Over the past year I have become more and more interested in the nature of technologies and precisely what we mean by the various terms we use to describe the kind of things I'm mostly involved in: technology-enhanced learning, learning technologies, e-learning, m-learning, etc, etc, etc. I've come to the opinion that pedagogies are as much learning technologies as learning management systems and SCORM-compliant learning objects. This perspective has some interesting consequences.

As an object, a stick is not a learning technology. It becomes one when we utilise some aspects of it and the things with which it interacts to achieve some goal. For instance, if we use it to point at something then we are using its thinness and pointiness, combined with our knowledge of how people understand such pointy things and the value of drawing attention to things in helping them to learn, to help achieve a learning goal. If we use it to rap the knuckles of an inattentive learner then we are using its length and hardness, combined with our knowledge of how such aspects can cause pain, combined with a belief (however misguided) that such an action can help people to learn, to help achieve a learning goal. It is the same physical tool, but in each case it is a different technology. Kauffman proves that even a humble screwdriver can become an infinite range of technologies, depending on how it is used. Arthur explains this: a technology is the orchestration of phenomena to some use or uses.

Almost always, a technology an assembly of other technologies. A car, for example, has technologies of seating, steering, listening to music, doors, engines, spark plugs and so on, each of which is composed of and deeply interrelated to a multitude of further technologies. It exists as part of a larger technology of road transportation that includes the road itself, road signs, driving techniques and rules of the road. This leads to another important point: that technologies do not have to directly involve physical nor even virtual tools. A driving procedure, a traffic law, a management process, a set of policies, a means of timetabling (whether computer-assisted or not) are all technologies. Indeed, as Franklin observes, there are even technologies of prayer.

Pedagogies (as distinct from the broader subject of pedagogy) are repeatable, observable, abstractable  processes. They are technologies. If we employ a process to help people to learn that we can abstract from its context, then we are using a learning technology. We orchestrate phenomena (e.g. that people seem to learn better if we give a top-down overview followed by details, or if we employ a learning cycle, or if we start simple and build up, or push learners a bit but not too much beyond what they already know) to achieve a purpose (learning). In this sense, anyone involved in education or training is doing technology-enhanced learning.

Most of the time, we use pedagogic technologies with other technologies - a stick, a whiteboard, a classroom, a timetabling process, email, a learning management system, whatever - as part of an assembly, the central goal of which is to aid learning. It's not just about learning though. In an educational system, there are other goals and other technologies with which we need to interoperate that move things beyond simply learning, from union agreements to government policies, from timetabling rules to classroom layouts and affordances, from professional accreditation requirements to library lending rules. This is not unusual: the radio in the car takes us beyond the simple technology of transport from A to B and the car-assembly also contains technologies that relate to conveying status, providing comfort and plenty more.

Pedagogies are 'soft' technologies: they are easy to change, malleable, flexible, responsive and essentially embedded in human social and internal behaviours. At least they have that affordance even if a given individual teacher might not take much advantage of that softness. When fitted with flexible technologies (Franklin call these 'holistic') they allow for great variety and adaptativity. When fitted with less flexible, harder technologies (Franklin calls these 'prescriptive technologies') the situation is rather different. Hard technologies tend towards rigidity. Their purposes are narrow and, typically, they embody soft technologies in a fixed design. For example, Franklin contrasts traditional weavers' tools with weaving machines in factories: the former enable creativity and flexibility,  and afford many opportunities to allow them to interoperate with different soft technologies. The latter embed the soft technologies in the process and offer very limited opportunities for their operators to deviate from the demands of the machine. As a rule, soft technologies enable, hard technologies constrain, at least from the perspectives of their users. This does not make hard technologies at all bad in principle as their purpose is nearly always to enable us to achieve many other things, sometimes better than an earlier technology, sometimes that simply couldn't have happened without them: a railway system is mostly a very hard technology with a very tight focus on a single fixed purpose, but it opens up fantastic opportunities for travellers. However, it cannot and never will enable travel beyond the rails.

Learning technologies vary greatly in the degree of softness and hardness contained within them. E-mail is a tool (indeed, with its surrounding assemblies, a technology) that is fairly neutral to pedagogy and can be used for anything from Socratic dialogue, to didactic instruction, from social-constructivist conversations to formal testing. With a rich range of support for multiple media and a transport mechanisms that makes it accessible to almost anyone with an Internet connection (and even some that don't, via cellphone gateways for instance) there are few pedagogies (pedagogic technologies) that it cannot support and interoperate with as part of a technology-enhanced learning assembly. Of course, some things are easier than others and if, say, we wished to provide multiple choice objective questions as part of our pedagogical process then it would need a lot of soft technology on the part of the teacher to support it, compared with a computerised alternative.

A computerised multiple choice question system, on the other hand, is at the harder end of the soft-hard continuum. While it may be used as part of a wide range of different kinds of assembly that include various different pedagogies, in itself it is quite constraining.

The learning management system (LMS) occupies a very interesting place in this schema. Typically, an LMS will provide a wide range of tools and technologies at various ends of the soft-hard continuum. In some ways it can be viewed as a toolset of specialised and less specialised tools, a mix of holistic and prescriptive technologies. This is good. Unfortunately, the LMS has a darker side. Almost all (if not all) grew up as a result of (naively) automating and computerising things that were done in classrooms and training rooms, whether or not these played a direct role in learning, including many parts that automated processes that were historically contingent and not much to do with common sense pedaogies and learning needs. Universities and schools emerged as a very sensible set of technologies to deal with a paucity of resources. They *had* to be centralised because there were only so many teachers to go round, only so many books. Lectures were a good idea in times of scarcity. Libraries were a very sensible way of dealing with the expense and scarcity of scholarly literature. Terms and semesters helped deal with students' needs to go back to their villages and join in the harvest. Classes and courses needed to fit these structures, to make efficient use of lecturers' and professors' time. Timetables were needed to make best use of these classes and scarce resources. Summative tests and written exams emerged as an efficient means to judge these modularised chunks of learning and enable interoperability between them and cater for external bodies that needed easy-to-compare assurances of competence. The list goes on: the educational machine is built the way it is built as much because of its history as its mission to aid learning. The popular LMSs (Blackboard, Moodle, WebCT, Sakai, Desire2Learn and so on) were designed to reflect and support that machine. As a result, the things that they automate are strongly constraining. I've previously reported on the fact that well over 99% of all courses at my other institution, the University of Brighton, kept the defaults of showing announcements as the entry point to a course despite the fact that, when questionned and forced to reflect on it, many of their authors thought this inappropriate. Similarly, I have written about the challenges of moving from a custom-designed LMS to Blackboard to support a program that did not maintain traditional institutional structures. Learning management systems offer suport internally for softer technologies but, at a structural level, tend to embed the soft technologies in the hard-wiring of the system itself. It would be far better to build systems from smaller, more flexible, interoperable parts, some soft, some hard, though that itself comes at a high price in terms of manageability, scaleability and learnability.

The need to insist on the primacy of pedagogies over other technologies that is so often expressed in educational literature is unsurprising because, by default, in our educational machines that is unlikely to happen: hard technologies and the systemic pressures of other parts of the machine relegate pedagogies to a supporting role in many cases, especially as they are much easier to change than the harder parts of the machine. However, it is a much more complex issue than that. As George Siemens notes, we cannot ignore context. Equally we must consider the role of other machine parts, the affordances and constraints of both soft and hard technological systems. We are engineering systems here and all the parts need to work together as an orchestrated whole to achieve whatever it is we are trying to achieve, of which learning is only the major aspect. Terry Anderson has a nice metaphor to describe that complexity - it's a dance.

However, it is not all about engineering. While good use of technologies (including pedagogies) can certainly improve the learning experience, the situation is analogous to that of music. It is possible to make great music with terrible tools and dubious technique  (some old recordings of blues musicians and skiffle bands make that abundantly clear). Similarly, some teachers can work miracles using the worst pedagogies and other technologies that are available: many of us can recall great and inspiring lectures that prove that point unequivocally. On the other hand, the most perfect instruments and flawless technique are not enough to make a great musician, even though the result is usually not at all bad to listen to. Similarly, we can make perfect use of pedagogies and other technologies and achieve perfectly satisfactory, if not great, results. Good tool use (including pedagogies) definitely helps, but it is not enough.

The vast majority of educational research describes the technologies - the pedagogies, tools, techniques and processes - not the art of teaching. It is not at all surprising therefore that the general consensus of aggregated literature is that there is not a great significant difference between online and face-to-face learning. In the first place, there tends to be little consideration of the interoperability between pedagogies and other technologies: using the same pedagogy in a different technological context can be as ridiculous as using a spark plug in a horse-drawn carriage, or attaching a jet engine to a mule (and the results equally both harrowing and funny). In the second place, relatively little research considers the magical hidden part of the iceberg, the elephant in the room that is the talent of the teacher, whether that role is mediated, distributed or direct. Process is easy to observe and report on, creative genius and art not so much.

Finally, it is worth noting that good communication channels can compensate greatly for poor uses of technology, whether pedagogies or otherwise. It is far easier to adapt soft technologies when they do not seem to be working than hard ones. Communication gives us the means to know when to do so. Several of the greatest teachers I know use terrible pedagogies and/or other technologies but the fact that they are able to respond to learners and adapt what they are doing to suit makes up for all their shortcomings. If there is no other reason to embed communication in the learning process (and there are many other reasons) the fact that we are imperfect tool users has got to be a clincher.