04.30.07
Scribd: Reading Between the Lines
Scribd is a website with an interesting marketing angle: they want to be the YouTube of documents. For long time readers of mutednoise that statement is a bit of a puzzler – after all, the web itself was first conceived as a set of linked documents. A quick scan through the comments on a recent Om Malik inquiry about the company’s ‘hotness’ turned up a number of people who are also scratching their heads.
Many are quick to point out that the comparisons to YouTube are especially prudent given the number of copyright violations – sheet music for the Beatles, Harry Potter, etc. But is being able to see an original copy of the ‘Dawn of the Dead’ script really what’s fueling the growth? I think there’s more at play here.
I worked for several years writing software for a document control company. What I gradually learned is that there is a large difference between a document and data. A document represents a fixed set of data at a particular moment in time. Having these fixed ’snapshots’ can be highly valuable: while data is ongoing and without a fixed state a document provides the moors so that comparison and analysis can happen.
On the web data is able to exist naturally. It is constantly being scraped, mutated, appended, rearranged, split up, and recombined. By freezing that data into a state (in this case a PDF document) an authoritative (but not necessarily correct) version of the data can be presented across multiple forums in the exact same identical way. In this way, too, Scribd is like YouTube: they make embedding an entire document on another website as easy as cutting an pasting code. The ’snapshot’ of data maintains its referential integrity.
Is that a subtle benefit? Hell yes. However, its one that seems to be getting lost in the poo-pooing of the service. Scribd is providing some real value. It’ll be worth watching to see just how long it takes the doubters to realize it.