Bookmark and Suspend Data issues

If you cannot change scorm versions I would suspect most tools will have the same problem, the data limit is the same for scorm 1.2.

Simple things to do are reset slides to initial state, remove any text entry questions or data entry. Use a complete trigger instead on sending quiz results. 

Look at the bookmarking data in scormcloud how much are you going over by? At least then you can see if it is possible to get inside the limits.

Also what LMS do you use it maybe possible to override the limit

I'm with Phil. Without seeing the file, tough to tell what might be going on. Questions tend to take up a lot of suspend data. Question and choice strings seem to be stored. It's pretty inefficient and can easily fill up your suspend data fast. 

Do you have a lot of questions in each of the files? 

I have no more than 20 questions and all are graded questions and no text entry. I sent my file to Phil Raymundo last week and tested things on Scorm Cloud.

Built in to the workbook are extra layers for resources ( not using the built in resources option) and I asked Phil if these could be causing the problem and he told me definitely not yet I've just deleted the extra layers after converting them to PDF's and now using the resources option it seems to have sorted the problem.

Switching the questions to freeform will send less data, however will mean the data will be less meaningful.

How much resume data are you using? Which LMS are you using?

It's been awhile since I looked at this. Questions take up A LOT of space. So those are definitely contributing. 

Extra layers could also contribute to the problem. Because the base layer and each layer has an option to resume or reset to initial state. That state information has to be retained somewhere. If the layers are complex, the state of each item also needs to be retained so it can add up fast. Unless you absolutely need to retain state, setting each slide to reset to initial state is a good practice for reducing your suspend data payload. I believe setting the slide to reset to initial state is inherited to the layers but would need to test to verify.

Hi Dean - 

Here is an example file that you can use if you run into the issue again. I placed a JS trigger on the master slide along with a display variable. This runs a check to the LMS on each slide entry that pulls up the length of the suspend data. 

SCORM_CallLMSGetValue("cmi.suspend_data").length;

You could call this up in the browser console to return the length as well.

In this simple example you'll see how goofy the suspend data construction can get. It'll flux a bit because it's compression and each change affects things. But you'll notice that the questions each add a little. Revisiting the questions balloons the suspend data a little bit. Even revisiting slides can have the suspend data adding 5 characters EACH time. 

By playing with the resume options on the slide you can experiment with the ones that heavily load your suspend data.

Good luck! I know how annoying this can be. 

Thanks Steve

Really appreciate your help. 

Dean

I uploaded several versions of the SCORM test file I posted above. Even with resume turned off for all slides, the suspend data string seems to grow about 10 characters with most slide changes. Simply toggling back and forth between the first two slides a number of times, I was able to inflate the suspend data to nearly 1000 in about a minute. That's 25% to the limit.

I'll submit this to see if we can see some changes to the way this is done. Not the first report of this type of an issue with painful suspend data behaviors. The more the better. 

Hi Steve

I've tried everything I can think. I've set it to reset to initial state, automatically decide and reset to saved state. I've removed all resource slides and saved them as PDF's to reduce file size.

The bookmark will still not function correctly.

I've added the file if you wouldn't mind having a look to see if it's anything glaringly obvious.

Kind Regards Dean

Hi Dean - 

Good/bad news. I wanted to confirm that the piece would actually bookmark and restore. I ran through the entire thing without answering the questions and I was able to get it to bookmark. That's the good news. The bad news, I ran through without answering any of the questions and the suspend data load was almost 3900. Here's some profiling. Again, without answering the questions in the first pass. This is the character count in the suspend data field when the slide is entered. So the slide data has not been committed yet.

Q1: 269

Q2: 482

Q3: 665

Q4: 883

Q5: 1054

Q6: 1270

Q7: 1440

Q8: 1555

Q9: 1648

Q10: 1834

Q11: 2055

Q12: 2271

Q13: 2464

You can see where this is going. Each question has a payload of about 200 characters. This is without answering. Let's compare with answering:

Q1: 262

Q2: 649

Q3: 924

Q4: 1236

You can see the payload is a little heavier with items selected. I got the first question right but got the rest wrong. The next challenge is what can we do about that? Since we don't have much control, we can try a few things:

1) Rechecking to make sure the restore isn't set for the questions. I don't know if this will mess up your results slide though. I'm seeing large numbers before the questions even appear. This isn't making sense.

2) We might be able to modify some Javascript where it's grabbing and stuffing the slide into the suspend data. This is a stretch... Not sure it's possible but I'll look. 

Can you send your source file to xpconcept at gmail.com? I'd like to take a look to see if there are some things we can do to lighten up that load a bit. 

It's probably filling up at about the 3/4 mark. Logically, you would think this shouldn't happen. Courses are built to have questions added. That there is an upper limit this low means something is wrong. I've built outputs for SCORM 1.2 that had 50 or more questions. We should be able to figure a way to a better configuration that won't fill up so fast.

Hi Steve

I'm clutching at straws. Each page had a hi res stock photo from Adobe which I've tried removing but still the same issue.

There is a small bit of js on each slide which brings up the highlighted resources button but I'm not sure that would take up much.

Regards Dean

Nope. The JS doesn't take up any space at all in suspend data. Nor the images. If you can, jet me your source .story file at xpconcept at gmail dot com and I'll have a peek. Bound to be a solution. Stuff shouldn't be this hard:) 

Hi Dean - 

A few things to try. One is changing the save state for each slide. I've illustrated how to do that in this quick recording. This could have side-effects you don't like. 

At the end of the video I caught an issue with extra stuff in the question stem. EVERYTHING you add to a question stem is added to suspend data for some goofy reason. 

This, for example is 221 characters. Because it's not that compressible, this will likely be between 150 and 221 characters in your suspend data. Most of that is unnecessary. 

Q1: Before carrying out a workplace Kaizen activity there are health and safety requirements that need to be considered.

Pick from the statements below which you think would need completing before starting the activity.

First trimming you could do is taking the question instructions out of the question stem and adding to another text block outside of the stem. This would give you a lot of savings per question. The other option is to take the ENTIRE thing out of the question stem and add it to a text block and instead leaving just the question identifier Q1, Q2, Q3, etc.. And move that identifier block off of the slide stage. 

This would save you a lot in your suspend data. Under normal circumstances this would have some consequence for accessibility but SL's output isn't tying those elements together correctly anyway so it doesn't matter:)

Hi Steve

I've just watched the video and wanted to thank you for your help.

I'll give everything a try and will let you know how it goes.

Kind Regards Dean

Great! You have a few options. Cleaning up the question stem is probably the best balance. Will give you between 100 and 230 characters a slide back. Might be enough. If not, removing the state storage for the slide and only allowing forward travel might be the way to go.

Hi Steve

I've reset each slide and removed the instructions from the boxes and everything seems to work ok now.

Many many thanks Dean

Hi Steve

I sent the zipped file to the customer and they replied with the bad news that it still isn't working properly. I then split the file into 2 parts and tested part 1 on my platform (Talent LMS) and also on Scorm Cloud and both times the bookmarking would not work correctly.

Just to recap: I removed the instructions from the question box, I've reduced the size of the questions where possible and I've reset everything to initial state as instructed but I'm still having issues.

Any idea what next?

Dean

Hi Dean - Can you provide more info on what the issues are? Is it bookmarking up to a point and then stopping? For example, when launched can the user go through the course a few slides or questions, quit and then resume or is that failing too?  Is this the only Storyline course that behaves this way?