Hierarchical navigation in a Hydra repository

[Richard Green]

Hi all

 

At Hull we have started serious work on implementing our promised IR head.  Much of what we want to do (and how to do it) is clear in our minds; there is one notable exception.

 

As well as a facet driven discovery system we need to give our users the option of finding material by drilling down through a hierarchical structure (maintained by isMemberOf relationships) such as we have in the current repository.  Such a structure behind the scenes also has advantages from the management point of view.  Last Friday we had a long meeting and finally decided on our definitive approach.  By Saturday we were sending each other e-mails admitting that we might have got it wrong!!!

 

My question, then, is this: how best to represent this drill down navigation to the user on-screen?  If anyone has done this (or is in the process of doing it) I’d be very grateful for an exchange of ideas.  I realise that the likes of Amazon have their own solutions to this but wonder if there are alternatives out there?

 

Richard

___________________________________________________________________

 

Richard Green

Consultant to Library & Learning Innovation, University of Hull

managing the CLIF, Hydrangea in Hull and Hydra (Hull) Projects

 

http://edocs.hull.ac.uk

http://www.hull.ac.uk/clif

http://hydrangeainhull.wordpress.com

https://wiki.duraspace.org/display/hydra

 

[Tom Laudeman]

Richard,

The AIMS group is creating functional requirements for Arrangement and
Description which is primarily a tool that in read-only mode looks and
acts like a desktop explorer or the Mac Finder. I think this is the
most common hierarchical representation system.

Amazon's navigation has some serious issues, and I'd be happy to list
those. Mostly, such a system depends on a comprehensive and normative
set of facets. It has a place in navigation, but such systems always
come up short for certain searches. (For example at Amazon, try to
find unlocked GSM cell phones without plans.)

Do you have a set of functional requirements for the IR hierarchical
navigation? What would be the issues with the desktop explorer
paradigm?

Cheers,
Tom

> *****************************************************************************************
> To view the terms under which this email is distributed, please go to
> http://www.hull.ac.uk/legal/email_disclaimer.html
> *****************************************************************************************
>

--
Tom Laudeman, AIMS programmer
tw...@virginia.edu
(434) 924-6732
http://www.people.virginia.edu/~twl8n/
http://laudeman.com/

[Richard Green]

Tom

Yes - the desktop explorer / Finder system has been considered and, indeed, may well form the basis of what we end up doing.

As to the Amazon issue you list - we would not be trying to "find" particular things, merely providing a drill down mechanism through background groupings (sets). Other search facilities would provide "find".

R

[Tom Laudeman]

Richard,

I don't know the schedule for Hypatia and its user interface, but you
could follow along and see if the results are suitable. I'm happy to
add you as viewers (and even editors) to the evolving Hypatia
arrangement and description functional requirements.

Drill down via facets requires detailed facets, and requires that
every object have a full set of facets. In my experience with Amazon
and Newegg, many items (and classes of items) are missing critical
facets. The navigation becomes impossible and frustrating. My
conclusion is that search is comprehensive, and facets are a
convenient alternate navigation or filter.

Do you envision the facet hierarchy having "other" or "unknown" for
items with a missing facet at a given level in the tree?

Cheers,
Tom

[Richard Green]

Tom

As I said in my first message, we're talking RDF in RELS-EXT as the basis for this, not hit-and-miss metadata.

R.

[Tom Laudeman]

Richard,

Right. I got "facet" stuck in my brain. The Hypatia A&D tools will
support an intellectual arrangement created by the archivist. This
arrangement will be a hierarchy of the entire collection. We are
implementation agnostic, but it will be the usual Fedora technology.

Is there a weakness in the desktop explorer metaphor as applies to
archival work?

Cheers,
Tom

[Tom Laudeman]

Richard, Simon, Chris,

Anyone with the link can view the Hypatia A&D functional requirements:

https://docs.google.com/document/d/1-dVn6pRLegKAtakpNIsT-5mqIyYuLJCebaBvQxI1TDQ/edit?hl=en&authkey=CI3SvuAC

This document is actively being worked on.

Cheers,
Tom

On Mon, Jan 31, 2011 at 5:04 AM, Richard Green <R.G...@hull.ac.uk> wrote:

[Mark A. Matienzo]

I should emphasize that the Hypatia requirements document is still
provisional and subject to substantial changes within the next few
weeks. Please direct any questions to me.

Mark A. Matienzo
Digital Archivist, Manuscripts and Archives
Yale University Library

[Richard Green]

"Is there a weakness in the desktop explorer metaphor as applies to archival work?"

This is for the IR, not necessarily for archival work. The metaphor is fine (indeed we have a tool in use for ingest to our existing system that represents a repository collection hierarchy in just that fashion), the main problem is lack of space - representing the hierarchy in Blacklight's narrow left-hand panel. For instance, see the attached mockup which would have to rely on 'hover' capability for some collections we have. Not ideal. So we launch a bigger popup (like with long facet lists)? Well, maybe - but that is why I wanted to canvass ideas from people working in the same space (and still do folks, please!!!).

[Tom Laudeman]

Richard,

I may have too many assumptions about your question. What is the
hierarchy a hierarchy of? Is it the intellectual arrangement of each
collection? Do you envision a human created hierarchy of all IR
deposited collections?

Thanks,
Tom

On Mon, Jan 31, 2011 at 5:04 AM, Richard Green <R.G...@hull.ac.uk> wrote:

[Tom Laudeman]

Richard,

Looking again at your example, the hierarchy appears to be script
generated based on donor-created facets. My concern would be that
donor's facets would be poorly conceived and incomplete. At both
Amazon and Newegg we find that incomplete facets make the tree
browsing very frustrating to use. Documents aren't found and facets
are misleading and incomplete.

Does any of this mirror your concerns?

-Tom

[Richard Green]

"Do you envision a human created hierarchy of all IR deposited collections?"

Yes, it is a human generated hierarchy representing an intellectual arrangement. None of it is machine generated.

R

-----Original Message-----
From: hydra...@googlegroups.com [mailto:hydra...@googlegroups.com] On Behalf Of Tom Laudeman
Sent: 01 February 2011 1:57 PM
To: hydra...@googlegroups.com
Cc: Simon W Lamb; Chris L Awre
Subject: Re: [hydra] Hierarchical navigation in a Hydra repository

[Bryan T]

hi all, 

I know this is a very old discussion but at California State University digital library system we have a need to implement tree (parent/child relation) facet for filtering and what you are discussing here seems to fit what we are trying to do so I wonder if there is a project in github that I can take a look. We are using hyrax and trying to have a filtering tree like

 Architecture Architecture

 Architecture Architecture: Architectural Engineering Architecture: Architectural History and Criticism  

Arts and Humanities Arts and Humanities

Arts and Humanities Arts and Humanities: African Languages and Societies  

thanks,

Bryan Tu

[Max Kadel]

This sounds like a Blacklight pivot facet, see documentation here - https://github.com/projectblacklight/blacklight/wiki/Configuration---Facet-Fields#pivot-facets. I don't have a specific repository that's implementing this, but I know when I've played with it it's been pretty straightforward.

Best,

Max

[Bryan T]

Max,

thanks for the tip.

Bryan

--
You received this message because you are subscribed to a topic in the Google Groups "samvera-tech" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/samvera-tech/59GvFLPw_dQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to samvera-tech...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/samvera-tech/3ee2986f-4686-4dd8-9463-d0e7337891b5n%40googlegroups.com.