Many of the learners that I worked with would use the word long as a pejorative, on its own. “That’s long!” would be a singularly damming indictment of some activity that they were asked to engage with. The learners I worked with almost always preferred heuristic solutions over analytical solutions to any challenge presented to them; short cuts in other words.
The shortest short cut that learners used for answering questions was, it seemed, to get someone who already knew the answers to the questions to give the answers to them. In a world where search engines do exactly that, give instant answers to questions, why wouldn’t this approach seem to be the obvious one? To a typical learner of today, a teacher is expected to be a sort of walking search engine that can provide answers, answers that need to be remembered for just the interval between being told them and being required to repeat them under test conditions. The learner’s expectation of the teacher is for the teacher to tell them what test questions will be asked and what the answers to those questions are.
I am of course exaggerating, but perhaps not all that much. Most learners were actually prepared to engage with procedures required to determine answers rather than just be given answers, but for a significant number of learners, only to a certain extent. The certain extent was that these procedures were broken down into a series of steps, where each step was effectively to ask the teacher ‘What do I do now?’ and then do that. If any step required reference to an earlier step than the immediately preceding step, that requirement would render the entire procedure long. The ideal was that each step, once completed, could be forgotten immediately thereafter.
From the perspective of the person undertaking the procedure, a discrete and acyclic set of simple steps would indeed seem to be the fastest and most efficient way of completing the procedure. The complication that this way of completing the procedure would be of little value in producing recollection of how to repeat the procedure or insight into how to complete a similar but non-identical procedure was not easily appreciated by learners, because that would involve them appreciating that they were completing the procedure not for the sake of completing it but as preparation for later repetition or adaptation, and to be thinking ahead about something that wasn’t even happening yet was most definitely being long. Whatever needed doing in the future would be done then, by the most direct and efficient short cut available.
From the perspective of having successfully learned something, it is obvious how sticking to immediate efficiencies all the time when undertaking tasks is hugely inefficient in the long run, because it turns out that there are many, many inter-dependencies between the tasks; completing each task as quickly and efficiently as possible minimises the opportunities to recognise task inter-dependencies as all attention is reserved for completing the current task only. Avoiding anything long ends up resulting in so many repetitions of short tasks that the total time expended to complete anything of significant complexity ends up being very, very long indeed.
Physics (and science generally, but not to quite the same extent) is almost entirely concerned with characterising inter-dependencies (which is what laws of nature more-or-less are), so learners who deprioritise inter-dependency characterisation are not just choosing to learn about physics in a drastically inefficient way, but to learn about it in a way that resists any sort of understanding of the subject other than as some sort of esoteric doctrine.
Some learners do engage with characterisation of inter-dependencies. They seem to have an intrinsic motivation to do this, and I think that these learners see that engagement as a kind of game.