Thoughts, links, and essays from Peter Merholz
petermescellany   petermemes


Archives before June 13, 2001

RSS Feed

Adaptive Path (my company!)

About peterme

Most of the Time
Oakland, CA

American history around the time of the Revolution, figuring out how to marry top-down task-based information architecture processes with bottom-up document-based ones, finding a good dentist in San Francisco Oakland
Designing the user experience (interaction design, information architecture, user research, etc.), cognitive science, ice cream, films and film theory, girls, commuter bicycling, coffee, travel, theoretical physics for laypeople, single malt scotch, fresh salmon nigiri, hanging out, comics formalism, applied complexity theory, Krispy Kreme donuts.

Click to see where I wander.

Wish list
Show me you love me by
buying me things.

Track updates of this page with Spyonit. Clickee here.

[Editor's note: began as a site of self-published essays, a la Stating The Obvious. This evolved (or devolved) towards link lists and shorter thoughtpieces. These essays are getting a tad old, but have some good ideas.]
Reader Favorites
Interface Design Recommended Reading List
Whose "My" Is It Anyway?
Frames: Information Vs. Application

Interface Design
Web Development
Movie Reviews

Thoughts on the definition and community of "information architecture." Posted on 11/16/2001.

Recently, the SIGIA-L mailing list, the popular stomping ground for information architects and folks practicing nearby, has exploded with discussion around two topics--What Is Information Architecture? and Where Does It Belong, Professionally?

These are questions I've spent a while avoiding, because they tend to go in circles, and I've opted to be more involved in thinking about methods, processes, and principles for doing this kind of work, not labels and affiliations.

For the past couple of months, though, I've been thinking about this more deeply, in part because terminology is starting to gel, in part because the conversation continues in a number of different groups, and, finally, because I'm thinking that it is important to start settling these discussions, both because it's important in our communication to folks who are NOT in this discipline, and because it would be a good thing to just move on.

Answering the first question helps address the second. "Defining The Damn Thing has taken up a lot of energy in the community. Such discussions often point to Richard Saul Wurman's coining of the title "information architect," in 1976 which roots IA in the field of information design and "design for understanding." Wurman's term, though, didn't catch on for twenty years, and I think that's an important fact--to me, it suggests that it was inappropriate for what he was labelling. So that while he might have been first, he wasn't necessarily right. I also believe that the main reason it didn't catch on was that there was already a perfectly good label--"information design." Why confuse matters?

Another inevitable thread in the discussion involves information architecture as any aspect of web design that isn't graphic design. This ends up including gathering requirements, user research, writing feature requirements, organizing content and functions, drawing up site architecture diagrams, creating prototypes and wireframes, and writing functional specifications. The problem with this definition is that folks who call themselves "interaction designers" do much of this, and have long before "information architect" became a popular job title. Why not leave the interaction design alone?

(For some thoughts on the "design for understanding" and "interaction design" tributaries in the IA river, I suggest reading Lillian Svec's "Information Architecture Practice at Sapient"(PDF) from the 2000 AIGA Advance for Design Summit.)

(For further thoughts on defining Interaction Design, I suggest reading this presentation given by Robert Reimann and Jodi Forlizzi and the 2001 Advance Summit.)

Yet another tangent often seeks to define information architecture in the interests of that particular information architect. This leads to the fallacy that information architecture is what information architects do. Self-appointed "information architects" do all manner of things that isn't information architecture, and folks with other titles (notably "product manager," "web designer," "web master") do information architecture.

Another fallacy arises when it information architecture is defined as what an information architect *wants* to do. Many IAs express frustration at being strait-jacketed into the online world, wanting their practice to extend to other media, or into organizational consulting, or anthropology, or whathaveyou. This is all well and good--IAs, shouldn't confine themselves... but that doesn't mean that all such activities are information architecture.

I argue for a more focused definition of information architecture, something of a cross between Argus' "Information architecture involves the design of organization, labeling, navigation, and searching systems to help people find and manage information more successfully," and Jesse James Garrett's "Stuctural design of the information space to facilitate intuitive access to content." The Argus definition comes from a library background, and Jesse's from journalism. Both, though, are focused on finding, managing, retrieving, accessing, and understanding content and/or information. And, I think, both implicitly acknowledge the Web/networked/online nature of the practice.

The reason being, "information architecture" was meaningless before the Web. The fact that RSW's definition didn't take hold until 1995 shows that. I think not acknowledging that dilutes any worthwhile definition of information architecture. I do think information architecture is a great label for what is a genuinely new problem, of designing and structuring online, networked, information spaces. I don't see any reason for us for information architecture to include information and interaction design--both already exist as happy distinct entities. Obviously, IA must integrate with these, and other disciplines, and any worthwhile study of information architecture will incorporate that. But, and let's be clear, they are *not* information architecture.

Now, I do think having an umbrella discipline and term is important, because it's clear to everyone that information architecture, as I'm defining it here, is a facet of a much larger activity of designing a complex, information-rich tool/space/whathaveyou for others' use. And I think we've already got that umbrella discipline -- "user experience." A term that's been around since at least 1991, and defined to address the design of all elements that touch a user of a designed system.

I think it's foolhardy to have information architecture == user experience (as many have argued), because then it glosses over the very real problem of designing information spaces to facilitate access and understanding, a problem which clearly deserves its own label.

All that said, it leads to wondering, "Where does Information Architecture Belong, Professionally?" And given the line of reasoning presented here, I would argue, very simply, ASIS&T, since it's devoted to library and information science and thus clearly bound with the issues of design and structure of information spaces. Interaction designers already have an amazingly professional society, ACM's SIGCHI. And those who want to look at the bigger, umbrella picture can find a home within the AIGA's Experience Design community. (I, like many others, have issues with ED being supported by the AIGA, an organization I feel otherwise no commonality with. But I've been involved with the ED's efforts for a few years, and they're approaching matters very intelligently.)

I'm done.

26 comments so far. Add a comment.

Previous entry: "Vernacular Thesauri."
Next entry: "OOH!"


So does anybody happen to know of a good article/book/site that explains (at least pretty well) definitions of all these related fields and how they compare and contrast?

For example: information architecture, information design, user experience, usability, interface design, interaction design, etc.

I'm just so confused on how these all relate (and even if they do)?

Originally, I thought I was interested in information architecture as the field I wanted to puruse when graduating. But now maybe it's more interaction design or interface design or... maybe I should just look for a job as a web designer or fry cook.
Posted by Scott Arkin @ 11/16/2001 11:55 AM PST [link to this comment]

The Web itself is so young that roles, definitions, and everything else continue to evolve. We haven't quite settled down at one place, which is why there's so much confusion.

I started out as an IA guy and gradually spread my wings. Personally, I feel that the reason there's so much disagreement is that IAs *want* to do more, but still want to remain IAs. So instead, they choose to expand the role's borders instead. I mean... c'mon, Information Architects don't design interfaces, for example.
Posted by MadMan @ 11/16/2001 12:08 PM PST [link to this comment]

Good book in a few months? - Jesse James Garrett's "The Elements of User Experience", currently underway for New Riders, is exactly what you're looking for (if I correctly understand what it's about).

For now check out his PDF that digs further into it...The Elements of User Experience
Posted by Jess @ 11/16/2001 03:47 PM PST [link to this comment]

Over three years since I read IAftWWW and I still hate the term. It still sounds as pretentious and silly as "design surgeon" or "user experience samurai".

I am well aware that this is my own personal idiosyncratic preference, but I really did think (and hope) the term would die with the dot.coms.

On the other hand, I have a lot of respect for the work done by some people who call themselves information architects. I just wonder if people will still be using the term in 10 years.
Posted by stewart @ 11/16/2001 06:01 PM PST [link to this comment]

The last AIGA Experience Design summit attempted to define the different roles involved in Experience Design: Information Architecture, Interaction Design, etc. They did well insome respects and not so well in others.

The thing I appreciate (having been an AIGA member for 10 year and an ASIST for only 1) is that the AIGA is working very hard to shift themselves and become a relevant organization to the communities they serve. Unfortunatley having been associated with Graphic design for so long - it will take awhile to be seen as relevant to the IA world. Personally, I feel they are more relevant to me than ASIST because they have been more inclusive and aware of the fact that this new community is diverse and comes from many backgrounds. I think the upcoming forum with SIGCHI at CHI in April is just the first visible step to reaching out.

The AIGA originally started as an organization for more Printing oriented folks and not designers. They have evolved before and will continue to evolve as it is appropriate to the community.
Posted by erin @ 11/17/2001 07:31 AM PST [link to this comment]

been there, and done that. yet here we are again.

btw, Jess, Jesse's book (and mine, for that matter) have both just begun being written, so look for them in 2003, I suspect. or maybe in time for next xmas. Until then, the polar bear is still the huggly-wuggly book for IA's. And if you reread it (as I am happily getting to do, for research) you'll see an IA that is narrowly defined as Peter demands, yet also requires a wider set of skills to be effective in that craft.

Perhaps you can have your cake and eat it too.
Posted by c @ 11/17/2001 10:12 AM PST [link to this comment]

Scott, it you're anywhere near Palo Alto, CA, you should consider taking Designing Experience from Stanford University. It starts next semester (Jan. 2002). Maybe I'll see you there!

By the way, Peter: that Reimann/Forlizzi presentation - NICE! Thanks for the link.
Posted by Brad Lauster @ 11/17/2001 11:14 AM PST [link to this comment]

To some degree the area that the IA role/skills, in the narrow sense, is being applied to is changing. The IA role of researching the understanding of information and the user's perceptions of that information so to build and set structures was initially set for Web sites. The uses of this information now are commonly streatching beyond Web sites to mobile devices, which require a narrowing of terms and explanation, and other machine communication (XML/RSS, etc.) for syndication and so on. The change from static Web pages to more broadly defined informaiton applications has a greater need for proper structuring of information for access and use. This information structuring largely falls into the IA's lap.

The term Information Architect is very apropos in this construct. The IA maps out the structure and support for the various uses of information as an architect sets out the plans for constructing buildings for various uses. Having an understanding of use helps mold the information strucutures and how the information can be used in a modular sense.
Posted by vanderwal @ 11/18/2001 09:25 AM PST [link to this comment]

My book is still slated for mid-to-late 2002. And yes, it will address all these questions (and many more!).
Posted by jjg @ 11/19/2001 03:01 PM PST [link to this comment]

> The term Information Architect is very apropos in this construct. The IA maps out the structure and support for the various uses of information as an architect sets out the plans for constructing buildings for various uses.

See, this is exactly where I disagree. Why not say that "The IA maps out the structure and support for the various uses of information as an ASIC designer sets out the plans for constructing integrated circuits for various uses."? Or "The IA maps out the structure and support for the various uses of information as a furniture designer sets out the plans for constructing chairs for various uses."?

Interpreted very loosely, the analogy makes sense, but very loose analogies aren't very illuminating. And at that level of generality many other analogies make just as much sense (after all, composers also make marks on paper which communicate implicit instructions for how people who get paid less than they do are supposed to carry out their jobs, but we wouldn't look for an analogy there).

Again, perhaps it is only because I have a profound (and perhaps somewhat fetishized) respect for what "real" architects do, but it seems totally audacious to claim that what IAs are just a species of architect who happen to work with "information" instead of wood or concrete, etc.

I was actually coming here to say something about the disanalogy between the kind of work that IAs do and the kind of work that went into the beautiful 1908 building around the corner (the proportions of the window panes in their copper-plated iron frame, the exquisite masonry, the wrought iron balconies, the dentils along the roof overhang, etc.) but I see that some sort of similar conversation was going on between Christina and Matt Jones (can't find the starting point) and Matt's final point bears repeating:

You should only say you've created an information architecture if Vitruvius could recognise your intent 2000 years later...

Hopefully I'm not coming off (merely) as a crank on this topic. I genuinely think the spatial metaphors are badly broken and if we begin our thinking in terms of "structures" which facilitate "navigation" thorugh "information space", we can't help but come up with designs which are saturated with spatial concepts.

But perception and cognition don't go on in a spatial framework (with certain exceptions which aren't trypically relevant to this conversation), and bits of information don't relate to each other spatially (concepts don't exist below or beside or to the west of one another). Call me Whorfian, but how we talk affects what we do. If our talk is wrong, our work will be too.
Posted by stewart @ 11/20/2001 01:06 PM PST [link to this comment]

The term architect is a broad term and system architects and building architects have the similar skills that they apply. Information architects also have similar skills that they apply at the base of what each of these professions do. Choosing terms to describe what we do is important. Most of the world does not know what we do as Information Architects and that is a shame. Either we wrapped ourselves in the wrong term that is based on analogy (as most term adoptions are) or we do have many common traits to architect and we face a marketing game ahead.

Much of the narrow term of IA is based on is information structures based on user's understanding and preconception of terms and ideas that are the basis for a structured site. Spatial relationships are as important to IA as it is to building architects. It is not, as you say building a room to the West, it is building the kitchen in proximity to the dining room. There is a flow through a Website as there is a house or office space that is planned for the users. These are not metaphors nor analogies they are terms of use.

If you look into your communication theory that existed far before the term IA (even RSW's usage in 1976) you will find research on information space. Part of this is an extension of Marshall McLuhan and his "Medium is the Message" thought process. This is where one's message fits in an information space of similar ideas or into what information space a message may fit. This is very close to the work IAs perform trying to place concepts in a site navigation. This is, in a near sense, a spatial relationship to other ideas and concepts.

IAs do help create the spaces with relationships to other spaces. Based on card sorting, metadata, relating information to expected areas in a site, etc. the IA is involved in a similar practice as a building architect, which is setting structure to an element to be built that combines components that are grouped together based on users expectations and understandings.

IAs in the narrow sense would only be drawing floor plans if they were building architects (yes, now THIS is an analogy).

In turning the discussion top to bottom, part of the problem with Stuarts understanding of building architect is that it is only the facade that he associates with the architect moniker. This association is one that many make, but as soon as somebody, like Christina states that RSW's IA book is to be included in a broad understanding of IA, many IAs have a fit, yet they are the same ones that only think of the building architect as the men and women that design the pretty bits of buildings. They have fallen into their own misunderstanding or more appropriately the misapplication of the term architect (sorry Peter this is your site and this is not meant as a poke in your eye, nor anybody else for that matter). IAs and building architects also map out the floor plans, ensure the building meets local code laws. Both architects try to build for the user of a site (not an analogy as it is the same term with similar meaning).

As we try to explain why we are IAs or why a project needs an IA those that are not familiar with the topic are going to turn to their common understanding of architect. Why do we not want to see this when our profession is based on understanding how the USER understands terms?

Personally, I leave the pretty bits to the design side of the firm and I work to set structure and map out the parts of information applications (including Web sites) that I usually end up building. Every building needs a plan as does every information application (Web site) and these are provided by architects.
Posted by vanderwal @ 11/20/2001 09:28 PM PST [link to this comment]

> If you look into your communication theory that existed far before the term IA (even RSW's usage in 1976) you will find research on information space.

If you have any pointers on that assertion, I'd love to see them. This is something I've researched fairly thoroughly, at least in the context of the precursors to the web: search Douglas Englebart's Augmenting Human Intellect: A Conceptual Framework (1962) or Vannevar Bush's As We May Think (1945) you will find no (relevant) occurances of "navigat*", "space", "move" or the other common terms of the spatial metaphors we deal with regularly.

(I haven't done the same for Nelson's stuff, only because I've never found much of the original Xanadu description online, but people tell me that he was concerned with spatial metaphors. I'll conveniently ignore that.)

The earliest occurance I've yet found for "navigation" (in the way we currently understand it) is in a 1985 paper entitled Reading and Writing the Electronic Book by Andries van Dam et al. I'm pretty convinced that, if Netscape chose to call their product something other than Navigator, "navigation" wouldn't've caught on the way it did. (Could be wrong about this though: Tim Berners Lee was using the word "navigation" to describe web usage back in 1990).

> part of the problem with Stuarts understanding of building architect is that it is only the facade that he associates with the architect moniker.

Sorry if I gave that impression. Growing up with architects, builders and developers, I understand very well that the architect's job deals with a lot more than the appearance of the exterior of the building. However, I don't think this helps much in settling the matter: architects are concerned with building materials, construction methods, time to build, cost to build, ("architect", by the way, is from the greek: master-builder) climate, physical connection to the surrounding environment, etc. The analogous (dodgy) concerns in web design —Oracle or SQLServer? DHTML or Flash? Perl or PHP? Budgeting, resourcing, and so on— these are handled by developers, programmers, designers, project managers, etc.

For some reason (again, all of this is my personal idiosyncratic view) "System Architect" and "Database Architect" don't bother me (I think I find the former acceptable partly because of the way it was described in the Mythical Man Month -- that's a much better analogy to an architect.) I don't think "web site architect" would bother me that much either, if it described a person who had powers and responsibilities in the development of a website equivalent to those an architect (typically) has in the development of a building or set of buildings.

(Peter, as an aside, is it possible to increase the width of this textarea? Holy smokes it is hard to type in here.)
Posted by Stewart @ 11/21/2001 02:02 AM PST [link to this comment]

Ah, the architect portion is not the insult as it is the co-mingling of architect and information. I have to agree in some respect in the same way that I agree with Jef Raskin in his "There is No Such Thing as Information Design". I consider my self in whole an information application developer and therefore it may be far more fitting that the portion or the development process labelled information architecture more accurately information application architecture.

I am not overly fond of Web-based labels as many of us developers, designers, and IAs have to consider and include in our work mobile applications/context and machine communication using XML, RSS, etc. The work of just Web development has expanded for many of us and what we are creating is information applications that have the similar and yet distinct architecting, design, and development.

I am having to dig farther into my library to track down the "information space" inception. With regard to spatial concepts and the Internet not only did Tim Berner's Lee us "navigation" early in the Web's development, but William Gibson gave us the term cyberspace around 1984. I am finding the base if "information space" in Osgood's Semantic Space which emanates from the late 1950s. I am going to have to dig further in my library to find the extension to information space, but Osgood's "semantic space" seems to be at the center of the answer.
Posted by vanderwal @ 11/21/2001 04:36 PM PST [link to this comment]

Hypertext in Context, Chapter 4 and Chapter 8 ties information space and spacial relationships in navigation, relationship of ideas,and even to Osgood in navigating the semantic space. This however is 1991, so there is more digging.
Posted by vanderwal @ 11/21/2001 08:04 PM PST [link to this comment]

I think that Peter's narrow definition of IA is a good one. It's bound to be controversial because it doesn't describe what most IAs do all day, which usually breaks down as "pure" IA plus interaction design plus whatever other information design a given project demands. So, many IAs will read his definition and scratch their heads a bit... I think this definition works well, as long as one doesn't think of it as a typical job description, but more as a theoretical definition.
Posted by anne @ 11/22/2001 10:18 AM PST [link to this comment]

McLuhan was writing about space and environments pertaining to electronic communications since at least The Medium is the Massage in 1967, and much earlier perhaps (as early as 1954's "Catholic Humanism and Modern Letters".

From the Massage: "Environments are not passive wrappings, but active processes [...]. But environments are invisible. Their ground-rules, pervasive structure, and overall patterns elude easy perception."

There are important differences, in that McLuhan spoke more of an event-field than a real analog to a physical space. But still, it is perfectly appropriate to talk of actual spaces as event-fields as well...

From "Catholic Humanism": "In ordinary human perception, men [sic] perform the miracle of recreating within themselves - in their interior faculties - the exterior world. [...] The exterior world in every instant of perception is interiorized and recreated in a new manner."

If information architecture is at least partially about creating instances that the intellect can process and interiorize (and so, recognize and navigate), then McLuhan certainly had something to say on the matter.
Posted by Michael Boyle @ 11/22/2001 10:31 AM PST [link to this comment]

Vanderwal: Thanks for the links, particularly to HiC. Lots of reading ahead :)

The "cyberspace" described in Neuromancer is very different from what the term came to mean in the Great Web Explosion. In Neuromancer, people in cyberspace move around in something like (as I remember it being described anyway) a "normal" 3d space. Their position determines their view of bunches of information which exist in the space, but as a web user I never have a sense of position, even though many visual designers spend most of their design time trying to convey one (this is the crux of my beef with the spatial metaphor).

(For further reading, Jamie King wrote a PhD thesis on [pdf] The Cultural Construction of Cyberspace which I have discussed with him, but never read. Shh. It sounds good though.)

What it comes down to for me is this: "real" navigation is a process of continual self-location relative to a system of coordinates. The hard part is locating oneself (hence all the historical hubbub about methods for determining longitude, development of the chronometer, etc). If I know where I am (on the map) and I know where I am going (on the map) then navigation is trivial: the map contains all the instructions.

But when I'm "navigating" a website, I remain in the same place (in my chair or wherever) and all the "you are here" bits of website (useful though they may be) don't make any difference to that because (a) there is no system of co-ordinates (b) I don't (typically) know "where" I am going.

"Navigation", when it comes to the web, is a process of determining what things to click on in order to cause X to be displayed on the screen, where X is whatever I am looking for. This is a lot more like (Spool and others) way-finding by smell than it is turning left or right and looking for 9th St. or travelling 162 miles NNE.

The relationship between objects on the web just isn't spatial. And there are lots of important relationship-types which aren't spatial ("father of", "more expensive than", etc.) More fruitful innovations will come with better metaphors. (Hopefully.)

Michael: the best antidote to McLuhan on perception is Stephen Toulmin but anyone from Wittgenstein to Dennett will do (I put a transcript of one of Toulmin's best and least published lectures — The Inwardness of Mental Life — online last year — wonderful and quick read).

I just don't believe in "interior faculties" or The Cartesian Theatre of the "Ideas" of the British Empiricists. There are (1) people (2) in the word. That's all. The idea that we need some (3) impenetrable subjective interior realm in order to explain perception, cognition — or even representation — is bogus. But that's a whole other argument ...
Posted by stewart @ 11/22/2001 05:07 PM PST [link to this comment]

I agree, Stewart, that an antidote to McLuhan on such matters is a Good Thing. For the record, I was simply trying to make the point that the idea of information as space goes back longer than some had written earlier in this thread.

I studied with a McLuhan scholar for years and although I think McL did have some insight, his constructions about perception as reconstructions of space seem very heavy and wholly unnecessary to me. We argued about it for years.

In any case, back to the central thread, I think that Information Architecture, as a profession and a term, is belittled by (oddly enough) concentrating too much on information.

It's clear to me that the important job is not about information itself, but information in context. And inseparable from context - and many different contexts, some under control and others not.

And when dealing with context, I agree with Stewart as well that architecture (at least I think I do), as a model, is too limiting in that it privileges the increased granularization of information, and doesn't address relationships between people and information explicitly enough.

This is important beyond job descriptions and such as well - I posit that the very blinders that were put on by early web/internet-era theorists - their too-dogged focus on information itself - impoverished the whole discourse, perhaps irreparably.

Anyhow. That too is a whole other argument.
Posted by Michael @ 11/22/2001 07:32 PM PST [link to this comment]

There seems to be two uses of the term architecture. One in the system architecture frame of reference, which would be the development of structural elements to support a given use. The other term use is a building architect which can extend beyond the structural elements to the design and materials used elements. The use of architect in a narrow context in the IA definition, which is based on a role in the broader User Experience development model. The broader building architect term use can easily encompass the full User Experience development model, but that still leaves us with the nasty construct of Information Architect, with could be rightfully construed by some to be a difficult proposition.

At the heart of what any architect does is first look at the relationship between structure and the users so to determine what the users want from the tool/space. Use and user seem to be somewhat inherently included in the term architect.

Michael I am curious about your "too dogged focus on information", where else should there be focus. It seems that you set up the point by stating there needs to be context, which is dead on. Having information without understanding context is dangerous and most developers and designers tend to understand this.

Michael I also agree that McLuhan does have insight, but he did not bring to use the term "information space", which is really dogging me. Tim Berners-Lee used the term to describe the inception of the World Wide Web, which may be why there is a cognitive relationship between the Web and spatial relationships. One of the elements that the Web broke was the linear relationship to information flow as hyperlinks allow the user to pursue tangential thoughts easily, which leads to a concept of an information space that houses related ideas. The idea is also analogous to footnotes that allow the reader (should they have all the related materials) to jump from one linear presentation of information and jump to another related source of information.

I am still looking for the use of "information space" prior to the Web and its use outside of technology. It may be that the term was used to explain Osgood's semantic space (a precursor the McLuhan and has been attributed as a foundation for some of McLuhans thoughts), as the usage I remember was quite similar. The term as I remember the use was applied in correlation to advertising, public relations, and marketing practices to explain the area of discourse for related ideas. If one wanted to communicate to a business audience the message would be targeted to the business information space, which was portrayed across media. The business information space would include CNBC (television), Fortune (monthly periodical), Wall Street Journal (daily newspaper), Marketplace (radio), etc. Now that an information space was chosen the medium could be chosen to best target. This is a simple use of information space. I still have not found it in written form.

Andrew Trelor posits a slightly different understanding of Information Spaces. This information space term usage... "Such a space is the location where the human mind interacts with information or communicates it to another. Location should be understood fairly broadly in this context, and not necessarily seen as tied to a particular physical place. Information spaces facilitate the storage and retrieval of data and information, processing of data into information, communication of information, navigation through structured information and the linking of different pieces of information. In a sense, such spaces are where information `exists'."

Stewart, I am in agreement with much of you discussion of the downfall of spatial metaphor for the Web as you have framed it. I disagree somewhat in that the Web does present spatial-like relationships. As we enter a Web site all we can see is what is right before us or presented to us. We make choices based on the clues we are presented with in search of information. The task seems to be very navigational and the problems you mention could be based on poor mapping of how to get from place to place. Our navigation is often by hyperlinks or leaps through connections that circumvent linear or quasi-linear navigation by using search engines like Google to group possibly similar elements. The use of the term space is not physical space or quasi-physical spatial representations (avatar worlds of Gibson's cyberspace), but possibly topological space or the definition that relates to an area designated for a particular use.
Posted by vanderwal @ 11/22/2001 09:57 PM PST [link to this comment]

Thomas, I think that relying on information is overly-reductionist and increases the risk that people miss the really interesting things that happen on the WWW.

Take a message board. What's more interesting, the information in each post, or the relationship between the people making the posts? On Random Tech Support Board, it may very well be the information. But what about on the WELL or Cafe Utne, where the same people may have continued many hundreds of discussions over a matter of years?

To reduce people to the information they leave in that case is obviously inappropriate, and some may argue beyond the scope of IA.

So, take a pirate ftp site (one of the mythical data havens). OK, the information is the focus - the app or whatever. But is it the most interesting thing? The most important? I don't think so - there too I see the relationship, and the overall context being the primary issue.

In general, I think information is theoretically inseparable from the relationships that surround it. So to place too much stock in the information as the natural unit of analysis is perhaps to cover over relationships between people, which I would argue is always the most interesting thing about any website - even a document archive or something "straight" like that.

I know I'm splitting hairs in this, and it may seem like a meaningless academic distinction (there is such a thing?). I don't think so, though. The web has potential beyond information transmission, and although we here may understand that implied in that are relationships of various kinds, I don't think that idea travels well into the general public.

Hence, we had to endure years and years of silly articles about an "information revolution" as if the information availability itself was the most important thing.

I prefer to think of relationships in the primary position, which can be about (and privilege the transmission of) information, but they can and are in fact about lots of things that don't rise to the level of being considered "information" (even if they are in a professional discourse). I just turn it around a bit.
Posted by Michael @ 11/23/2001 10:38 AM PST [link to this comment]

Michael, in part I was hoping this was the case you were making as I completely agree that information with out context and a sense of relation is less valuable. It is something like wearing a uniform without an organization. A uniform gives a sense of structure, context, expectation, and belonging, yet a uniform without an organization leads to confusion and improper understanding.

I have been working with an organization for the last two years that has been trying to provide an data and information repository. Much of this resource was being assembled from external entities. Myself and others have advised the organization that the first step to making this data and information was understanding the information and providing the users accessing this resource using the Web tools to give understanding and context. The client believe those that advised them they did not need to understand the information and good programming could work around the lack of understanding. Needless to say there was a figurative train wreck.

An information architects goal is to keep these type of situations from happening. The IA should be providing understanding and context for the users.

One of the great things about the information revolution is it is taking place on a platform (the Web) that facilitates the addition of relationship and context to the user. One can click a hyperlink to find the source of the information to add depth and understanding where the information grew out of. One of the problems I find with this depth and breadth of information is that much of it points to other lateral information sources that echo the same of similar message. I like having the ability to find the fountainhead of the idea and the steps of growth and permeations that idea has taken through time. This lack of historical direction has been part of my frustration in trying to dig out pre-technical and non-technical understandings of "information space".
Posted by vanderwal @ 11/23/2001 02:47 PM PST [link to this comment]

With respect to the question of navigation and hypermedia (comment 11), the idea of "navigation" and spatial hypertext was already central to the first hypertext conference (November 1987). Jeff Conklin's 1987 review paper in IEEE Computer, highly influential at the time it appeared, comes close to making spatial navigation part of the definition of hypertext; my own 1988 paper on orientation aids for hypertext "Bookmark and Compass", was nearly heretical in arguing that hypertext maps should be hand-drawn rather than system generated.

From 1987 through 1991, a central concern of hypertext research was "the navigation problem", the fear that readers would get lost in hypertext space unless the architecture was simple amd the signage conscpicuous. This anxiety began to abate in the early 90's, and was essentially quashed by 1993 although one still sees ritual invocation og the Navigation Problem, especially in student papers.

Early spatial hypertext systems include Intermedia, Aquanet, arguably NoteCards, and Storyspace. All appeared before 1990. "Navigator" had nothing to do with it.
Posted by mark bernstein @ 11/23/2001 03:32 PM PST [link to this comment]

> "Navigator" had nothing to do with it.

I'm sure the Netscape product marketing group had little impact on the terminology used by academics (or the niche commercial "hypertext community"), but their 50 (100?) million installs had a lot more impact on everyday language than Intermedia, Aquanet, etc.

What I meant was, if Netscape called their product "Webster" or something like that, it'd be less likely that people like my mother (god love her, a person who, if her expertise with computers was described as a percentage, that percentage would be negative) would use the word "navigation" to refer to groups of links in a left-hand column of a web page.

Out of curiosity, given that there is always more than one possible representational diagram for a given information "structure" (or, given that Storyspace users can rearrange their maps any way they please), how does space come into things at all? (Could we just replace the spatial language with talk about "sets", "members", etc.)

In these cases, the metaphor to "real" space seems even weaker than physicists talking about "phase space" or CS people talking about "state space".

(Googling around, I found a paper on Spatial Hypertext: An Alternative to Navigational and Semantic Links which perhaps answers some of my questions.)
Posted by stewart @ 11/23/2001 06:36 PM PST [link to this comment]

In my searches for information space in a non-technical arena I found many articles that help shape the term space as it relates to information. Two articles have stood out and have decent references: Andrew Treloar's "Information Spaces and Affordances on the Internet" from 1994 and Gregory B. Newby's "Building Information Space ".

Stewart, your query about replacing spatial language with other terms and term types has piqued my interest. I will have to sleep on this one as I pro missed my wife I would not be up half the night pulling books off the shelf and digging through the recesses of the Internet (if there could be far reaches of the Internet).
Posted by vanderwal @ 11/23/2001 09:02 PM PST [link to this comment]

Hey all, hi, how ya doin'? Good, good to hear that. Glad to see you all here. I feel like I should contribute something. But I don't know what to say. Except Michael's post reminded me of something I wrote a couple years ago on Communication Design. Go here and scroll down to August 10.

And remember, you're beautiful. Don't go changing.
Posted by peterme @ 11/23/2001 09:42 PM PST [link to this comment]

(I think he's drunk.)
Posted by Stewart @ 11/23/2001 11:47 PM PST [link to this comment]

Add A New Comment:


E-Mail (optional)

Homepage (optional)

Comments Now with a bigger box for text entry! Whee!

All contents of are © 1998 - 2002 Peter Merholz.