Home arrow Services arrow News arrow Joomla arrow Joomla Usability Sucks

Joomla Usability Sucks

Published on Monday, 14 September 2009
Joomla 1.6 development has been progressing at a furious pace over the last few months, past the Alpha, and towards a more stable Beta product.

But how good is it? Does it actually provide an enterprise level CMS product at the end of the day? Is the new edition actually really any value-added benefit for enterprise and government level business? Or as some have said, will it continue to be relegated to being a "hobby" project that only gets used because there are a lot of evangelists for it, and it has a lot of third party extensions attempting to meet people's needs?
Numerous documents have been produced internally to the Joomla team, all adding layers of complexity in order to navigate workflows based on a rewritten PHPGacl framework inside the core code for end users.

Absalom Media has been doing extensive research in parallel to this in regards to other workflow solutions. Now we can finally present our findings.

Over the last few months Absalom Media has been engaging in auditing of the new ACL / Access Control System which controls workflow and processing for the administration of a Joomla site. Workflows are needed within the enterprise in order to meet, control and segregate information flow by an organisation, both in terms of editing and publishing authority, and in terms of security to view sensitive information, either over a VPN, or for a member or customer.


The aim of this research was to test the validity of a visual representation within Joomla 1.6 in comparison to the usual table-based model represented within this document for an enterprise's workflow. Enterprise level workflow, in the real world builds Absalom Media has been involved with, have more than one single stream workflow behaviour, passing up from one staff member in a linear fashion to the final publisher once authority has been approved. This aspect of enterprise level workflow management means multiple levels of authority to publish, multiple editors across multiple departments all involved in passing editorial oversight to something being published. It was hypothesized that people would respond favourably to a mindmap as an ACL layout, understanding it's function without being intentionally exposed to the rest of the Joomla system. This would be a fair and reasonable hypothesis due to visual representations of business models being seen extensively across business systems through such tools as Powerpoint. There is, of course, the old proverb: "A picture is worth a thousand words".


The two mindmap visual layouts generated previously by Absalom Media were deidentified from being part of Joomla by cropping them to focus on the ACL actions and the mindmaps themselves. They were then submitted to 5 Second Test, and set to display for 5 seconds, with the end users reporting up to 5 things they remembered from each layout. Each layout was to be tracked for two months. The Site Layout test and User Layout test were assessed independently, not as a whole. All users tracked in these tests were also deidentified. No information was given as to what they were meant to be rating in order to preserve impartiality on these tests.  Raw organic data was more important than potentially biasing the users through education.

Deintification was required primarily to remove any biases had towards or against the Joomla product. As stated previously, there are a lot of Joomla evangelists which explains its high usage by small to medium web teams and businesses.

The 5 Second Usability tests were then advertised on 3 distinct groups: Whirlpool Australia's web development forum, the Australian Web Industry Association forum and via Twitter. This allowed a fairly wide technical and business spread of users from Australia, UK and the US. These tests were never advertised on the main Joomla forum, and that, in some ways, allowed a more impartial view on potential functionality.


Over the course of 2 months, a sum total of 161 responses were recorded across all tests. Of all the received data, only 3 involved junk data, with one more test attempting to attack the test analyst. The Site Layout test recieved 59 responses, the User Layout test 102 responses.  There were 47 unique keywords (including variants) recorded within the Site Layout test, representing an 80% effective response rate to that test. There were 70 unique keywords (including variants) recorded within the User Layout test, representing an 69% effective response rate to that test.

The top 10 keywords for the Site Layout are as follows:

Keyword Mention %
solutions 34 57.63%
home  28 47.46%
technology 25 42.37%
access levels 6 10.17%
publish 6 10.17%
site 4 6.78%
view  4 6.78%
create  4 6.78%
approve 3 5.08%
chart 3 5.08%

The top 10 keywords for the User Layout are as follows:

Keyword Mention %
web manager 36 51.43%
hr 33 47.14%
legal office 27 38.57%
levels 25 35.71%
warehouse 15 21.43%
cat 7 10.00%
groups 7 10.00%
marketing 7 10.00%
edit 6 8.57%
users 5 7.14%

Both results tended to follow the distributed mean bell curve in regards to user behaviour. This is a good thing. It means we can remove all the variant responses and focus on the key aspects of this data.

Further data analysis of the variant data showed only 1 respondent recognised this as a test based on Joomla, and that one user in the US has personal or business issues with the test creator. The latter was found out through demographic analysis of language present in the responses.


It would be fair to say that the results confirmed the hypothesis. This is primarily seen through the level of responses listing key elements inside each mindmap, and the editing/publishing functionality that is present with it. Most users tended to see both the editing buttons and the layout as one concise unit, considering the spread.

This has ramifications for the way Joomla is designed in terms of a backend interface. Having excessive tabling and more and more functionality pressed inside the ACL structure as per this document may actually end up confusing users. In order to maximise enterprise level functionality, and increase the potential for Joomla to be considered an enterprise level content management system, it is recommended that a "less is more" approach be taken. Through visualisation frameworks readily available to plug into Joomla's Moo and JQuery heavy backend, it would be the shortest path to an effective management solution.

Absalom Media is open to discuss with the Joomla development team in order to further this and leverage the best industry practices to transition Joomla into the enterprise.
blog comments powered by Disqus

“Intelligent discontent is the mainspring of civilization.”


Welcome to Absalom Media

Delivering web accessible websites

Absalom Media seeks to deliver industry best practice website usability and design for your budget.

As we are upgrading systems, login has been disabled.

Account management

As we are upgrading systems, account management has been temporarily disabled

What are you looking for?


Join us on Facebook

Follow us on Twitter

Talk to us on Skype

Generated in 0.36204 Seconds