1. Soft, exposed nature of the ed design article itself. Ideas are the stuff that you deal in: abstract, malleable and open to interpretation in a field where mistranslation is standard practice. You face a long and treacherous implementation process before you or anyone else really knows whether your idea is any good. Good luck!
   
2. Collaborative ecology of education project work. Caring about things being done well pulls you in two opposing directions:  on the one hand trying to extend control over other people and on the other trying to do everything yourself because control is just too hard. Both are dead ends of exhaustion and burn out. You can't afford to do other people's thinking for them and you don't get much from them when you do. You need people to think for themselves and to share their thinking with you. You need systems of conversation not control. And you need people and management structures able to make sense of this subtle distinction.
3. The extended timeframe of educational innovation.  Educational design ideas do not come perfectly formed on first iteration. It's not like eggs from a chicken. It may take more than one project to get where you really need to go. Having people run off with your ideas when they're only half done is an occupational hazard. Having them come up with their own half-cooked version without the slightest glance at yours is another.
4. The diversity of the field. The paradox is that the breadth of ed design world makes it very easy to get into a narrow rut without even knowing you're there. It's hard to think about the field as a whole in any coherent, intelligible way and much easier to focus on one or two things that you learn to do pretty well if you do them enough times . . . over and over again.
5. Narrow, short term focus of most real life educational resource projects. You still need to get something built that keeps your customers satisfied and your salary paid. The trick is to find a way of keeping up with vital short-term goals without losing the sense of broader issues that enable you to exist as a designer. Another trick is finding people to work for who are likewise able to fit short-term and long-term into some sort of coherent package. Especially at management  level. 
6. The experiential basis of educational design expertise. You learn by doing rather than from rules and formula. You only grow through experience that is diverse, interconnected and focused around a sustained long term agenda. The experience that you get is precisely the opposite: stereotyped, fragmented and ad-hoc.
7. Uninspiring  track record of educational resource development generally. The low uptake of new educational ideas and poor shelf-life of most educational resources have generated whole bodies of literature devoted to their commentary and critique. Discussion of these issues tends to focus mainly on teachers and teacher resistance, but the under-developed state of the majority of 'innovative' education resources is sufficient explanation for the unenthusiatic reception in most cases. The poor quality level of the industry as a whole should normally be a spur for improvement, but more often becomes an excuse for continuing with business as usual. If we’ve got away with it so far, why change?
   
8. Planning, budgeting and staffing decisions taken without regard to any of the above, as though education resource development was just another production process, no different from the one that turns chickens into chicken cubes.

I take 'educational design' to be the set of practices involved in constructing representations of how to support learning in particular cases. This distinguishes design from development - the practices of turning these representations into real support for learning (materials, task specifications, tools, etc). It distinguishes design for particular educational applications from the broad consideration of learning in general. It focuses on practice rather than theory, while recognising that practice embodies experiential and theoretical knowledge. Within this framework, it can be seen as a reworking of instructional design, but without the narrow pedagogical repertoire that the term 'instruction' is often taken to connote (Goodyear, 2005).

This is great . . . .  but the language is a bit refined for my taste. It's not for general consumption. It's also an odd fit for something that supposed to be all about practice rather than theory. As well as that, there are some difficulties in the way the boundary is drawn with educational resource development.  The distinction between 'construction of representations' and the construction of resources embodying those representations is hard to see when the bulk of the resources are representational in nature and when the representations only exist in the form of specific resources. Some tweaking needed.

DEFINTION B:

Educational design is the work of designing educational resources.

A definition in these terms is easily understood but also potentially misleading. Just because you've designed something called an 'educational resource' doesn't mean there's anything particularly educational about the design. Unfortunately.

DEFINITION C:

‘a planned process of making curriculum decisions about how best to support student learning in some defined area’ (McNaught, 2002).

This is close to Goodyear's definition, but coming from a different perspective. Where Goodyear talks about educational design as being about representations of how to support learning, McNaught's definition is about decisions on how to do so. Both agree that educational design is concerned with particular cases. This suggests a new definition drawn from the common elements of both Goodyear and McNaught.

DEFINITION D:

Educational design is the work of finding ways to support learning in particular cases.

This is getting too broad however. It looks too much like a definition of applied educational research - which is certainly part of the educational design picture but not the whole. It emphasises the ideal aim of educational design at the expense of the context of practice, whose constraints are only represented in a vague, remote sense by the reference to 'particular cases'. Support for learning is not the only requirement that educational design needs to meet. Limits of time, skills, money and technologies and other resources are critical particularities in the particular cases that ed design deals with. The definition needs to reflect this.

DEFINTION E:

Educational design is the work of finding ways to support particular learning needs using particular resources.

E is my preferred definition for the moment. It presents educational design work as a matching of ends and means, a problem-solving type of activity. This puts it on a similar conceptual basis to other design professions operating in a problem-solving mode: graphic design, software development, engineering etc. The definition also helps sort out the boundary with educational resource development which is raised as an issue in definitions A and B, but not resolved. Definition E makes it clear that educational design deals with resources primarily from the perspective of how they are used, not how they are created. Maybe someone else will think of something better than this - but for the moment I'm sticking with it.