Preview Playtime: The Homepage

Preview Playtime: The Homepage

It’s time for us to highlight another new update featured in the preview of the new Europeana portal – go to preview.europeana.eu to see it. 

One of the biggest improvements to the new portal is the homepage. We’ve made changes so that it’s easier for you to identify Europeana’s treasures and get on with exploring them.

The blog section has been improved visually with images for each blog post, because we know when it comes to the web the majority of you prefer images over text. That’s why we’ve also enhanced the Pinterest section to include more than one pin and made it into a mini carousel. The idea is that as soon as you get to Europeana, you can see exactly what you want to see.

We really want to know what you think, so as always please use the comment section below. If you encounter any problems with the preview, we’d appreciate if you could fill out our bug reporting form – this will help us make further improvements before the launch.

4 thoughts on “Preview Playtime: The Homepage

  1. Dear David

    I’ve been watching a bit preview.europeana.eu and I have some issues that I send to you for if they were useful.

    I’ve taken as an example the work ‘Erik Satie Journée 11h: Alexandre Tharaud, Eric Le Sage’ [http://preview.europeana.eu/portal2/record//03707/08D3AB689D2DD0FD89AB648AA64FDEC1648C7D42.html?start=6&query=planos+satie&startPage=1&rows=24]. In this record, autogenerated tags show the term http://www.eionet.europa.eu/gemet/concept?ns=1&cp=6279 [label: plan. A scheme of action, a method of proceeding thought out in advance. (Source: GOOD)]

    This term ‘plan’ has no relation to the work of Satie, I only find that ‘piano’ is the Italian term for ‘plan’. In fact, this is the only occurrence of the word ‘piano’ in GEMET, as the Italian term for ‘plan’; GEMET does not have the word ‘piano’ for the musical instrument (piano in English, French, Spanish.., pianoforte in Italian). Of course it is obvious that GEMET has the enormous advantage of being multilingual, but may be is a too high level thesaurus for Europeana as the more specific term related to this work of Satie’s that one can find is ‘music’ (http://www.eionet.europa.eu/GEMET/concept?cp=5449&langcode=en&ns=1).

    Other autogenerated tags are ‘photograph’ and ‘music’ in different languages. Neither photograph nor music have the corresponding URI as it is the case of ‘plan’ http://www.eionet.europa.eu/gemet/concept?ns=1&cp=6279. It is understood the inclusion of ‘music’ in the autogenerated tags, but not ‘photograph’.

    Besides that, this work can’t be retrived searching ‘photograph + Satie’, nor ‘musik + Satie’ but yes with ‘plaan + Satie’ or ‘planos + Satie’. The behavior of these autogenerated tags in searching it’s not clear for me, I can’t say whether or not autogenerated tags are used in search or it’s being used other mechanism as is done in searching for Satie that retrives Satie, Satis, Sati… It seems that the idea is to facilitate a multilingual search based in this autogenerated tags and sounds a very good idea that Europeana could retrive ‘music + Satie’ (or whatever) in any European language. Maybe it is an ongoing indexing process.

    Broader concepts related to ‘music’ are http://www.eionet.europa.eu/gemet/concept?ns=1&cp=1921 [Culture (Society)] and
    http://www.eionet.europa.eu/gemet/concept?ns=1&cp=1912 [Cultural Heritage]; Broader concepts related to ‘plan’ are http://www.eionet.europa.eu/gemet/concept?ns=1&cp=2889 [Environmental Planning] and Broader concepts of ‘photograph’ are http://www.eionet.europa.eu/gemet/concept?ns=1&cp=2284 [documentation] and http://www.eionet.europa.eu/gemet/concept?ns=1&cp=4303 [information]

    However, broader terms seems not be involved in search at all, which seems quite logical to retrieve only most relevant results, none of these ‘autogenerated tags’ allows to retrieve this work of Satie [eg, search All fields ‘documentation + Satie’] nor appear to participate to obtain “Similar content”. The role of broader terms it is not clear and confuses a little.

    Can it be that the fact that a term is not in GEMET result in many records that don’t have autogenerated tags? [E.g., http://preview.europeana.eu/portal2/record//2022701/0CEE04C9DF98C43970F4919F4FED3CF1A8113A46.html?start=1&query=arquitectura&startPage=1&rows=24 has as subjects ‘civil architecture’ and ‘military architecture’, terms that do not exist in GEMET. But instead, http://preview.europeana.eu/portal2/record//07927/1EC45DCEB203D7D819235C5FBAC63B844167B96A.html?start=11&query=Sketches+in+Egypt&startPage=1&qf=TYPE:IMAGE&qf=TYPE:TEXT&rows=24 has among other subjects ‘Architektur’ that is on GEMET, but there are no ‘What autogenerated tags’]. Surely it is a ongoing process that will progress over time and that more and more records will have autogenerated tags.

    I hope these comments will be useful
    Best regards

  2. Hi Xavier,
    And thank you for your detailed issue report!

    A quick response:
    We’re aware that our enrichments includes a number of false positives. Often these are caused by the same word having different meanings in different languages, like piano meaning plan in Italian.

    Our plan (!) is to change the semantic enrichment to be more strict and be aware of the language of the literal when creating a corresponding GEMET resource. So in this case, with the language of the metadata record being in French a match would not be made with the Italian language label for the English word plan. Changing this may take some time because right now we’re focusing on more fundamental issues in creating a large scale ingestion process for EDM compliant metadata. We’ll have to deal with semantic enrichments later.

    I need to check with some colleagues more well versed in our enrichment to figure out why the record has received the labels for music but not the URI. That seems like a clear bug to me.

    That http://preview.europeana.eu/portal2/record//07927/1EC45DCEB203D7D819235C5FBAC63B844167B96A.html hasn’t been enriched with the resources for Archaeology and Architecture also seems like a bug to me. I will report it onwards.

    Finally, you’re very correct that a major part of the purpose of the semantic enrichments is to improve discoverability. Both by creating synonyms and by including labels in multiple languages and then indexing them for search in Europeana (portal and API) and also including them in our search engine optimization efforts (sitemap and semantic markup).

    Broader labels are if I recall correctly not indexed for search in Europeana (but are crawled by search engines). We may however in the future experiment with using them as top level facets or as broad subject based browse entry points (Music, Photography, Books, Archaeology, etc.) to Europeana.

    And finally, GEMET. When we began experimenting with semantic enrichment it was one of very few openly licenced and SKOSified ontologies. Its domain is not perfect for us but we needed to start with something and on balance I think that was the right decision.

    Again, thanks for the bug reports and if you’d like to discuss semantic enrichment please don’t hesitate to mail me directly at david.haskiya at kb.nl

    Best regards,
    David

  3. Thank you for your details in how to use the information given by Europeana.It is a european offer to all citizens of european nations and further to the humanity to save and difuse knowledge and information of the european civilazation

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.