Danny Sullivan weighs in on URL shorteners, blasting ow.ly for its framed content that deprives the organization that created the content from page views and, thus, ad revenue. Top two: bit.ly and tr.im (I use both). New to me: cli.gs.

On Monday, Twitter reported that it is managing 50 million tweets a day (600 tweets per second). In 2007, the average (?) was 5,000 times a day and by 2009 hit 35 million a day.

Twitter Reports 50 Million Tweets Per Day

ComTweets (@comtweets) is a (free) service that organizes a Twitter community around a common email address, like Facebook organizes networks. The stated goal is to facilitate “easy discovery and communications between coworkers.” This is not unlikeYammer’s goal, but Yammer’s conversations are viewable only by people in the network (common email addresses). Of course, this means adding yet another social networking account to your plate.

Launched in May 2009, ComTweets lets “the world” see everyone who has elected to affiliate their Twitter account with their employer. And it is very easy to set up a new network on ComTweets: you simply sign up with your work email address. The first person to sign up gets to set up the organizational account, like so.

When you associate your Twitter account with an organization/network, ComTweets asks if this is a personal or organizational account. The question caused me to stumble, as I first thought “this” referenced the ComTweets network. Then I realized that ComTweets was asking about the Twitter account that I was associating with the email address. That’s when I knew I was to select “personal.”

Consequently ComTweets appends “employee” beneath my Twitter avatar when it displays my tweets. Note: it does not display @replies.

It’s possible that ComTweets could have two networks for UW, as we have two different sets of syntax for the same email address — uw.edu and u.washington.edu. We’re probably not the only large network with an issue like this, evidenced by this June blog postexplaining how to merge networks.

If your website or blog allows javascript, you can easily display tweets from the aggregated feed.

There are a few large-ish communities on ComTweets but in the main the numbers are small:

• CNN: 20 twitters, 21,915 tweets, 5,892,228 reaches
• Comcast: 27 twitters, 7,726 tweets, 51,929 reaches
• Google: 76 twitters, 35,898 tweets, 1,831,262 reaches
• IBM: 153 twitters, 68,875 tweets, 112,776 reaches
• Microsoft: 276 twitters, 183,891 tweets, 570,308 reaches
• Sapient Interactive: 1 twitters, 149 tweets, 126 reaches
• Temple University: 1 twitters, 59 tweets, 248 reaches
• Virgin Media: 4 twitters, 8,528 tweets, 1,269 reaches

What do you think? Useful or not?

This post first appeared at WiredPen.

The latest “celebrity death” to make the rounds on Twitter happened mid-day Thursday and was fueled by mainstream Canadian press. The “Drudge-ification” of North American news seems complete.

What’s more interesting than the Twitterstream is how the media treated their stories after the false report (AKA rumor) was outed.

Trendistic

The Gordon Lightfoot story was relatively short-lived and did not reach stratospheric numbers on Twitter (about 0.25% of all tweets). Some have used the short-life as an example of Twitter’s ability to self-correct. I don’t think that’s really the case; people were still making “he’s dead” retweets that contained links to the updated CANWEST story. That’s tweeting before thinking, a behavior that we need to haze.

Case Number 1: Keep the link, update the story

CANWEST was widely criticized for running the initial story reporting that Gordon Lightfoot had died. But rather than kill the link to the story, the media organization continued to update the story.

And the URL remained the same.

This is a key best practice. The revised story appeared on all CANWEST websites, so if someone clicked a link in a tweet that said ‘Gordon Lightfoot is dead’ — the result would be the news that the singer/songwriter was very much alive. Unfortunately, lots of people retweet without checking out the link that’s in the tweet.

Case Number 2: Kill the link, pretend it never happened

The converse happend at CBC Radio3. Maybe their software doesn’t allow revisions; if that’s the case, it’s time for application surgery (or burial).

An Incomplete Paper (erh, electronic) Trail

These tweets were harvested using Twitter’s advanced search. I searched for “Lightfoot and dead” as well as “Lightfoot and not dead.” I searched for “Lightfoot and died” as well as “Lightfoot and not died.” I also have what I think is a comprehensive archive of tweets with links that go back to 11.23 AM, the earliest Tweet that I found.

This post first appeared at WiredPen

Much has been made of Twitter’s role in helping raise money for victims of the Haitian earthquake. Less has been written about how Twitter makes it so easy to spread false information and the related need for digital literacy. I tackled these issues at several posts at WiredPen:

• Beware of Twitter Memes; How To Really Help Haiti
• A Case For Digital Media Literacy: Tracking Down A Meme
• More On AA/Haiti Twitter Meme

Prior to an interview on Friday about Twitter’s “staying power,” Caitlin Murphy (KIRO intern, UW grad) asked me that question. In preparation for our meeting, I tweeted: “Why do you like Twitter? Which is your favorite, Twitter or Facebook?” I also looked at some of the answers to the question “What would you say to a friend who asked you about Twitter?” from our spring survey.

Your Take-Aways

Here’s what some of y’all said on Twitter:

From Paolo Mottola (@paolojr):
Twitter is my social RSS feed, but Facebook is where I socialize. Facebook>Twitter IMO

From Tarja Kallinen (@tkallinen):
I like twitter because it’s easy (with iPhone), instant, brief, just main points. I like Facebook too but it’s more labor-intensive
and
One more thing about FB vs Twitter: on FB you’re communicating w/ friends-feels inappropriate to be too opinionated or political.

From James Roberts (@jrnoded):
twitter by a long [shot] — no hidden agendas (well maybe) but just conversation no fluff.

From Shannon Huggins (@shuggins_6):
Depends on what I want to do.

From Walt Boyes (@waltboyes)
twitter for fast breaking stuff, facebook for most everything else.

From Elaine Nelson (@epersonae)
re twitter vs facebook: twitter, definitely. it doesn’t try to absorb my whole online life.

And some of our anonymous survey participants (data to be released in 2010 with the Twitter book):

• [Twitter] connects you to your choice of social nervous system.
• Ultimately the value in Twitter is specifically due to the length restriction of messages.
• [Twitter is] the main way I keep in touch with friends, meet new people, get my news, follow what’s going on in Seattle…
• Twitter, from the view point of a television news person, offers a chance to communicate with the community and put a human face onto an industry often maligned as a “faceless corporate monster.”

My Take-Aways

Those comments mirror my biases:

• Brevity is good. (Count the number of characters in each of the 10 Commandments in the Old Testament — only two are more than 140 and I can easily edit those two to fewer than 140 characters without affecting meaning.)
• Simple is good. (I’m now using NutshellMail so that I don’t have to deal with the Facebook interface.)
• Openness is good. (I used to think of tweets as being ephemeral, but now that I’m getting Facebook updates via NutshellMail, I realize that it is the Facebook status that is truly ephemeral, if you have friended more than a few people).
• Single-purpose is good. (Generally speaking, I prefer tools designed for a job than a Swiss-army knife.)

When Caitlin asked “how do people use Twitter?” I laughed and replied, “how do people use the phone?”

I was being slightly facetious, but only slightly. For example, it wasn’t that long ago that one of my best friends from college kept a cellphone for “emergencies only” (think flat tire). At the same point in time, my mother was using a cellphone for its “weekends free” long distance plan: it was how she kept in touch with her sisters.

Twitter is a communications tool, a tool that can be used in a myriad of ways, just like the phone. Just like “the mail.” Just like faxes. Just like email. Just like “blogs.” Just like printing presses!

We talked about the differences between Twitter and Facebook (openness, reciprocity) and the current tension arising from Mark Zuckerman’s efforts to strong-arm members into being more “public” with their information (“user-generated content”).

On the “where do you think Twitter will be in five years” question (“staying power”) … I reminded her that five years ago we didn’t have YouTube (founded in February 2005) and yet by May 2006 its viewership has passed CNN. Instead, I told her that what we are calling “the real time” (public) web is what is here to stay.

How we are going to be accessing the real-time web in five years? That’s where my crystal ball gets cloudy. How clear is yours?

Related: Tim O’Reilly, Why I Love Twitter

This post first appeared at WiredPen

TweetDeck released version 0.32.0 today, a desktop application update that includes the new Twitter retweet feature. On the plus side, TweetDeck makes it easy to choose between sending an edited retweet and a new retweet. On the negative side, TweetDeck does not bump a tweet if you are already following a person who has been retweeted.

The developers thought first of the account holder author: when you click the TweetDeck retweet icon, TweetDeck asks you if you want to just retweet (the new feature) or “edit then retweet.” If you change your mind, you must make the second choice, then highlight/delete. And you can make this decision permanent in settings.

However, unlike Tweetelator (iPhone only), TweetDeck does not “bump” the new retweet. Why is this a big deal? As I wrote two weeks ago:

One of the advantages (to the original author) of RTs is that each becomes a new tweet, a new instance. This increases the chance that someone will “see” the original tweet.

Think of a retweet as “bumping” an item “up” in time. But that’s not how the new Twitter.com interface works. On low volume accounts, this isn’t a big issue. But on moderate– to high-volume accounts (measured by number followed), it is. Here’s why.

Twitter doesn’t “bump” the old tweet — it merely changes the “retweeted by” count. So if you read your Tweets on theTwitter.com site and missed a tweet the first time around, you’ll miss it each-and-every time it’s retweeted … because it will remain in “history,” far below the fold.

TweetDeck desktop does, however, flag a new retweet and does so in a way that provides more visual information to the reader than the Twitter web interface. Note that the timestamp is that of the original tweet, not the retweet, which looks very odd in your timeline.

TweetDeck is following the paradigm established by Twitter. If you are reading Twitter via the web interface, you’ll find that Twitter does not “bump” a tweet if you are already following someone who has been retweeted since you read the original tweet. Moreover, because that Tweet is “in history” (so to speak), that tweet will never show that it has been retweeted. This oversight remains a fatal flaw in this system, in my opinion.

This post first appeared at WiredPen

Almost two weeks ago, I urged early recipients of the Twitter retweet link to be cautious with its use, because most third party clients seemed unable to display these retweets.

Today I’m repeating the caution, and it’s not only because of spotty third party client implementation. It’s also because these new retweets don’t become “real time” in the Twitter.com timeline for tweets made by anyone you are already following.

High Level Summary
One of the advantages (to the original author) of RTs is that each becomes a new tweet, a new instance. This increases the chance that someone will “see” the original tweet.

Think of a retweet as “bumping” an item “up” in time. But that’s not how the new Twitter.com interface works. On low volume accounts, this isn’t a big issue. But on moderate– to high-volume accounts (measured by number followed), it is. Here’s why.

Twitter doesn’t “bump” the old tweet — it merely changes the “retweeted by” count. So if you read your Tweets on the Twitter.com site and missed a tweet the first time around, you’ll miss it each-and-every time it’s retweeted … because it will remain in “history,” far below the fold.

The third party clients that have implemented the new retweet feature, however, appear to be treating the retweet like a new instance. In other words, they are treating them like the “old” retweets. The problem: some popular third party clients haven’t yet implemented the feature.

1. Twitter.com — The new RT feature does not “bump” tweets, ie, these new retweets are not “real time” for tweets made by anyone you are already following. I think this is a fatal flaw.
2. TweetDeck — both desktop and iPhone application — is still not displaying retweets sent via the “retweet link” in the Twitter web interface.
3. Seesmic Desktop is still not displaying retweets sent via the “retweet link” in the Twitter web interface.
4. Brizzly, a beta web interface for managing your Twitter and Facebook accounts, is displaying the new retweets
5. Tweetelator and Tweetie2 (iphone apps) are still displaying the new retweets

Methodology
In order to test how well clients render the new retweets, I “bracketed” the new retweet with an old-fashioned copy-and-paste one — either immediately before or immediately after using the Twitter retweet link. I sent retweets from all three of my accounts: @kegill, @kegill_uw, @kathygill. Each account follows the other two; this is great for testing!

1. Twitter.com
Even though I do not follow Howard Rheingold from my @kegill_uw account, the retweet showed up in that timeline on Twitter.com, with his avatar. Note that the timestamp relates to when Howard tweeted, not when I retweeted. This is a major change and is a little jarring if you’re looking at tweets before and after it that were sent seconds ago. Twitter should add the time stamp after “you”. This time-stamp business has other ramifications that I will explain momentarily.

On both of my low-volume accounts, the Twitter.com interface shows both original and test retweets in real-time. But on my high volume account, @kegill, not a single test tweet (there were four) was displayed in real-time. Why not?

For example, the @kathygill timeline shows multiple retweets — both the “test” retweets as well as tweets retweeted by the account and by other accounts. This is a very very low volume account (follows 18 mostly low-volume accounts).

Next, see an example of the @kegill timeline, with the @eMarketer test retweet showing at the top. In this instance, the test retweet was made after the new retweet, as you can see from the prior screen capture.

The original tweet, with the eMarketer avatar, did not appear in my @kegill real-time timeline on Twitter.com, although I retweeted it, using the new feature, from both @kathygill and @kegill_uw. I went 5 minutes back in time looking for it, even though the tweets were sent seconds apart.

This is a big problem with the new feature.

Because @kegill already follows @eMarketer, that tweet had already been displayed in my timeline. The new retweet doesn’t “bump” the tweet in time, it merely changes the “how many people RTed” data on the original tweet. This means retweets aren’t “real-time” for anyone who is already following the account you retweet. I think this is a fatal flaw.

However, if someone that I do not follow is retweeted by someone that I do follow, that retweet shows up in real-time, even when it’s really old (in this case, 48 hours old):

One of the advantages of the new feature, however, is public metrics. Twitter added a “Retweets” link beneath favorites (right hand navigation). It’s a handy way to see how many people retweet the things you do (or your tweets) as well as who they are.

2 and 3. TweetDeck and Seesmic

I made several attempts to see if TweetDeck would display a tweet that was retweeted via the new web interface link. Not a single one was displayed on TweetDeck, whether iPhone app or desktop app. I do have the latest version of TweetDeck on both the desktop and iPhone.

Here are sample screenshots — they show the “test” retweet (copy-and-paste). I guess you’ll have to take my word for it when I saw the original did not show on any screen.

In this screenshot from the desktop, the “test” retweet was sent immediately after the original. Howard Rheingold’s tweet, without my “test” label, did not show up in TweetDeck.

The same is true for TweetDeck iPhone. In this instance, the “test” retweet was sent immediately prior to the original.

Seesmic Desktop does not display the new retweets, either. What follows are three screen captures — one from Twitter.com, showing a retweet; one from Seesmic, showing that it is missing; and one from Tweetdeck, showing that it’s missing.

4. Brizzly
Brizzly is in beta (just ask if you’d like an invitation) and is doing a great job of rendering new retweets.

4. iPhone application Tweetelator
There’s no change since my last report; Tweetelator is still displaying the new retweets. Moreover, it marks them with a little flag:

Conclusion
First, the new retweet link is designed to make it easier for people to retweet. However, there’s nothing to stop old-fashioned copy-and-paste if you’re using the Twitter.com web interface. Keep an eye on how your third-party clients implement this feature — will they retain their current “RT” process or adopt the new? My vote is for the former.

Second, one advantage of using the retweet link is the Retweets summary link, just below Favorites in the right-hand navigation. This feature makes public statistics about each retweet — just how many people retweeted and who they are. I’m not convinced this advantage outweighs the limitations of retweets not being real-time, however, for moderate– to high-volume accounts.

Therefore, until Twitter displays retweets in real-time in the timeline (something that I doubt they will do) and until all third party clients display retweets (hopefully like Tweetelator and Brizzly, in real-time), I again urge caution if you are using Twitter for branding or influence. Not only is there an issue with everyone being able to see these retweets (think Tweetdeck) but there is the issue of their possibly not being real-time on the Twitter.com interface.

This post first appeared at WiredPen

For those of you who are in the Twitter “retweet” beta test, I have a word of advice: proceed cautiously.

Here’s why. Currently, retweets that are executed via the Twitter web “retweet link” are visible to your followers who are using theweb interface to read your tweets but are not visible to popular third party clients.

Let me say that a different way: popular third party applications are currently not displaying RTs executed via the Twitter web interface “retweet” link. In other words, these retweets are MIA in a follower’s timeline in popular third-party clients. The exception: Tweetie2.

This may be a “cart before the horse” problem. In other words, third party applications may not have had a chance to integrate the new API. Or it may be that Twitter privileged Tweetie2 developers. I don’t know.

But if most of your followers read your tweets using third party desktop clients, this “bug” could have a serious impact on your personal retweet rate in the short-term. Experiment wisely. And mindfully.

The Story In Pictures

People who are part of the beta test will see a familiar-looking “alert” on their Twitter home page:

Sending A Retweet With The New Feature

Twitter has made it very easy to send a retweet via the web interface. This may be an attempt to make the web interface a more pleasant experience for people who follow a lot of accounts. Certainly, the ability to easily retweet has been a mainstay of third-party Twitter clients.

(1) Read your tweets via the web interface. When you find a tweet that seems interesting enough to share with your followers, mouse-over. You’ll see the “retweet” link to the right of the familiar “reply” link.

(2) After you retweet, this is what the tweet looks like from your home page: it bears the avatar of the original account (instead of your avatar) and includes information about who retweeted it as a tagline. Thus, this new feature might make it easier to discover interesting people to follow.

versus

(3) On your profile page, the visual difference starts with an icon instead of initials (RT). Again, the Twitter ID that shows is the original author, not the person who retweeted. Note that there is an “undo” option. I haven’t tested it to see how long this “delete” feature takes.

(4) In addition, Twitter tells you how many other people have retweeted a specific tweet. That’s an incentive, of sorts, to use the new feature.

Receiving A Retweet With The New Feature

If you are in the beta test, when you receive a retweet from someone else in the beta-test, you’ll see an alert when you are reading your tweets from the web. But if you aren’t reading them from the web, you won’t see those retweets in Seesmic, Tweetdeck or TwitBirdPro.

(1) If you are in the beta test, Twitter gives you a heads-up to explain why you are seeing a new avatar in your timeline. This is a smart move for the part of the Twitter community that follows a small-ish number of accounts; for them, the new avatar might be visually jarring.

(2) If you are not in the beta test, your retweets look exactly like they always have, in the web interface. (This is how my “new retweet” tweet looks in my kegill_uw account. Yes, I follow myself there.)

(3) However, the problem comes for your followers who do not use the web to read your Tweets. The Twitter-powered retweet simply falls into a black hole.

First, see the Barbara Clements retweet in context (the tweets before and after it, web interface). Then look at Seemsic, from the desktop, and Tweetdeck and TwitBirdPro, from the iPhone. Notice that the Barbara Clements retweet is MIA in all three instances.

(4) One exception appears to be Tweetie2. My @romensko retweet from the new interface does show up in my kegill_uw account in Tweetie2, just like it did on the web interface.

So there you have it.

Be judicious in your use of the new retweet link if you think most of your followers read your tweets from a third party client, unless that client is Tweetie2. I’ll update this post as I test more clients.

Update: 10.30 pm Saturday
Here is the “base” tweet we’re looking for:

And how that retweet “looks” at the kegill_uw account, in context:

(1) No Go: Twitscoop:

(2) No Go: Twitterrific:

(3) No Go: TwitBirdPro
There was an update for this application at the iTunes store, but it didn’t enable this functionality.

This post first appeared at WiredPen

Not at the OK Corral but at the Web 2.0 Summit. Microsoft and Bing face off against Google; the fight is over Twitter, as Microsoft has Facebook in its pocket. (Remember the $240 million investment.) Marshall Kirkpatrick at ReadWriteWeb sorts through the search fight that came to a head today.

Most Facebook users want “privacy” (there’s the reciprocity thing) and keep their content protected. Most Twitter users want connections and keep their content public (ie, not protected). I think it will be more difficult to convince Facebook folks to “open up” than Twitter folks to update their Google profiles, if by updating the profile they’ll get better (more contextual, more meaningful) search results.