The Great Folder vs. Search Debate

Back in the summer, there was a fun running debate that started about folders, but ended up being a discussion that was really about the “pre-organization” of taxonomy against the value of metadata-driven search.

Chris Riley started it with his post entitled Folders are the New F-Word. It’s a seven-minute video (in which he appears to bleep the word “folder” to make a point), and it’s worth watching. He sums the argument up in the intro:

Folders only give you the perception of organization. Real organization happens with meta-data. […] If you truly want to organize your content, meta-data allows you to slice and dice information in any way you need at any given time.

In the video he gets even more serious:

The solution to [bleep] …well, in my perfect world, you disable them.

(Incidentally, Google didn’t help this back when they released Gmail. They even dissed folders directly in the Gmail philosophy).

Search, don’t sort

With that, Google instantly started weaning a large segment of email users off folders and all other pre-organization schemes. I personally haven’t labeled anything in Gmail in years. (I have 27,000-some-odd emails just floating free in my inbox.)

Laurence Hart then jumped in with The Truth About Folders where he made an important point:

People are used to folders. They used them in DOS and then Windows. Before “My Documents”, people were creating their own mechanisms to save and track their documents.

Folders – or, more generally, parent-child hierarchies of data – are natural. People understand and accept them. We’re used to deductive searching – getting more and more specific while we’re on an information scent for something. Folders support this.

(Laurence also pointed out that information professionals have been arguing this point for years. He pointed to this post on his own site five years back focusing on the same thing. It’s a perennial topic.)

This brings us to Sean Hederman and his post The Truth About Folders: A Rebuttal. Sean is a vendor, so he has a bias (which he admits), but he proceeds to respond to every one of Laurence’s points about folder-based organization structures.

Specifically, he has a clear point about how people are used to search, just as well as they’re used to folders.

By that logic we shouldn’t use search engines to access the Web, we should organise it into folders instead, just like Yahoo used to do. You see, people are also used to search engines, they use them every day of the week. In fact, more and more people are using search based idioms rather than folder based idioms to access their systems. Look at the search box built into the Windows 7 menu, and every Windows Explorer window.

Here’s the thing –

When you think about it, the folder structure (or…geography) is one form of metadata. It’s really a browsable form – it’s metadata that leads to other metadata. (Oh…dear God…it’s a taxonomy…)

There’s so, so many places I can go with this, but at the end of the day, why can’t a folder structure just be considered another form of metadata? Putting a file “in” a folder – how is that different from applying a tag? If a file can be in more than one folder via symlink or something and your tag structure is hierarchical, then this metaphor is even more applicable.

In fact, give me a hierarchical set of tags with a decent API and I’ll build an interface around it that looks all the world like a folder structure. I can have your users browsing through “tags” that may as well be “folders,” for all they know. The only difference would be a native ability for an item to appear in more than one place easily. Tags not hierarchical? That’s fine – how many folder organization systems are only one level deep?

In the end, why can’t both systems live together in harmony? Complaining about people sorting things into folders but pretending that they’ll be diligent enough to valiantly apply metadata is probably a little optimistic, isn’t it?