 |
 |
On 17 Apr 2011, at 22:55, Pat Lockley wrote: > On 17/04/2011 11:05, Alex Lydiate wrote: >> At which point, I guess, we're talking again of an aggregation of resources. This was the number one topic in Learning Objects over the previous decade so there is a lot of prior work. OER is principally about licensing. OER's aren't necessarily Learning Objects. >> We would also be talking, I suppose, of prescribing that most other resources are stored in Git, those that it can handle, whilst those it can't are elsewhere? Which is not really part of the whole 'publish anywhere' tip, unless what we end up putting into Git is a collection of description files. >> >> I think I am not diverted from the OAI-ORE trail :) > > Maybe if we step back we can visualise this in a different way. > > OAI-ORE is good for describing a finished project, and git is handy for working projects. > > So lets say you want to start making an OER tomorrow. I use Github a lot - I'm quite a fan. But then again I'm a software engineer. I think one of the big problems with ed tech has been trying to make teachers think and work like developers. Wikis are a more natural fit for writing - especially collaborating on open textbooks. Which is a another kind of OER again. Then again I've co-written papers using Word docs in Dropbox, Google Docs, and Mediawiki. Lots of options. Ultimately whenever I've made materials for teaching with I've ended up using Powerpoint..! Can OER tech move towards teachers and learners existing practice, rather than require changing practice using the tech? > A git would be a great place to go to get a finished product, either for remixing or using as is. But neither Git or OAI-ORE are great as a "recipe book" ( - hat tip to CSAP-OER2 at Birmingham). Let says I want to see all pictures used in sociology OERs (might be a good place to start, build other services from) - so ideally you want some sort of multiview system (i think this might be called a lattice file system). > > So ideally you'd submit descriptor files (and maybe file size rather than files?), which could populate as an OER, and wider user-configurable views? > > This way you could record works in progress, ingredients, and the finished products / working projects too. > > So it's kinda OAI-ORE, as long as you can have 1 file aggregations, and 1 file which changes. S