[ckan-discuss] Topic list (vocabulary/valuelist) forclassifyingdatasets .

Thu Jun 14 10:32:14 BST 2012

Hi Pascal & David,

I agree on the importance of having a value list that offers unambiguous definitions. And international translations are definitely a bonus. I also agree with you and David that the number number of topics that are displayed to a user should be limited. A maximum of around 12 *displayed to the visitor* would be ideal: I agree with David on that. The EUROVOC offers 22 topics. However, I guess this will not be such a big problem, since it is likely that not all topics will be filled when a dataportal contains fewer than say, 5000 datasets (and for larger dataportals, having 22 topics may be justifiable). Furthermore, for large data portals a visitor will most likely perform a search on a specific keyword, and then filter down based on topics (using faceted search). It is unlikely that a specific keyword will have hits in each topic, so the topics list (facets) displayed will be less than the full 22 topics. For example: if a visitor would search on "Car", one would expect hits in the  TRANSPORT or ENVIRONMENT (etc) topics, but not in the POLITICS (etc) topics, limiting the number of topics displayed to the visitor.

Regarding your comment about tagging datasets with 72 GEOGRAPHY: I think having a geo component in a dataset does not automatically classify it in that category. For many geo datasets 48 TRANSPORT, 52 ENVIRONMENT or 56 AGRICULTURE, FORESTRY AND FISHERIES will make more sense. 

@Pascal: So from your response: do I gather you agree that Eurovoc is the best option to use? 

I am not sure if I follow what you mean when you say:
"we could implement a select dropdown that offers the subset to the end user but offer him the ability to describe its dataset with the top-level of this vocabulary". Could you elaborate?

Regards,

Edo Plantinga
data.overheid.nl

 -----Oorspronkelijk bericht-----
Van: p.romain at cg33.fr [mailto:p.romain at cg33.fr] 
Verzonden: woensdag 13 juni 2012 12:34
Aan: Plantinga, Edo
CC: ckan-discuss at lists.okfn.org
Onderwerp: Re: [ckan-discuss] Topic list (vocabulary/valuelist) forclassifyingdatasets .

	Hi Edo,

	I totally agree with you. Eurovoc is a more suitable vocabulary but what matters is the possible alignement those vocabularies offer.
	For example the term gezondheidszorg http://www.eionet.europa.eu/gemet/concept?cp=3866&langcode=nl&ns=1 <http://www.eionet.europa.eu/gemet/concept?cp=3866&langcode=nl&ns=1> refers to an exact match in Eurovoc, Agrovoc and to a close match to dbpedia.
	No one today would use the eurovoc 72 geography category to describe an opendata dataset including geolocalised data and the issue we are facing in my opinion is the number of thematic entries we offer to the end user : too many and we lose them, too little and we might confuse them.

	As a response to this issue we could implement a select dropdown that offers the subset to the end user but offer him the ability to describe its dataset with the top-level of this vocabulary   
	That's the way we currently go with our work on Ckan. We could probably share some code in that matter, couldn't we ?

	Best,
	Pascal Romain
	Chef de projet informatique documentaire
	Service Projets Etudes Conseils
	Direction des Systèmes d'Information
	05 56 99 33 33 poste 6643
	@datalocale
	  <http://www.gironde.fr/> 

	De :        "Plantinga, Edo" <Edo.Plantinga at koop.wmrijk.nl>
	A :        <ckan-discuss at lists.okfn.org>
	Date :        13/06/2012 12:15
	Objet :        Re: [ckan-discuss] Topic list (vocabulary/valuelist) for        classifyingdatasets .
	Envoyé par :        ckan-discuss-bounces at lists.okfn.org

________________________________

	Hi Pascal,

	Thank you for your swift response. 
	Looking at the GEMET vocabulary, it seems to be limited to geographic
	datasets. Naturally, many open datasets are geographic datasets.
	However, there are also many datasets outside this domain. How would you
	classify the datasets on the EU portal that are now under the current
	categories Finance and Budgeting, Education and Communication, Economy
	and Industry, Social Questions 
	Population & Health? The GEMET vocabulary does not seem to be suitable
	for this. 
	A subset of the EUROVOC (while still not perfect, looking at the
	datasets we have), seems to be a better fit for datasets outside the geo
	domain, in my opinion. I can see a fairly big overlap between the
	current EU portal topics and the EUROVOC topics, by the way.

	These are the EUROVOC categories I am referring to, to avoid confusion.
	04 POLITICS  08 INTERNATIONAL RELATIONS  10 EUROPEAN COMMUNITIES  12
	LAW  16 ECONOMICS  20 TRADE  24 FINANCE  28 SOCIAL QUESTIONS  32
	EDUCATION AND COMMUNICATIONS  36 SCIENCE  40 BUSINESS AND COMPETITION
	44 EMPLOYMENT AND WORKING CONDITIONS  48 TRANSPORT  52 ENVIRONMENT  56
	AGRICULTURE, FORESTRY AND FISHERIES  60 AGRI-FOODSTUFFS  64 PRODUCTION,
	TECHNOLOGY AND RESEARCH  66 ENERGY  68 INDUSTRY  72 GEOGRAPHY  76
	INTERNATIONAL ORGANISATIONS 

	The advantage of EUROVOC is that there are subcategories that can be
	used for classifying datasets that do not fall into an obvious category.
	For example: under what category should a hospital quality dataset fall?
	Searching on 'Hospital' on the EUROVOC site gives a search result of
	health policies of "MT 2841 HEALTH", so according to EUROVOC this falls
	under "28 SOCIAL QUESTIONS". Still not perfect, but hey, no
	catagorization system will be.

	I'd be interested to hear your take on this.

	Best regards,

	Edo Plantinga - Data.overheid.nl
	---------------------------------------------------------------
	Bezoekt u binnenkort een locatie van de Rijksoverheid?

	Dan dient u in het bezit te zijn van een geldige Rijkspas of een geldig identiteitsbewijs (paspoort, nationale identiteitskaart, rijbewijs of vreemdelingendocument). Indien u bij controle geen geldig identiteitsbewijs kunt tonen, wordt de toegang geweigerd. Legitimatiebewijzen van andere organisaties worden niet geaccepteerd.

	Dit bericht kan informatie bevatten die niet voor u is bestemd. Indien u niet de geadresseerde bent of dit bericht abusievelijk aan u is toegezonden, wordt u verzocht dat aan de afzender te melden en het bericht te verwijderen. De Staat aanvaardt geen aansprakelijkheid voor schade, van welke aard dan ook, die verband houdt met risico's verbonden aan het elektronisch verzenden van berichten.

	This message may contain information that is not intended for you. If you are not the addressee or if this message was sent to you by mistake, you are requested to inform the sender and delete the message. The State accepts no liability for damage of any kind resulting from the risk inherent in the electronic transmission of messages.

	_______________________________________________
	ckan-discuss mailing list
	ckan-discuss at lists.okfn.org
	http://lists.okfn.org/mailman/listinfo/ckan-discuss <http://lists.okfn.org/mailman/listinfo/ckan-discuss> 

	__________________________________________________________________

	Ce message et toutes les pièces jointes sont confidentiels et établis à l'intention exclusive de ses destinataires. Ce message ne constitue pas un document officiel. Seuls les documents revêtus de la signature du Président du Conseil Général ou d'un de ses délégataires sont de nature à engager le Département. 
	Toute utilisation ou diffusion non autorisée est interdite. Tout message électronique est susceptible d'altération et le Département de la Gironde décline toute responsabilité au titre de ce message s'il a été altéré, déformé, falsifié.

---------------------------------------------------------------
Bezoekt u binnenkort een locatie van de Rijksoverheid?

Dan dient u in het bezit te zijn van een geldige Rijkspas of een geldig identiteitsbewijs (paspoort, nationale identiteitskaart, rijbewijs of vreemdelingendocument). Indien u bij controle geen geldig identiteitsbewijs kunt tonen, wordt de toegang geweigerd. Legitimatiebewijzen van andere organisaties worden niet geaccepteerd.

Dit bericht kan informatie bevatten die niet voor u is bestemd. Indien u niet de geadresseerde bent of dit bericht abusievelijk aan u is toegezonden, wordt u verzocht dat aan de afzender te melden en het bericht te verwijderen. De Staat aanvaardt geen aansprakelijkheid voor schade, van welke aard dan ook, die verband houdt met risico's verbonden aan het elektronisch verzenden van berichten.

This message may contain information that is not intended for you. If you are not the addressee or if this message was sent to you by mistake, you are requested to inform the sender and delete the message. The State accepts no liability for damage of any kind resulting from the risk inherent in the electronic transmission of messages.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/ckan-discuss/attachments/20120614/d4d9d77f/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 7254 bytes
Desc: ATT1407859.jpg
URL: <http://lists.okfn.org/pipermail/ckan-discuss/attachments/20120614/d4d9d77f/attachment-0001.jpeg>