TOPIC 2

Dear participants, The official start of the e-Conference will take place at 13:00 (GMT +0.00). During that time, the first questions that will elaborate on each topic will be published. To be able to participate in the initial discussions, we kindly ask you to use the time left, to familiarize yourselves with some of the resources and reading material that are provided in the first page of the Forum (folder: RESOURCES). With kind regards, Nikos Palavitsinis

Initiating the discussion on Topic 2, we would like to provide you with a set of questions that focus on various aspects of the proposed topic. As you will see, some of the questions are more generic, whereas others focus on specific stakeholder groups (i.e. content/course creators, users of the content, etc.). The idea here is to start elaborating on these topics, clarifying and delving into the characteristics of a quality learning resource. To facilitate the discussion, we would like to ask you to start replying by indicating the question number addressed (i.e. Q1), so that our colleagues that read you answers can easily identify the question you refer to. 1. Have you described in the past, your learning resources using metadata (description, title, keywords, tags, etc)? Which are the metadata elements that you mainly use? 2. Do you feel that providing metadata for resources is useful? What are the incentives that drive you, to provide the metadata? 3. What constitutes high quality metadata for a resource? Is it completeness for all metadata elements? Is it the clarity and correctness of the language used? Other aspects? 4. What are the benefits you see in providing metadata for learning resources in practice? With kind regards, On behalf of Topic 2 Moderators, Jackie Wickham & Stephanie Taylor Nikos Palavitsinis

[quote="Lisa-Cespedes"] Providing quality metadata: Is the gain worth the effort? The topic above, is refined with the questions that follow in the post of the Moderator of Topic 2. You can either reply to this general topic, or address one of the other questions. [/quote] I think providing quality metadata has much benefits that one might think of, but as long as a standard is used I am sure the quality will be kept alive.

[quote="nikospalavitsinis"] Initiating the discussion on Topic 2, we would like to provide you with a set of questions that focus on various aspects of the proposed topic. As you will see, some of the questions are more generic, whereas others focus on specific stakeholder groups (i.e. content/course creators, users of the content, etc.). The idea here is to start elaborating on these topics, clarifying and delving into the characteristics of a quality learning resource. To facilitate the discussion, we would like to ask you to start replying by indicating the question number addressed (i.e. Q1), so that our colleagues that read you answers can easily identify the question you refer to. 1. Have you described in the past, your learning resources using metadata (description, title, keywords, tags, etc)? Which are the metadata elements that you mainly use? 2. Do you feel that providing metadata for resources is useful? What are the incentives that drive you, to provide the metadata? 3. What constitutes high quality metadata for a resource? Is it completeness for all metadata elements? Is it the clarity and correctness of the language used? Other aspects? 4. What are the benefits you see in providing metadata for learning resources in practice? With kind regards, On behalf of Topic 2 Moderators, Jackie Wickham & Stephanie Taylor Nikos Palavitsinis [/quote] Hi Nikos, let me answer the questions you posed, 1. I am using a metadata to construct and I am tagging them using the acceptable lom standard. 2. Providing metadata to your resources is useful if you want to search them and if you want to make them searchable in a federated repository structure. The incentive...a good will that someone on the other side of the world is making ones resource available and for the sake of giving a better learning environments for students or any other clients of the subject matter 3. The quality of a metadata, I believe, is not completeness in its adherence to the standard of metadata used but the correctness and using the minimum, at least, acceptable tags from the standard. The language is also important but the process is time consuming 4. It will make your works and your organizations' reputation in providing learning materials to a bigger level. You will get more comments and directions other that the recognition you get for your resources

[quote="ameha"] [quote="Lisa-Cespedes"] Providing quality metadata: Is the gain worth the effort? The topic above, is refined with the questions that follow in the post of the Moderator of Topic 2. You can either reply to this general topic, or address one of the other questions. [/quote] I think providing quality metadata has much benefits that one might think of, but as long as a standard is used I am sure the quality will be kept alive. [/quote] That's a good point about using standards. I'm reminded of the late, great Groucho Marx who said: "These are my principles, and if you don't like them...well, I have others". It's perhaps the same with standards - if you don't like one, have a look at the others. A colleague of mine at the University of Nottingham recently posted a very entertaining and sensilble blog about the plethora of standards and has suggested a useful way forward: http://blogs.nottingham.ac.uk/learningtechnology/2010/10/05/babelona-the-tower-of/ There are two main purposes for metadata - to enable people to discover things and to enable machines to talk to each other. (Well perhaps three if you count metadata that supports preservation and administration - or is that four?). There does seem to be a gap in the metadata world - we don't seem to have any way of evaluating metadata shemas in terms of their usefulness in connection with these two purposes and, dare I say it, their return on investmment. Does anyone know of anything or have any ideas about this?

[quote="ameha"] Hi Nikos, let me answer the questions you posed, 1. I am using a metadata to construct and I am tagging them using the acceptable lom standard. 2. Providing metadata to your resources is useful if you want to search them and if you want to make them searchable in a federated repository structure. The incentive...a good will that someone on the other side of the world is making ones resource available and for the sake of giving a better learning environments for students or any other clients of the subject matter 3. The quality of a metadata, I believe, is not completeness in its adherence to the standard of metadata used but the correctness and using the minimum, at least, acceptable tags from the standard. The language is also important but the process is time consuming 4. It will make your works and your organizations' reputation in providing learning materials to a bigger level. You will get more comments and directions other that the recognition you get for your resources [/quote] Dear Ameha, Reading your opinion on the questions asked, it feels like you are already familiar with the concepts of metadata and you are also capable of working with them. I only have one question, leaving the ground to Jackie and other colleagues to follow up on your answers. Do your colleagues, or people you work with, in other institutions, share your thoughts on the necessity of metadata? In general, what is your perception of people's attitude towards using metadata to annotate resources?

Hi, The main question of this topic is actually the central theme of all the people like me that work in systems supporting and exploiting metadata. The question is : where is the value in metadata? Here you have a paper I coauthored some time ago about this critical issue: http://portal.acm.org/citation.cfm?id=1389548 However, we actually did not provide answers to the question but only more questions. Just as for other Information System resources, metadata needs to be cost-justified, be it with cost-benefit analysis, strategic resource theories as the resource-based view of the firm, or any other kind of effectiveness or efficiency related argument. The problem is that there are few reports on this issue, and our understanding of the value of metadata is still in its inception. More empirical studies and in-depth analysis is for sure required! All the best, Miguel-Angel

[quote="msicilia"] Hi, The main question of this topic is actually the central theme of all the people like me that work in systems supporting and exploiting metadata. The question is : where is the value in metadata? Here you have a paper I coauthored some time ago about this critical issue: http://portal.acm.org/citation.cfm?id=1389548 However, we actually did not provide answers to the question but only more questions. Just as for other Information System resources, metadata needs to be cost-justified, be it with cost-benefit analysis, strategic resource theories as the resource-based view of the firm, or any other kind of effectiveness or efficiency related argument. The problem is that there are few reports on this issue, and our understanding of the value of metadata is still in its inception. More empirical studies and in-depth analysis is for sure required! All the best, Miguel-Angel [/quote] Hi Miguel-Angel This looks like a really interesting article but I can't access it :cry: Is it possible to also publish it in an open access way or is this restricted by the publishers. I'd really like to read the full text. Best wishes Jackie

Q1. Have you described in the past, your learning resources using metadata (description, title, keywords, tags, etc)? Which are the metadata elements that you mainly use? I have already described quite a few learning resources by using metadata. In the repository tool that we use for the specific purpose the metadata fields mentioned (title, description and keywords) are obligatory. In addition, we use metadata for describing areas such as copyright status, educational context, target age range, classification of the resource, contribution information etc. Q2. Do you feel that providing metadata for resources is useful? What are the incentives that drive you, to provide the metadata? Describing resources with metadata is a dirty, time-consuming job... but it pays well! I mean that it's a hard task but it surely helps at a later step, for indexing and retrieving the resources. It's much easier this way! Q3. What constitutes high quality metadata for a resource? Is it completeness for all metadata elements? Is it the clarity and correctness of the language used? Other aspects? I find all the aforementioned aspects equally important. I believe that a partially-described resource can only be partially helpful. If improper language is used, then the resource will also suffer... there should be some general guidelines that should be followed, that should contain both the completeness of the metadata as well as the quality of the used metadata. This guidelines could be implemented in the form of a quality assurance schema. Q4. What are the benefits you see in providing metadata for learning resources in practice? As I mentioned in my reply to Q2, the description of the resources with metadata elements really benefits the indexing and retrieval of the resources, especially when we are talking about a large number of resources found in the same repository!

[quote="jackiewickham"] There are two main purposes for metadata - to enable people to discover things and to enable machines to talk to each other. (Well perhaps three if you count metadata that supports preservation and administration - or is that four?).... [/quote] Well, in my opinion, from a teacher's perspective the most important thing about metadata is that it helps me to select the resources I need without inspecting all the elements in veeeeeeeeeeeeery long lists of educational resources. But if you want to get it like this, your metadata must be good, descriptive and apropriate for everyone. If also multilingual, you will guarantee the ssame benefits for a wider audience. Metadata descriptions like "A photo of an organic farm" are useless, so if this is what we are prepared to put in, let's forget about. Instead, comprehensive, detailed, standard-complinat metadata are pricesless!! Regards, Salvador

1&4. Yes, I have described in the past learning resources using metadata and most probably will have to do it again in the future. Describing it manually is a time consuming activity, but useful for the obvious reasons: indexing and retrieving, but most of all sharing&exchanging resources of different repositories. 2. On the other hand, I have few times tried to find specific resources using their metadata descriptions and realised that many times this was not helpful. I think that one element that is usually misleading for me is the title of the resource. In most cases this didn't offer the sufficient information that a title should and thus made the resource look unattractive/ useless. 3. So when talking about quality it should not necessarily refer to completeness (there are so many elements that are not that important when searching for a resource, such as size, format etc), but to relevance. When the results of a search engine are more relevant than the ones of an educational portal.. you start to worry. Good metadata is objective metadata. :)

Q1. I have indead described in the past learning resources by using metadata. I mostly use the following metadata elements: title, description, keywords, tags, educational context, topic, IPR etc. Q2. I agree with other colleagues that using metadata for describing learning resources with metadata is a hard and time-consuming task but it's very useful for indexing and retrieving the learning resources. It is something that we have to do for the targeted users and it has great value. Q3. Both aspects are very important. A learning resource described only partially in clear and correct language has the same problem with a learning resource fully described using improper language. They will be hard to retrieve. Q4. As I mentioned before the main benefits are the positive effects on the indexing and retrieval of the resources.

It looks like the need for metadata is quite clear, given the benefits from semantic activities such as categorization, indexation, sortation, etc. However, as it occurs with cars and roads where the same tire isn't suitable for any kind of road, the quality of metadata depends on how useful it is for the architecture and the characteristics of the systems that are going to process it. For example, while uncontrolled vocabularies seem to work fine for collaborative tagging, they aren't suitable for formal categorization. Therefore, achieving quality metadata could be a reasonably effort demanding task if it is adjusted to the consumer goals and requirements. Best regards, Leonardo Lezcano

My answers are following in the next lines: Q1) Various Metadata standards are available for the description of a learning resource and most appropriate should be used in each case. I have already applied some of them (Bublin Core, IEEE LOM) to the detailed description of learning resources into the bioagro repository (www.bioagro.gr) or the Organic.edunet repository (http://portal-organic.edunet.eu). The most common use metadata elements are the following: like title, author, description, language, keywords, tags, copyright status, format. Q2) Metadata elements are valuable for a Learning repository with a huge number of learning resources. User is able to search for the most suitable learning resource by checking only some of them metadata. Additionally, the cooperation between different learning repositories can be benefited, increasing the access to more qualified and up-to-date information Q3) A reusable and acceptable learning resource should be consisted of high quality metadata. In order to assure the high quality of the metadata, an evaluation process of the learning resource description should be followed, like fulfillment of all the fields of metadata, precise/accurate description of metadata elements, detailed description of the learning resource (metadata value: description). Of course, there is a variation concerning the importance of the metadata elements for the description of a learning resource. Title, description, language, author, keywords and copyright status are the key metadata for the description and search of learning resource. Also the language of the metadata should be taken in concern. Multillinguality into other languages, except from the initial language, is an added value for the quality of the learning resource. Q4) some first thoughts are the following: 1. Categorization of the learning resources per topic/country /audience etc 2. Facilitation of the search of qualified learning resources 3. Easiest way to exchange educational material 4. Provision of the improvement of the already existed / build of new learning resources With kind regards Charalampos Thanopoulos

[quote="ssalonso"] [quote="jackiewickham"] There are two main purposes for metadata - to enable people to discover things and to enable machines to talk to each other. (Well perhaps three if you count metadata that supports preservation and administration - or is that four?).... [/quote] Well, in my opinion, from a teacher's perspective the most important thing about metadata is that it helps me to select the resources I need without inspecting all the elements in veeeeeeeeeeeeery long lists of educational resources. But if you want to get it like this, your metadata must be good, descriptive and apropriate for everyone. If also multilingual, you will guarantee the ssame benefits for a wider audience. Metadata descriptions like "A photo of an organic farm" are useless, so if this is what we are prepared to put in, let's forget about. Instead, comprehensive, detailed, standard-complinat metadata are pricesless!! Regards, Salvador [/quote] Thanks Salvador. It's really important to get the user's view - that's the point of creating the metadata in the first place. It would be interesting to find out from you which of the metadata elements are the most useful. Earlier this year, I did some work on a multi-institutional project in the UK which aimed to "free" educational resources created in universities (our project covered medicine, dentistry and veterinary science). Not surprisingly, I worked on the metadata workstream alongside a colleague, David Davies, from Warwick University. We ran a survey and focus groups to ask teachers what metadata was important for them in discovery and assessment of resources. http://www.rsp.ac.uk/events/SummerSchool2010/Presentations/Resource%20discovery.pdf There were some interesting results. For example, "When choosing a learning resource, how influenced are you by those who created it? There was a 50:50 split between those who thought it important and those who didn't. In the focus groups, there was a strong view that the teachers wanted to evaluate their own resources - so the key metadata was good keywords and a short but high quality description.

Dear colleagues, Q1. Yes, I do, usually IEEE LOM. The most used elements would be title, description, keywords and location. Q2. I think it's very useful, as it can help to understand the resource and its purpose without being obliged to study it. So, we can filter resources by their metadata, and select those which would appear to be more fitted to our goals. Also, the use of metadata can ease the implementation of advanced search mechanisms. Q3. I think the most important would be the correctness of the metadata. Having lots of metadata annotated in an inaccurate way can end up being counterproductive. Q4. Increasing the reusability of a given resource and, so, its value for both teachers and students. Best regards, Ramiro

[quote="jackiewickham"] [quote="ssalonso"] [quote="jackiewickham"] There are two main purposes for metadata - to enable people to discover things and to enable machines to talk to each other. (Well perhaps three if you count metadata that supports preservation and administration - or is that four?).... [/quote] Well, in my opinion, from a teacher's perspective the most important thing about metadata is that it helps me to select the resources I need without inspecting all the elements in veeeeeeeeeeeeery long lists of educational resources. But if you want to get it like this, your metadata must be good, descriptive and apropriate for everyone. If also multilingual, you will guarantee the ssame benefits for a wider audience. Metadata descriptions like "A photo of an organic farm" are useless, so if this is what we are prepared to put in, let's forget about. Instead, comprehensive, detailed, standard-complinat metadata are pricesless!! Regards, Salvador [/quote] Thanks Salvador. It's really important to get the user's view - that's the point of creating the metadata in the first place. It would be interesting to find out from you which of the metadata elements are the most useful. Earlier this year, I did some work on a multi-institutional project in the UK which aimed to "free" educational resources created in universities (our project covered medicine, dentistry and veterinary science). Not surprisingly, I worked on the metadata workstream alongside a colleague, David Davies, from Warwick University. We ran a survey and focus groups to ask teachers what metadata was important for them in discovery and assessment of resources. http://www.rsp.ac.uk/events/SummerSchool2010/Presentations/Resource%20discovery.pdf There were some interesting results. For example, "When choosing a learning resource, how influenced are you by those who created it? There was a 50:50 split between those who thought it important and those who didn't. In the focus groups, there was a strong view that the teachers wanted to evaluate their own resources - so the key metadata was good keywords and a short but high quality description. [/quote] I attended an interesting meeting in the UK this week - "What Metadata Is Realy Useful?" http://wiki.cetis.ac.uk/Event:_what_metadata_is_really_required Dr Bob Strunz of the Irish National Learning Repository outlined their approach to deciding on what metadata was needed for learning objects deposited in their repository. Basically, they began by asking users (teachers and learners) what they wanted to be able to do with items - e.g. how they wanted to search, how and why they would be looking for material. From this, some use case scenarios were created and then worked through to create a basic framework of 'essential' metadata. The other point Bob made was that, for this repository, no-one was being paid to enter metadata, so it was essential that the right balance between simplicity and good metadata was maintained.

Dear colleagues, On Monday 25/10, Mrs Amee Evans Godwin, from the Institute for the Study of Knowledge Management in Education, gave a Keynote, addressing some of the main questions of the e-Conference. More information on the Keynote and ISKME activities, can be found in the Resources section, HERE To replay the Keynote, please click HERE and choose the "Replay" option in the popup page. With kind regards, Nikos Palavitsinis