The purpose of Task 3.1 is not just to provide a translation of terms but to conceptualize the cultural context of each country. Keeping this in mind, partners are asked to provide one or two short paragraphs describing the challenges they encountered in undertaking the translation of the CMM Multicultural extensions. D3.1 will have a greater breadth and depth when we incorporate these descriptions outlining cultural differences. (Please upload your views on or before Wednesday 15 th April 2009. Thank you.)
PARTNER and COUNTRY: NIS-SU, Bulgaria
|
Can you mention below any instance or instances when the translation could not capture nuances in your local context accurately enough?
NO.
|
Which element or elements were the most difficult to translate. Why?
NO, it was easy.
|
If you wish to make any other comments on this matter, feel free to write them here.
Translation of CMM metadata tags and descriptions was easy and straightforward.
|
PARTNER and COUNTRY: ITD - Italy
|
Can you mention below any instance or instances when the translation could not capture nuances in your local context accurately enough?
Drawing down the Italian CMM Multicultural extension, a moot point was how to manage the concept of “learning object”, whether keeping this English expression in the Italian version of CMM, or translating it into an Italian expression.
In fact, the literal translation of “learning object” into Italian (“Oggetto di apprendimento”) doesn’t capture the real meaning of the English expression and other Italian expressions don’t cover all the nuances of the original concept. In addition, the “learning object” English expression is fairly used in Italy to identify reusable learning resources. But, in our view, this expression is more used by experts in the e-learning field (online trainers, online academic Professors, technicians, etc.), while traditional school and academic teachers are not so confident with it. In addition, sometimes, the “learning object” concept is intended in a negative perspective, being identified only with decontextualized chunks of contents for self-learning which don’t support any kind of collaborative learning.
So we decided to use the “risorsa educativa digitale” (literally “digital educational resource”) Italian expression, even if it limits the set of the educational resources to the “digital” ones (which is true for the Share.TEC project, but not for the LOM standard [IEEE, 2002]).
|
Which element or elements were the most difficult to translate. Why?
We encountered some problems when translating the vocabulary of values associated to the “10.5.2 - Didactic Strategy” element. As a matter of fact, the literal translation of some expressions provided in the list of didactic strategies is not so telling and sometimes long circumlocutions are needed to explain the concept. For this reason, Italian educators often use English expressions to talk about didactic strategies. Anyway, we choose to translate as better as possible into Italian expressions all the values in the vocabulary.
____________________________________________________
The CMM is based on the LOM v1.0 Schema [IEEE, 2002] and derives from it a lot of technical expressions.
We think that technical terms (especially those ones located in “Size” and “Data type” columns), such as for example “Langstring” or “LangCode” shouldn’t be translated, because they are only for internal use.
Other LOM-specific expression has been hard to translate. In particular, we had some problems to identify the slight difference in the meaning of:
-
“entry” (1.1.2 and 3.1.2),
-
“metadata record” (3 and 3.1.2),
-
“metadata instance” (3.2 and 3.2.2).
We considered:
-
“entry”, as a specific string characterizing a learning object or a metadata record;
-
“metadata record”, as the complete set of metadata associated to a learning object;
-
“metadata instance”, as the specific set of metadata provided in a particular moment by a particular contributor.
These terms are not easy to be translated into Italian: in this case, we choose to adopt a literal translation approach, even if the resulting Italian expressions are not so “user-friendly”.
|
.
Drawing down the Italian CMM Multicultural extension, we realized that the educational resources that are going to be managed by the Share.TEC system are referenced as “learning objects” in categories 1 to 9 (according to LOM schema) and as “digital contents” in category 10 (according to TEO). We think that a unique term should be adopted and that the consortium should find a shared opinion about it. We will clarify this point in D2.3.
|
PARTNER and COUNTRY: UVa Team, Spain
|
Can you mention below any instance or instances when the translation could not capture nuances in your local context accurately enough?
We have not found any problem when we have made the translation of the CMM Multicultural Extension Guidelines to Spanish. We think that our cultural context nuances are integrated in the direct translation from the original document.
|
Which element or elements were the most difficult to translate. Why?
There have not been any special difficulties when we have translated the terms. We can only point out two terms that are not too much clear:
We can find this terms in 2.3.1, related to the role:
-
Terminator (En): finalizador (Sp) (Entidad que realizó el objeto de aprendizaje)
-
Iniciator (En): Iniciador (Sp)
These two terms are not so explicit, but we have not found better terms to translate it.
|
|
PARTNER and COUNTRY: SU, Sweden
|
Can you mention below any instance or instances when the translation could not capture nuances in your local context accurately enough?
e.g. “value space” was hard to find a suitable solution for in Swedish because we are uncertain of the term´s meaning and use.
|
Which element or elements were the most difficult to translate. Why?
-
|
If you wish to make any other comments on this matter, feel free to write them here.
It is not unusual within the Swedish context of this field to use terms and words in English. Some terms are more used in English than in Swedish and translation is sometimes not necessary. Some concepts would be easier to understand and handle if we keep the English terms. The Manager of Cataloguing at Stockholm University claims that e.g. “namespace” is used within the Swedish context.
|
APPENDIX C
Discussion after the delivery of the first version of D3.1
ITD: The impression we get is that we generally agree on the result we want to reach, but have different ways of looking at the steps and the tasks that will help us get there. Basically we're dealing here with 3 components – the Common Metadata Model (CMM) and the Multicultural Metadata Model (MMM) and the interface/application layer and their interrelations. We’ll address some of OUNL’s points on these first.
Common Metadata Model
It seems that we conceive a broader role for the CMM than OUNL does. Our understanding is that for OUNL, the CMM is primarily a shipping vehicle for metadata exchange, mainly via harvesting. For ITD, the CMM is also the formal specification of (meta)data, in that it informs the design of the repository schema.
OUNL: Neither does the user need to have ever heard of LOM or the CMM.
ITD: Right (nor of the MMM, for that matter). These are back office concepts. There was never any intention to explicitly present them to the user as unitary concepts: no point.
Multicultural Metadata Model
OUNL: So we can agree that the MMM should focus on translating the vocabularies rather than the metadata element names? In that case, I would suggest to do so in a separate document that can accompany the CMM, but that is not part of the CMM.
ITD: There has never been any suggestion that multilingual or multicultural information of any kind would be part of the CMM. Any such information pertains to the MMM and/or the application layer. This is true regardless of whether that information comes via partners or from official LOM translations. We wonder whether the concern here is actually about Del 2.3 containing/reporting both the CMM and the MMM (which does not make them part of each other of course). This was a technical requirement from the commission, not a conceptual choice. Returning to MMM’s focus, ITD’s idea is that the MMM should contain both the element names and the vocabs, with the multicultural effort obviously focused on the latter. Even with the wizard-type QandA interface that OUNL describes in place of plain electronic forms, the MMM will still provide significant support for user interface localization: the wizard questions will need to be expressed in the user languages, and this design-time activity can draw from the MMM as a reference.
OUNL: For trivial fields (that most people understand), like author of document, we can provide translations, but these are more intended for the end user interface.
ITD: Exactly.
OUNL: I would however limit to the fields that an average teacher educator knows (title, abstract, etc..). Bad examples would be relation, contribute, catalog, URI, ...
ITD: Sure. What finally determines whether examples are good and bad will be feedback from users. Where translations of these terms would actually be useful is in the *process* of localizing the RICK interface: goodies may well pass through as translations, while baddies get paraphrased in a question, say. In any case we need the MMM as a basis for this task, including translation of element names. This is especially true for all the elements in Sections 9 and 10, which will capture the multicultural dimension in describing digital contents.
OUNL: So to be clear we will not create translations like English -> Dutch; LOM -> lom; lom.general -> lom.algemeen; lom.general.title -> lom.algemeen.titel; But we will focus on translations like: vocational training -> beroepsonderwijs
ITD: Again we agree.
OUNL: Apart from that, the user interface can be translated. And we can translation labels or questions like: Allow commercial uses of your work? However, such localizations are not part of a metadata model.
ITD: If we changed the MMM’s name to, say, Multicultural Extension would anything significant change? It would still refer to the CMM, while remaining distinct from it.
OUNL: Teachers preferred tools like Slideshare or YouTube, where publishing did not require entering metadata. So, maybe the partners will prefer to insert the minimum number of metadata.
ITD: Fine but we don’t want to limit the expressiveness of our description language a priori. Some users, albeit a minority, may wish to describe their resources more thoroughly, especially if we give them strong incentives to do so. As we’ve said on previous occasions, it’s a matter of a having a flexible/personalised interface.
CENEC: The sections 9 and 10 of CMM are mandatory or recommended? The minimal instance of LOM excludes the fields of these sections, in respect of standard LOM and for a greater compatibility with others applications. So, what happens if many of our metadata does not contain a value for these fields?
ITD: All elements in sections 9 and10 are either recommended or optional. There are no mandatory elements there. As mentioned above, encouraging users to provide full, culturally characterised descriptions should depend on the incentives they have to do so: these may be extrinsic (rewards) or intrinsic (desire to make one’s contents more findable, desire to contribute, sense of belonging to a/the community, etc.)
CENEC: In this case how do we deal with multicultural aspects?
ITD: Of course, if an element is unspecified it either remains so or gets some default value (which is probably culturally independent).
UVa: The MMM has not been considered enough in the use case analysis and validation. During the workshop, we may try to get concrete feedback from the users on this general issue.
ITD: We also expect to get feedback on what has not been included in CMM/MMM, such as the possibility of including annotations in the metadata.
NIS-SU: Research on multicultural aspects should have been done even before making the very first TEO.
ITD: We see the development process as one of iterative refinement. Once the first prototype is evaluated in Dublin, we might want to revise some design aspects. In this respect, it is not really important whether you start bottom up or top down or whatever; what really maters is that you capture the various perspectives accurately. In a way, TEO, CMM, MMM and Co are half fabricates in the process of building the system.
CENEC: We would had to develop also the use cases before.
ITD: The above applies here too.
|
Сподели с приятели: |
