louisrosenfeld.com logotype

Home > Bloug Archive

Jan 03, 2002: Disposable Information Architecture

A couple of Blougs ago, I mentioned the term "disposable information architecture". I think it sounds kind of cool; who knows, maybe it'll catch on and become a meme?

Hah! Just did!

Anyway, an example of disposable IA would probably be appreciated. Let's say you're designing the information architecture for a blog. You plan on implementing a search system for your blog entries at some point, but don't have the time just yet. So you decide to set up a browsable index page to serve as the site's archive.

Because you don't have that many blog entries yet, it doesn't really make much sense to create something elaborate. Probably not many of your blog's readers will bother using the archive anyway. Or maybe you don't have the time or money to throw at it. After all, it's just a blog!

So you create a fairly simple-minded archive page that you know won't scale forever. It's the best you can do for now, and you know you'll improve on it later.

Voila: disposable information architecture!

Perhaps a better term would be "transitional information architecture," but you get the point: sometimes just good enough will do. If you're in a hurry and don't have much money to spend, it's OK to take a middling approach. Just be sure that it's a stop along the way to something better, not a permanent part of your site's architecture.

email this entry

Comment: vanderwal (Jan 4, 2002)

I like the "transitional information architecture" idea. I have been witnessing more than just personal sites go through this morphing process. Some of these transitional elements have shown up on sites that are incorporating applications in phases. Many of the architectural features are there in the plans, but the buttons and or links have not been implemented. This does seem to be an alternative to links to "coming soon pages" (that may not arrive).

On the personal site front some of us have incorporated category/classification links to related information to suffice for search at this point. The classification of information can also be repurposed to augment a search which will be implemented on that rainy weekend off in the future.

Comment: Dan Kapusta (Jan 5, 2002)

Wow Lou, the timing of this post is perfect. You just described the current state of my blog (http://inmyexperience.com). I'm in the process of transitioning from a global archive in favor of time and category based browsing. (And text searching will come later :)

On the concept of disposable/transitional IA, you have to expect to deal with this in two ways. One for content driven sites and the other for web apps. And we can look at the semantic diff in disposable versus transitional to help out.

Where a content driven site will have salient elements that need to be dealt with in real time, Web apps tend to get re-architected when changes become unavoidable. You can 'just make a change' to your blog (transitional) but when you release new features in a web app, it usually has larger implications in terms of interaction design and then you tend to re-architect (disposable).

Plan for the future so you can transition instead of dispose.

Comment: mini-d (Jan 5, 2002)

nice one Lou, I believe in this... "transition" word is perfect to define information architecture... in fact i'm avoiding a complex archive section now because it has only 4 months long... 84 posts covering 10 categories... but using just 4 categories... i think it's not ready for a massive and complex archive structure... :) anyways this post made me feel identified... =)

Comment: Eric Scheid (Jan 5, 2002)

I've been thinking on the same topic for some time now - personally I was inspired in part by Peter Morville's article [http://semanticstudios.com/publications/semantics/speed.html] *The Speed of Information Architecture*, and also partly by endless experiences of having to patch up various legacy Information Architectures so they work as needed.

I jotted some notes on this concept back in November.. [http://IAwiki.net/DesignForDestruction] ... which I've updated with some design principles which may make this strategy more successful.

Comment: Lou (Jan 5, 2002)

Great comments folks! Eric, I really like what you put up on the IAwiki. There is one point I'd take exception with: "don't worry about scaling".

I think you should *always* worry about scaling. Even if you have no idea how much content you'll have to work with initially, it's important to keep an eye on the rate of content growth (in fact, if you're really not sure, that's even more reason to do it).

More importantly, we often do have at least a hunch about scalar issues. That hunch might help us design a solution that is more transitional than disposable. So we might know, for example, that a classification scheme won't scale for our blogs in the long run; so why bother developing one now? Etc.

Comment: Eric Scheid (Jan 6, 2002)

Semantics. When I say "worry" I mean "endlessly imagining a bunch of negativity over which you have no control".

I do agree though that we should always *acknowledge* that scaling may be an issue, and to keep an eye on it. Just don't *worry* about it.

Comment: Lou (Jan 6, 2002)

Sort of "don't worry, be conscious"?

Comment: Dale Mead (Jan 7, 2002)

I think that IA's tend to create disposable information architectures almost any time we take over a website. Real IA takes a lot of time and work, but that shouldn't mean that the site users have to suffer while I do that work. Almost invariably, I can make major improvements right away just because of my experience. That gives me the time,and more importantly, the credibility, to do the proper IA work.

The problem with disposable IA, as with disposable plastic plates, is that if the quality is too good, people might decide not to throw it away.

Comment: Edward Vielmetti (Jan 9, 2002)

In the "software tools" approach to building complex systems out of smaller parts, there's a tradition of building components that do one well-defined task and then chaining these components together to build the full solution. The first version of the resulting system has been built and tested interactively and solves 80% of the problem in 20% of the time, and then you figure out whether to live with it as-is or refine it or throw it away and start over.

Measure never, cut twice,


Comment: Samantha Bailey (Jan 15, 2002)

I think there may be more room for the idea of transitional/disposable IA in some mediums than in others--in a blog, it may make a lot of sense, especially as that's an agile (and somewhat transient) medium. For corporate websites I think that Dale has nailed the issue perfectly in his comment about the dangers of "disposable" becoming permanent or semi-permanent by being "good enough." The cycle to get projects from approval to implementation is so long (loooong) at my company, that almost any undertaking will be in place for 6-12-18 months before even pretty limited changes/fixes will "replace" it. I think we're not agile enough for disposable IA... Although this is also making me think about ways to speed up the development cycle--maybe disposable IA would be the next logical step after rapid prototyping? hmmm.

Comment: Andrew Hinton (Jan 18, 2002)

I really dig how two actual architects deal(t) with these issues. Bucky Fuller and Rem Koolhaas... Fuller was big on re-usable, inter-universal parts that could be reconfigured on the fly (extreme object-oriented architecture, I guess). Those Lego-like houses are amazing.
And Koolhaas, even though he's a little too uber-hip these days, has championed the idea of non-designed spaces, where human systems just make do however they can. The Internet, as it moves into being more of a global medium (rather than a medium for the comfortable classes), will see a shift toward fewer cathedrals and more bazaars. (That's my fortune-telling for the day.)
[See this summary (PDF) for a kind of overview of this perspective: http://www.m20m.com/workshop2001/papers/shannon.pdf]

Comment: Andrew Hinton (Jan 18, 2002)

oops. the bracket screwed that link up. here it is again:

Add a Comment:



URL (optional, but must include http://)

Required: Name, email, and comment.
Want to mention a linked URL? Include http:// before the address.
Want to include bold or italics? Sorry; just use *asterisks* instead.

DAYENU ); } else { // so comments are closed on this entry... print(<<< I_SAID_DAYENU
Comments are now closed for this entry.

Comment spam has forced me to close comment functionality for older entries. However, if you have something vital to add concerning this entry (or its associated comments), please email your sage insights to me (lou [at] louisrosenfeld dot com). I'll make sure your comments are added to the conversation. Sorry for the inconvenience.