Another Certified LMS
10.31.2007 - Tim Martin
One message I've been delivering to our customers and prospects a lot lately is this... partial SCORM conformance might actually be worse than no SCORM conformance. If you're going to bother with SCORM at all, be sure that you finish the job so that you can reap the benefits of interoperability and compatibility.
One customer who obviously took that to heart was ELogic Learning. ELogic has licensed two of our products, the SCORM Engine and SCORM Untethered. The news of the day, though, is that ELogic's eSSential version 8.22 was recently certified by ADL. Check out the formal certification notice here. Completing an LMS certification is no small task, and we congratulate ELogic on their accomplishment.
We're also excited to note that no fewer than three of the first twelve LMS's certified for SCORM 2004 3rd Edition use the SCORM Engine as their delivery mechanism.
2:18 PM
0 comments
SCORM Stats
Curious about the adoption of SCORM 2004, I recently conducted an informal study of the packages uploaded to SCORM Test Track. Here's what I looked at and what I concluded.
SCORM Versions
I first looked at the distribution of SCORM versions over time. The data in the chart above represents the percentage of courses adhering to a given version imported during each of the last 18 months. As you can see from the pie chart, there is about a 50/50 split of between the number of SCORM 1.2 courses and SCORM 2004 courses. This split has remained fairly consistent overtime. Empirically we know that the SCORM Test Track is primarily used by advanced SCORM users and mostly as a tool for testing sequencing. This suggests that the actual usage of SCORM 2004 is probably lower than 50% in the overall market.
SCORM Versions By User
To get a better feel for the actual adoption of SCORM 1.2 vs SCORM 2004, I took a different slant on the first set of data. It occurred to me that we have some power users of the SCORM Test Track that might slant the data one way or the other if they are use a particular version exclusively. The chart above shows the percentage of users that have uploaded at least one course of a given standard. This angle shows that 43% of users have uploaded SCORM 2004 content.
The conclusion I draw from this data is one which we could have guessed. SCORM 2004 is gaining significant adoption, but SCORM 1.2 is still the industry workhorse. That is not unexpected or even inherently bad. SCORM 1.2 gained favor as the de facto industry standard because it works. SCORM 2004 is better, but it will take a while for it to surpass the adoption of SCORM 1.2.
Users
SCORM Test Track just signed up its 3000th user. The number of people discovering SCORM Test Track has steadily grown since we launched the offering a year and a half ago. Finding 3000 users with virtually no promotion means that there are a lot of people out there using SCORM. And, at least 1290 (3000*43%) are using SCORM 2004.
Realizing the -ilities?
One of the big promises of SCORM 2004 is that it would take us closer to realizing some of the ADL "-ilities". Particularly, it would enable content authors to break content down into smaller chunks which theoretically enables more reuse than the traditional SCORM 1.2 model of a single monolithic SCO.
It appears that we're making progress. While there are still some single SCO SCORM 2004 courses, the trend has shifted. In SCORM 1.2 we saw 71% of courses uploaded as a single SCO. In SCORM 2004, we only see 35%. (And who was the guy who uploaded a 5000 SCO course? I really hope he was just trying to crash our server, if not, I really sympathize with his learners!...and I'm happy to report that the package imported successfully!)
SCORM 2004 also promised to increase adaptability, or "The ability to tailor instruction to individual and organizational needs". A good indicator of whether content developers are using SCORM 2004 to increase adaptability can be found in their use of SCORM 2004 sequencing rules. The chart above indicates that about half of all SCORM 2004 courses use some form of deliberate sequencing strategy. This chart doesn't take into account the fact that 35% of the uploaded SCORM 2004 courses contained only a single SCO. Since single-SCO courses don't need any sequencing rules, we can ascertain that roughly 70% of all multi-SCO courses use some form of deliberate sequencing strategy.
10:19 AM
0 comments
Remediation Is Hard - Part 1
10.26.2007 - Mike Rustici Retry All Resets Everything
Many SCORM 2004 sequencing strategies require remediation upon failure. These strategies often call for only material which was not previously mastered to be required on subsequent attempts. The most straightforward approach to implementing such a strategy would call for storing mastered competencies in global objectives and then issuing a retryAll to trigger remediation. Unfortunately, unless the global objectives are global to the system (a dangerous practice), when a retryAll request is issued, a new attempt begins on the root of the activity tree and all of the global objectives are reset making them unavailable as input into the remediation sequencing strategy.
Solution(s)
Option 1 (DANGEROUS) - Set objectives global to system = true. Since the global objectives now have a broader scope, a new attempt on the package does not trigger a reset of all the objectives. There is peril ahead though should objectives ever experience naming collisions (i.e. there is nothing that says that another course can't use the same objective identifier as you do), not to mention the chronic pain in the neck of deployment testing (have you ever tried to reset global objectives in a production LMS?).
Option 2 - Create a shell inside the course that can be retried without initiating a new attempt on the root. For example:
-Course Root ---Remediation Shell - is visible = false, post condition rule, if not satisfied, retry -----SCO 1 -----SCO 2
Basically, moving remediation down a level will preserve the state of global objectives since it does not create a new attempt on the course as a whole.
10:04 AM
0 comments
SCORM 2004 XSD Files
10.24.2007 - Mike Rustici
I recently completed development of a rather complex SCORM 2004 sequencing template for use by the Army in ATSC and TRADOC. It was working great in several environments, but imagine my surprise when it wouldn't pass validation against the XML schemas.
A long investigation turned up the culprit, a faulty XSD document. Specifically the imsss_v1p0objective.xsd file is missing an attribute of "maxOccurs='unbounded'" on the mapInfo element. In English (or at least SCORM-glish), this means that a local sequencing objective can only map to one global objective. This is clearly against the spec as defined in CAM 5.1.7.1.2.
Hmmm, how could this be. I've been using these same XSD files ever since SCORM 2004 3rd Edition came out. I have no idea where I originally downloaded them from, but it must have been some authoritative source. With a little digging I found at least one place on the ADL site that is distributing a copy of the faulty XSD (the MSCE Linear example). By and large though, the correct versions are being distributed, it looks like the faulty version just managed its way into a few selected downloads.
To be on the safe side, I re-downloaded all the standard SCORM 2004 3rd edition XSD files from THE authoritative sources for each set. If you would like to do the same, here is where you can get them from (yes, I agree, it would be nice to have them all in one place).
IMS Content Packaging and IMS Simple Sequencing - available from the IMS site at www.imsglobal.org. You have to download them separately under the sections for each specification and you will probably need to register.
IEEE LOM - The default instance of the IEEE LOM metadata XSD files can be found at http://standards.ieee.org/reading/ieee/downloads/LOM/lomv1.0/xsd/. You don't have to login, but unfortunately, it appears that you have to download each file separately. Other profiles of LOM are also acceptable for SCORM conformant packages.
ADL Extensions - ADL has specific extensions to IMS Content Packaging and IMS Simple Sequencing that have their own set of XSD files. These files can be downloaded from http://www.adlnet.gov/scorm/20043ED/Documentation.aspx. You will need the "content packaging", "sequencing" and "navigation" XSD files.
12:12 PM
0 comments
Learning 2007::Tuesday Afternoon General Session
10.23.2007 - Tim Martin
Back at it this afternoon... (quotes are loose)
John Howley, legal expertise
"Tell your lawyer what you need from them." This is the great incentive to get your lawyer up to speed on the things you require.
We have developed in the common law tradition, in which judges were driving law creation based on existing law. Recent developments have pushed us away from that toward legislation only. Interesting... everything I hear today asks that judges not legislate. Is it inherently bad?
"Get away from the lawyer as risk avoider."
Scott McPherson, CIO for Florida's Legislature and Florida's Pandemic Coordinator
A pandemic is a novel virus, to which humans have no immunity
A pandemic requires sustained human to human transmission
He gives some compelling reasons to fear a pandemic
The first risk he mentions is that we haven't had a pandemic in so long
I acknowledge that this has little to do with learning, and less to do with SCORM, but I'm fascinated, so you get it here.
From Eisenhower, "The plan is useless, it's the planning that's important."
Wayne Hodgins
Here comes the relevance... Wayne is going to talk about the future for SCORM and LETSI.
Adoption exceeds expectations dramatically
The commonality of the problems people encounter is surprisingly high
Learning Education and Training Systems Interoperability (LETSI) is a federation of organizations loosely coupled to shepherd SCORM forward
Elliott would love...
An open standard API that allowed for plugging things together... Editor's Note: This is a great idea, but incredibly difficult. In this case, the devil is absolutely in the details. But, it's a great place to start.
Could you maintain your learning profile on a thumbdrive?
What should we expect?
Q1: LETSI becomes a formal organization
How do we prioritize the needs?
Announcing: The Global Learning Expedition
Wayne is going south from San Francisco in a boat... literally
More or less, this a focus for Elliott and Wayne on global learning
Well, the relevance level came up slightly. Stream of consciousness blogging is what it is...
3:35 PM
0 comments
Learning 2007::Session 646
Session 646 was titled "What if Amazon and Google Partnered to Build an LMS?" Good thought, and something we're thinking about all the time. Google and Amazon are doing some things really well, and we'd love piggy back on many of those concepts.
Thoughts from the group (and these aren't new to us):
iGoogle as personalizable, "LMS" homepage
Context aware content suggestions/recommendations for learners
Profiles. How can we establish a profile on an individual and use that effectively.
Commission based virtual classroom. Editor's Note: This one really resonates for me. Democratization of content production and offering is fascinating to me. I want a brilliant content producer (like Jenny Zhu of ChinesePod) to be able to deliver content to people directly.
How do you incent learners to take the content?
Tie all of this into "communities of practice"
Access to a system of mentors, or an SME network?
Amazon/Google are trustworthy, innovative, easy to use. LMS's fail in this regard today.
Usage statistics on the content itself... which pieces of content are useful and used.
They want to stop thinking about SCORM and AICC (I know someone who can help with this...)
So, the most interesting thoughts and questions to me...
The best analogy I heard for a training department of the future was that of the "newsroom". Newsrooms are looking at all the stories, structuring them, validating them... Training departments need to be doing the same things.
How big does an organization have to be in order to make use of "the wisdom of the crowds" for creation of content? Can a 500 person company effectively generate content at the bottom levels and then confirm its validity? Can we please open up LMS's and centralize them like Google and Amazon so that usage data and information has a massive audience?
11:56 AM
0 comments
Learning 2007::Tuesday Morning General Session
Don Tapscott (Wikinomics) talked this morning, and spoke on his vision of the future. This is a guy with tremendous credibility in predicting the future. The most interesting thing he shared (to me) was this:
Technology, to kids, is like air. It's just there.
I'm pretty comfortable with technology, but it did occur to me, perhaps for the first time, that those who are coming behind me are more comfortable with it than I am. I think I like that.
11:51 AM
0 comments
Learning 2007::Session 320
10.22.2007 - Tim Martin
Yup, I skipped my 200 level session for the blog... it was basically more of the same from my 100 session.
Richard Culatta led the session, demonstrating how freely available tools could be used to create content iteratively and quickly. This is a concept near and dear to my heart. Blogs are one of many tools that I believe can be aggregated to create great learning content. I'm fascinated by the tools of the web as well (youtube, flickr, etc).
Richard used several free tools (Thinkature, Fauxto, etc) to similar effect. Here's the problem:
The process is disjointed. I can grasp the leap between the tools, figure how they fit together. But it's not immediately apparent. In the course of the session, it was obvious that the tools are not entirely ready, and they certainly aren't tightly integrated. How do you put it together elegantly? How do you best aggregate these tools?
1:32 PM
0 comments
Learning 2007::Session 108
Session 108::Special Session for Learning Suppliers, Vendors, and Freelancers.
Elliott Masie and Cushing Anderson got us (learning vendors) together to give us some of their insights on the industry. Again, more thinking follows, but the biggest thought for me was this (again, paraphrased)
We need to find a way to allow for the creation of contextualized, user created and edited content.
User created isn't new. It's not done well yet, but we're all talking about that already. What's exciting here is thinking about how users (the source of much expertise) can actually contextualize and edit the content itself. This is so counter to the technical aspects of SCORM. This is why we're spending a lot of time thinking about breaking down further barriers between online learning systems.
Further comments from Elliott that caught my attention:
Users would love to buy content to which they can add the last mile.
The level of innovation from us (vendors) appears to be lacking.
Elliott implied that we were innovating, but not marketing it well. I question that. I think the level of innovation in eLearning is lacking. And, we need to market that better when we do it.
12:53 PM
0 comments
Learning 2007::Morning Session
Sorry for the loss of momentum... lack of connectivity is a blogging killer.
We got started this morning at 8am with another excellent general session. Again, great guests.
Dan Pink, author of A Whole New Mind, was certainly interesting. He had a number of interesting things to say, discussing the need for play, design, symphony, and other items. Facts, for example, are easy to come by. He shared an example (paraphrased below)
Imagine a librarian at Cambridge and a 15 year old in Tanzania, circal 1990. If asked to find the GDP of Belgium in the prior year, the competition would have been a laugher.
Jump forward 15 years... The same competition is a completely different story. As is well established, the internet has democratized facts. It is no longer the facts that are differentiating individuals and companies.
The skill that he seemed to say was most important was the one that was "difficult to outsource, difficult to teach".
Dan was followed by Jane McGonigal who was fascinating as well. Jane does really cool stuff around Alternate Reality, which some might simply call simulations, but are games of a different ilk. Check out "World Without Oil", an imagined scenario where the impacts of an oil shortage are played out in real life. Posts around the internet, puppet masters governing the situation, it's a wild expansion of learning... This is where I start to get excited.
I think there are ways where technology can improve learning in a simple, linear fashion. Say, for example, direct grading of a test with statistics for a teacher (rather than requiring the teacher to grade and aggregate data).
But the things McGonigal talked about are non-linear improvements to learning. New stuff, wild stuff. I don't know, 3 hours after she talked, exactly what I'll ever do with this sort of Alternate Reality game, but I do like that I'll be thinking about it.
12:26 PM
0 comments
Learning 2007::Opening Session
10.21.2007 - Tim Martin
Part of the excitement about attending a conference like Learning 2007 is that it isn't about SCORM and the related technologies exclusively.
Highlights of the session for me:
Elliott, as he is wont to do, sent us to talk to two people we didn't know. In each case, when I told them that we (Rustici Software) help people with SCORM, they both said, "We need help with that!"
Pat, who I spoke to first, indicated her biggest desire it to allow her experts, her SME's, to be able to produce content. Get the technology out of the way and let those people with the expertise put together the content.
Dr. "Butch" Rosser does some interesting work. More or less, he's figured out that games and simulations (fancy word for games) help surgeons achieve at higher levels. Dr. Rosser seemed to me a unique combination of surgeon, teacher, researcher, and Southern Baptist preacher.
As part of Learning 2007, the Masie's are donating a malaria net on behalf of each attendee to Malaria No More. At 2087 attendees, that's $21,o00 worth of mosquito netting. Pretty cool.
Jenny Zhu from ChinesePod.com was fascinating, and I never would have guessed it. Jenny's site is a great example of how content will be created. We, at Rustici Software, need to be producing the tools that will accommodate, discover, and distribute this content for learning. We're working on it.
To this point (early in the conference) Jenny and her content are the enduring image from this conference for me.
Doug Lynch spoke about ROI. While I recognize the importance of this... I got distracted with this blog post... Funny, though, was when he referred to Elliott as the "Willy Wonka of Learning", noting that if everyone came along with Elliott, we'd be a lot farther along.
8:04 PM
0 comments
Learning 2007::Getting Started
A quick note of warning... several posts are forthcoming on Learning 2007, a conference put on by Elliott Masie and his crew. Each year, I attend several conferences related to SCORM and the eLearning industry. Some are exciting, others are not.
Put simply, this is a conference I'm attending because I want to learn, not because I want to sell. I'm excited to be here for a bunch of reasons, but here are a couple...
Elliott Masie and his crew seem to really live this learning stuff... They will do everything they can to keep it interesting... engaging... I'm really looking forward to it.
We (Rustici Software) always looking to see what's coming, and this is a great place to do that.
Bobby Flay is going to be making dinner... Somehow, I'm told, this will relate to learning. Me? I don't really care... I just think Bobby Flay is cool.
I'll do my best to post a few interesting tidbits throughout the conference.
If you're here at Learning 2007 and you keep up with this blog, certainly seek me out. I'd love to talk to you about what you're learning.
11:47 AM
0 comments
It's Getting a Bit Crowded In Here
10.19.2007 - Mike Rustici
Three Mondays ago, we got a resume from a guy down in Orlando named Lewis Vance. We liked his brains and we liked his vibe.
Two Mondays ago, "Andy" Vance stopped in to interview after spending the weekend driving the tractor on his Dad's Tennessee farm. Yes "Andy" is "Lewis" and no, "Andy" doesn't derive from "Lewis" in any possible form or fashion.
Andy came in at noon. We offered him a job by 4:00. He accepted by 4:01.
We like to move quickly.
So does Andy.
One Monday later, Andy had packed up and moved to Nashville to start work.
A week later, Andy's already made me feel stupid on at least three occasions. That doesn't happen often, but it makes me happy.
Think you can make me feel stupid? We're still looking for a few more great developers. Rustici Software is a great place to work, but it's looking like it's about time for us to start shopping around for some more space!
3:20 PM
0 comments
SCORM Crossroads
I just returned from the ADL Technical Working Group meeting where we tried to set forth a road map for the future of SCORM. With the stabalization of SCORM 2004 and the transition of SCORM stewardship to LETSI, there was lots to talk about and it was great to see that we had a packed house. Participants weighed in from all sorts of different industries and geographies. It was one of the most productive, yet least conclusive meetings of the TWG.
The biggest conclusion of the meeting was that we are at a crossroads. We're at at an amazing point of inflection where SCORM accomplished many of it's original goals (interoperability chief among them) and it's time for us to figure out what new goals are worth striving for. Do we focus on further enabling the original SCORM "-ilities"? These were first defined over a decade ago, are they still relevant in today's world of "Web 2.0", informal learning and social collaboration? If not, what is relevant. Does the answer depend on your specific community of practice? SCORM has taken off and is now the de facto standard for learning, education and training. Yet, it's vast users all have different needs. The needs of K-12 education are quite different from bleeding-edge military simulations which is a world apart from corporate compliance training. Yet all of these communities (and many more) are leveraging SCORM. How does SCORM evolve to meet all these needs and yet still retain interoperabiltiy?
This meeting marks the first gathering of this group since the passing of two invaluable contributors to SCORM and the world of online learning. Phil Dodds and Claude Ostyn were both visionaries in the field and tireless supporters of ADL and SCORM. I never had the priviledge of working closely with Phil, but I could tell that part of his genius was an uncanny ability to cut right to the heart of an issue and decide the best course of action, not only technically, but also to support the business case. Claude was always a passionate advocate for the learner who truly believed in the power of online learning to change the way we learn and ulimately affect the lives of those who might not otherwise have access to an education. Both of them would have left a legacy which will be long remembered and appreciated by the ADL community.
Over the next few months, SCORM will evolve over two tracks.
The first track is looking at any remaining interopability issues with SCORM 2004. In December, a subcommittee of the TWG will make a recommendation as to whether there are enough loose ends to warrant a SCORM 2004 4th Edition, or whether ADL should just publish some best practice guidelines to clarify some lingering issues. Part of this analysis will include a side-by-side comparison of several certified LMS's to get a first hand look at subtle differences that might affect interoperability.
The second track is taking a look at SCORM from a broader perspective to decide what "SCORM 2.0" will look like and how LETSI will manage its requirements gathering process. ADL and LETSI are looking for feedback from any interested stakeholders. If you have thoughts, opions, requests or requirements, now is the time to speak up and make your voice heard. What are you using SCORM for now and how would you like to use it in the future? What problems are you facing? What do you want to do that you can't today? What are your current pain points? Thing outside the box, what other problems face online learning and how might SCORM address those? If you will be attending the Learning 2007 event next week, stop by the "Lunch with LETSI" session on Monday. Or, feel free to post your comments to this blog and we'll make sure to send them on to the right folks at ADL and LETSI.
10:34 AM
0 comments
Sequencing Tidbit #2 - Redundancies
10.12.2007 - Mike Rustici
"Redundancies"....There's got to be a great quote from Office Space that should go here, but 30 seconds on Google didn't turn it up, so sorry, no snappy quip to start the morning.
Because "Not Ain't Always Not" in SCORM 2004 sequencing and navigation, people tend to overly explicitly define sequencing rule conditions which leads to redundant rules.
For instance, the following rule set is redundant:
If the activity has achieved a status of "not satisfied" then, by definition, its objective status must be known, thus the first rule condition is redundant.
There are essentially three sequencing states relating to satisfaction: unknown, not satisfied (failed) and satisfied (passed). They are represented like this in sequencing rule conditions:
Given that multiple conditions often need to be specified to include the unknown state, does listing redundant rules as described above make things clearer or more confusing? What do you think?
8:25 AM
0 comments
Sequencing Tidbit #2 - Not Ain't Always Not
10.05.2007 - Mike Rustici
"It depends on what your definition of not isn't"
Us programmer geeks have a very well defined preconceived notion of what "not" means. If I say:
var person = Pick_A_Random_Person_Off_The_Street();
if ( NOT ( person.Thinks_Mike_Is_Cool() ) ) then print "mike is lame" else print "holy cow"
In almost all cases that program will print "mike is lame". Only in the rare case that somebody explicitly states that I am cool will it print "holy cow". In most cases where the random person off the street has absolutely no knowledge of my coolness (my coolness is unknown), the program will still print "mike is lame" because NOT means "do this anytime the following statement is not true".
Simple, right, no no. SCORM 2004 Sequencing and Navigation takes a different slant on the meaning of the not operator. Instead of not meaning "any time the following statement is not true", in SCORM S&N it means "any time the opposite of the following statement is true". Notice the subtle difference, in SCORM the not operator covers the opposite case but not the unknown case. To write code equivalent to the code above using the SCORM definition of the not operator, you would have to write it like this:
if ((NOT(person.Thinks_Mike_Is_Cool())) OR person.Does_Not_Know_Mike() ) then print "mike is lame" else print "holy cow"
In SCORM, NOT doesn't cover all the cases where mike isn't cool (person doesn't know mike or person thinks mike is lame), it only covers the explicit opposite case of coolness (lameness).
Ok, enough abstract, self-depricating analogies. How does this apply to real sequencing rules. Say you want to write a simple rule that says "unless the user satisfied an objective, disable the activity". Seems straight forward enough. Using traditional logic, you would write something like this:
But that doesn't do what you probably want it to. What that really says is "if the user attempted the objective and failed it, then disable the activity". If the user never attempted the objective, then this activity will still be enabled. To achieve what you're after, you'll need code like this:
That code says, if the objective has an unknown status (has not been attempted) or if it was attempted and was not satisfied, then disable the activity.
In summary, the NOT operator does not cover the unknown state. Sequencing rule conditions need to explicitly account for the unknown state.
HOME
