Dr Karl Stolley

A number of people have expressed interest in having a collection of everything that was Tweeted at Computers and Writing 2009 with the #cw09 hashtag. However, because there were more than 1500 Tweets from the conference (!), it is not possible for any one person to archive all of the Tweets.

If we are going to have such an archive, it will take a community effort. And fast–we have less than a day or two to do this.

First, here’s what you should do (I’ll explain why next): go to your favorite, feed-enabled Web browser and enter this URL, replacing USERNAME with your Twitter username and PAGE with a number starting with 1 and increasing it by one until you no longer have any results left:

http://search.twitter.com/search.atom?q=%23cw09&rpp=100&from=USERNAME&page=PAGE

Each time you run this query, right-click and choose View Page Source; in a good browser, you will see the raw XML (and not the styled HTML page that browsers like Safari and Firefox generate for RSS). Your ATOM/RSS will look something like this (my most recent 100 #cw09 tweets).

Copy that entire chunk of source by right-clicking and choosing Select All, and paste it into a good text editor: either WordPad on Windows (*not* Notepad) or TextEdit on Mac. (In TextEdit, choose Format > Make Plain Text if it’s trying to save as RTF.)

Save the file as like username-cw09-page-1.rss.txt for the first page, username-cw09-page-2.rss.txt for the second page, and so on. Because of the way XML works, do not paste all of your results into a single file; keep a separate file for each time you up the page number in the URL.

We’ll have to figure out as a community where to share and post these files, and how to (eventually) get them into a searchable, sharable database as an SQL file. But for now, just be sure to grab your Tweets, save them, and keep them safe.

The reason this is necessary is that Twitter will only return 1500 results for a given search (like #cw09), and it will only return results that are less than roughly a week old. So that’s why we have to do this individually, and do it fast. After sometime on Thursday of this week, the Tweets from during the conference will rapidly begin to disappear from the searchable Tweet stream.

4 comments.

I was fascinated by the #CNNfail Tweetstream Sunday night. The “Tweetstream” is nothing more than the realtime updates to a search for a particular word or phrase on Twitter at search.twitter.com. but when I posted about it to Facebook from Twitter via Selective Twitter Updates, some of my non-tweeting friends reported being unable to decipher what one of them described as Twitter’s “jargon.”

Well, the Tweetstream is not really plagued by what I would call jargon. Rather, Twitter users have developed their own shorthand for communicating within Twitter’s hallmark 140-character post limit. (And it is important to note that the conventions I’m listing here have emerged from the community’s broader rhetorical practice, and not as features designed by Twitter the website/service itself.)

Here’s a tweet, or actually a retweet, of mine roughly as it would have appeared in the #CNNfail Tweetstream:

karlstolley RT @timleberecht: How #CNN fail(ed) – screenshots from news sites worldwide http://bit.ly/bqgHN #iranelection #cnnfail #fb

• So, the first item that appears is the username of the person who’s doing the tweeting; no big mystery there.
• Next up is RT, or “retweet,” which is a sort of citation mechanism in Twitter that, among other things, cause tweets to be passed on beyond one’s own immediate group of followers.
• @username, which is the syntax for both referring to and directing messages publicly to another Twitter user, based on her username. In the Tweet above, I’m retweeting a tweet written by @timlebrecht, whose Tweet was just one that I spotted in the Tweetstream.
• Tim also uses a hashtag, #CNN, in the body of his post; hashtags are another Twitter convention and can be used on words in the message (as #CNN is) or at the end (or even the beginning) of a post, as #iranelection and #cnnfail are. I also added the #fb hashtag so that this particular Tweet would update my Facebook status, as is the convention with the Selective Twitter Updates app.
• http://bit.ly/bqgHN, which is the URL that Tim was pointing to; bit.ly is a URL shortening service, like is.gd and tinyurl.com, which keep the URLs short (but which also have problems of their own in terms of long-term viability/accuracy).

Which is all just to say that RT, @username, and #hashtags are nothing more than metadata about the actual content of the post. True, they are visually noisy to newcomers, but such is the nature of new forms of digital expression.

But obviously, even with this explanatory list in mind, there is no shortage of noise on the Tweetstream. My own experience reading it, especially on a stream that was moving as fast as #CNNfail was Sunday night, is that the colorization brought on by hyperlinked hashtags and usernames helps me visually sort through to what I want to read: too many usernames, as in a string of retweets, or too many hashtags lumped at the beginning or end of a tweet, and I’m not likely to read carefully. I’ll look for the black text of tweets that will likely be more informational and, with luck, something that hasn’t been tweeted already.

Write the first comment.