Daniel's Place

I was recently looking through my expense reports for the year so far and found an interesting anomaly in my Automotive expenditures:

Washington liquor control board = automotive expense?

As you can see, a payment to the Washington liquor control board was marked as an Automobile expense. Normally, drinking and driving should not have anything to do with each other.

Is Money trying to tell me something?

So if you want to make money on the internet, there are typically two major choices:

1. Charge people for a service that should be free (bad idea)
2. Show people relevant, tasteful, non-invasive ads (slightly better idea)

Since I’m not a genius, I’ll settle for slightly better.

The question: How would you like your ads delivered? Click here to respond to the survey.

Yesterday was historic, we saw the inauguration of a visionary new national executive and @scobleeffect overtook @variables in terms of followers. Like all historical events, the caveat is that neither Matt or I did any work toward this goal. This growth was organic and presumably a delayed effect of @scobleeffect’s followers’ scripts.

It should be noted that @variables has far more useful content posted to his Twitter account than @scobleeffect. A quick conclusion is that many of these auto-follow scripts do not attempt to assess the credibility of those that they follow. The idea for a list is born.

I found a great inspirational article in 43folders called Photography, and the Tolerance for Courageous Sucking as I was clicking around the Photography section of Alltop.

I won’t go any further into summarizing the article than by saying that its author uses a short journey for a lost shoe to justify being that weird guy who always carries around that silly looking oversized camera. I was so inspired, I stuffed my 35mm Contax into my messenger bag and swore that, in addition to taking my normal photography outings, I would use life as the outing and photograph it.

We’ll see how long that lasts.

It’s been a while and updates are probably in order. Since the first post, the plans to exploit the API for fast following failed and our research has forked into two separate and competitive threads. Let’s see where things go.

Matt (@variables) has developed his own system for selecting and following people that is largely subjective compared to the original Scoble Effect designs. He also initiated the competitive spirit and was able to pull far ahead with 570 followers as of the time of this posting. Further trend information can be found Twitterholic.

My designs have remained roughly the same though I give myself little time to implement. As of writing this post, we have determined that working in the browser is much more effective (albeit click-intensive) than working through the API (which limits the number of requests). On a side note, the accounts we registered in quick succession all came under suspension and scrutiny as we tried using the API to follow our influentials. Many of those accounts have been suspended and API calls do not work.

Lesson: Don’t abuse Twitter, they’re not totally naive.

As I click through all the names on the list I am finding very many interesting people and projects. Many of these I am noting down and will probably follow on my personal account.

My next steps after following that massive list will be:

• Find and profile each of the major influentials in brief
• Tweet for a little while to add legitimacy to the account
• Come up with stuff to say to the influentials based on profile and context
• Say it
• Watch the stats

Wish me luck!

I love Twitter. I recently found TweetDeck. So I’ll write about it.

TweetDeck is a wonderful Twitter client written in Adobe Air. The design is pretty good and the interface fairly intuitive. With TweetDeck it’s easy to wade through a tsunami of tweets and post replies quickly. I can also store search queries as feeds within TweetDeck to get an immediate pulse on what is happening around the Twitterverse. These are all good things and very good reasons to use TweetDeck. Betas, however, have their quirks.

Beta is better than nothing. And a good time for feedback! So here is some feedback for the next release:

• The buttons and icons are too small – Seriously guys, I have 20/15 vision and I have no idea what those buttons do until I mouse over.
• No easy way to scroll sideways – The application was designed for vertical and (lots of) horizontal scrolling but in its current state I can only use the mouse wheel to scroll in one direction.
• Color scheme contributes little to utility – I have not found a way to color code tweets or feeds individually. If feeds are color coded, users can more easily make sense of the context of horizontal feeds.
• Memory - TweetDeck is to memory as a volcanic eruption is to carbon emissions. This may be Adobe’s fault, this may be TweetDeck’s fault. One fact remains: Collect your garbage! This simple app should not be consuming more memory than my browser.

TweetDeck's memory footprint is larger than my browser!

All around, TweetDeck is a great application. I spend less time navigating and more time tweeting. Hopefully they improve on those few points outlined above.

I strongly recommend that anyone who uses Twitter regularly give TweetDeck a try!

Matt Crowley (@variables) and I were sitting in our usual Starbucks one day, discussing the implications of tools like Twitter and their ability to influence the masses. Having recently tinkered (documentation coming soon) with the Twitter API, we decided it was time to try doing something useful.

The Idea

A while back, Matt replied to one of Scoble’s (@scobleizer) posts. Scoble then posted an @ reply to Matt. Shortly after that, Matt’s popularity in  Twitter followers jumped by about 20 users. A significant increase in reach and influence. Obviously the effect of having an “infulential” like Scoble take notice and respond has an effect on a person’s twitter popularity. Matt  coined the term “The Scoble Effect.”

An obvious question immediately came to mind: What can we do with existing popular influencers to gain more influence and increase the legitimacy of opinions we post?

Hypothesis: By increasing the number of Twitter followers you have, you increase your influence in the Twitter community.

That’s great. How do we get many followers at once and how do we make our posts more legitimate?

Ever followed one of the twitter bigwigs, the kind with tons of followers (think @scobleizer, @pistachio, @guykawasaki)? They followed you back almost immediately. These powerful and influential figures are using auto-follow scripts like SocialToo and Tweet Later to add an impersonal touch when expanding their influence.

Gaining influence by working directly with the influencers is contrary to Guy Kawasaki’s first rule of Using Twitter as a Twool: Forget the “influentials.” Perhaps we are braving new territory. Maybe we’ll be wrong. Let’s do some research:

Research

There are several questions we think we should address to get this project off the ground:

• Who are the influentials? These are the people we want following.
• Who is auto-following? Which of them will auto-follow us?
• What can we say to these folks? We don’t have to speak to each influential, but we should come up with some well thought out and targeted questions and thoughts that will catch attention and get @ replies.
• How do we measure success?
• How long before we lose interest?
  

Who are the influentials?

Twitterholic is a great place for Twitter statistics. We take the top 1,000 Twitter users with the highest friend count. Follow each of those top 1,000 users.

Who has auto-follow enabled?

Follow the influentials from the above question. Check back a little later to see who is auto-following. Given that these influentials have so many followers, it stands to reason that they would auto-follow instead of manually reacting to each of 10,000 follow notifications.

Follow notifications come through email; so we’ll get an email when their auto-follow bot does its job and follows us back.

What can we say to these folks?

This will require special thought based on the users in question. Look for an update to this post or updates in the next post.

How do we measure success?

If one of the influentials @ replies us and our follower count jumps, that’s success. In short, we’ll consider The Scoble Effect worth more research.

How long before we lose interest?

Given products we’ve created in the past and our historical attention spans: one week. On the Twitter timeline, this is ancient history.

I recently finished Donald Norman’s Emotional Design: Why We Love (or Hate) Everyday Things. Indeed an interesting read and one that has me thinking about the world (more so than before) from the designer’s perspective. It would seem that most designers are sadistic.

The first two thirds of the book are an interesting and informative look at designing products for aesthetics, use, and the owner’s sense of pride (called visceral, behavioral, and reflective respectively). I would absolutely recommend the book to anyone who designs products with commercial intent.

The last part of the book seemed to be an exposition of the visionary variety. Norman spoke at length about the design of robots from the human and robot perspective. While this is an interesting study, I think Norman was wrong about robotics at the time of the book’s writing. Furthermore, I am not sure what to make of the concept of giving robots emotions.

The larger debate on instillation of emotion into artificial technology is beyond the scope of this post, but certainly valid for a future post.

One of the larger problems that should be solved before we move on to having robots make high level emotional decisions should be standardization of communication. How many different ways are there for me to schedule an item in my datebook? Assign the identity “apple” to a picture of an apple? Log an identity into different websites (e.g. Facebook Connect, OpenID, Live ID)? In addition to the definition of these standards, we need implementation.

Many businesses are starting to adopt XBRL for reporting financial statements. A step in the right direction.

I say that we should establish a clear methodology for saying an orange is an orange, “Let’s have Burgermaster for lunch at 12:00pm next Tuesday,” and that I really am Daniel Jurek writing these words. Once that’s done, we can tackle larger problems, like whether my washing machine should try to understand why I am upset at the Seattle weather.

</tangent>

Emotional Design is an excellent book and certainly worth the read if you find that your products continually frustrate your users. Norman gives many examples of good visceral, behavioral, and reflective design.