Thought I'd start a thread on new interaction ideas with a rough idea of how they'd look in full and then compacted form and maybe get comments as to whether worth it and how easy it would be to make i.e.
component: imageSlideOut
full
image button(s) which on click would slide out (make appear on IE) a text block to the right, stacked vertically
compacted (i.e. layout: left or right or mobile)
image button with pop up text block with close button
Hi Adam,
Personally, I like the idea of sharing opinions on new interaction ideas.
As a designer, my first questions would be what role would you see this component playing? Is there a key gap it can fill that existing components can't? (though to be honest, variety is also good to keep people engaged)
If you're happy to put one up, it would be great to see a mock up, to help visualise it.
Hi,
yapp - I like the idea of sharing interaction ideas. Even on a brainstorming basis - without restrictions in thinking.
I am not a Coder - more a non Tech Guy - and I am always impressed what can be done with todays possibilities. But quite often I have ideas for (especially) my eLearning Course creations which I am unable to produce with my limited skills.
So this Theat could be (if it works...) a good source of inspiration for the Developers here to push this great project to new levels.
Right now I am working with this framework to produce a course which was completely in Flash - so I have to reproduce it to be multi Plattform compatible.... Lots of Videos and interactions with Videos (Keyframe triggered Text Boxes and Voice Overs and so on).
I know that there are JavaScript Solutions to controll Video - but this is far away from my Skill Level.
But Video is great media for eLearning - so if there would be a solution... GREAT.
Maybe I'll come up with more ideas.
Cheers
Thomas
As Thomas pointed out, the ability to use basic triggers with video would be a useful feature, not least for the purpose of creating engaging and memorable branching scenarios where learners get to make decisions and explore the consequences of their choices etc.
For anyone looking to use the Adapt framework for language learning, flip/flashcards would be an invaluable interaction type. Example: http://quizlet.com/14020293/flashcards
Finally, Drag and drop sequencing interactions with text and/or graphic elements (like those found in Storyline, Captivate etc) would potentially make for a more engaging way for learners to explore and test their understanding of procedural/process-oriented knowledge.
Best,
Bjorn
Drag and drop would be great though I would be interested to see how this would work in a mobile device such as a smart phone.
<quote>Finally, Drag and drop sequencing interactions with text and/or graphic elements (like those found in Storyline, Captivate etc) would potentially make for a more engaging way for learners to explore and test their understanding of procedural/process-oriented knowledge.</quote>
Kind regards,
Mark
Hey,
Drag and drop unfortunately is a difficult component to implement. On mobile devices you quickly run into height issues. We've prototyped a few ideas and would like any other ideas on how this could be displayed differently on mobile devices.
Also, we've just posted this on another forum post:
https://community.adaptlearning.org/mod/forum/discuss.php?d=183
We're looking forward community members to give ideas for future components and we will build some when new developers come on board.
Thanks,
Daryl
Hi Mark -
We are working on a matching template which would change itself to drop down when we go on smaller screen sizes. The component is nearly ready just trying to find a way to tight couple these two parts together. Currently our matching template calls the drop down template developed by the core team but reverse is not happening so we are working on that.
These templates are ready for testing and implementation -
https://github.com/aniketdharia/adapt-clickandlearn
https://github.com/aniketdharia/adapt-flipCard
Cheers!
Aniket
@Thomas Schmidt I'm doing the same as you with converting a flash course and i've skipped around the video triggered events by creating them as a stand alone video with the animations/texts embedded in the video as especially with a mobile view the video element would be the only one on screen
Hi,
hmm - yeah thats an approach I'am thinking about too. I have some good skills in video editing and AfterEffects - so this would be a way.
So for my project, beside the Videos as a "Full Screen" elements, there will be pages with 'just' text and some graphics / photos, I will use the interactions and stuff which is already availiable in the framework - and I will add on some Tests and Quiz Functions to make this course worth to be imported in my moodle LMS with the SCORM capabilities.
But for the future - and to have easy ways to add new stuff to my course / change things - it would be cool the have interactions from the framework to trigger Videoeffects... I dont want to reedit my AfterEffect Files for small changes in Textboxes for example.
By the way - how will you embed the FullScreen Video in a article? One article, one block and one component? Does this work?
Cheers
Thomas
yes it seems to work as article-block- full component but need to play around in how it first displays in different browsers as noticed some odd effects when I was testing on another issue the other day when the video is first drawn. Unfortunately didn't have chance to investigate and test at the time.
Hiya,
Any chance anyone can help me with a block idea? Basically, I'm a non-techie, who can only do the verybasic stuff, and have been looking for ways to 'personalise' learning for different users through Adapt.
Please be as critical as you like :)
FYI - this is just my personal idea and has no link with the Adapt team.
Hi Nicola,
We have developed Role Selector in the internal branch of Adapt before we donated the code, and I'll be looking for opportunities to refactor this code and make available to the OS project.
In summary, you can swap out components, blocks, articles and even pages depending on the choice made by the learner at the menu level.
Hi Paul,
A role selector is a useful feature for dividing content - but it still divides people into large groups.
What I'm talking about here is a more finer grained approach - a way to really get each learner to identify with the content at a level deeper than the menu, and throughout the course. It's not just branching - it's fully responsive to the individual learner.
Say the content was an unmarked question, e.g. What's your biggest leadership challenge?
Each person would choose an option from a given list.
e.g.
- Difficult conversations
- Capability conversations
- Change management
- Performance management
The content below the block then re-draws and re-renders according to what the learner selects.
This is just one way it could be used.
Hey Nicola,
This is a great idea and something that Adapt is fully capable of doing due to it's flexible/live rendering. HTML is great at displaying content so an extension (I think clearing up a component and extension is needed - will put that below) like this would be great.
I could see it being used with levels in Maths or English. As the user moves through the course they have a filtered version of the course based upon their results.
Keep the ideas coming everyone!
Brief mention about component/interactions and extensions
Components/interactions are called Components in Adapt - we should keep it clear that we are talking about Components and not interactions. This will stop people being confused between the naming conventions.
Extensions - These are extra features like an assessment, tracking, trickle, blockSlider. These enhance Adapt's core and are not components.
Thanks,
Daryl
Hiya,
Role Selector is probably a misleading term, Nicola. Essentially it swaps out one component, block etc for others based on the choice of the user. Isn't that what we're saying here, but rather than making that choice at the menu it's made within the preceding block? In essence like a little sub menu for the block or article?
Also, just flagging based on our experience of using it its best to make this content non-mandatory when it comes to tracking (obviously not everyone will necessarily see everything).
Hi Paul,
I take your point about role selector being a misleading term as there are many ways you can divide content before the learner even starts (e.g. by the time they've got available) and we should be exploring them more.
There's great value in reorganising content at a menu level - but I think there can also be equal value in reorganising content as the learner progresses and gets a feel for the course.
My original design was me trying to repurpose existing functionality (non technie strikes again). I will take the design back to the drawing board. I think there is potential for both mandatory and non-mandatory uses.
Sounds like I'm talking more 'functionality' than 'component' or 'block'. I will keep you posted.
Hi
I really like this idea - it allows learner-led learning which is great for on-job development, and could be good for enabling students to address specific issues.
It would also allow a diagnostic test to suggest a route through a set of material, though giving a prescription not always welcome, so it may be useful for learners to be able to consent or amend the route.
Hi,
I'd like to see an Open Input Component, that allows a user to enter text into an edit box on a page, in a block.
Then that text could replayed back to the user on another page somewhere else, well technically any where on that course.
It's very simple but effective, you gather the thoughts of the user, then replay what they wrote somewhere else, usually comparing it against best practice.
There could potentially be several of them in a course, so there needs to be a way to uniquely identify each input box and match it to the output, where input1 displays in output1 and input2 displays in output2.
Hi Mark,
I like open inputs too, Do you find learners get a bit miffed when we can't save their responses between sessions though due to the 4K available in Suspend Data? I wonder whether we could provide the option to save this information via cookies /device storage? I don't know if there would be issues with that but if we could it would help address the one grumble I get about those question types.
If you were publishing for SCORM 2004 you could use the 'cmi.comments_from_learner' data model element to store this info - you can have up to 250 comments of 4000 characters each, if I'm reading the spec correctly.
It's slight mis-use of this data model element, which is really 'intended to provide feedback about the SCO or the learning experience with the SCO from a specific learner'.
Whilst the SCORM spec sternly warns you that 'using this data model element for other purposes may adversely affect interoperability' I'm sure you'll be fine. I won't tell if you don't ;-)
If SCORM 2004 isn't an option you could have a config option to fall back to localstorage/cookies
Hi Paul,
In the past we have used cmi.suspend_data where moodle has a limitation of 4096 chars.
We have *cough* modified Moodle to make that restriction go away read about the issue and solution here
https://moodle.org/mod/forum/discuss.php?d=252143#p1093368
It's a nice to have to store and retrieve the data between sessions, but not a must have.
How about a geolocation-based component that only completes when it detects you are in a specific location? Would have to be mobile-only of course ;-)
OK it's not an interaction, but something I would like to see is some method allowing authors to use variables in the course text.
For example, when running content from an LMS, the user's name is made available to the course via the SCORM API.
It would be really nice to be able to easily use this within the text e.g.
'Hello [FORENAME], welcome to this course on...'
Really like the idea of using variables, Matt.
What other information can the course access through the SCORM API (names, date, time left to complete...?)
Sadly, not much more than that.
You can tell whether they're visiting the course for the first time or have previously completed it - but that's not really a variable, more a condition.
Still could be nice for added personalisation e.g. having the menu text say 'Welcome, Nicola!' on first entry then 'Welcome back, Nicola!' on subsequent visits.
You can tell what score they attained in a previous attempt at the course.
You can also see how much time in total (i.e. across all sessions) the user has spent on the course up to now, but I can't see how that would be that useful/informative!
I'll have a think and let you know if there's anything else.
The main downside is that a large part of the data model in SCORM 1.2 is optional so you've got no guarantee that the LMS will support it. Note that the ones I have referred to are all mandatory.
The availability of variables would be great.
- cmi.core.student_name
Would be great for personalization of the course.
- cmi.core.student_id
Would be great if you need to have a link to some other activity outside of the course that still needs to know who the user is.
- cmi.core.lesson_status
Could perhaps be nice to know. If you want show something special to the user after he/she is passed.
- cmi.core.entry
Would be nice for that "Welcome back message"
The ability to save/retrieve information in cmi.suspend_data would be great
The ability to save information in cmi.interactions would be great. Support for this in all questiontypes and the ability to select wheter a component should save this information or not.
BR
Martin
Great idea Matt, I would definitely support using variables. Like Nicola, I would also be interested in learning what other information courses could access through the SCORM API.
BR
Mark
Not strictly a component, but how about a search field, if Adapt doesn't have the functionality already?
Say I'm a learner, I've gone through the course but I've got this burning piece of information that I can't quite remember. It would be great if, instead of jumping straight on a search engine, we could have a search field to go through a list of results and either:
- populate a list of text results
- populate the components themselves (maybe excluding tracked quiz questions?)
Hi Nicola,
In the internal branch of Adapt we developed a search feature which would return a list of blocks, ordered by relevance which the learner click on to then navigate directly to content (content yet to be visited or that was locked in some way was returned, but the learner couldn’t navigate to it via the results list).
We achieved the ordering of the results based on relevance by allowing up to five keywords to be added to a block, with the first word carrying more weight than the second and so on when the search algorithm was applied. This, plus the number of words searched for and/or the absence of keywords is what was used to return the results list and order them by relevance.
We didn’t adopt an approach where we would search the entire JSON because we believed that we could return more meaningful results on a given subject with the approach described above. Also, we didn’t tag meta data to components because A) you’re going to take the learner to a block, not a component anyway and B) I don’t think that you should be dealing with massively different subject matter in a single block anyway, so the metadata should apply regardless.
I’d be keen to carry over this feature to the open source branch but be good to hear if there are differing views on the functionality we should look to adopt or whether this approach works for most people.
Hi Paul,
Thanks for that.
I'm all for functionality that helps us create a more user-friendly learning experience. Personally, I think this would be a great feature for open-source.
