How will social media change the study of history?

[Beet]

The contemporary field of history, as opposed to say, archaeology, is inherently tied to the written word, which has provided its source data. Contemporary social media is a revolution in the written word and an increase in quantity of source data by orders of magnitude, on par with the spread of the printing press. Thus, one would expect that it will also revolutionize the study of history.

Not only is the quantity of source data exploded, but the nature is revolutionized. For example, for a historian studying a 19th century debate over the abolition of slavery, she would have to track exchanges of lengthy pamphlets or letters, but would not have access to informal, spoken-word day-to-day conversations. Yet on social media, such conversations occur on a regular basis, for example, in Twitter exchanges. It is also much easier to see which writings are popular and to put a quantifiable measure on it, as well as numerous other metrics such as the speed of conversations, and so on.

For example, a future historian studying the spread or social characteristics of supporters of Tea Party or OWS movements would have access to troves of much more complete data, than for example, someone studying the characteristics of Bucktail supporters or locofocos in antebellum America; or even much more recent movements such as the U.S. environmental movement in the 1960s and 1970s.

Edit: It should be of note as well, that not only the means of study but the object of study has changed. Historians of the future will not only have greater metadata about conversations, but the conversations themselves will be taking place over electronics rather than through voice.

[Filuwaúrdjan]

The contemporary field of history, as opposed to say, archaeology, is inherently tied to the written word, which has provided its source data.

Broadly speaking correct, though there has been an upsurge in interest in oral history in recent decades (although much of what has been produced this way is of questionable value) and there's also now film and so on.

You must be logged in to read this quote.

Well it would mean new sources to use of course (would they be used though? Even most of the reams of paper documents produced during the bureaucracy booms in the 19th and 20th centuries have never been read after being deposited in an archive...), but there's no reason to expect the actual methods of historical enquiry to alter. Most of it is really just applied logic (or at least it ought to be).

Unfortunately there are some fairly large problems with electronic documents and communications. The great advantage of paper (and vellum before it: actually it is even truer of vellum) is that it lasts for centuries. The files used for storing electronic data don't and even the internet turns out to be a bit sh!t at preserving things. What may look like a remarkable opportunity could easily turn out to be a massive nightmare.

[Beet]

Well it would mean new sources to use of course (would they be used though? Even most of the reams of paper documents produced during the bureaucracy booms in the 19th and 20th centuries have never been read after being deposited in an archive...), but there's no reason to expect the actual methods of historical enquiry to alter. Most of it is really just applied logic (or at least it ought to be).

Well, things like textual analysis can be performed on electronic archives but not paper ones, for example. Paper archives also don't have things like metadata. Previously, if I wanted to find the frequency of mentions of a particular politician by date in all publications over a 20-year period, I would have to either manually count or OCR every document and search, then categorize the date of each publication. But that still wouldn't allow me to see which references to that politician appeared in reference to, or because of, another reference.

With electronic sources in structured form, I can simply write a script to gather metadata such as date, time, document type, location, author, etc. as well as trace by sources through hyperlinks, previous conversations by the same profile, retweets, etc. This allows me to more easily make casual relations such as, "This cluster of activist centered around X, Y, and Z organization started talking about Occupying Wall Street before anyone else, and they influenced A, B, and C," and do so in a quantifiable manner.

You must be logged in to read this quote.

That may be true of the unstructured web, but I guarantee that companies such as Facebook, Twitter, Tumblr, Google, etc. have stored their data somewhere from almost the beginning of their operation, even if you have "deleted" it. Of course, it's a real question of how much of this data is accessible by historians or other academics, or whether they'd immediately know what to do with it; but I have little doubt that the data exists, and is accessible by someone, and it's only a matter of time before it's opened up, piece by piece, and only a matter of time after that before every bit of intelligence is extracted out of it, by people who do know what they are doing. I wouldn't be surprised if a script 80 or 100 years from now were to be analyzing this very post I am typing, obtained from Google's (permanent) cache.