Category Archives: blogging

Twitter opens analytics platform to public [TL/DR: images get more engagement]

I briefly logged into Twitter’s free analytics service again today, prompted by a conversation on (you guessed it) Twitter about the demographics of an account’s followers and the news that it was now open to all.

Today, any Twitter user can log in and access the online dashboard and see what Twitter says about how people are interacting with your tweets, among other insights.

I was glad to see that dashboard is definitely working better now than when Twitter first gave me partial access. (I could see follower demographics but not impressions). I know that some people may see these stats as fake-ish numbers, but I wish Tumblr, Pinterest, Vine, Instagram & Google+ offered similar free dashboards for their users — certainly, it would be great if Facebook did for people who turned on the Follow feature.

What did I learn?

digiphile-Twitter-follower-demographics-august-2014First, looking at the highest impression number (155,000 impressions on this tweet) I was reminded that the concept of “free speech zones” remains controversial in the United States, and that tweeting about them can result in a different kind “engagements” than RTs or Favorites: angry @replies from lots of strangers.

This is particularly true if combined with a journalist embroiled in controversy over a misidentification of ammunition and the #Ferguson hashtag.

Second, the gender numbers in the demographics of my followers continues to be heavily skewed toward men (81% vs 19%), a situation that has endured more or less ever since the beginning of 2010, when Twitter began recommending me to new users in its technology vertical.

I invite and welcome any and all women who like to follow me to do so here, if you’re interested in the sorts of things I tweet about, just as I do on Facebook or other social networks.

digiphile-engagement-twitter-august-2014Finally, what Twitter Media and News staff had already told people who are listening is backed up by what they’re showing me: including pictures, maps and graphics in your tweets will raises your “engagement” numbers, at least as measured by people resharing tweets, favoriting them, @mentioning or @replying to them.

I’ve intentionally done that more over the latter half of August, and it shows up in the data.

It takes longer to find the right image for a tweet but the effort can pay off.

Adding that to the process reminds me of how I described Twitter back in 2008: a distributed microblogging platform.

While a few tweets may still be produced and received as simple, humble text messages, as in 2006, many more are much more complicated, and have been for some time.

Back in 2010, the map of a tweet already looked like this under the hood, with some 30 lines of meta data.

raffi-anatomy-of-a-tweet

Years later, updates to the platform are much more complex, with integrated cards, videos and pictures. As Twitter rolls out e-commerce from within tweets, I wonder if better dashboards for sales, subscriptions and other conversions might be on the way for the social media company’s customers, if not, perhaps, all of its users.

2 Comments

Filed under article, blogging, microsharing, social bookmarking, social media, Twitter

Reflections on online cruelty and kindness

This morning, I read an interesting reflection on dealing with online cruelty in the New York Times by Stephanie Rosenbloom:

In the virtual world, anonymity and invisibility help us feel uninhibited. Some people are inspired to behave with greater kindness; others unleash their dark side. Trolls, who some researchers think could be mentally unbalanced, say the kinds of things that do not warrant deep introspection; their singular goal is to elicit pain. But then there are those people whose comments, while nasty, present an opportunity to learn something about ourselves.

Easier said than done. Social scientists say we tend to fixate on the negative. However, there are ways to game psychological realities. Doing so requires understanding that you are ultimately in charge. “Nobody makes you feel anything,” said Professor Suler, adding that you are responsible for how you interpret and react to negative comments. The key is managing what psychologists refer to as involuntary attention.

When I checked her reference, I found that Rosenbloom made an error with her citation of research, along with failing to link to it: the 2011 report on teens, kindness and cruelty on social networking sites by the Pew Research’s Internet and Life Project she cited found that a vast majority of young people (88%) had “seen someone be mean or cruel to another person on a social network site,” not 69%. That percentage refers to a happier statistic: “69% of social media-using teens think that peers are mostly kind to each other on social network sites.

On that count, I’m glad the author chose to end with a reflection on kindness and the psychology involved with focusing on positive comments and compliments, as opposed to the negative ones. Anyone who wants to see how a positive feedback loop works should look at how Justin Levy’s friends & networks are supporting him, or how dozens and dozens of friends, family and strangers supported me when I lost my beloved greyhound this week.

I’m not sure about the New York Times editor’s summary — that the “Web encourages bad behavior,” through anonymity and lack of consequences.

I think that what we see online reflects what humans are, as a mirror, and that what we see on social media (which is really what is discussed here, not the World Wide Web) is 
1) a function of what the platforms allow, with respect to the architecture of participation, and
2) what the community norms established there are.

Compare newspapers’ online comments, YouTube comments and Twitter to what you find in the comments at Ars Technica, BoingBoing or even, dare I say it, in the blogs or public profiles I moderate. As Anil Dash has observed, the people who create and maintain online forums and platforms bear responsibility for what happens there:

It’s a surprisingly delicate balance to allow robust debate and disagreement on politics, current events, technology choices, or even sports (hello, tribalism) while guiding conversations away from cruelty, anger, or even hatred, whether we lead a classroom or an online discussion. The comments we allow to stand offline or online largely determine the culture of the class, town hall or thread they’re made within:

While people bear always responsibility for their own cruel actions or words, it’s incumbent upon those of us who host conversations or share our thoughts publicly online to try to respond with empathy, kindness and understanding where we can, and with polite but resolute moderation when others do not respond to those tactics or attack our friends and communities.

[IMAGE SOURCE: Amanda Lenhart, Pew Research Center]

2 Comments

Filed under article, blogging, friends, microsharing, personal, research, social bookmarking, social media, technology, Uncategorized

4 tips for social media optimization on Facebook from Robert Scoble

SocialMediaNews1 (1)

NYU journalism professor Jay Rosen has been trying to trick Facebook’s algorithm by starting his updates with “You guys! Exciting personal news: I’m moving to New York!” and “Big news in my personal life. I’m engaged!” and “I have some exciting personal news. A new job!”

What Rosen is experimenting with here is social media optimization (SMO), or the art and science of getting your updates seen on Facebook and other platforms. In 2014, SMO is still something of a dark art, but in an age when people are using social media to discover news, getting seen there is now as important to media, marketers and public officials who want to find immense audiences online as search engine optimization has become over the past decade.

Without this hack, Rosen says, “Facebook won’t show my posts to nearly all my subscribers.”

When they are shown, they get good engagement. The hack puts them in front of people who never received a thing from me, despite subscribing to me.

Robert [Scoble] is convinced that’s because I used some humor and sounded like a person. But it isn’t. It’s because I used some idiot phrases (“exciting personal news”) the algorithm responds to. That gives my posts a chance to be seen. When they are seen people engage with them. When people engage with them they are seen by more people.

Robert [Scoble] and Dave [Winer] are telling me it won’t work for long and I am sure that’s true. Meanwhile, I have many people telling me they never saw my posts and now they do.

Rosen’s tactic prompted Rackspace’s startup liason officer, Robert Scoble, a power user of social media platforms, to make several suggestions for crafting Facebook updates that get seen in the newsfeed. Here’s a short, paraphrased summary of those tips, courtesy of Robert Scoble’s comments on an update in Rosen’s feed.

1) Short, one paragraph updates often get more engagement than updates with a photo.

2) One photo in an update often gets more engagement than an update with multiple photos.

3) Including a call to action with a URL like “Click here for insight on open government, technology and society: http://e-pluribusunum.com” in an update leads to 80% more clicks.

4) Sort your friends into lists and then remove friends who don’t engage from those lists. (This is different from unfriending them.)

I’m both personally and professionally interested in this advice: I have 129,000 subscribers to my public Facebook updates but am quite certain that the vast majority never see what they subscribed to when they clicked the “Like” button. While the rare Facebook update blows up, many are just never seen.

Questions for you:

1) Have you tried these techniques? If so, what came of it?

2) What other SMO tips do you use?

3) Do you use Facebook lists?

Two cautionary notes

First, just as blackhat SEO leads Google to penalize people for gimmicks, Facebook could flag pages and profiles that overuse them, leading to account issues.

Second, marketers, social journalists and community managers that find success with SMO experiments should take note of a recent Pew Research Center survey on social media and the news, the source of the graphics on this page, which found that “visitors who come to a news site through Facebook or search display have far lower engagement with that outlet than those who come to that news website directly.” That means that SMO doesn’t replace SEO for publishers, or the need to create great stories and interactive content that stands on its own.

SocialMediaNews8

Finally, it’s worth noting that social networks are full of people. Ultimately, the best way to “optimize” your interactions on Facebook or elsewhere is to be a human, not a marketer. Meaningfully communicating with other humans is going to require a different strategy than crafting headlines and URLs that highly relevant to search engines.

Hat Tip: Post by Jay Rosen.

1 Comment

Filed under article, blogging, journalism, microsharing, social media, technology

When speech becomes text, what happens to writing?

downey

I successfully put down the baby for her late morning nap half an a hour ago. After running quietly around in sock feet trying to do things while she was out cold, I sat down to answer email and messages. As I entered this post into WordPress, she awoke again.)

It’s not easy to respond quickly and at volume using one hand or thumb, though I’ve gotten much better at both over the past five months with a baby daughter.

Over that time, I’ve been struck by how good the voice recognition in iOS on my iPhone has become. I’ve been able to successfully dictate a rough draft of a long article into the email interface and respond to any number of inbound inquiries that way.

That said, neither the soft keyboard nor voice-to-text on the device are a substitute yet for the 15″ keyboard in my MacBook Pro when I want to write at length.

It’s mostly a matter of numbers: I can still type away at more than eighty words per minute on the full-size keyboard, far faster than I can produce accurate text through any method on my smartphone.

Capturing and sharing anything other than text on the powerful device, however, has become trivially easy, from images to video to audio recordings.

The process of “writing” has long since escaped the boundaries of tabulas, slate and papyrus, moving from pens and paper to explode onto typewriters, personal computers and tablets.

Today, I’m thinking about how the bards of today will  be able to reclaim the oldest form of storytelling — the spoken word — and apply it in a new context.

As we enter the next decade of rapidly improving gestural and tactile interfaces for connected mobile devices, I wonder how long until the generations that preceded me will be able to leave decades of experience with keyboards behind and simply speak naturally to connected devices to share what they thinking or seeing with family, friends and coworkers.

Economist Paul Krugman seemed to be thinking about something similar this morning, in a blog post on “techno-optimism”, when he commented on the differences between economic and technological stagnation:

…I know it doesn’t show in the productivity numbers yet, but anyone who tracks technology has a strong sense that something big has been happening the past few years, that seemingly intractable problems — like speech recognition, adequate translation, self-driving cars, etc. — are suddenly becoming tractable. Basically, smart machines are getting much better at interacting with the natural environment in all its complexity. And that suggests that Skynet will soon kill us all a real transformative leap is somewhere over the horizon, maybe not this decade, but this generation.

Still, what do I know? But Brynjolfsson and McAfee have a new book — not yet out, but I have a manuscript — making this point with many examples and a lot of analysis.

There remain big questions about how the benefits of this technological surge, if it’s coming, will be distributed. But I think this kind of thing has to be taken into account when we try to imagine the future; I’m a great Gordon admirer, but his techniques necessarily involve extrapolating from the past, and aren’t well suited to picking up what could be a major inflection point.

That future feels much closer this morning.

[Image Credit: Navneet Alang, "Sci-Fi Fantasies, Real-Life Disappointments"]

8 Comments

Filed under blogging, journalism, research, scifi, technology

Beware sexy honeybots spear phishing on social media

220px-Robin_SageIf your connected life includes access to sensitive, proprietary or confidential information, be thoughtful about who you friend, follow or connect to online.

When fake femme fatale can dupe the IT guys at a government agency, you could also be spear phished.

If this all sounds familiar, you might be thinking of “Robin Sage,” when another fictitious femme fatale fooled security analysts, defense contractors and members of the military and intelligence agencies around the DC area.

Everything is new again.

[Image Credit: Wikipedia]

Leave a comment

Filed under blogging, security, social media, technology

What went wrong at Healthcare.gov?

20131017-101720.jpg

Folks, I’m going to be on the Kojo Nnamdi Show on Monday and need your help.

1) What are the best explanations of what went wrong at Healthcare.gov? This digest by Charles Ornstein is a start but I’d love more.

2) What are the best papers you’ve read about federal contracting? Where would you point people to understand how contracting works, why there are so many rules about how technology can be acquired and how this system needs to change/is changing?

Who do you think has best answered the question of “what went wrong at Healthcare .gov” amongst the national media and expert technologists?

In addition to the links above , I’d add:

What else should people read?

12/1/2013 Update: After two months of intense scrutiny, the tensions and troubles behind Healthcare.gov have been well-documented by investigative journalists at The Washington Post, The New York Times, The Wall Street Journal, ProPublica and NPR News.

No single issue led to the Healthcare.gov’s failure at relaunch on October 1. Rather, a combination of procurement problems, poor work by a key contractor, bad management skills, insularity and political sensitivity led to a bug-laden website with a broken backend.

How well is Healthcare.gov working today? Better, at least on the front-end, as detailed in an operational progress report released on December 1st. Lost in that update on the administration’s “big fix, however, was a detail released in a December 2nd post on improved window shopping at Healthcare.gov, published on the Department of Health and Human Services blog (emphasis is mine):

Over the last several weeks, we’ve made a number of changes to improve the accuracy of the “834” messages to issuers. The team, working with issuers, determined that more than 80 percent of 834 production errors were due to a bug that prevented a Social Security number from being included in the application, which in turn caused the system not to generate an 834. That bug has been fixed. Other issues related to the remaining 834 production issues have either been fixed or are in testing so that the fixes can be deployed soon.

In other words, when the Healthcare.gov marketplace launched, a single programming error meant that enrollment data being sent to insurers was invalid. That’s not just a bug: it’s a fundamental shortfall in meeting the requirements for a functional software application of this sort.

Leave a comment

Filed under blogging, journalism, technology