By Sill Pontes, Senior Open Source Instructional Designer,
and Gilney Carvalho, Software Developer
Isn’t it the dream of every teacher and instructional designer to make people’s eyes shine? To allow everyone, regardless of who they are, a path for happiness and enlightenment? So it was no wonder why the Open Source interactive standard H5P was met with such high praise.
The sheer amount of features and interactivity types is magnificent. Plus its learning curve is the least steep I’ve ever encountered for authoring tools. Sure, over time you start looking at it more rigorously and begin to nitpick a little. Não é grande coisa, no big deal!
And then, you get the Moodle 3.8 and the native integration with H5P. Now you have us seriously doubting.
Now, I’m sure the first rebuttals to this article are going to point out that we’re on the early stages of the integration, and how these things need time to properly mature. We wholeheartedly understand. We need to remember and accept that new integrations or solutions need time to “mature” and correct possible failures. The problem is, there’s vast and flourishing Moodle communities around the world that don’t get a seat at the table. Moodle has done a lot to be an inclusive project, at least in terms of its features, and there are initiatives such as the Moodle Users Association, but access to them can still be prohibitive on developing countries. In our particular case, we feel that we need to voice our concerns with the community any way we can. These are the issues we’ve been discussed the Moodle Brasil community in the past weeks.
At the very least, we hope our opinion help avoid some of the dark fates we have witnessed in the past. And there’s plenty. Many of us have been in the fight for decades now.
Until today, the H5P plugin has been very successful at least until 3.7. In the last 12 months we have witnessed the plugin growing with more intensity and quality, many new features, beautiful and wonderful HTML.
The more veterans among us are deeply fond of H5P because it allowed us to get rid of SCORM packages. SCORM is the sand in the Havaianas of the instructional designer. Complicated to update, even to fix a typo; disappointed level of compatibility (despite the marketing efforts), useless without an internet connection; and don’t get us started on its security vulnerabilities or its far-fetched mobile and social learning possibilities.
We can claim to have been early adopters of H5P, having workshops in Brazil MoodleMoots since 2017. We realized its potential and we learned to trust the team behind it. Many technologies look great at first sight, but you can easily see the shortfalls when updates stop coming in steadily, or if the community doesn’t grow at a healthy pace. Of course, the same goes for Moodle, whose success on these dimensions are widely known.
The integration of H5P in Moodle 3.8 made our hearts pound with high expectations for 2020, and as such, with any passion we ended up overlooking some clearly outstanding issues. We ignore them and looked the other way in the name of passion. But the passion fades and the problems stay. Ignoring them would only be a disservice to the Open Source learning movement.
As companies and freelancers we face a complex challenge: We need to defend and promote the technologies at which we excel, but we need assurance that any outstanding issues and concerns will be properly addressed.
Out for our blood
A curious development covered the national papers and even made international headlines in 2015, when a team of anthropologists finally agreed to return the blood samples they dishonestly took by "tricking" children of the Amazonian Yanomami tribe in 1967, which was used for genetic testing without their knowledge or permission.
Our first concern with H5P could sound too purist or radical. I particularly don't like to feel stuck inside a plugin, an app or a piece of software. That feeling, "stuck," can take many forms: In the sense of insecurity, in the sense of high monthly fees, in the sense that the software has been discontinued. What to do now? I've been burned more often than my pride would let me admit by fantastic-looking apps that at first release a very limited "freemium" version, and then expect us to pay for a premium version that is supposed to have all the missing features.
Sure, it's fair to pay for a complete version. As long as the version is complete. We are often responsible to guide our clients on the purchase of educational technology. So you can imagine the horrors of having offered a product that failed to deliver, or just straight misrepresented itself.
In my madness and daydreaming I am wondering: Will any of this happen to H5P? It is a difficult question to answer, and its open source license could even muddy the waters further. For the record, my fears do not stem from the H5P team or the technology. We believe H5P has shown nothing but maturity and baggage, and barring a cataclysmic event, the interactivity standard is here in the hybrid education market to stay. And yet, I believe there is a lot of room to grow, a lot of opportunity to capitalize on, and a lot of high expectations on which to deliver. If it doesn't, the disappointment would be enough to put me to tears all month. From the bottom of my heart, I hope the intention of the H5P project is to generate more autonomy and not to get us "stuck."
I'm still in love with H5P as I was since we first started, but now I'm wondering about our future.
Too green for prime time?
Now let's focus on our second concern: If you do not have a lot of experience with the integration of technologies, brace for a world of confusion.
It is clear that the H5P baked into Moodle 3.8 is "consumer only," so you should not expect to be able to install Moodle and go straight to your first Interactive video, snappy Drag and Drop or Memory game. There is not even an H5P Activity. All you can do is insert it on the Moodle HTML editor. So there goes our plan to have the built-in H5P linked to the Gradebook.
What I wasn't aware of until late in the game is that you can only include an H5P activity (H5P toolbar icon) if you have the Atto editor activated. That prevents the instructor from dealing with code and making sure it is placed properly. Atto is Moodle's default text editor, but many prefer the simplicity and sturdiness of TinyMCE, the default text editor of several other open source projects. So you cannot take Atto's availability for granted.
The rest of integration proposal (it should still be considered as such) doesn't make much sense either, and at every turn the lingering sensation is immaturity. The official developer documentation on H5P (fairly out of date) is straightforward in its own incompleteness as well as on the "MVP" status of the integration. Its introduction states:
«Since there are different approaches, in this document we will explain the approach that we believe is most suitable for Moodle 3.8 given the complex context.»
We just don't have the confidence the approach was the right one.
A scenario that's come up among many of us is the following: We're running Moodle 3.7 with the latest H5P plugin. Things are smooth. But then it comes the decision to upgrade, and we and our clients get the question: Is the H5P without its plugin any good?
Here's a practical example. This is a result of comparative tests using H5P on Moodle 3.7 and 3.8. We have an H5P Interactive video uploaded to our Moodle. We're not afraid of a little bit of HTML, and its inclusion (only on the Atto editor) is in fact beautifully simple.
We switch from HTML to WSYIWYG view. So far so good.
We already had some H5P objects on our site. Grabbing them for the Atto editor by either embedding them or using the toolbar button is straightforward, bonito.
And even if you have the .H5P file on your desktop, there's no trouble and the loading is fast.
Now I'm looking at it. Nit-picking me. Should I capitalize the heading or go for a darker font? Or you know, a meaningful change due to an update on the topic at hand. Say, a new law has passed that affects some parts of the content.
Let's try to make an edit:
So now I have to go back to where I started, my Moodle 3.7 site, to access an H5P editor. Then I save and download the package to then upload on Moodle. Wait a minute… The flashbacks…
SCORM: My death was… Greatly exaggerated
Missing out no matter which path you take.
It was at this very moment that I remembered the SCORMs of life… And the passion is over!
Alright, I have bit of a flair for the dramatic. It's only to get my point across.
Perhaps our most urgent concern is the development path Moodle and H5P would take in this integration. It is clear it will take a long time to have the full H5P experience inside Moodle. Which leaves us with a lot of questions.
- Will the implementation focus on supporting H5P natively? So will the H5P plugin stop development? In the mean time, will some of the known issues with the H5P plugin (among many others, the link to xAPI listeners and LRS) remain open?
- Would the existing plugin functionality be compromised by trying to extend the integration into a limited framework?
- Will the Moodle plugin ultimately stay around?
- Will there be any harmony between the built-in and the plugin components, especially as next Moodle versions arise? The proposed mockups, while on a very early stage, appear to be a noticeable depart from the UX in the H5P experience.
Only few right answers will ease our concerns!
The developer documentation for the H5P - Moodle integration offers no comfort. It argues that
«The integration of the H5P activity with editor decreases the range of Moodle users that benefit from it since it does not allow the use of H5P content anywhere and limits it to the specific use of the activity.»
Basically, it confirms that a built-in H5P editor is far away on the horizion. The doc does on to say that even if they do try to provide editing options, the project risks to always remain behind the regular H5P editor. The icing of the cake is something no SCORM survivor wants to hear: "There is a serious security problem regarding the qualifications that we cannot solve from Moodle, so we would have an activity with unreliable claims."
What is there to do? Words of caution and action
It seems the answer is obvious and has been with us since the beginning. Put simply: Do not stop using the H5P plugin in Moodle. As long as it continues to be developed, your designs and your students will be spared from the complications listed below. And yes, some of the small performance benefits of running an immutable H5P activity added through the Atto editor in Moodle.
It might be easier for you to influence the development of Moodle and H5P if you have the ability to hear from them if you are near their close circles. For the Moodle Brasil community, trying to get Moodle's attention has been a challenge despite the size and vibrance of our local community. (We hold two jam-packed Moots every year.) On the upside, we have started engaging with the H5P team, and so far there seems to be a great atmosphere for collaborating with us in more proactive ways.
Here are other avenues worth exploring.
- Support H5P development. Either sign up for H5P.com to get a dedicated platform to edit and host content to display on any compatible LMS or content platform, or consider joining the H5P Supporter Network. (Admittedly, the lower tier could still be pricey for us and other countries.)
- Be vocal on the issues. We think more voices should be heard, from many angles. We welcome similar posts or even responses. There are ongoing conversations on the Moodle Tracker and the Moodle Forums. On a more productive level, submit your issues on the H5P GitHub. And why not, provide a fix to an issue and "pull request".
- Stay current. Testing has taken place at h5p.moodledemo.net, (use 's1' or 't1' and 'test' as username and password, respectively) where you can compare the functionality of both H5P avenues.
- Sign up for global H5PCon 2020 in Sao Paulo. Kidding! It will take place in Wisconsin starting May 18. Early bird rates are available until January 31.
Stay tuned, tests things thoroughly and do report your results. We are not sure what will happen, at least today the plugin is there. I think that would be it for now, we will wait for the news during the work.
Sill Pontes and Gilney Carvalho. TODRAW.COM.BR