Wednesday, March 21, 2012

Following the Discussion

Over the time that online discussions have been around, there has been an evolution in the way we read things. Near the dawn of "social media" the metaphor required that you first find the right discussion, and then you could see what people say. (This was the way it worked in the BBS and "discussion forum" days.)

The current fashion is to start with the people (presumably that you know and trust) and then delve your way into the discussions (which, presumably would be the ones of interest to you). We progressed from "who's talking about my topic" to "what topics are on my friends' minds?"

The mechanism that allows us to use the topic or idea to discover people, and of course to explore the ideas themselves is the #hashtag. Here are some of the things we've learned about the use (and abuse) of the hashtag.

Using the hashtag on Twitter - Here it's pretty easy to track a hashtag, and in a way it's just about the only way to tie in to a specific discussion. Using twitter search, or a tool like Tweetdeck that allows you to set up monitor columns using a specific search criterion, you can easily see what's being said about a topic -- as long as the content contributors religiously use the tag.

Using the hashtag on Google+ - The search for a hashtag is also straightforward and effective at Google+. What comes back is a news stream that contains posts that contain the tag. The result is an easily read collection of posts. It is even possible to use a more complex search string to aggregate elements of a conversation once commonly used terms become apparent from inspection of the content. (To see an example, try this search -- generated by using the search terms "#occupy" and "#OWS")

Preserving the results of a hashtag search on Twitter - This is the tricky part.

We had held a fair amount of hope for Tweetdoc, which promises to generate a PDF document from the result of a search -- but we found that our results were not complete and that the document generated was a little bit awkward.

We used a hashtag that is somewhat specialized (#48taos was used to refer to a specific recent event, Filmapalooza) and we asked for a document that contained all mentions of the tag from Feb 15, 2012 through Mar 15, 2012. This covered the primary activity range of traffic on that tag.  Tweetdoc returned only 10 results (we discovered in an independent search that there should be 47).

The resulting document has a cover page with statistics and thumbnail profile photos of the discussion participants, but then a vast expanse of white space. The actual twitter updates appear, beginning on the second page. We fear that some readers may glance at the document and having reached the presumption that no results were returned, abandon reading before they can get far enough in to see the actual conversation.

We found our favorite approach to this task to come from the combination of two useful tools. Using Social Mention and Evernote, we were able to create a document that met our objectives. The workflow looks like this:

  1. Use Social Mention to perform the search on a specific hashtag. (We suggest using their advanced search to narrow and refine the results display. In our example, we used #48taos, limited the search to Microblogs, and asked for 100 results per page.)
  2. Use the Evernote Web Clipper to clip the page to a note. (You will need to set up a free Evernote account, and download the web clipper extension for your browser. Trust us, it's free and non-toxic.)
  3. Print the resultant note to a PDF output device. (Most modern systems either have an option in the browser to export a document to PDF, or a printer device that generates a PDF on the local drive.)

It takes more steps and you may have to do a little setup first, but the final result is that you can have a document to share that aggregates the conversation that surrounds a specific hashtag.

Here's an example of our efforts on behalf of Filmapalooza. (Twitter updates from Filmapalooza)

And here's what people suggested that Twitter might have been like prior to the invention of pie!




No comments:

Post a Comment

Reduce to Dashboard

When developers use DataWeave, they often come to rely on the reduce() function to fill in any gaps left by the standard Core library. Altho...