 |
 |
Hi Alex. I happened to be on a call with Lorna, talking about these bids, when your message came in. So please consider this a reply from both of us. Thanks for taking the time to write such a detailed response, this is exactly the type of discussion that we need to help us evaluate the proposals. We have been moving away from standards such as the LOM over the past 10 years or so. Lorna, John Roberston and I wrote a position paper for an ADL repositories and registries summit describing this, see <http://blogs.cetis.ac.uk/lmc/2010/04/16/then-and-now/>. I guess we can debate what approach works best where, but one thing I need to correct you on is the TAACCCT mandate for SCORM, since this was amended (see <http://www.doleta.gov/grants/pdf/SGA-DFA-10-03-AMENDMENT1_2-8-11.pdf>). best regards, Phil On 13/04/2011 15:25, Alex Lydiate wrote: > Hi Lorna, > > Thank you for the question - its a very interesting topic. > > Certainly, I think the creation of meta data is a complex concern. My > view is that formal metadata, adhering to common ontologies, is quite > necessary for access of any resource to be at its widest, accessible > to both human and software agents. However, it is certainly true that > attempting to collect detailed, formal data on point of entry from a > single human agent is unsustainable - I think this is also what you > are saying that resource developers feel - as the process will either > be too time-consuming for the individual, or if suitably streamlined > the data is likely to be incomplete or inconsistent. > > As a solution to this problem I see it best that meta data in fact be > collected at point of entry by a combination of both human and > software agents. For example, where as a 'description' field is > naturally suitable only to be completed by a human agent, populating > the sort of data required by LOM's Technical object is best achieved > via a software agent reading the resource and populating the data > behind the scenes; the human agent need only give the URI of the > resource, the software can do the legwork. Such technical data is > imperative for other agents to then be able to disaggregate the OER > and reuse it's constituent individual resources with a question along > the lines of 'Point me towards all the mp3 resources within a certain > category of OERs which are under 3 minutes in length'. If OERs were > presented according to our proposal of using OAI-ORI as the > aggregation descriptor and IEEE LOM as the individual resource > descriptor, such things are eminently doable. > > As I understand it, the traction of the LOM standard is very > established and growing, not least due to the US Government's > mandating of SCORM as part of their recent announcement of massive > funding to TAACCCT. A debate about SCORM is no where near the scope > of this discussion, but the power that gives LOM in terms of > establishing further traction I think is undeniable. Regardless of > this, to my knowledge there exists no better standard to describe the > individual resources within an OER, it is certainly very well defined > and very well established. > > However, the main objective of our proposal is to implement the > OAI-ORE standard as an aggregation descriptor; should a better > descriptor of the individual resources be proposed we would welcome it > to be appraised against the value of LOM during the design period of > the project, if we were successful. > > In terms of the 'distributed, publish anywhere' environment, this is > certainly an environment we embraced in our design of the OSTRICH > distributed repository. Or 'referatory'. I'm not quite sure which is > better :) Anyway, we do not intend to store the individual resources > on our servers, as we feel this would be fundamentally unscalable. > Instead, we refer to the resources at their own location, which might > be anywhere at all on the Web. However, we have then a clear need to > be able to describe these resources to both human and software agents, > a need which we feel is common to many, and hence our proposal. We > believe the it supports a distributed environment. > > Alex > > On 13/04/11 13:58, Lorna M Campbell wrote: >> I must admit I didn't read the OAI-ORE proposal as being "a >> repository". I've been interested in the potential affordances of >> using OAI-ORE for managing OER aggregation for some time, >> particularly in relation to connecting resources to distributed >> comments, ratings, rankings, and other paradata type stuff, so I'm >> quite intrigued by this bid. >> >> I do have one query for Alex though, I'm not entirely sure about the >> applicability of LOM to OERs. Certainly LOM could be used to describe >> OERs but I wonder how likely it is that this standard will gain >> traction in the distributed, publish anywhere type of information >> environment that has emerged over the last few years. We know that >> resource developers tend to see the creation of formal metadata as >> problematic which I think is part of the reason that there has been a >> move towards tagging and capturing more informal, or less structured, >> resources descriptions. >> >> Thoughts? >> >> All the best >> Lorna >> >> >> On 13 Apr 2011, at 12:03, Alex Lydiate wrote: >> >>> The 'ORE one', ie ours, is not a repository. We've got one of those >>> already - http://ostrich.bath.ac.uk , wouldn't wish to make another. >>> >>> It is rather a proposal to implement the OAI-ORE standard for the >>> purpose of presenting OERs as aggregations of resources, as resuable, >>> disaggregable objects of complex types. >>> >>> >>> >>> On 13/04/11 11:45, Scott Wilson wrote: >>>> #1 I really like the "Cut and Paste Reuse Tracking" proposal - its >>>> novel but also connected to "real world" use. I suspect its >>>> actually the most technically challenging in reality but well worth >>>> pursuing. >>>> >>>> #2 I don't think OER Bookmarking is viable as it really would have >>>> to be "another delicious". HOWEVER - If it were instead a proposal >>>> to try to work with Yahoo! and other sites to offering bookmarking >>>> to get them to integrate better licensing into their sites, that >>>> would potentially be a far more interesting proposition though >>>> obviously with less predictable outcomes. (Another option might be >>>> to work with oEmbed. That probably goes for #1 too) >>>> >>>> #3 I think vocab management tools is a non-starter as vocabularies >>>> are principally a system-wide management and sustainability problem >>>> not a technical problem. >>>> >>>> #4 And finally, the ORE one is a repository. >>>> >>>> --- >>>> S >>>> >>> >>> -- >>> Alex Lydiate >>> Software& Systems Developer >>> LTEO - WH5.39 >>> University of Bath >>> 01225 383576 >> -- >> Lorna M. Campbell >> JISC CETIS Assistant Director >> University of Strathclyde >> Glasgow >> Email: [log in to unmask] >> Phone: +44141 548 3072 >> Skype: lorna120768 >> >> The University of Strathclyde is a charitable body, registered in >> Scotland, number SC015263. >> > > -- Please note new email address: [log in to unmask] -- Heriot-Watt University is a Scottish charity registered under charity number SC000278.