Everything Buckets

10th of February, 2009

As a long-time, dedicated YoJimbo user I feel compelled to reply to Alex Payne’s case against Everything Buckets:

If you search for “productivity” or “organization” software for the Mac, you’ll find variations on a particular type of application. These applications claim to be “your outboard brain” or “your digital filing cabinet” or similar. They go by many names: Yojimbo, Together, ShoveBox, Evernote, DEVONthink. There may be differences in their implementation and appearance, but these applications are all of the same sinister ilk. They are Everything Buckets.

I think Alex has the concept wrong. YoJimbo is not a productivity tool but a remembering tool. Indirectly it can lead to productivity because you’re not searching through bunch of different applications or mediums looking for information. Primarily it’s a tool to alleviate the burden on your brain to remember thousands of tiny pieces of information.

It could be considered a productivity tool when compared to Alex’s alternative: meticulously organising every piece of information into an appropriate structure.

I share Alex’s concern about proprietary data storage and non-interopeable formats like RTF. It bothers me that YoJimbo’s only export format is RTF but it does offer mass export functionality and while full of junk, RTF is readable as plain text.

They’re virtual scrapbooks, applying a lightweight organization system to (often) unrelated data of varying types.

This is the issue, tiny pieces of unrelated information with varying levels of importance — from insignificant to vital — all in different formats, all requiring unique structures. Do I create an organisation structure to store my tax file number, of which there is only one? Do I create a structure for the one line shell script I use to shh tunnel through a proxy server? Or do I throw them into an everything bucket, find it with a quick search when or if I ever need it? It’s information unrelated to anything else, with it’s own unique structure, it’s taken seconds to record and takes seconds to retrieve. What’s the problem? It’s not machine readable? There isn’t a situation that requires it.

They’re poor filesystems, poor text editors, poor databases, poor to-do lists, poor calendars, poor address books, poor bookmark managers, and poor password managers.

I agree. That’s why I use a TextMate for text editing, Things for to-do lists, iCal for calendars, Address Book for addresses, etc. Everything buckets are for those pieces of information that don’t fit within these obvious structures.

One of my Rules For Computing Happiness is: “do not use software that does many things poorly.” Everything Buckets violate this rule up, down, and sideways.

Everything Buckets don’t break this rule, they do one thing well, that is collect unstructured, unrelated, small pieces of information so your brain doesn’t have to.

So while you, the user, are presented with the illusion of “Googling” for that phone number you threw into your Everything Bucket, your computer is constantly hauling ass to make up for the fact that you couldn’t be bothered to put the phone number in your virtual address book.

I store phone numbers in Address Book, a beautiful, structured, single purpose application. Just the kind Alex champions. Yet, without fail I will use search to locate information within Address Book. Computers are great at searching, you could almost argue it’s what they’re best at. The telephone book is extremely structured but how slow and painful is locating a number within it?

Beside that, why do I give a fuck how much work my computer has to do to make my life easier? Well:

“What do I care how much work my computer does?” you say, until the next time the Spotlight indexer kicks in and your Mac grinds to a halt.

Spotlight indexing on my 8-year-old 1.5ghz PowerBook G4 is fast. On my relatively new 2.4ghz Core2Duo iMac it’s literally invisible, I’ve never seen or noticed Spotlight indexing. Yet almost every piece of information on my computer is seconds away with a Spotlight search.

Your computer wants your data to be structured. Throw it a bone.

We have a massive surplus of computing power, the trick is writing software to take advantage of it. Spotlight’s ability to take messy, unstructured information and make it instantly findable sounds to me like the perfect utilisation of fast, modern computers.