Adapt Learning Community: Workflow Discussion

Skip to main content

HI

Lets use this thread to collate ideas on what workflow functions we'd like to see within Adapt. These are functions that aid the production process including asset management, version control, review and feedback, publishing etc.

Thanks for splitting this out, Al.

Workflow is one of the key aspects, which I think can make this tool very powerful. We have chosen to go for a server based authoring tool and to build the architecture so it can accommodate workflow plug-ins. The idea is that we will supply some standard workflow functionality but that anyone can also extend and change this as they see fit by writing a different plug-in (and sharing this with the community).

I am in the process of revisiting this concept document about the authoring tool. My plan is to write up a sentence for each of the bubbles to explain what they mean.

The things most relevant to this discussion are in the bottom half of the document, namely, the 'user types bubble' (which identifies the system actors) and the 'workflow' and 'user actions' bubbles.

We have identified the following actors:

Content Editor
Reviewer
L&D Manager
Guest User
Developer
Demonstrator / Trainer
Administrator

We have identified the following key workflow stages:

Plan
Build
Review
Release
Maintain
Archive

I will try to get the document with the concept diagram and explanations done before the end of the week.

Show parent | Reply

Server-based is great as it makes it much easier for different people to collaborate on the e-learning projects.

Possibilities for one person to work on texts while another works on media on the same e-learning "page" or "module" for instance is great.

I have a lot to read up about but I like the way this seems to be headed....

Review in the system with comments from reviewers connected to the correct content ("Page", "module") is a very good way to handle customer input.

Liking the Versioning, Snapshots and Rollback part that I see in the PDF a lot.

There is a little ring that says Export. What kind of exports have you planned?

Are there plans for instance for exporting speakerdata (text , soundfile and metadata) that can be used in a sound studio to create the soundfiles for the course.

How about exporting texts for translation?

Another export that we have in our own system and that is very good to have is exporting the answers to questions.

Show parent | Reply

Hi Martin,

thanks. yes, we are very excited about the scope of extension a server side tool offers us, especially with regards to the workflow.

The first planned exports are all using the standard Adapt output (i.e. they are an e-learning module with the functionality that the Adapt framework offers - demo to follow soon) but have different tracking options. Out thoughts on this are:

no tracking: a standard, interactive package without any tracking functionality
SCORM 1.2: same package but with SCORM tracking
Tin Can: same as above but Tin Can instead of SCORM

Texts for translation is a very good additional thought, thank you. This would work especially well, if we were able to also import the translated files. The translation agencies we work with tend to deal in XML so that may be the best option for the output format.

We are building in other language capabilities from the start so this would be a great addition.

Show parent | Reply

Same here in Sweden. We export text as XML, send it to the translation agency. After we get the XML back we then import it to our current tool. This saves a lot of time on every project that needs translation.

As Adapt is built on responsive HTML5 the fact that different languages take up different amount of space on the screen should be a non-issue.

I will be very happy if this feature is implemented.

Show parent | Reply

Hi Sven,

Perhaps its too early but I'm wondering if you should also consider including the Caliper framework being developed by IMS?

Although it should interoperable with any RDF triple, it promises the entity model that is missing from xAPI. IMS are expecting 100% adoption in LTI partners, so its worth thinking about.

http://www.imsglobal.org/IMSLearningAnalyticsWP.pdf

http://developers.imsglobal.org/catalog.html

Show parent | Reply

Hi Jason, thanks. It would be great to get a sense of whether the existing IMS standards are used much in education. On the commercial (and largely private sector focused) e-learning production side I haven't come across any real demand for this. when tracking (SCORM) isn't required we usually just ship a zip file without schemes or appliccable standards.

Any views are very welcome.

PS thanks for pointing this out, though. I'll read up on the new standard.

Show parent | Reply

Hey,

Currently Adapt exports its data in JSON format. I did some research around translation companies before we moved away from XML - the outcome was that most translation companies work with JSON files. They don't charge extra which is also nice.

One feature I would love to see and something that we're planning on putting into our folder structure is the concept of having one course with multiple language assets (including images and data). There would also be a global assets folder for images/videos used across all the languages.

Another is to put in RTL language support. We've done a few tests with this and due to the nature of HTML and it's support for RTL we got some good results.

Thanks,

Daryl

Show parent | Reply

One benefit of an export option is that it can export only the text you want translated.

Currently the JSON data holds configuration information as well - and it's not so obvious when strings are for configuration only and should not be translated. So you might well find you get all instances of 'true' translated to 'vrai' ;-)

In XML you can separate the two with a simple rule like only ever putting configurations strings in attributes and on-screen text as node values within CDATA tags - but if you're going to put in an 'export for translation' you might as well ensure the output only contains text that should be translated.

Equally there's actually a specific XML format (XLIFF) that the localization industry uses.

I would say that translating the JSON is probably a good-enough-for-now solution and that adding in proper export/import functionality should be something for future development, ideally after some conversations with translation companies to find out how best to do this so that they can translate as easily (and therefore cheaply) as possible.

One smart feature I've seen in a CMS is that you can do the whole thing as a round-trip within the application itself i.e. you click a 'translate this course' button, select your language, you're then presented with a list of companies/prices, you choose one and the necessary files are sent off in the background to your chosen company... and they submit them back through the app which then notifies you when the files are ready for review.

Show parent | Reply

On the RTL front - I strongly recommend that this is done by someone with experience with working on RTL text. It's harder than you think, believe me!

Show parent | Reply

An early incarnation of Adapt has been used to produce a course in Hebrew (which is RTL) and by all accounts it was reasonably straightforward. It might be worth getting Hertzel, the guy who produced this course, involved in discussions.

Show parent | Reply

I know an option to export 'speakerdata' would be a popular option amongst PMs here at Kineo.

Show parent | Reply

HI

I wonder whether it would be good to split out the developer role to ID and Developer. I think to have a simplified developer environment with limited editing/design functions, that IDs can use to create high-level course structures based on page templates with placeholders, a comments function to explain ideas, combined with the other importing and exporting functions could be really useful.

Reviewers (clients) could then add comments and once agreed it could move into development. If we could maintain comments an audit trail of comments/amends even better.

Show parent | Reply

Hi Al,

I think we considered the ID functionality under the 'Content Editor' role/actor in the early workshops, albeit that terminology doesn't make it obvious.

There are a list of user stories, which are currently a bit lost in one of the architecture threads. I have a task to split them out. In the meantime, please follow the link.

Hope this helps! It would be great to add to the document and put it into the version control repository.

Show parent | Reply

Hi Al,

I agree, be great if we could have something which allowed us to produce elementary wireframe type layout/scopes that could be commented on by client. We scope like this for Adapt with Balsamic and we’ve found many benefits, but the two biggest are that the client gets early oversight of layout/treatment that is more immediate and easier to understand and we also use them as graphics briefs for the art directors.

I’ve uploaded one below to show the type of information we include. Also be great to have something which allows for the creation of a top level structure which you could use as a course schematic which you then built upon and allowed for the moving of pages or even blocks from one location to another.

Cheers,

Paul

Scope wireframe

Show parent | Reply

I haven't seen this discussed before but I might have missed something.

An easy way of selecting what content in a course that is "active" would be great.

In eStudio we have a tickbox for chapter, page and element so you can "remember" for instance a chapter.

This is very useful for testing different ways of presenting the same content or for having a course with most content the same for everyone but one or two chapters that are specific to specific groups.

Show parent | Reply