Hi everyone,
I was wondering if anyone had any comments on the http://www.vivoweb.org project and it's relationship ship to Drupal. VIVO in a nutshell, is a web application provides a semantic database that can be searched that contains information about scientists, their research interests, affiliations, publications, grants, etc. VIVO is in deployment at several universities and is being expanded through a $12.2m stimulus grant from the National Center for Research Resources (NCRR) of the National Institutes of Health (NIH). see more at http://www.vivoweb.org/about/faq/about-project
It appears to be a java application that runs on Tomcat and used Jena/SDB as a triple store and apache lucene for searches. I've downloaded and looked at it, and it is amazing that they are using semantics for it, but, the application they have created to view and manipulate the RDF data needs a lot of work. Their security model is too simple and the editing screens jarring. No doubt they will work on it and it will improve it in the future.
That said, they have created their own rdf ontology for the VIVO project. --> http://vivoweb.org/ontology/core and I thought perhaps their ontology could be brought into Drupal (7) and the functionality of their application could be matched and surpassed quickly. The resulting application being compatible with the other VIVO sites since they would all talk the same ontology, as well as, the rest of the semantic web being RDF. The Drupal version of VIVO would then inherit the myriad of modules available in Drupal and far more robust CMS capabilities. Thoughts anyone?
Erich Bremer
erich@ebremer.com
http://www.ebremer.com
Comments
Thanks for bringing this to
Thanks for bringing this to everyone's attention, I saw this in the comment on my blog post too, but didn't have a chance to respond since I've been at SemTech.
I looked at VIVO when I was thinking about Universities in RDF. Definitely some really cool stuff going on with the VIVO project and it would be great to bring some of that thinking to Drupal.
Unfortunately, I couldn't find documentation for the ontology and the formatting of their OWL document is unreadable. Have you found human readable ontology documentation somewhere?
Hi Lin, I hope your
Hi Lin,
I hope your presentation went/goes well. I'm hoping I can make it to SemTech next year.
I'll see if I can find something more readable on their ontology. I'm still studying what they have created. They are having their first annual VIVO National Conference on August 12 & 13, 2010 at the New York Hall of Science that I'm going to attend and see what they have to say. (http://conferences.dce.ufl.edu/vivo/) The VIVO/RDF in Universities concept is fantastic, I have need to build something like this for Stony Brook/SUNY right now. I'm just not so happy with the current state of VIVO at the moment which is making me long for Drupal 7 :-)
Vivo at Cornell
Although I'm not doing any development on Vivo my boss (whose office is next to mine) developed the initial system. There's actually a meeting going on right now a couple of doors away with all of the Vivo developers at Cornell (the room is full). The Vivoweb project is huge, and there is a lot of work in progress so I wouldn't judge Vivo based on the state it's in right now. As I have been doing some work with Vivo and Drupal 6 integration I have been following what they're doing fairly closely. In any case, I'll point those working on Vivo here at Cornell to this thread so that they can't answer any questions directly.
Do the Vivo devs at Cornell interact with the Ontolog Forum?
Off topic a bit but trying to build synergies http://ontolog.cim3.net/cgi-bin/wiki.pl/
Good to hear there is more
Good to hear there is more going on. Do you have a roadmap yet that can be looked at? - E
Vivo roadmap
I don't know where to find the Vivoweb roadmap personally but I'm sure I can find one next time I talk to my boss. BTW, there is a poster across the hall from me that has sort of an org chart of all the people working on Vivoweb. I asked the person whose office it's next to how many people are involved with Vivoweb and she told me about 100.
This looks a promizing work -
This looks a promizing work
- Opening in Protégé, it is extensively self-explicative
- their web site http://vivoweb.org is Drupal!
Great to hear that it is
Great to hear that it is readable in Protege. However, if Web developers are going to implement it in Web applications, it needs to really be "on the web", which means also human readable on the Web.
I agree with with you Lin, it
I agree with with you Lin, it would be helpful to have a human-readable version readily available. We have a semantic-divide between those who understand semantics and those who don't even know it exists yet. A human-readable version will help those who don't understand the technology to appreciate what is covered in the ontology. And the lazy people who do understand the technology, to not have to bother to load it into Protege to get a quick glance. ;-)
Graphic view
For lazy ontologists, I extracted a graph from the Vivo ontology
http://www.controlchainmanagement.org/node/1330
It's great to see discussion
It's great to see discussion about VIVO starting here. Like fereira, I've been loosely involved with the VIVO project at Cornell, where much of the development is taking place. Being quite familiar with the application and the project in general, I would say it's scope is beyond the capabilities of Drupal in its current state, and Drupal as a replacement for the application just isn't feasible. However, there are huge possibilities for integration between the two systems.
The way I see it (which is definitely oversimplified), the VIVO software handles all the data management. This includes ontology management and content management. And in VIVO's case, content management involves not just creating/editing content, but also huge content ingests from various data sources throughout an institution, all being mapped to the VIVO ontology. The VIVO software will provide a very nice frontend for its content, which will be focused on the main goal of the project: to enable scientists to find each other and collaborate.
Now there obviously endless other possibilities for all that semantically rich content. While I don't see it mentioned on the VIVO website, I know that all published content will follow Linked Data principles. And while it's not clear whether a SPARQL endpoint will be included, I imagine setting one up alongside a VIVO installation will be straightforward. The content will be accessible.
Drupal's already in a good position to start using this content, particularly with all the great new stuff in D7. Lin's SPARQL Views project is definitely going to open some doors too. Similar to how DBpedia has acted as a "lab rat" in many of Drupal's Semantic Web efforts, I think VIVO content will provide a great means for experimenting with semantically-driven sites and applications. (sorry VIVO people, didn't mean to call you lab rats)
For an example of the kind of content I'm talking about, take a look at Cornell's early VIVO site: http://vivo.cornell.edu
Drupal and VIVO
I agree that this is a useful discussion. We are looking into integrating VIVO with our faculty information system at my university, and I'm separately interested in Drupal 7's support for semantic web having set up some Drupal sites in the past. I will be attending the Drupal Camp Colorado this weekend and that first VIVO conference in August. I plan to learn much more and meet others with similar interests. I encourage everyone that can attend the VIVO conf.
I'd previously found Protégé useful in understanding the VIVO Core ontology, but I'm used to debugging C/C++ and Perl code ;) That said I appreciate efforts to make the ontology more accessible, as well as exploring solutions for share extensions that aren't pulled back into the core. I recently had a good discussion with a local developer working on an open source project to support this, and also came across OntoWiki which looks interesting. I look forward to future communication/collaboration. If you will be at either conference please drop me a line.
How about posting your screencast
Hi Miles,
How about posting your last screencast of the work you've been doing integrating vivo with Drupal. I bet there are a few year that would like to see it.
Here's one of the early integrations of vivo data with Drupal that Miles worked on:
http://cals-experts.mannlib.cornell.edu/
It's mostly a proof-of-concept system that is not in production but it essentially uses the FeedApi/Feed element mapper/CCK modules to create nodes and taxonomies from vivo data. Both Miles and I have since done some development using the Feeds module to essentially do the same thing.
There is also a Google Summer
There is also a Google Summer of Code project to enable easy import of RDF content using Views and the Feed API. Depending on how this works, SPARQL Views may be able to get the SPARQL content in and then automatically feed into nodes.
http://groups.drupal.org/node/57223
Recent integration work
John:
When I have things in better shape I'll certainly be posting details. Don't want to claim to have something useful when it's not yet usable :)
However, the work I've been doing with Drupal 6 really doesn't make use of the richness of VIVO content -- the semantics get stripped away and only raw content is being saved. Other than fetching content via SPARQL, I'm not doing anything yet that leverages semantic web tools.
Lin:
Thanks for pointing out the GSOC project, I hadn't seen it yet. Feeds is such a flexible and well-designed module, linking it to Views could lead to amazing things.
VIVO - Drupal synergies
As the development coordinator for the VIVO team at Cornell and the larger project, I can confirm what Miles has indicated -- that there are many synergies between VIVO and Drupal. He has been doing considerable work to facilitate data feeds from VIVO to Drupal via SPARQL queries, including detecting the properties present in results of a SPARQL query and setting up the fields of a custom node type -- I hope he'll describe this further on this forum soon.
As Miles indicates, VIVO does have a web front end to complement its ontology and content creation/ingest/editing functionality. The web front end will be improving dramatically in upcoming releases with a much cleaner MVC architecture, top-level navigation, better page-level content display control, and visualizations. At the same time, however, we fully recognize that a platform like Drupal will offer much broader functionality for content management and delivery, as well as benefit from a larger development community. We have been exploring VIVO-Drupal pathways in order to make co-evolution easier and help provide developers a full spectrum of semantic tools.
On linked data (http://linkeddata.org), the Cornell site is still being migrated from an earlier ontology to the latest code base and ontology, but the nascent University of Florida VIVO (http://vivo.ufl.edu/) does respond to linked data requests -- try entering the URI of the project director (http://vivo.ufl.edu/individual/n25562) in the Marbles RDF browser at http://marbles.sourceforge.net/ (note that Marbles is a research instance so may not be available at all times). Miles is also looking at ways to bring RDF from linked data requests to Drupal as conveniently as the results of SPARQL queries.
The VIVO development team at the University of Florida is also exploring Drupal as the front end delivery platform for VIVO data at UF -- I hope they will post here when they feel they have something to show.
I'll also put in a shameless plug for the upcoming VIVO conference (http://conferences.dce.ufl.edu/vivo/) on August 12 & 13 in NYC -- many of the developers and ontologists including the entire Cornell team will be there and eager to discuss how VIVO could be improved by closer ties to other open source projects including Drupal. Note also the conference has a competition for applications using data coming from VIVO systems to support science with entries due July 31 -- look under the Call for Submissions.
VIVO Drupal integration
I attended the VIVO Conference in NYC and heard that there was a poster on Drupal integration, but need to track down who did it. I'll post info when I figure it out.
VIVO+Drupal poster
August 2010 conference materials are indexed here:
http://www.vivoweb.org/conference2010
The poster you reference is available here:
http://www.vivoweb.org/files/MilesWorthington_VIVO_conference_poster-web...
The poster's author is Miles Worthington, an Interface Designer at Cornell University. Per the link below it looks like he was on the design team for their Mann Library web site which is built on Drupal. You might be able to get ahold of Miles via the library staff contact resources (he is not listed, but maybe one of his teammates is) or a little googling.
http://www.mannlib.cornell.edu/about-us/about-site
VIVO+Drupal poster
Miles isn't listed on the library staff page because he doesn't work for Cornell anymore. He's currently in Thailand but still does some consulting work for us. He also follows this group so he can speak for himself. BTW, the staff directory on that site is not built from a feed from Vivo but I'm hoping to get that changed. I developed the service and initial module that produces the staff directory on the site (the data comes from another source).
That was me
kuniholm:
Apologies, I meant to post something here about my poster at the conference.
What I've been working on is an import system for mapping VIVO RDF to Drupal components such as CCK and Taxonomy. The D6 module I've been working on is set of Feeds module plugins that use the ARC2 libraries for handling RDF. I don't have the module in a useable state yet, but hope to in the next few weeks.
I've posted a PDF of my poster for anyone interested. It's more of a general overview than a technical explanation. Please feel free to send me any thoughts or questions!
Thanks!
Miles, I appreciate the link to your poster. Did we talk at the conference?
Maybe we can work together on this, we could help you, or perhaps you'd be willing to let me pick your brain on whether your approach is appropriate to our situation. At the Open Prosthetics Project (here's our current social net http://openprosthetics.ning.com/), we've chosen the Open Scholar Drupal implementation for future development for a variety of reasons, many of which you cite in your poster (we're in the process of setting up a test instance). We're very interested in VIVO, mostly to involve academics. We would like to be able to use Drupal to manage at least a set of VIVO users, as well as to extend the available information on other VIVO users who would be willing to join our social network using their external VIVO profiles. I think it might be useful to extend any VIVO information for the purposes of adding Web 2.0 content, e.g. a rating or review of a paper. We'd also like to maintain triples referring to objects that VIVO might not be concerned with, say commercial medical devices, in order to create links to patents and papers on the devices, product reviews, or places to buy them. We're partnering with a couple of other organizations, and want to be able to take advantage of each others work.
One example of something that is of interest to all of us is accessibility. This came up in a discussion that we had at the conference regarded the Section 508 compliance of VIVO, something that everyone had either not thought about, or wanted to think about later. I pointed out that Drupal already has a couple of related options (http://drupal.org/project/accessible_content http://drupal.org/project/accessible), as well as a bunch of accessible themes (http://openconcept.ca/blog/mgifford/accessible_drupal_six_themes). The existence of such work is an advantage of Drupal as a mature CMS with an active community.
Anyway, thanks again for the poster, and please contact me at jon**at-openprosthetics-dot-org if you would like to further discuss this.
Best,
Jon Kuniholm
research networking and profiles
Hello Miles,
My name is Eric Meeks, I'm a technical architect at UCSF and I believe we talked at the VIVO conference last year. I'll be attending the one this year in DC and presenting some work we have done to integrate RDF into OpenSocial.
But a different note. We do not run VIVO at UCSF but instead run a similar open source product called 'Profiles' that was developed by Griffin Weber at Harvard, who is a member of the VIVO technical advisory board. The current version of Profiles has an XML based API which we are going to use to create a Drupal module which will allow us to pull in Profiles content into Drupal based sites (there are many Drupal sites at UCSF).
However, the future versions of Profiles will produce VIVO based RDF. You can already see this in action at Harvard: http://connects.catalyst.harvard.edu/profiles/profile/person/32213/viewa...
Once we have this version of the code we will want to change our Drupal module to consume RDF rather than the current proprietary XML, as an RDF based module offers a more reusable solution. In fact, it sounds you may have already built this :)
It may be good to talk sometime about all this.
Thanks
Eric Meeks
http://profiles.ucsf.edu/ProfileDetails.aspx?Person=5138614
Hello Eric, I'll be at this
Hello Eric,
I'll be at this year's VIVO conference as well, so we should definitely talk there. Looking forward to learning more about OpenSocial.
In case you didn't already find it, the RDFimporter module was the result of work I was doing with VIVO last year. The screencasts linked on that sandbox page give a pretty good overview of how it's used. The module hasn't seen much attention from me in recent months, but Alex Dorsk did great some work to get the module functioning under Drupal 7. Both D6 and D7 branches are included in the git repository.
During the conference I'm hoping to speak with different institutions about what they're looking to accomplish with Drupal. My own use cases have been fairly narrow, so I'd like get a sense for what kind of features and workflow would be beneficial for others.
If your group already has a system in place for consuming Profiles XML, you might want to look into the ARC2 library which powers many of Drupal's RDF-related projects. ARC makes handling RDF surprisingly straightforward.
See you in a few weeks!
See you in DC!
Thanks for the quick response! We will check out the RDFImporter module and the ARC2 Library. Very happy to talk to you about OpenSocial. We've been able to add a number of extensions to Profiles with OpenSocial and we'd like to see those extensions work with VIVO as well. Our work with integrating OpenSocial with RDF is something we find very exciting & I plan to have some demos ready for the conference.
Eric
Drupal / Vivo
Greetings, all. Could anyone direct me to a place where I can find out more about Drupal / Vivo integration over the last year?
Best wishes
Jen