There are very few technologies that are wholly hard or wholly soft, at least when viewed as an assembly. It is the proportion of soft/hard parts that make a particular assembly softer or harder. That’s how come replacement can make things harder and aggregation can make things softer. When we replace, we take something from the technology assembly that was formerly flexible and negotiable and make it less so. When we aggregate, we make no changes to the assembly that was there before but we increase the adjacent possible and thus enable more human choice and active shaping of the technology. To help clarify, here is a simple example of how we might harden and soften an assessment activity in a course:

1. soft: Tell the learners to pass their work to you when they are ready. You will tell them what you think
2. harder: Tell the learners to pass their work to you when they are ready. You will grade them according to a set of criteria
3. harder: Tell the learners to pass their work to you on or before a certain date and time. You will grade them according to a set of criteria
4. harder: Tell the learners to pass their work to you on or before a certain date and time in PDF format. You will grade them according to a set of criteria
5. harder: Tell the learners to pass their work to you on or before a certain date and time in PDF format, presenting you with no more than 1500 words. You will grade them according to a set of criteria
6. harder: Tell the learners to pass their work to you on or before a certain date and time in PDF format, presenting you with no more than 1500 words, structured with an introduction, literature review, discussion and conclusion, with correct citations in APA format . You will grade them according to a set of criteria.
7. harder: Give the learners a templates set of questions that they must answer in essay form with fixed word limits. You will grade them according to a set of criteria.
8. harder: Put the whole thing in an LMS and use an LSA-based automated marking system to provide feedback and results, according to criteria programmed into the machine. 
9. Put the whole thing in an LMS and use an LSA-based automated marking system to provide feedback and results, according to criteria programmed into the machine. Late submissions receive an automatic mark of zero. Results are automatically added to a student record sheet without teacher intervention.

Notice that, each time, we replace one flexible part of the technology with another, less flexible part. Logic would suggest that we should simply reverse that procedure to soften things again but, once we start down this kind of path we tend to create a set of path dependencies and  systemic interdepenencies and patterns that affect not only the technology we are looking at but the technologies of which it is a part and those with which it interacts. Typically, this kind of pattern would be accompanied by regulations, processes, tools, learner expectations and policies that would make it trickier to reverse than it was to create. Also, we would seldom want to throw away everything we have gained by hardening and start afresh. If that’s not the case then all is well, we can modify things back to a softer state if that’s what we need. And that would be lovely. But what if we wanted to keep some of the good things from the harder system while enabling alternative and more flexible approaches where needed? Or if other systems (e.g. the pedagogies of the class, the regulations of the course etc) had been built around the harder system? There is another way, using aggregation, that provides a more flexible and simpler-to adapt method of softening. Here is the beginnings of a similar list that reverses parts of the hardness by adding extra technologies:

1. softer: Put the whole thing in an LMS and use an LSA-based automated marking system to provide feedback and results, according to criteria programmed into the machine. Late submissions receive an automatic mark of zero.  Results are automatically added to a student record sheet without teacher intervention. The results produced by the system are treated as provisional. Students must print their records and take them to the tutor for a signature before they are officially recognised. The tutor may manually amend the marks, and sign the amendment to avoid fraud. The tutor may also manually mark an assignment that has received zero for late submission, if the student provides an acceptable excuse
2. softer: Put the whole thing in an LMS and use an LSA-based automated marking system to provide feedback and results, according to criteria programmed into the machine. Late submissions receive an automatic mark of zero.  Results are automatically added to a student record sheet without teacher intervention.The results produced by the system are treated as provisional. Students must print their records and take them to the tutor for a signature before they are officially recognised. The tutor may manually amend the marks, and sign the amendment to avoid fraud. The tutor may also manually mark an assignment that has received zero for late submission, if the student provides an acceptable excuseBefore the tutor will sign it, he or she is optionally given a reflective document produced by the student that describes the process. The document gives the student a chance to explain answers in ways that were not possible using the automated form.
3. softer: Put the whole thing in an LMS and use an LSA-based automated marking system to provide feedback and results, according to criteria programmed into the machine. Late submissions receive an automatic mark of zero.  Results are automatically added to a student record sheet without teacher intervention.The results produced by the system are treated as provisional. Students must print their records and take them to the tutor for a signature before they are officially recognised. The tutor may manually amend the marks, and sign the amendment to avoid fraud. The tutor may also manually mark an assignment that has received zero for late submission, if the student provides an acceptable excuseBefore the tutor will sign it, he or she is optionally given a reflective document produced by the student that describes the process. The document gives the student a chance to explain answers in ways that were not possible using the automated form. If the student wishes, he or she may submit an essay-form version of the answers, formatted according to recognised academic standards. The tutor may choose to treat that as the submission instead of the original automated work.
4. etc

Note that, though it looks much more complex and full of redundancies this can, with a sufficiently technically advanced component-based system, be at least partially automated  without too much hassle. Using this kind of approach we can selectively soften different parts of the system without necessarily having to roll back the whole thing and lose the hard parts that we value. But soft is indeed hard. The paragraphs describing the technologies are now longer because we are now combining more technologies to achieve the same (but more attuned to our needs) result, and introducing redundancies and complexities for teacher and students that may make it unworthwhile. That’s the trade-off, of course, the reason we need to think about just how soft or hard we would like our systems to be. There are no generalisable right answers here, but our technologies must be sufficiently malleable to allow us to shift the amount of softness and hardness in them according to our ever-changing needs. Without such malleability, we risk being slaves to the machine or floundering in unnecessarily inefficient complexities and substandard tools that are a poor fit with what we want to do

I finished the first synchronous sesssion for my bit of the #change MOOC a little while ago. Self referentially, I was a victim of a hard technology of my own devising, a set of slides that formed the background to my talk that, as I talked through them, I realised were in quite the wrong order.

Did that stop me?

Not a bit of it!

Interestingly, however, I could have done so had I had my wits about me. I had found the option where Blackboard Collaborate did give me a menu of available slides to select in addition to the usual back/forward controls. By adding this hard technology to the existing hard technology the makers of the tool had softened it, allowing me to take a non-linear path had I chosen to do so.

But, of course, soft is hard.

The effort of managing the talk, following the chat and grappling with the tools gave me insufficient time to think through a more appropriate path so I took the hard, easy one. As Garrison and Baynton showed in 1987, increasing pacing also increases structure – it is much harder to change the structure in real time than over a longer period. This suggests that ‘hard and fast’ make a better pairing than one might at first think.

It will be interesting to continue the conversation on Friday at 10am MST – see http://change.mooc.ca/cgi-bin/page.cgi?event=45 for announcements about how to get to that.

The blog post for day 2 of my week at the Change11 MOOC…

Yesterday I made a quick and dirty four-minute  video in which I explore the ways that a stick can become a technology because of how it can be combined with soft processes and used as part of an orchestration of phenomena to some use. It can thus become many technologies. On reflection, and looking at the video, I realise that it was a mistake to describe the stick itself as a soft technology it is not. The stick is a part of a great many (probably an infinite number) of soft technologies.

I think that this cuts to the heart of a great many of the mistakes that we make when we talk about learning technologies. We often make the assumption that, because the same thing is involved from one context to the next – a learning management system, a discussion forum, email, a whiteboard, a classroom, a teaching method, etc – that we are talking about the same technology. We are not.

When I use something like this site (the Landing) in different classes or even in different parts of a class, it is just part of an assembly that is, in each case, a different technology. I am orchestrating different phenomena in different ways for different purposes, even though the tool is the same, just like the stick. Similarly, it is a different technology for me, now, writing this, than it is for you, reading this. For me, I am using its ability to share content with a public audience. For you, it is a means to display content. Different uses, different technologies, even though we are both using precisely the same underlying set of tools. It is a very unhelpful generalisation to think of only the tool, not its use and not the things that we are assembling with it for that use to occur.

Churchill said ‘we shape our dwellings and afterwards our dwellings shape our lives’, a sentiment echoed by McLuhan whose take on it was ‘we shape our tools and thereafter our tools shape us’. That’s the starting point for the theme for my bit of the Change11 MOOC. Ignoring the elephant in the room for the moment (because that’s mostly what we do), there are big questions about the kinds of dwellings and tools should we be using and designing to help us to learn, what are the effects of our choices?

It’s partly about softness and hardness. Harder technologies are constraining: learning management systems, for example, work because they deliberately limit the number of choices we have to make, performing some of the orchestration of phenomena on our behalf so that we don’t have to. That’s why they are relatively easy to use, at least when compared with alternatives such as adapting existing tools or building our own programs. There is a strong risk that they can therefore stifle creativity. However, constraints can also drive creativity. I was reminded of this while touring art exhibitions in Vancouver yesterday, where I saw some images that drew their form from the medium on which they were painted, using the natural whorls and knots of the wood grain to structure the picture.

As Stravinsky said, “The more constraints one imposes, the more one frees one’s self. And the arbitrariness of the constraint serves only to obtain precision of execution.” Bradley Messer seems to have taken this message to heart and, though I do not love all of the pictures that result from that, there is no doubt that something creative has emerged from the constraints.

The trouble is, the constraints of our learning technologies are far from arbitrary and are often the result of turning soft and flexible processes observed in traditional teaching practices into something harder. This is not like the knots in the wood, where arbitrary forms can guide our creativity. This is the solidifying of history, the intentional creation of path dependencies that can entrap us in pedagogies and methods used by our forebears that may be less than perfect for our current needs. We need to think far more carefully about those constraints, not to take them as a given, to find ways of using them but not being bound by them. Or maybe we need to build our learning technologies more like wood, more organically, more arbitrarily. 

Excellent talk on the confirmation biases and filter bubbles that make the Internet a less diverse space than it might be.

Address of the bookmark: http://www.ted.com/talks/ethan_zuckerman.html

Donald Clark giving one of his marvellous signature incendiary talks about how to avoid bad ways of teaching (lecturing in particular) with Twitter accompaniment. The tweets are almost as interesting (though not in a good way) as the talk itself. He was well and truly tweckled. A great talk and a fascinating glimpse of the power of backchannels.

Address of the bookmark:

This is a very fine article that provides excellent arguments and evidence about the fundamental dangers of rewarding results rather than effort in learners, and the extreme silliness of learning styles. The notion of a ‘learning quotient’ rather than an ‘intelligent quotient’ is a nice distinction. The idea of fixed abilities and ways of thinking is highly antagonistic to learning – telling people they are good at something or that they are not good at something, or that they learn in one way and not another reduces motivation to become more than they are. We are not static and unchangable beings. To learn is to change, and the best and most rewarding learning is that in which we learn to learn. While there are certainly limits to what an individual can expect to be able to do in a single lifetime and we are all better at some things than others, with enough effort and an appropriate way of going about it, we can almost always learn more, learn to learn differently, and expand our horizons further. 

There are some interesting follow-up comments too: some are predictably dumb, but someone had to make them. There is a particularly heinous opener in which the author claims that rewarding results and punishing failure is what it is like in the real world so learners had better get used to it. This is wrong on so many levels, but mainly because the author completely misses the point that it is about becoming competent, which is not at all the same thing as using that competence once it is achieved. It is possible to learn about reward/punishment systems too, and how to cope with them, but that’s a separate problem, and simply punishing or rewarding every result, without further reflection on and examination of what that means, is definitely not the best way to go about it. However, though there are a few more comments of this nature, the conversation explores a lot of the issues in some depth in a way that extends the depth of the original article considerably. Well worth reading.

Address of the bookmark: http://mindshift.kqed.org/2011/11/can-everyone-be-smart-at-everything/?utm_medium=referral&utm_source=pulsenews

Interesting article overviewing a range of uses of analytics (including learning analytics) raising some slightly fuzzily-articulated but very real concerns about the use of statistics to guide behaviour.

Address of the bookmark: http://www.readwriteweb.com/archives/ipad_apps_for_kids_measure_increase_learning.php?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+readwriteweb+%28ReadWriteWeb%29

Social digital storytelling radio station

Address of the bookmark: http://ds106.us/

I’m sitting here listening to Barbara Means giving the keynote on blended learning in K-12 in the US at E-Learn 2011. The audience is a little thin on the ground this morning as the conference is in Waikiki this year. The people on the beach missed a wonderful traditional chant and greeting – wonderful.

Barbara is telling us that, in meta-studies and in quasi-experimental studies, blended learning is equally or (often) more successful than face to face equivalents. And, the second point, as she observes, is that course completion rates are lower, of course (so the ones that survive are better). For the latter insight it is reasonable to consider the ‘e’ element of it – one absolutely distinctive feature of online learning is that it is much easier to ignore, so less motivated or well-organised students or those who have not learned online learning skills are at a disadvantage (something Barbera rightly observes). For the former, it is not at all.

The problem with these kinds of studies is a failure to understand and cater for the nature of the technologies they are examining. If we are looking at outcomes, it is almost nothing to do with whether people use online teaching or not, it’s about how the technologies (notably including pedagogies as well as organisational systems and the physical and virtual systems) fit together and how it all relates to things like motivation, time on task and the passion of the teacher, which together account for a good 80% of the reason for success or failure in most educational interventions. 

It ain’t what you do, it’s the way that you do it. A bad teacher is a bad teacher no matter what model is used (though good technologies such as sensible pedagogies, amongst others, can reduce the harm), while a good teacher, as long as the physical and organisational constraints are not too limiting, is good no matter what model is used. Technics matter, but caring, art, sensitivity and skill matter more. Good teachers can and often do use bad pedagogies and other bad technologies and yet they still succeed. Similarly, teaching things that people want to learn, when they want to learn them is more useful than almost any other factor. 

Some interesting insights into what happens when technologies are incompatible – like insane state legislation requiring students to sit in class when there is nothing for them to do there because they are working online, or when students learn to learn using one technology but no one notices that the same approach doesn’t work when you use another. Inevitably, she is talking about flipping the classroom (a trendy name for what good face-to-face teachers have done for millenia) with some encouraging stats to suggest that an approach of teaching rather than telling is becoming more fashionable as a result.