Choose your own adventure… er… learning path

There’s a lot to be said that’s positive about our declining economy. (Yeah, it’s an odd beginning. But hang in there with me.) In the publishing industry, for example, we’re having to be a lot more careful about what products we release. We’re also having to be more efficient in our products; people want more for less.

As part of that, we’ve been doing a lot of thinking about non-book products. O’Reilly has offered books through Safari for a long time, and there’s certainly the O’Reilly School of Technology. But what about folks that want traditional book topics, like JavaScript or PHP & MySQL, in a non-book form? These people don’t necessarily want or need course credits, but they expect more than a PDF-style view of a book.

When you start to think about effective learning in an online medium, one the very first concerns you’ve got to deal with is sequencing. Web users don’t live in a page-by-page world. John Lewis calls this the “disclosure sequence” in a recent post on learning paths:

In our view, the most important aspect of the design of a training course is the “disclosure sequence”: the sequence in which the topics of the subject are disclosed to the learner. This is the central issue on which all other aspects of the course design depend.

If John and his colleagues are right, there are two critical realizations here:

  1. Order matters, but web users are not comfortable with linearity. People on the Web tend to skip around, looking for just the piece of information they want.
  2. Again, though, order matters! Just because web users may want more freedom doesn’t necessitate that as learning experience designers, we have to give those users complete freedom.

So how do you split the difference? Can learning really be broken down into small enough segments that web users feel they have sufficient control (I want to jump to “closures” in “Ruby”), while still allowing teachers and educators an ability to guide a learner through that smaller segment (You need to learn about Ruby’s handling of memory, then move on to closure syntax, and then examine the pattern in play within a closure)?

One key aspect to solving this problem (or at least approaching it intelligently) is to reorganize in terms of concrete, functional goals, rather than lexical or technical topics. John Lewis again has useful observations on what you can achieve when grouping things according to functionality:

In general, a more effective guide to designing a disclosure sequence is to choose a path through the goals which the learner already has, beginning with those which can be achieved simply and progressing to those which are more complex to achieve. It is evidently not always easy to find these; but if such a path can be found, then the requirement to motivate learners along the way tends to be easier because they are achieving parts of the objective along the way, while building on what they already know.

So here’s my question; it’s a two-parter.

  1. If/when you’re learning online, do you expect complete control? Do you expect the ability to move anywhere within a course and/or topic, even to specific steps within a process? Or are you satisfied with topical/functional control, and then are willing to follow a process within that subset?
  2. Does all this seem silly to you, and you think learning online is really just a fancy term for a good Google search?
  • 1. I expect as much control as i need at that skill level..Meaning if i am a novice and am looking to just learn about the topic don’t mind lesser amount of control.. else will find a resource with a level of control that i am comfortable with….
    2.I assume this is redundant.. the answer to the first question has to answer this one…

  • When learning a new language, I like the topical/functional flow but when I’m looking for specifics a Google search is indeed the way to go.

  • As a novice, order matters, yes. But as an expert, dependencies matter. The shift from novice toward expert is difficult to assess, and by the time you’ve figured out how to assess which I, the current user, am, I’ve moved on to a Google search to try to find what I’m looking for.

    The reason I want full control is that I have significant prior knowledge that allows me (intellectually) to bypass a lot of content. If I’m not allowed (technologically) to bypass that content, then I become quickly frustrated.

    You want the service to match the user’s mental model of the domain: weak (possibly inaccurate) mental models need structure to help them define the domain, while strong (and accurate) mental models need the freedom to cast about and find the information they need.

    So, what I’m imagining at the moment is an interface that (a) provides a structure that is (b) not enforced or mandated and that (c) provides transparent indications of the dependencies within the content so that I can (d) backtrack wherever my prior knowledge may not suffice.

    Books have a and b. Wikis (arguably) have b, c and d, although not necessarily at the level of detail that’s most useful for learning (I need just-in-time information… not just more information).

    For example, if you’re reading along and reach some information that relies on linear algebra for the explanation, there should be a way to link out to the _particular topic_ in linear algebra that relates. Ideally, I’d like the option to see those as tree structures/webs, too, in order to help me visualize the domain(s).

    Read top-down, those trees offer a path; bottom-up, the offer the dependencies I’m seeking. As webs, there are many paths through the topics.

  • George Spivey

    Linearity is nice but when you’re not even sure what you need it’s cumbersome. My ideal learning tool would a 3 part window with history on the left, hyperlinked material for review in the middle and a column on the right suggesting related info. Bookmarks and notes would be nice too.

  • As others have mentioned the key is that novices tend to need more guidance than more advanced learners. The sticking point is that novices frequently don’t see themselves as novices. Some novices try to skip material that they think they know but don’t. With in person education it’s generally fairly easy to tell the two apart, in a book or automated online it will be a bit trickier.

  • The best way of learning (same for working) for me is always the ability to stay in “flow” (or in the zone, in another often-used term).

    Therefore, first, the topics should be small enough to chew on, not so big that I have bang my head against it for hours (or even 45 minutes). Small progresses are, just like unit-tests, encouraging and helps me stay in flow.

    Second, topics are getting more and more inter-connected, and those topics are usually not on a single ‘sequence’ branch. It’s quite easy to get carried away and lost. so sometimes I don’t think the word “sequence” is good enough, and that’s why I always use freemind to keep track the related-points encountered or to drill down later.

    So, a map, like those found in many history books, on which bite-sized-topics are connected with recommended paths to follow and I can mark, link, and annotate, is very much welcome :-)

  • On a daily basis, I would say online learning is just a fancy term for a good Google search. The reason being that most regular requirements of new knowledge are just a minor broadening of what you already understand. Often you are aware of a basic concept but are looking for specifics.

    For example, I don’t consider myself a bonafide programmer but I am perfectly comfortable editing code and adding the occasional modification into PHP code in various CMSs I work with while designing sites for people. Having had some basic C and intermediate Pascal exposure in school (wrote my own rudimentary text adventure), I understand the big picture of how things work, but I am often ignorant of the specific syntax I need to achieve an effect. Google to the rescue!

    The thing with this sort of learning is that it is very fragmentary and is meant to meet an immediate need. I usually forget the newly found solution in an hour or two and it’s only after I have re-found the same solution a few times in different contexts that it becomes part of my knowledege. When I am learning on my own I consider that freedom of movement through a topic a necessity, and the extended learning time-frame is an acceptable side effect of focussing on random, often un-connected bits of information.

    Sometimes though, you can’t learn completely on your own because the topic is too far removed from your present knowledge and experience, or you are looking to build a more comprehensive understanding. This is when you need a teacher, not necessarily in physical presence but at least in spirit. In a world where virtually all the facts are freely accessible, what is a teacher if not a guide who rations your exposure to the right facts in the right sequence so as to best improve your real understanding of a subject?

    Your point of lexical vs functional learning comes to the fore here, because when I want to really learn PHP beyond the slip-shod tinkering I currently dabble in, I’m not goin to read quick snippets of code. No, I’m going to pick up a nice multi part tutorial on how to build a simple CMS system in PHP and work through it. If I’m smart, I’ll stick to the sequence and let my knowledge build up natuarally from one part to the next while I tackle the very real thrill of creating something functional.

    In this sort of deeper learning I think process is important, even if only on a topical subset level rather than through an entire body of knowledge. Choose-you-own-adventure books are an interesting novelty and a great way to kill some time, I own a few myself. But, the vast majority of books I own tell a story and they tell the story the writer intended to tell, in the sequence he intended to tell it. Needless to say as human beings, it is the stories we remember.


  • Vahe Katros

    You’ve said some things like

    “People on the Web tend to skip around, looking for just the piece of information they want.”

    That make me feel that your audience segmentation could be more robust.

    The design, you know, will vary for experts and beginners, for Usage patterns (“I need an answer in 5 hours, 5 days, 5 months). Motivations (personal, project, exec overview) Lifestyles (pick-em), Painpoints (? what are the headaches for using your stuff) Unfulfilled needs (study the customers who you’ve lost or recruit people who love what you do, and hate what you do)

    There’s a lot to be said that’s positive about our declining economy. It means you can spend time learning so you can do it right and different and transformational.

    Tap your community – I know folks that would kill for your access, use it, it’s a win-win.

  • I have used Safari as part of my ACM subscription.

    I’ll start with my experience and then move on to your questions:

    • I found the restricted set of books through ACM a surprise and a little anoying.
    • I Wanted to relate with the site as a web site, and found the linear path annoying, to the point that I bought books instead.
    • The page by page metaphor felt wrong for the medium, chunking the content by the “conceptual unit” would be an easier way to navigate.
    • I wanted to share content, deep link into the site and/or email it to a collegue and while it was possible, It wasn’t easy. I would want to be able to deep link from a blog and have an easy way to give other consumers access to the content. How to allow that while maintaining a business model is a hard problem.
    • The fact that outgoing links weren’t in the text was offputting, making the content seem outdated.

    So, your answer is:
    Yes I expect complete control, possibly to the extent of pingbacks to referring pages of content elsewhere.
    I also want the option of taking a path to learn the topic, but they should be optional and easy to follow, examples from the past are the Java trails tutorials that I remember from the early days.
    I want the content to be a first class citizen of the web with all that goes with that.
    2. Search inside the books is needed but doesn’t need to be by Google, i think you actually need BOTH, you need to include the content from Google/Web into the context of reading the neobook.

  • Learning online is very useful. I do not think it’s just a fancy term for a goog Google search.

    Even though a few Google searches will come up with all the information about any topic, for a person who encounters the topic for the first time, the plethora of information can be a bit overwhelming. What such users need is small relevant bits of information, and a little guidance here and there. And yes, such users would like to have the information disclosed to them in a certain order, but at the same time retain the ability to jump around from a “table of contents” page if they need to look for something out of order.

    I have used the web for learning and for teaching, and I personally different preferences for different topics. If I am familiar with the subject and am looking at some specific information, then a pre-defined disclosure sequence would be a hindrance. However, if I am totally unfamiliar with a topic, then I will appreciate a disclosure sequence. But at the same time I will also appreciate the ability to jump around, just in case I want to learn out of order.

    This is just my opinion, but for online learning, a pre-defined disclosure sequence, with the ability to override it is what would be most useful.

    I have also recently tried experimenting with the model of a blog as a course medium. So far several people have shown interest, but they have yet to start working on the tasks. Here’s the link to the experimental “Java Insights 101” course:

    Parag Shah

  • Jeff Solar

    To me, the key is the ability to go backwards (or sideways), rather than ahead. If I already know the material, I find it easy to skim through to the next section. If I’m struggling, I need to be able to find additional information, or perhaps go back to an earlier topic to refresh my understanding of that content.

    The great advantage to electronic media is the ability to provide an easy method to find that additional information without losing the original “page”, vs. hardcopy where the user must page through to find the needed information.

  • C. Creasy

    A list of desired “electronic learning tool” features


    1. A detailed Table of Contents (TOC)
    An electronic document or web site might have the TOC as bookmarks in a PDF or as a sidebar in a web site which when clicked on or hovered over expose greater detail (the chapter/section outlines).

    2. Chapter/section outlines
    Those outlines in the TOC should be repeated at the beginning of each chapter (clickable to the subsection).

    3. A thorough Index
    An Index is a great feature for reference as well as for learning. If you think you know the material in a book, just take a look at the Index to see if you understand all of the terms. And, of course, a good Index IS a search.


    4. Pop-up descriptions (with or without links).

    5. Short video demonstrations which are thoroughly described (and indexed) to let the user know if they want/need to watch them.


    6. A way to mark your progress (e.g., a reader’s bookmark when taking a break from a book).

    7. A way to mark a section for future reference (a reader’s post-it or dog-eared corner flagging a book’s page)

    8. A way to find important material quickly within a section (a reader’s highlighted or underscored text in a book).

    9. A way to add notes/questions (a reader’s comments in the margins of a book).


    10. When material is updated, the user needs a way to access all of the updated material for a quick review. [Will the specific electronic learning/reference tool be purchased or will there be a subscription where all updates are “free” as long as the subscription lasts? How will User Annotation port to the updated version?]

  • Char James-Tanny

    Most of C Creasy’s list matches what many authoring tools today can provide. Numbers 1-3 (TOC, chapter outline, index) are a given, and most online docs also include full-text search. You can choose to follow the TOC in the order the author has set up, or you can use the index, search, and hyperlinks to jump around to the specific topic you need.

    Popup descriptions are generally created with glossary terms. And video is still iffy…it depends on the subject and the development environment (including the author’s experience creating videos).

    User annotation stuff is currently pretty weak. Some online docs include a place to store Favorites (which could be used for links to current place and future references). However, annotations usually require writing something to the user’s system and that’s typically frowned on (other than cookies). It’s possible, but harder to implement.

    And updates are again dependent on the author/development environment. Sometimes release notes (or a “what’s new with this release” section) are included, but many times, updates are what they are…replacements for the existing content. And annotations may not make it through the update…it depends on how they’re implemented. (If the author can change the underlying reference to a specific piece of content, whatever it might be, then annotations lose their frame of reference following the upgrade.)

    And no, Google searches cannot be used in place of good online learning docs. Instead of one document, you end up with many, many links.

  • I used O’Reilly learning products when I was studying for my CCNA. Used techniques I learned when I was in college (both undergrad and grad school). Namely, break things down into “digestable” chunks that can be read and understood in no more than 45-minute sessions.

    Your materials provided a CD-ROM along with the book… The CD-ROM provided interactive test questions on each chapter–and then an overall test for the whole book. The book was handy for reading when I was commuting to and from work. The CD came in handy to review what I had read. The combination helped me to pass the CCNA exam.

    I also used “online” resources to get to information that may not have been clear to me. Wikipedia was good–because it allowed side-tracking to other information sources and references.

    I won’t use a book for reference if it isn’t indexed well… or if there’s no Table of contents.

    Websites that aren’t referenced may also be questionable for their accuracy.

    It’s a dilemma that all content providers should be aware of.

    Interesting discussion point.

  • As a developer of courseware for an online university, I find that the most important aspect of developing a formal course is to have very clearly stated, measurable learning objectives. This way the user can see at a glance whether they will be able to learn what they need from the course.

    Moreover, if you clearly map those learning objectives to the modules or sections of the course, this will provide even better clarity and help the user find for themselves those things they may or may not need to learn to solve the problem at hand.

    Of course, my experience is with classes, not something where a discrete knowledge unit is needed, but I wonder if online teaching systems could be scaled to smaller levels of granularity, making it easier to do the kind of learning that might be attempted with a “quick Google search”.

  • Eder Andres


    mmmm, Creating a training course is something dificult. Looking in Microsoft web site, I found an interesting page taking about which competences a student (high school) must have.

    I’m impressed with the Education Competency Wheel.

    And all 37 the competecies

    (Currently, I’m reading the Individual Excelence)

    However, Is it possible to adapt them to a university training course? mmm it seems a hardous task, but the result could be great.

    Last but not least, techniques for self-motivation are suseful too and can be considered in the beginning of a course.

  • The courses and degree programs I’ve worked on are at the college level, and would be taken after general ed courses have been completed. But I am aware of other outfits that have successfully taught pretty much every subject out there in a distance learning environment.

    The competency issue is important, too, because the competencies are specifically aimed at employment. What if you’re not looking for a new career or even enhance your current one? What if you just need to remember how to do something you’ve forgotten?

    Motivation is a big issue, too. Student retention is the usually one of the biggest problems faced by online schools. Unfortunately, (or fortunately) the most effective means of keeping students is making sure they hear from their instructors. But, that’s for courses and degrees. It’s not the case if you are learning something small and focused.

  • I pick up a technology say Javascript and go through complete books and video tutorial series. But I hate spending too much time learning syntax and those features which are rarely used. Books which may teach you less but give a solid foundation like “headfirst” series are my personal favorites.
    I dont end up cramming the whole tech from the online tutorials, instead I start coding asap and learn as I code.

  • Certainly I’d want to be able to do a search and jump to a specific node “(“Closures in Ruby”), and what would be a rare and extremely helpful feature is a list of links to prerequisites needed to understand that node (“Ruby Memory Management”).

  • I think a good idea is to have pre-requisite/assumed prior knowledge.

    And then designing topics to be stand alone as possible. Also, a rating system (beginner, intermediate, advanced) is pretty clear.

    There should be a correlation between pre-requisites and levels (beginner etc).

    So advanced shouldn’t mean complex, it should mean “requires more prior knowledge than beginner or intermediate”.

    If performance/learning objectives are action (doing something) based, then it follows that the more you’re required to be able to do before you can learn a new thing, the more advanced the topic is.

    Also, keeping reference docs (wiki for example) separate from guided learning resources (tutorials, screencasts, etc) is a good idea. The two are differentiated by purpose: one is a lexicon, the other is a learning experience that might refer you to the lexicon/wiki for reference info.