Two new projects
I have two projects in the design stage I'd like to share and get some feedback on. One is for work and one for fun.
For fun, I have a writing site that's been open since 2000 that I'm converting from a CMS I wrote in coldFusion to Drupal. I need a way for individual users to create a story and then open it up for other users (of their choice) to access and edit with revision control. It's meant for large collaborative writing projects.
My first attempt is called 'collab' that can be connected into any existing content type you want to allow users to share. It works with the buddylist module. Sharable nodes get an extra tab (along with view, edit & revisions) for collaboration. This pulls up your buddylist with a set of checkboxes. You simply check the box to share the story with that buddy. Then they can edit and access revision control. It then adds that user as an author on all displays. Does anyone know if something like this already exists and I'm re-inventing the wheel? I could find anything like it. The book module is similar, but doesn't really do want I'm looking for.
For work, the library I'm currently working with uses a Follett system, but is going to be switching to Koha at some point down the road. I need to get them an OPAC online as quickly as possible. Follett's webcollection plus doesn't seem to be a pretty or flexible creature. So I'm looking at several options. But either way, I'm thinking Drupal for library usage needs something like an OPAC abstraction layer similar to Drupal's database abstraction layer. A set of generalized commands to manipulate and query the OPAC that can be wired through to any ILS with various .inc files. (Also like Mark's generalized authentication module). This is in the VERY, VERY early stages of development.
Any thoughts, features, comments on this would be great.
Phil
This work is licensed under a Creative Commons Attribution- NonCommercial- ShareAlike 2.5 License.
Several approaches
It looks like someone else is interested in integrating Koha and Drupal (http://drupal.org/node/117160) as well.
I agree, a way to mash up Drupal and OPACs effectively would be great. There are several approaches. The first is to use Z39.50 to connect to library systems. Work has already been done on this, for example http://drupal.org/project/z3950 (and my OPAC client mentioned in the PHP/YAZ thread would be another place to start as well). The problem with this approach is the overhead of Z39.50 and the vagaries of how well it is supported by various library system vendors.
The second approach is to load the records from a library system into Drupal. Scriblio (http://about.scriblio.net/) pioneered this with Wordpress. A related approach is to use Solr to index records dumped from a library system and integrate the Solr search into Drupal. I have seen a prototype of this approach and it looks good. I'll contact the developer to make sure he sees this thread.
Some commercial systems have proprietary APIs that could be used to provide a Drupal front end to their catalogues (see http://www.aadl.org/ for example, which uses one of these proprietary APIs to do just that), but they are all different from one another.
In any approach, we'd need to integrate individual records from the library system into Drupal so users could take advantage of Drupal's node system and all the wonderful modules that use a node-based data model. Having a user search a library catalogue from within a Drupal website but not enabling them to save records, comment on them, rate them, apply tags to them, etc. seems a bit disappointing.
This is an awesome idea and I encourage others to pitch in their $0.02 so we can hammer out some functional requirements, at least.
Solrpac
not exactly what you're looking for, but we're currently dumping records, indexing them in solr, and using a custom drupal module as an experimental opac: http://beta.lib.muohio.edu/.
it's our plan to release the source, but i want to make sure i've got "real/institutional" approval to do so, before i just start opening up a svn repo to the world.
send an email to rob[dot]casson[at]gmail, and i can create some personal accounts; you'll need a login to do the tagging, but will still be able to mark/export records, view others' tags, etc...even uses unAPI for zotero support ;)
would love any feedback, and i'll keep my eye on this thread...nice to hear that others are thinking along similar lines,
rob
Drupal and Solrpac
I should have guessed Rob was behind the Drupal OPAC stuff - awesome work and always a little ahead of the curve :-)
This is great stuff Rob - UPEI will be moving away from our Sirsi Unicorn system (planned in 2008) and I have been looking at your test OPAC as a great replacement for that part of the whole ILS thing. We are using Drupal as the front-end for a series of Fedora repositories with Lucene as the indexer, so this might work very nicely for us. Using one Drupal interface to search our OPAC and the Repository at the same time would also solve some interesting problems.
My main question on this kind of integration is how you plan to add/edit the records? Are you think of this as an adjunct to a full-blown ILS? I would like to consider it as a replacement, so would be interested in your thoughts on this.
Great work Rob...
Mark
SolrPac to be presented at ALA Midwinter
Just saw Thomas Dowling's email announcement on web4lib about this. The University of Virginia's Project Blacklight and Villanova's VuFind (both non-Drupal) are also on the bill.
Meeting details quoted from the email: "The LITA Next Generation Catalog Interest Group will meet on Sunday, January 13, from 4:00 to 6:00pm in the Hilton Garden Inn (conveniently located just outside the convention center), in Salon A/B."
OPAC questions???
Rob,
When you say "dumping records", how do you accomplish this? Is dumping automated within your ILS? How often does this need to be done in order to keep the front end fresh?
Thanks
Phil
re: 'collab'
re: 'collab', I think that Organic Groups module and its kin is supposed to achieve the kind of on-the-fly access control you imply:
http://drupal.org/handbook/modules/og
You could create organic groups that create only 'book' content for revision control, etc.
The apparent downside is that it doesn't play nice with other modules that manage access.
OG
I tried OG out and it did cause more problems than it solved. Apparently Drupal 6 is implementing this kind of control with a patch to the node.module, but the documentation wasn't very clear.
I found making a module to share arbitrary nodes wasn't really possible without directly hacking the node.module (don't want to do that right off the bat), so I've switched to building a specific node type that allows the kind of sharing I need.
Thanks
Phil
Millennium integration module
A short while ago I released an Alpha version of a Drupal module that sucks (well, hopefully, unsucks) records from your run-of-the-mill Millennium WebOpac via the MARC view option on every record, and maps most information into taxonomies and CCK fields.
See:
http://drupal.org/project/millennium
Post new comment