David Scotson

The Ilias LMS apparently has QTI import and export working (as of version 3.2). Since it's GPL'd and PHP it's probably worth a look, though I think knowing German would help too.

There is some brief mention of this in their forum.

I believe this is caused by the character set declared for the page. If a browser detects characters outside the range of the specified character set in a form they will either replace or delete them. Some browsers replace them with question marks (?) others replace them with HTML entities.

I'm a big fan of the Glossary, Activity and Resource linking but I think they can be a bit intrusive when they link repeatedly to the same term.

The preg_replace function that is used to make these replacements takes a limit parameter that controls the number of times that the matched words are replaced with the hyperlink. I propose setting this to 1, so that only the first occurance of a word in each processed section will be linked (e.g. once per forum post, glossary entry, resource etc.).

current:

$text = preg_replace('/('.$list_of_words_cp.')/is', $href_tag_begin.'$1'.$href_tag_end,$text);

proposed:

$text = preg_replace('/('.$list_of_words_cp.')/is', $href_tag_begin.'$1'.$href_tag_end,$text, 1);

Does anyone have any thoughts on this? Obviously this limit doesn't make sense for all filters but are there any others apart from the three mentioned above that I should change?

Somewhat related to this topic:

Evaluation of Distance-Learning Environments: Impact of Usability on Student Performance

I seem to be struggling to say anything on this topic without offending someone, but apparently I've outdone myself here. I apologise, but hopefully can explain that I did not meant to cause offence.

The bikeshed thing is an empirical observation of the social side of open source software, and if you read the link you'll see that, despite the informal tone of the piece, it is actually a positive encouragement for people to push forward and make changes despite the fact that small (though not trivial!) changes can often create the greatest amount of discussion.

The nipple remark was, I thought, a common yet colourful way to concisely sum up why it is unwise to follow the standard application of 'intuitive', meaning inborn, instinctive knowledge, to highly complex software situations. It is better suited to describing natural phenomenon such as breast-feeding, where the young child truly responds intuitively without having to learn from experience or be taught what to do. When people use the term to describe computer interfaces they do not generally intend to suggest that a child raised by wolves would have no problem understanding it, and if two different people claim that two competing options are both more 'intuitive' then there isn't really any constructive way to proceed unless surveys or research is conducted. That, of course, is an admission that the 'intuitiveness' lies in the relationship between an individual and the software, not in the software itself.

With that warning on 'intuitive' and 'learnable vs. usable', and the general warning that seemingly small changes can often lead to unexpectedly heated debate, I had hoped to avoid arguments and offence, as people argue about what is more 'usable' or 'intuitive' without a shared reference or context for these terms. Obviously that intention has somewhat backfired.

Please (anyone) feel free to rephrase the entries to better reflect this or delete them if they are unsalvagable.