Difference between revisions of "Library Usage Instructions"

Revision as of 20:32, 11 July 2017

Add a Resource to the Library

• Link to add a resource - http://knoesis.org/add-library/ (This interface requires a login. Please contact Kno.e.sis Web Support in case you do not have a login/have lost your login)
• Once Logged in, you have two options to add your library:
  • Either use one of these four options to auto-populate your new library entry:
    • PubMed Lookup - enter a PubMed ID and click "Populate using PubMed"
    • Paste RIS Record - Paste RIS Record of your library in the text field and click "populate using RES record"
    • Paste BibTex Record- Paste BibTex Record of your library in the text field and click "populate using BibTex"
    • DOI Lookup - Enter a DOI name in the given format and populate the entry. NOTE: You need to register with CrossRef and then enter your CrossRef UserID in your profile before using DOI Lookup.
  • Or select one of the publication type from the drop-down menu available below "Next" button and enter the library details manually. Please verify your library details before publishing.

• You can upload any external file you need to attach with the library entry in the "Related Files" section.

• Please read the instructions on each of the forms carefully before adding your resource.

• You can click on "save" button available on the bottom of the page to save and publish your library.

Edit a Resource in the Library

• In order to change some information in a resource you would have to use the edit interface.
• Start with a resource in the library. As an example http://knoesis.wright.edu/library/resource.php?id=780
• Now click on the link “Edit this page”. You would see 2 buttons
  • Edit_Entry
    • Login is required for this functionality
    • Will show all the information for this resource to be edited
    • You should see a message on submitting the form which would say “Publication edited successfully!!"
  • Delete_Entry
    • Login is required for this functionality
    • Will delete all the triples from the RDF store for this resource
    • You should see a message on submitting the form which would say “Publication deleted successfully!!”

Important Note

• For all the forms after submission please click the link “Jump Reload To Library” in order to see the changes in the front end at http://knoesis.wright.edu/library
• It might happen that your browser might cache the page http://knoesis.wright.edu/library and hence you would not see the new resource added. It is better to use the a different browser in such a case.

Other Important Information About Library Code

• We used forms for adding data/records to the database. The forms were built using JSP. On submit data was converted to RDF using JENA and was written to the database (I think we were using RDF Sesame version 2.4.1).

• The main page (Simile exhibit) requires the data to be in Exhibit JSON format. In order to convert RDF data (stored in the database), we used simile babel service (http://service.simile-widgets.org/babel/)

• The problem with Simile exhibit is that it loads all the data in the JSON at once. This overloads some browsers like Internet explorer. In order to over come this issue Ajith wrote some scripts that would create JSON - smaller chunks of data (i think by reading the big JSON converted using babel).

• The other pages (individual records E.g - http://knoesis.org/library/resource.php?id=1690) are PHP pages, that get all their content from the JSON itself.

Questions About LIbrary Code

So here are some high level steps -

• We used forms for adding data/records to the database. The forms were built using JSP. On submit data was converted to RDF using JENA and was written to the database (I think we were using RDF Sesame version 2.4.1).

• The main page (Simile exhibit) requires the data to be in Exhibit JSON format. In order to convert RDF data (stored in the database), we used simile babel service (http://service.simile-widgets.org/babel/)

• The problem with Simile exhibit is that it loads all the data in the JSON at once. This overloads some browsers like Internet explorer. In order to over come this issue Ajith wrote some scripts that would create JSON - smaller chunks of data (i think by reading the big JSON converted using babel).

• The other pages (individual records E.g - http://knoesis.org/library/resource.php?id=1690) are PHP pages, that get all their content from the JSON itself.

Recommendation -

• If possible and if time permits I would recommend moving away from Simile Exhibit. The user group is not very active, the system is buggy and very difficult to work with (it needs JSON is a specific format which means the ontology needs to be very simple to be able to convert JSON and rendered on UI). Naveen suggested using http://flamenco.berkeley.edu/ (Abzooba used it for an internal product) and was very happy with the tool

• Would also recommend moving away from RDF Simile. Great database back in the days but has the same issues as 1 with the newer versions. Allegrograph, Stardog, Virtuoso are better choices (ordered by popularity)