Training

In this section you can find training materials produced by ATHENA on LIDO and the ATHENA ingester.

Training material provided in the ATHENA Training Workshop, Rome, 18-19 January and Berlin, 20-21 January 2010

LIDO

ATHENA INGESTER

UPDATINGS ON LIDO AND THE ATHENA INGESTER

  • 11 June 2010, Ljubljana (VIDEO_1) (VIDEO_2)

  • The LIDO v0.9 as currently implemented in the ATHENA ingester can be found at:

    • LIDO v0.9 Specification Document ()

    • LIDO v0.9 XML Schema Definition ()

  • The now applicable version LIDO v1.0 can be found at:

  • LIDO v1.0 Specification Document ()
  • LIDO v1.0 XML Schema Definition ()

 

Basic rules for mapping

by Regine Stein

  • Map first the mandatory elements, make sure that the respective fields in your source data are filled in.
  • Populate indexing elements if possible, and only in second step display elements (people tend apparently to fill in only display elements because it seems easier but is usually not appropriate)
  • Analyze your source data for event-mediated information: look for implicite event types in field names and for qualifier elements
  • Think of your data outside of your own context - add implicite information: e.g. the repository name, if you map the inventory number

Guidelines on  the use of the ATHENA ingester 

by NTUA

The Athena Tool can be reached at the following url:

http://athena.image.ntua.gr/athena (RESERVED TO THE ATHENA CONTENT PROVIDERS FOR THE INGESTION OF THEIR CONTENTS)


User & Organization Registration

New users can create an account through the "Register" page.

You can apply to an existing organization during registration or through user profile (for returning users).
If your organization is not registered in the system and you will be acting as administrator, you should register leaving the "Organization" field blank. This will grant you administrative rights to create your own organization and manage all its users and data.

If you are acting as coordinator for organizations in your country, we advise that you create an organization that will serve as the aggregator e.g. "Ministry of Culture". Administrators of respective organizations that will be monitored by the aggregator can subsequently set it as the 'parent' (alternatively, an aggregator can create all the sub-organizations himself and assign users/administrators to them).

When a user applies for an organization the administrator is notified through email in order to validate and assign appropriate role. More info on available user roles & rights can be found in the tool's home page.

Make sure that athena-admin@image.ntua.gr is not marked as spam by your email software!


Data Import

The tool can only handle xml representations of metadata. They can be uploaded in single file (containing one or more items) or multiple ones bundled in a ZIP archive (rar compression is not currently supported), or can be harvested from an OAI repository. Make sure that the files you are trying to import contain valid xml.

Characters like "<" and "&" are illegal in XML elements. In text and attribute values, you need to escape those ASCII characters using the following encodings:

' '
" "
< <
> >
& &


Overview

Use this page to browse all imports made for the organization/s you have access to. Follow the indicated steps (1.define item root, 2.define mappings, 3.transform your items) for each import dataset. Mappings can be edited and reused.

You can switch between import and item view using the options provided and you can also preview the mappings and final transformation by using the item view.


Mapping

Documentation and tutorial material for the reference model (LIDO) and the functionality of the mappings editor can be obtained from respective links in the user home page (login required) at http://athena.image.ntua.gr/


Transformation and OAI harvesting

Both processes can be time consuming and may also be queued for completion depending on the system's load. You can logout and check back later to see status updates.


General


Please make sure you use Firefox browser to get the full functionality (the tool has not yet been verified for all browsers).

Also note that until March the 1st the tool will be used for testing purposes. The process of actual ingestion will begin on March 1st, so until then changes in the tool can occur and data may be deleted for database rebuild purposes.

Any technical issues, questions, suggestions and feature requests should be directed to
athena-helpdesk@amitie.it (if you are not registered, please send the relevant registration form)


Guidelines for publication on Europeana (25 October 2010)

AGGREGATION PROCEDURE

  • First of all, the partners who have not done yet, MUST immediately register in the ATHENA ingestion system (http://athena.image.ntua.gr/athena). For any question about the procedure, please e-mail to Gordon McKenna (CCing Valentina Vassallo and Marzia Piccininno).
  • As soon as possible, every partner SHOULD UPLOAD AND MAP TO LIDO a sample of data that covers all the elements in XML file (even a small set of records is sufficient for the test). 
  • When the data sample has been uploaded, the partner SHOULD INFORM Valentina and Gordon (CCing Marzia). For any problem with the uploading/mapping, please send an e-mail to Gordon (CCing Valentina and Marzia) who will review data in order to make the partner able to proceed with the final uploading.
  • Before the publishing procedure, ASK FOR FINAL AUTHORIZATION to Gordon (CCing Valentina and Marzia).
  • After the authorization, the partner WILL PRESS THE PUBLISHING BUTTON (please inform Valentina, Gordon and Marzia) in order to activate the harvesting process by Europeana.
  • If there are changes respect to the previous mapping, the partner MUST ASK AGAIN FOR THE VALIDATION. If there are no changes, partners CAN press the publishing button but they MUST inform Valentina, Gordon and Marzia of the new situation.
  • It is important that if the partners have collections not yet ready to be uploaded, SHOULD programme periodic upload in order not to overcrowd the system and the procedure in the last period of the project.

OTHER GUIDELINES

  • Partners SHOULD supply a link to a thumbnail preview which is preferably the same width or slightly larger than the requirements of Europeana (currently 200 pixel wide). Partners SHOULD NOT supply a link to a very large image (i.e. much wider than 200 pixels). If the thumbnail is larger than the requirements, Europeana WILL  REDUCE it. If smaller, Europeana WILL NOT RESIZE it (It is recommended that content providers look into possibilities to supply bigger size source images for image caching if possible).
  • Partners SHOULD NOT supply a link to a ‘dummy' thumbnail that indicates that there is no image or that no image can be displayed for IPR reasons. If this in not easily possible, partners SHOULD supply  the URL of the ‘dummy' thumbnail to Gordon.
  • Europeana WILL NOT ACCEPT records without showing a digital object (image, etc.) on the partner's site. However, these records CAN BE UPLOADED on the ATHENA tool.


For any question of general interest, please send your requests to the technical list.


Clarifications concerning the ATHENA server (that can help the content ingestion):

 

  • Your imports are visible to your organization alone; only published ones matter, you can keep or delete the rest at will and even have the same import twice and more.
  • For each provider, Europeana accesses one merged set of ESE items; there is not information about imports.
  • Every time you publish/un-publish, your ESE set is updated.
  • To un-publish follow the same procedure; drag and drop, from right to left part.
  • Published imports cannot be altered (e.g. re-transformed with an updated mapping) unless un-published first.
  • Update of the ATHENA OAI repository (currently performed only manually) means that NTUA replace your old ESE set with the current one.
  • There is a duplicate detection mechanism that will not add identical ESE items in your published set even if they exist in different imports. Moreover, Europeana also checks for duplicates based on some key elements (id, shownAt, shownBy etc)
  • If two ESE records for the same object are not identical (in case of updated metadata in a newer import) then you will have both in your ESE set; unless you un-publish old import and publish only new one.
  • Advice: you should always have all the items you want published in your ESE set so that Europeana can easily delete your old submission (thumbnails remain cached) and replace it with the current one.