louisrosenfeld.com logotype

Home > Bloug Archive

May 24, 2003: IA and CM; or, What I Shoulda Said...

Tony Byrne interviewed me in the latest CMSwatch which, if you don't already know it, is a great site for keeping up with what's happening in the world of content management (CM). The interview was done by phone, so immediately after hanging up I was thinking about all these things I should have said (or explained more eloquently) about the relationship of IA and CM. So here are a few thoughts:

Field vs. Industry?

Though it's emerging as a field, CM is a really an industry, dominated by CMS vendors. IA is a field, and there really isn't a class of software vendors who consider themselves the "IA industry". CMS vendors have invested significant moolah into polishing and spreading the CM message to clients and analysts alike, and have helped finance CM conferences as sponsors and exhibitors. IA, without these commercial benefactors, barely scrapes by, dependent on the largesse of established organizations like ASIS&T to provide infrastructure for its meetings, and on the sweat equity embodied in the voluntary efforts of AIfIA and individuals alike.

On the other hand, because IA practitioners have no choice but to be technology agnostic, we have no blood on our hands. This ultimately will help IA, as we won't be associated with tool vendors' marketing lies and disastrous implementations. And scraping by without corporate benefactors means that IAs have no choice but to be more communitarian than CM practitioners. I haven't yet come across a CM association, industry conference, and the other trappings of a professional community.

How might these groups fit together? It's not so obvious. Maybe the IA Summits should begin developing programs that are more marketable to the CM folks--just getting us all in the same room from time to time would help. Maybe IA practitioners should develop and publicize an "IA Report Card" that details how well CMS vendors support basic IA needs, such as integration with search tools and metadata management. Maybe CM folks should start asking their managers to add IAs to their CMS implementation teams. I don't know; the fit isn't clear, but it seems well worth exploring.

Who is the User?

When I spoke at JoAnn Hackos' CM Strategies conference last month, I had to be careful not to use the term "users," which for CM means content authors and editors. Instead I used the term "end users," a group that, according to many at the event, doesn't receive enough attention in the CM world. I got the sense that many in the CM world aren't familiar with even the most basic user research methods. Conversely, I was impressed by the detailed discussions of workflow and process, and refreshed by the event's focus on fine-grained issues of determining content objects and developing content models. The IA world is just starting to scratch the surface of bottom-up information architecture, and has a lot to learn from CM.

It's exciting to envision how this could all come together. Under the guise of a typical CMS implementation, CM folks might develop processes to support workflow, and the administrative metadata needed to manage content objects. Meanwhile, IAs might develop descriptive metadata that enable "end user" searching and browsing of those content objects. Both specialties could jointly spec content objects that balance the needs of both CMS managers and end users. IAs might determine the logic to link content objects for improved contextual navigation. And I'm probably leaving out a bunch of other ways the two specialties could partner. But I have to believe the result would be far better than most current CMS implementations.

Who Has the Fattest Books?

If you compare the top four selling books in each area, you'll see how much fatter those CM titles are:

Popular CM books average length: 554pp

Popular IA books average length: 299pp Here's a gratuitous amateurish chart version:

amateurish chart of book fatness

Which proves absolutely nothing. But it does suggest that CM is a more mature field--at least in terms of the methodological detail its books offer. Then again, it's arguable that IA has more interdisciplinary roots, finding much of its deep methodology in books on HCI, ethnography, information science, and other areas.

Enough Already

I'm sure there are many other overlaps, differences, and ways to collaborate, but I'm out of time. Hope some of you will pick up this thread; the fields are crossing paths increasingly, and it's worthwhile to explore how we might better understand each other and work together.

email this entry

Comment: JoAnn Hackos (May 27, 2003)

Hi Lou,
> I want to thank you once again for an excellent presentation at the CM
> Strategies conference. You did a good job of tying together to two activities.
> Sounds like you've considered the differences in more depth for the interview.
> Very interesting idea.
> I find that it's extremely difficult to get technical publications people or
> even business enterprise people to connect their content-management goals to
> the needs of the users. We were working with the IT organization (8,000
> people) at a major financial institution. We began the engagement by asking
> for access to the internal users of the information. They could not understand
> why we wanted to know how people used something. All of their thinking was
> focused on the internal constructs for the information resources. We finally
> withdrew from the engagement because there was such distance between what they
> were doing and their actual needs.
> I recently encouraged a team we are coaching through their Information Model
> to hold a meeting with their customer service people to define a
> troubleshooting information type. The meeting was successful (with us on the
> call) but they were reluctant to investigate further. The customer service
> people are the primary users of what troubleshooting information exists. In
> fact, they write most of it for end customers. The technical writers want to
> define everything formally from their own experience, rather than discover the
> customer requirements.
> Of course, Web site typically get designed the same way. Siloed areas designed
> by individual departments based upon their own views of the information. I
> once spent nearly 8 hours convincing Compaq's marketing organization that they
> should not organize the Web site by sales regions. They wanted to take
> customers through three to five successive screens of selections about what
> kind of customers they were before they would let them view any product
> information.
> [Are you in the US or Europe? Are you in the eastern or western region? Which
> state do you live in? etc, etc.]

At all levels, we need to ask people to focus on the user experience in addition to the author experience. Without a user perspective, we're just creating very expensive file servers.
> JoAnn

Comment: Dharmaraj (May 28, 2003)

Thanks for a very illuminating interview with CMS Watch and the corresponding *Bloug*. I have been following IA for sometimes now. And still reading the Polar bear.

IA seems like a much more broader field than CM in that apart from bringing in the end-user perspective, it can also help in categorizing content. But then thats a lot of hats for an IA consultant to wear - the website administrator/designer, the User experience, the content editor/manager, - a sort of the juggler!

I mean if I come from the website admin role, data/content to me is essentially where it is placed in the page. As a content guy I might think only in terms of how much detailed the content is and as an UI designer I will think in terms of which content should get priority and within how many clicks should the end-user reach the data, etc.

Where does one draw the line? I mean at times, the IA consultant profile seems, to me, to be at cross-purpose! :-)

Comment: Donna Fritzsche (Jun 2, 2003)

Hi Lou,
Thanks for the interesting comments, here are some thoughts along the same line.

Content Management is a process. It combines skills, tools, and techniques used in the fields of writing, editing, advertising, journalism, web design, and project management. Another more formal way to think about the steps involved can be found here: http://www.content-management-software.co.uk cm_process.htm

Content Management is also a field/an emerging discipline. There are professionals who facilitate the above described process. They call themselves Content Management Specialists, Content Strategists, Web Editors, Knowledge Managers, etc.

Content Management Systems are tools! They were created out of an observed need to make the web publishing process more efficient. I suspect that in most cases it was programmers who saw the initial possibilities. Over time, some of the top content management tools made it to the top and became players in what is known as the Content Management System Industry. Many small developerment firms also have their own version of Content Management tools.

IA’s need to be able to work within the constraints of a variety of technological tools. This is especially true of Content Management Systems. CM systems impose certain technical and pragmatic constraints on the IA process and IA deliverables. Any given CMS will make certain functionality easy to carry out, more intuitive, etc. Similarly, some of the different attributes of CM systems can make certain IA recommendations difficult to implement or maintain. It is for this reason that IAs should become knowledgeable about the different CM systems and we should form opinions about which features work well and under what circumstances. For any given project, our IA recommendations should be compatible with the CMS that we are working with. In fact, someone with IA knowledge should probably be part of the CMS selection process.

As Lou states, IAs should try to positively influence the decisions of the vendors. And vendors should consider the implications of their designs on our process and deliverables. I think some shared sessions at the different conferences would be a good thing. There also is some crossover on the different listservs. Additionally, some of the best discussions will probably take place on-the-job between IAs and CMS representatives and programmers.

Overtime, I think that teams composed of IAs, Content Management Specialists and Technologists will come to recognize which CMS tools will work best in a given situation. Hopefully, we will continue to see a variety of CMS tools so that the right tool can be chosen for the job.


- Donna

Comment: Lou (Jun 3, 2003)

Great ideas! But...

How do we make this happen?

Comment: Donna Fritzsche (Jun 5, 2003)


I think that there are three important areas to address: education, communication and collaboration.

Expanding on earlier suggestions, here are a couple ideas that we might try out.
(Possibly they could be sponsored by AIfIA.)

1.) A Two-Day educational CMS workshop held in conjunction with the IA Summit.

We could invite different CMS Vendors to participate. I would suggest that each vendor present for at least 2 hours. One hour is just enough time to give an overview, but doesn’t give enough time to get into the nitty-gritty issues. It is my gut feeling that it is in the details of the implementation that the real integration issues and IA implications will come out.

As opposed to a free form session, I think it would be good if we had either a set of predefined and carefully considered questions for the vendors to answer or if we actually created a set of “IA benchmarks”.

Benchmark suites have been used in supercomputing for years. The performance measures vary depending on the type of problem that is being computed. Similarly, the performance of different CMS suites will vary depending on the qualities and architecture of a given website. We might, for instance, create 4 benchmarks which will exercise, explain, and challenge the different CMS packages.

For starters, I’ll toss out four types of benchmarks that I can see as being relevant
1) a basic corporate or informational Internet site
2) a subset of an Enterprise-wide Intranet
3) a subset of an E-commerce site
4) a subset of an Ezine/Publishing site

While the point of benchmarks in the supercomputing world is to highlight and compare quantifiable performance measures, I see the point of the IA benchmarks to be educational in nature. They would facilitate discussion and collaboration. They would allow the vendors to show off and explain their strengths in a variety of contexts. Benchmarks would improve learnability and provide for consistency in discussions and comparisons.

2) The AIfIA could sponsor a yearly award for best CMS implementation. In the spirit of collaboration, it could be asked that a team consisting of a content manager, an IA, a technologist, and a vendor, write a paper describing the success of their CMS implementation from each of their different perspectives.

If there is interest, I would be willing to participate in a team that moved these ideas forward. Is anyone else interested? Does anyone else have any other ideas or feedback?



Comment: Victor Lombardi (Jun 6, 2003)


Great ideas. I think many IAs would benefit from seeing a concrete representation of a CMS, i.e. actual templates and administration screens in the various software packages. Beyond that, I'm afraid vendors would throw a lot of marketing at us. Perhaps we could balance them with some CMS experts, people like Hackos, Rockley, Boiko, Byrne, and Warren. And while it's a popular topic, basic on the most recent Summit I'd say 2 days might be too long/too expensive for the IA crowd.

Regarding what could be taught (and possibly also rewarded in a contest), I'd personally like to see much more specific items that get into the nitty gritty of CMS, like information modeling, how well CMS let's us use content chunks or facets, and the design/implementation of templates and admin screens.

Comment: Donna Fritzsche (Jun 6, 2003)

Hi Victor,
I like the idea of adding outside CMS experts.

I agree with the educational topics you mentioned. Part of my thinking of using the benchmarks is that if we strategically designed them, they could elicit examples of the type of functionality we are interested in seeing explained.

For instance we could design a set of sample intranet pages that required content chunks to be distributed contextually … or metadata that was fairly complex. So if a vendor chose to illustrate their tool by using one of our benchmarks, then part of that illustration would be to explain how they distribute content chunks or use the metadata.

I would also like to see detailed examples of how metadata attributes and vocabularies are entered, integrated and used, how dynamic publishing occurs, and what is the process of going from Site Architecture and Wireframe to the CMS implementation. What happens seamlessly? What doesn’t? I realize that vendors aren’t going to go into details on the gotchyas, but at least in asking some of the right questions we will be pointing them in a direction that supports our needs as users.

Thanks for the feedback!


Comment: Lou (Jun 6, 2003)

I heard Victor give a talk this week on content management; in it he mentioned a couple interesting fragments regarding CM and IA that I wanted to share here.

One is the survey of CMS problems from the IA perspective that Victor ran for AIfIA earlier this year. Check it out here:
Good numbers, and especially interesting comments to boot.

The other is simply the thread of gripes with CMS that were posted to the SIGIA-L list in February:

I really like Donna and Victor's ideas, and would love to see an IA/CM/CMS initiative of sorts, ideally sponsored by AIfIA. Donna, what do you say; interested in putting it together? :-)

Comment: David Heller (Jun 9, 2003)

Lou asked me yesterday to chime in here as a Vendor representative. I will add that I am not speaking as a Documentum employee, but rather as someone who has the experience of developing on both sides of the screen when it comes to CMS and IA.

I think there is an aspect to all this that is missing from Lou's discussion. There are two areas of IA that need to be discussed and they are equally important and in my mind they are both "end-users". When it comes to CMS (and btw from what I can tell we are really ONLY talking about Web Content Management, which quite honest contributes to a small fraction of CMS uses) there are a host of audiences all of whom require to be thought about as end-users. They all require the same level of user research and the research needs to be considered as intersecting because in the end they all effect each other.

IAs need to be equally deployed for developers, content managers and content authors as much as for content consumers. I don't think that you can separate the IAs for these areas from each other because how an author conceives of his or her world will effect how a consumer get their information.

I do also think that Donna's distinction of CMS as tool is an important one. It is also a tool made for all possible scenarios. This is liberating but also problematic. How can someone make 1 tool that is equally good for a magazine as it is for a catalog as it is for an intranet, etc. etc.? The vendors then are only part of the story and to concentrate on them is actually a mistake. The system integrators (SIs) are the real "tool designers". Documentum very much thinks of our product not as a solution, but as a platform for solutions. SIs are the ones who build the solutions. I have many times gone onsight to see how a customer is using our products to find a whole list of areas that could have been easily improved by the SI who did the installation and configuration of the product. Many of conflicting requirements because they are partnered with a poor application server vendor that can't do X,Y,Z w/ our product (which is why we don't recommend that combination); others have had a customer buy into a CM vendor and used all their budget on teh product b/c they didn't realize the overall cost of ownership and setup and thus didn't have enough resources to support a real SI engagement.

But back to the IA. I have heard from IAs that the tools aren't there in CM products and then I ask what is missing and I have not been given a real answer to the problem.

CM products have taxonomy services both manual and auto (and combined). There are data dictionary capabilities and just using the platform capabilities of the product you can build almost any facetted navigation system you can imagine.

Does it do it out of the box? Of course not. It can't. The permutations of possible solutions are massive and quite honestly unpredictable for a CM vendor to address.

Back to Donna's initial proposal. I think the real problem is in multiple areas:
1. IAs need to be better educated in CM product possibilities: How to use databases, how to use and implement XML, etc.
2. SIs need to employ multi-tracked user centered approaches to build better CM installations and deploy better published output.
3. CM vendors needs IA to help them to better understand their product.
4. Customers need to better understand how IA can effect their enterprise along these multiple tracks.

That being said, I would suggest the following:
1. AIfIA and ASIS&T should co-sponsor a summit of CM Vendors & SIs together to discuss CM from their perspectives. There are some great case studies already available (meaning to be written, not alraedy published) with succesfully installed systems that have great IA for all involved. A summit like this would be 2 days. It would have limited IA involvement (the IAs should be subsidized) and the SIs and the Vendors should pay the brunt of the cost. I would be willing to funnel any discussion about such an event to the right people at my organization to get things rolling. I Assure you if DCTM is involved and we can get one other vendor (I think we can get IBM b/c they are a partner of ours) then we can work on the SIs we are partnered with many and make such an event work.
2. I think that we can create an education tutorial for IAs at the next IA summit "Concepts in WCM". What are templates? What are the basic structures of a CM product (lets face it they all do teh same thing)? etc. etc. In this same regard, I think a panel of companies who have success stories and reality cases would be a good edition to the summit. I proposed this last year, but didn't make the cut. I have a lot of customers like BBC who said they might be interested in participating in such an event. I also have a host of subscribers (200 or so) who belong to a list devoted to IA and CMS that no one is talking on (ia-cms@yahoogroups.com). I thin it important for this panel to include both consumers of content and contributors of content in their studies.
3. A speaking tour by Lou or Victor, an SI and a CMS vendor would be a good step too. Awareness and education are key.

Ok, this has been long. ;)

I also think that content sites need more focus on CMS. No one is creating a content site w/o some type of CMS (home grown, open source, or vendor). We need to make CMS a priority of understanding by the designer community in general and IA community specifically.

Ok, I'm done (for now). Lou, this was a great little thread to get people thinking about this topic. A quick next step might be to have a column on CMS on Boxes & Arrows just like they now have one on Interaction Design and Documention.

-- dave

Comment: Donna M. Fritzsche (Jun 9, 2003)

Hi everyone:
In response to Lou's question:

"I really like Donna and Victor's ideas, and would love to see an IA/CM/CMS initiative of sorts, ideally sponsored by AIfIA. Donna, what do you say; interested in putting it together? :-) "

Id like to say , yes I would be interested in putting it together. I think it would be appropriate to have it fall under the AIfIA umbrella.

(I didn't respond right away because I had to consider what it would involve time-wise, etc.)!

I think it would be a very worthwhile and challenging effort and one that I would enjoy be parting of. If this turns into an event, I probably would be in favor of piggy-backing off of the IA summit - for pragmatic reasons!

It sounds like Dave has some good ideas - I haven't fully digested them yet but I will reply back at some later point.

Lets keep talking about the possibilities!


Comment: Dave Bauer (Jun 18, 2003)

If you want to see professional community in action in the CMS industry check out Open Source Content Management http://oscom.org The recent conference at Harvard included no sales or marketing. It was a very cooperative atmosphere.

Comment: Lou (Jun 18, 2003)

Thanks Dave; this looks really promising.

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.