[
https://issues.apache.org/jira/browse/JENA-420?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13612138#comment-13612138
]
Andy Seaborne commented on JENA-420:
------------------------------------
Stephen - thanks for starting this off. The UI for Fuseki would certainly
benefit from a complete rethink and rewrite.
I wonder if we can apply a "separation of concerns" to help break this down
into different independent strands. Then maybe we can make progress on
different strands at different speeds. These different views also relate to
security; the core server needs to be able to be run without risk of letting
functions from one view leak into another.
* user view
* publisher view
* operations view
The user view, (1), (2), (3) which is concerned with how the end-user makes
queries and handles the results. If it is very sophisticated, including
user-state management for stored queries or preferences, this could be seen as
a separate application running on top of the core server. The Resource
Explorer, navigating RDF and datasets, could be a whole project on it's own.
There is a very simple form in "xml-to-html-links.xsl" which puts clickable
links into the XSL on XML results but it there is a whole lot more that could
be done. Not sure how (4) fits in as a user view - maybe only in the
sophisticated, separate application.
At the same time, it would be nice to have a better, simple interface that
ships with the server.
The publisher view, concerned with the update and management of the data
itself. As well as SPARQL Update/upload/GSP, there is logging of what actions
were carried out when, backup (shared with operations?), some aspect of the
statistics (queries logged per dataset).
There is the operator view, concerned with the running of the server ((5), part
(4)?) and even that splits into monitoring and into administration of
databases. Integration (document how to, and/or provide the little adaptors
needed) with ganglia and nagios would be great.
For statistics, (4), Solr has a nice HTML page for display its stats (which I
think is driven by JMX).
> Web-based User Interface for Jena Fuseki
> ----------------------------------------
>
> Key: JENA-420
> URL: https://issues.apache.org/jira/browse/JENA-420
> Project: Apache Jena
> Issue Type: New Feature
> Components: Fuseki
> Reporter: Stephen Allen
> Labels: gsoc2013, java, javascript, rdf, sparql, velocity, web
> Attachments: fuseki1.png, fuseki2.png, fuseki3.png, fuseki4.png
>
>
> Fuseki could use an overhaul / rewrite of the web interface. It would be
> great to design a nice web-based GUI interface for both querying/exploring
> the data and for managing Fuseki itself.
> Tools and skills that may be required:
> Java, HTML+CSS, Javascript, Velocity (templates used server-side in place of
> JSP)
> A few ideas to get started:
> 1) Resource Explorer
> This interface allows the user to select any resource from either a list of
> all triples in a graph, or from the results of a SPARQL query. After
> selecting the resource, they can view all of the statements that reference
> the resource.
> 2) Improved Query Input
> It would be great to enhance the query input form for both queries and
> updates. Some features here that could be useful would be an IDE-like input
> box that performed syntax highlighting, automatic indentation, and easy
> commenting out of selected lines of text. Additionally being able to
> pre-populate the input box with the user's commonly used prefixes would be
> great (stored in a cookie or some such).
> 3) Improved Query Results
> Better display of query results (especially SELECT queries). Already
> mentioned was a tie-in to the Resource Explorer, so a user can easily see the
> statements about some interesting resource in the results. But it would also
> be great to layout and format the result table better. Even display the
> results on the same page as the query input box using AJAX. Or allow the
> table to be resorted after the query results have been returned. Also
> display the total number of rows.
> 4) Query Management
> View and manage currently running queries. Display how long queries have
> been executing, and allow the user to cancel queries that may be taking too
> long.
> 5) Fuseki Management
> There are many tasks here that would be great. Expose JVM properties via JMX
> in a nice UI. Things like CPU/Memory usage, disk space remaining, etc.
> Allow initiation of back-up tasks. Allow the user to download a snapshot of
> the repository. Shut down the server. Possibly add and remove datasets
> dynamically.
> Attached are screenshots indicating the current UI and the workflow for
> issuing a query. The idea would be to overhaul and replace most of this!
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
