We tried out all of our little kernel hacks on a Compaq iPaq device to which we had root access. Those were some fun projects, and at one point I found myself on an IRC channel dedicated to helping Linux newbies playing with iPaq’s.

Twitter feed circa 1740 (or IRC chat capture, I couldn't quite tell)

I don’t really remember what I was looking for or what brought me there, but what stayed with me was the memory of utter bewilderment and incomprehensibility of this mode of communication. I understood IM; I understood older, 90’s style chatrooms, and I was for all intents and purposes a technically inclined person, a kernel hacker even.

IRC to me was something arcane. This feeling was reinforced by bizarre server messages, cryptic lingo, strange interfaces, and routine and nonchalant rudeness heaped onto newbies. The more I think about it the more negative a user experience I am able to recall. Now here we are in 2010, and I have this conversation with a friend, who is a Twitter virgin, explaining what Twitter is all about (Conversation was shortened and paraphrased, identities of those involved protected.):

me: these are microblogs. every message is like an SMS, just 140 characters long. i.e. there are a lot of small messages on different subjects.

friend: clearly i don’t get it :)

me: here is an example: http://twitter.com/ibagrak i know it’s complicated :)

friend: it looks like random ravings, crazy talk

me: thanks :) context is everything. people are exchanging messages, and without context it’s really hard to understand.

friend: is there no better place where messages like this can be exchanged? i mean, this is completely out in the open, right?

me: yes, exactly

friend: it seems people who tweet do so either out of loneliness or sheer boredom. i can’t see any other reason.

<end scene>

So there you have it…

I think the two experiences have more than enough in common. Despite the fact that Twitter and IRC are two applications designed for two ends of technically savvy spectrum, I would argue that neither one offers an optimal experience for its audience and new users especially. Yes, both are immensely popular, but as long as the latter conversation takes place how could anyone argue about Twitter entering mainstream?

I don’t want to knock Twitter (or IRC). Not all apps are designed for Normals. In fact, great apps can deliver great value without being designed for Normals, and I think Twitter is a living proof that. But I also think there are simple things that Twitter and IRC could do, really minor cosmetic stuff in view of everything else that’s been done, that could vastly improve user experience.

But the point is even if Twitters of this world choose not to budge, it doesn’t matter. It only matters that we choose to do it right in our application.


Reflections on tweeting and blogging


I’ve already mentioned here how the value of Twitter was lost on me at first. I now want to talk about why I think Twitter makes sense and where it fits in.

First let’s separate the users by what they do with the application. There are two roles any one participant can play. At any given point in time they are either a reader or a writer. This is true for any publish/subscribe set up, so we can now consider the application or any other traditional medium (like human conversations) from the reader/writer perspective.

It’s no surprise that the effort required to create a single tweet is extremely small, so the barrier to entry for a writer on Twitter is also extremely low. A writer can generate a tweet in a matter of seconds. Not only can this be done rapidly, but it’s also be done on the go. In effect this drives up the number of writers in the system.

Now compare that to a typical blog writer. Maintaining a blog requires a huge, and often contiguous allocation of time. A decent blog post probably takes 30 minutes to 2 hours to create, that’s several orders magnitude more than what is required for creating a tweet.

I believe this is to be the main limiting factor depressing the number of blog writers. There are just not many people with this level of dedication and amount of free time to blog in a sustainable manner and to be successful in having a sizable following of readers.

So, yes, Twitter has a lot more writers, which in my view compensates for the relative shortness of messages. But it also drives up the barrier to entry for the reader since the reader must now process more disparate pieces of information. Whereas blog readers’ time is distributed among a handful of blogs with are updated daily, tweet readers’ time and attention must be distributed over a much large set of tweets which appear every minute.

So how does Twitter solve the reader problem? Well, Twitter makes it really easy to filter tweets and go from one tweet to another. The filtering thing for blogs is too coarse-grained and even if keyword searches are used they can only get us so far since it’s often difficult to express the filter in terms of keywords only. On Twitter, every person is a filter. Collectively the people you follow are a distributed content filtration system. As a reader, you tune this filtration system by selecting the people you follow.

In essence I see two fundamental differences of tweeting vs. blogging:

  1. High ratio of writers to readers, and less overhead for writing
  2. More natural person-oriented filtering capacity for the readers

Then what about blogs? Have their relevance been gobbled up by our shortening attention spans as readers and our laziness as writers?

I like to think of it as snacking versus having a complete N course dinner. You get your calories just the same, but you when you sit down to dinner what you eat is hopefully more balanced, nuanced, and thought out. It’s a more wholesome and a more lasting and altogether memorable experience.

If we now turn to the evolutionary component of this (which incidentally I know very little about, so I am opening myself up to ridicule here), humans are conditioned by evolution to like stories. They offer a lesson that can be internalized that that can prove beneficial or even vital later on down the road.

Likewise our survival often depends on short messages or alerts that tell us, roughly speaking, when danger or reward is on its way. I am obviously oversimplifying this since there are many different types of short messages that we could possibly be conditioned to care about besides the danger and reward ones.

The psychological divide between blogs and tweets, from the reader perspective,  is that the former offer lasting lessons and the latter succinct smoke signals for what may be right around the corner. Obviously, both must represent some value to the recipient.

My recent experience as a content consumer


Over the last month or so I’ve noticed a dramatic change in my online behavior.  For some time now I’ve been a regular Google Reader user.  I’ve set up the RSS feeds from all the excellent blogs and newspapers on the subjects that I am interested in and the bloggers I’ve come to expect quality content from. I had the feeds organized by categories like “work”, “books”, “news” and “tech”. My usual pattern was to log in Google Reader scan the different feeds and catch up on what was going.

I’ve also started paying more attention to the people in my Google Reader social network. I actually have no clue how these people got in there, since I definitely don’t remember adding them by hand. I think what Google did was to add people I’ve exchanged emails with and who had a Google Reader “profile” automatically. Actually, if someone can shed some light on this, I would appreciate it.

Anyways the people were good sources of interesting articles that they read and starred. These would appear in their “inboxes” on my Google Reader page. Since I had at some point a conversation with them (or was friends with them) we naturally shared some interests, and I enjoyed reading what they enjoyed reading.

The main problem with Google Reader was that I was always way way behind, literally thousands of items behind, and that depressed me. Every time I logged in, I saw how far behind I was, and I chastised myself for not logging in more and spending more time reading this great content that was just sitting there waiting for me.

Enter twitter. I actually joined twitter some time ago, and I really hated it at first. The thing I hated the most was the artificially imposed 140 character barrier. I hated the internal lingo of RT, @, # and whatever other weirdness that imposed itself on all its users. It was yet another jargon to get used to, and I am not a big fan of jargons. So I was quiet for a few months on twitter, not really getting what it was all about.

My first two tweets were:

  1. Still trying to figure out why I only have 140 letters to express myself. Wtf is this, some kind of joke? Livejournal? I repeat myself
  2. Haiku: another social network // what do i care? // i’ll just use it for bitching

And these are very indicative of the user experience I was having at the time.

So at this point you are probably asking yourself: is there a point to this post? Yes, in fact, I was about to make one. The point is that I finally got the value of twitter about a month ago, and the value is that I get quality content only from people that have a history of delivering — not from a site that’s hit or miss but from a live person who has a consistent set of interests and acts as a screen, a filter over many different sources of information. If the person is proven, so is their content, and the more interests you share with them the better.

The particular subject that I became interested in was startups. I wanted to get some insights into what the startup community was talking about and what things were trending, and in span of several weeks I read more quality pieces than I could have ever spotted if I simply signed up for the RSS feeds of a handful of happening startup blogs. The signal to noise ration is insanely high when you have the power of collective human filtering at your fingertips.And that IS the value of twitter to me.

Oh yeah, someone’s been talking too much about his girlfriend and not so much about startups? Bam! – just unfollow him/her. It’s that simple. That’s the learning/pruning part of the algorithm, and it yields immediate results.

So here I am telling you about all this stuff. I made a point (sort of), but is that somehow related to the app we are building? Incredibly it is. No, I mean the twitter switch was not pre-meditated, it just happened naturally after we decided we were going to build something. But the point is that the core assumption of our future product is that people are fundamentally more interesting to watch than sites and that people are far better sources of content than RSS or content editors.

Everyone has already done their homework and built their network of connections in real life. Whether we consciously control the process or not we pick the people who are like us and who share our values and beliefs. More importantly, they often share our interests. The only thing we need to facilitate is delivering relevant social content from your network to you, also making it easier for your network to create this content. And that’s effectively what we are doing.

So coincidence or not, but I do feel we are onto something here..