Info architecture work that sometimes makes my head hurt

  • Most info architects agree – planning for reuse is harder than conditional text. But even conditional text can be difficult, especially if there are multiple conditions that overlap. The winner of the “most conditional text” contest was this commenter on my talk.bmc post with 64 conditions in a FrameMaker document.
  • I still struggle with topic authoring – but I’m finally “over” separating content from format. Whew! That only took a couple of years. This week I’m chunking of information using the rule of “seven plus or minus two.” That doesn’t usually make my head hurt, until I start coming up with all sorts of scenarios (maybe the user wants to set up their web site pricing for a DVD sale in the month of March!) and then I find myself writing too many topics.
  • I also read Jon Udell’s great post about potential reasons why del.icio.us hasn’t really gone mainstream, Discovering versus teaching social information management.  I think my own tag merging and pruning best practices need work.  My favorite lines are from the comments, such as “people need to both realize that they can do that database query, and that they can refer to the results using a stable URL. I’m coming to believe that both those operations are still way beyond the capabilities of mainstream web users.”
  • And finally, inline linking versus grouping links together. Usability studies and experts disagree on the correct way to link. I’m not sure I have the answers yet either. Better keep studying and linking.

What are some aspects of information architecture that are making your head hurt lately?

5 Comments

  • ffeathers
    February 8, 2008 - 9:52 pm | Permalink

    Hallo Anne,
    ‘What are the aspects of IA making your head hurt lately?’
    Good question! My pet headache is the structuring of information so that people can find it again. Let’s call it retrievability. I’m not a big fan of searches, because they never seem to work, no matter now much smarts you put into them. There’s always the niggling worry that the search has missed the most relevant chunk of information, just because it was written around a synonym rather than the term you searched for.

    After a few years as a book indexer (yes, I used to create the indexes at the back of books, and no, computers don’t do that!) it’s really clear to me that everyone’s internal classification engine is slightly different. So when you’re creating an index, you have to put in lots of cross references (‘x see y’ or ‘x see also y’).

    Now we’ve got searches, indexes, tags (aka labels) and tables of contents. Add wikis, where everyone creates pages just wherever they happen to be at the time the whim hit them…

    So maybe there isn’t an answer. You can never be sure that you’ll find just exactly the piece of information you were looking for. But you *can* be sure that you’ll find *lots* of information.

    And the best filter you have, once we’ve found that information, is still the internal processor in your own head.

    So maybe that’s where the next revolution will be. Not in the search or index itself, but in something that smooshes it all together and squeezes some sense out of the other end.

    Bit of a long comment. Sorry :)

  • Dee Elling
    February 10, 2008 - 5:27 pm | Permalink

    To-DITA-or-Not-to-DITA is making my head hurt. I find many proponents who want to migrate all sorts of content to DITA format. But what I need to find are end users who LOVE using a DITA help system or website or document, and why.

    One reason I am exploring this is that I work with a pre-DITA, DITA-like implementation of a Help system and many users dislike it. Now, there’s no direct evidence that they dislike it because of its information architecture. Their complaints are more about not easily finding the information they need, as discussed in ffeathers’ previous comment. Is there any evidence that DITA helps “findability” or “retrievability”?

    Our users had fairly strong opinions about in-line vs. link lists. They like to see links in context. We now do both.

    Thanks Anne, -Dee

  • February 11, 2008 - 3:24 pm | Permalink

    Hi Sarah and Dee -
    Thanks for the comments. You’d think we have lots of headaches but really these are fun puzzles to solve.

    Sarah, how interesting that you are/were a book indexer! What a valuable skill to draw from. I just read a post on the Yahoo DITA User’s Group from Bob Doyle suggesting that metadata should let you tag “synonyms (USE and USED FOR) for equivalence relations, hierarchical broader term (BT) and narrower term (NT), and related terms (RT), for arbitrary associations” – wouldn’t it be great if the metadata guidelines helped offer suggestions to writers for thinking of those other terms. Of course, it still wouldn’t help me decide whether to use “registration” or “enrollment” in the title of the topic I’m currently writing. :)

    Dee – It’s so great that your users have definite opinions about inline linking, then you can just add those in like you say. I’m working on finding out more findability and retrievability studies for DITA. Michael Priestly probably has ideas along these lines. I’ll let you know if I find out more! To DITA or not to DITA, that is an excellent question. Gordon Mclean and Scott Nesbitt have been asking that question as well. I’ve got a post in draft form linking to their posts – I’ll post it soon.

  • February 12, 2008 - 11:08 am | Permalink

    Dee, I’ve been there too, as Anne mentions. But I’ll hold off until she posts her thoughts (mine are on my blog).

  • February 28, 2008 - 5:57 pm | Permalink

    Your comment on conditional text and Author-it is interesting. I’ve looked at what they say about it and it’s making my head hurt. After I get my new PC set up this weekend, I’m going to install the new update and try to figure it out. It’s the main issue I have with the product right now, though there are some others.

    I still have a very strong gut feeling tht DITA is the way to go. Perhaps that’s because I was brainwashed by JoAnn Hackos and her crew at the CMS conference in 2006, but I feel that going to a product that isn’t based on DITA is a mistake in the long run.

  • Leave a Reply