Embracing the Tiger
When both Ming and Jon
Udell point to the same blog on
the same day, ya gotta pay attention. Actually, they both point to Leslie Michael
Orchard's riff on Charles Miller's complaint that
he's as tired as George
Carlin, having to catalogue all his "stuff" on
his computer:
"I no longer want to know where my files are stored. I no longer
care. I have hordes of directories on my various computers called stuff,
downloads and documents, and the effort that it would take to organise
them into a proper hierarchy is just not worth it. The hierarchical filesystem
is a really wonderful thing for programmers and websites, but it just doesn't
cut it for personal use. I no longer care where my files are stored."
Adding to the outcry, my friend Tom Raddemann pointed out today, "With
GigaHertz CPU's, I almost hear the processor laughing at me as I struggle
to do what
it can do better."
Who could argue with that? It's crazy to need to drill down into an arbitrary
structure, either to save it originally or to find it later. Charles wants
his OS to save files by asking for a simple string to remember it by, for example, "Foocom
project plan". But I'm sure that the tiny hint we're willing to provide
at the Save moment is not what we really want.
Ming says,
I think that's
what I want too (he's thought about this before). The thing
is that the world we live in is no longer hierarchical. Any piece of
information
fits into
a bunch
of different structures in different ways, depending on what I'm trying
to do. If I go and drop the item in a file in a folder in a filing cabinet,
in the place that seems logical at the time, chances are I won't find it
next time I'm looking for it. So, yes, maybe there is no good way of easily
storing it multi-dimensionally. Maybe the best is to store some concise
information about the information (which is called metadata), such as date,
person, relations to projects, interests, etc. and then leave it up to
an efficient search engine to find things by those keys later on.
So
there's an argument to be made for structure. Of course, as we start to
add a little structure, being human, we quickly make it hierarchical and
start down that
slippery slope of hierarchical data totalitarianism we all resent so much.
(Shouldn't we have people who take care of these things for us?) Where's
the intersection of good sense, ease of use and a satisfying way to really be
on top of our stuff. I suggest those are not exclusive. Jon thinks it needs
to be in the operating system:
Adding more Ptolemaic circles like that won't really help. Leslie's
right: helper apps aren't the answer. The OS needs to be deeply aware of
various namespaces -- the Mac's systemwide Address Book is a great step
in that direction -- and then surface them into a common completion UI.
Maybe the answer is to assign the tags when you're working with the content,
not in that moment when you know you don't want to lose whatever you're working
on.
Several years ago, I developed a system called MindShare to handle this problem
for workgroups and their stuff. The challenge then and now is to have a bulletproof
way to describe whatever might need to be found later. At that time, we
didn't have the benefit of XML, which is about to become the storage system
for all our stuff.
Steal This Idea
But we did find a bulletproof topology for assigning metadata to business
content strings. MindShare was based on the idea that, if something is worth
keeping at all, it should
be
available
fortuitously
when
we're looking
for
things like
it but
may not even
remember this item specifically. The universal
topology for everything we need to keep track of is the IPIA coordinate
system. IPIA says that the meaningful text strings in any file, correspondence,
meeting,
call, etc. can be classified unequivocally as an Issue, Promise,
Idea
or Appointment. You'll never mistake an issue string
from a promise received string.
And obviously our world is defined by promises payable and promises receivable.
Making them explicit is a Good Thing.
Example You get an email or open a web page or write a letter.
A series of widgets surround the message:
Naturally, the system knows who the email is from, when it arrived, etc. and,
as Charles suggests, provides those metadata tags as it can. Since the system
already knows all your contacts and appointments, new ones can be added by
clicking, typing or dragging them.
If something worth noticing is mentioned, it is always an Issue, Idea, a Promise
Made, a Promise Received or an appointment, a special kind of mutual promise.
Just highlight the text string and click the options. If your file or content
doesn't deserve all this scrutiny, then don't do it. But, whatever you highlight,
drag, click or, maybe, type, you can be sure your Model 2004 4GH XML-o-matic
CPU will not require you to know where the hell your stuff is.
Marc Canter has been urging us to embrace MOM—a
Media Object Model, that might look like this:
Someone, probably us, will add the text namespace options to Xpertweb
transaction forms. But we'll never do it at the system level. Since the IPIA
namespace is as
old as
the Agora. I hope someone applies it.
11:48:11 PM 
|
