Tell me more ×
Libraries & Information Science Stack Exchange is a question and answer site for librarians and library professionals. It's 100% free, no registration required.
up vote 4 down vote favorite
3

I am in the process of building a web app and I know now I will need an electronic library of some sort external to the web app, so I am now looking for an open source library to plugin in to the webapp. Can anyone that has any experience in this arena please suggest some options. Greenstone seems to be the best I found so far (and the only one that seems to be built with to serve solely as a digital library), but I do not want to commit myself to just one.
Edit:
After some research I have found that what I actually want/need is a document management system with built in faceted search and the ability to manage and audit documents and probably relate them in some way. At this point I have only been feature skimming but I have also found: Knowledge Tree, E-Prints and 1 or two others that seem to match most of my criteria on one way or another. Though Knowledge tree claims to have an open source edition, I have been unable to find it so far to download for testing.

share|improve this question
2  
What do you mean by an "electronic library" or "digital library"? – anarchivist Jul 19 '12 at 14:48
I guess what I am after (after doing some research) is a document management system. – Dark Star1 Jul 19 '12 at 15:34
2  
It really depends on just what sort of functionality you need to to do, and how large it's going to need to scale. There's everything from OODT, Fedora, DSpace, etc. You could even use a cloud service like S3 if you need it mostly just for storage. Or you could use something like iRODS as middleware to access multiple storage systems. – Joe Jul 19 '12 at 15:54
It would also seem to depend on what you mean by documents here. If it is really a lot of text docs something like dokuwiki might do the trick too. dokuwiki.org/dokuwiki – Trevor Owens Jul 20 '12 at 22:22
@TrevorOwens pdf docs with images, and the possibilities to add short videos in the future. – Dark Star1 Jul 22 '12 at 16:52

7 Answers

up vote 3 down vote

This a a rather broad question which may lead to several interesting hints and discussion but no concrete answer. What makes a "good" system is subjective unless your clearly name criteria on which to judge. If you look for a document management system(*) you may look for a wide bunch of functions such as storage, retrieval, metadata management, integration with other services... I bet that you don't want the all-in-one system suitable for every purpose (which may be Fedora) but for a dedicated, easy-to-use system that best fits your needs. For instance faceted search is a property of retrieval only: you can get faceted search with a search engine which does not manage any digital objects at all (maybe the discovery interface VuFind fits your needs?.

Can you please try to further describe your particular needs and re-edit the question, including the title?

Some comments on your current list of criteria:

  • faceting
  • full-text searching indexing & highlighting

This are properties of the retrieval and discovery interface. What do you mean by full text highlighting? Result snippets of full text as known by Google?

  • open source

Clear criteria unless you distinguish open source and free software.

  • semantic relations
  • tailored for DMS and lightweight (e.g. not drupal...)

Very buzzy and fuzzy, I have no clue what you actually ask for.

(*) I added a link to the Wikipedia article to clarify what we are talking about. Unfortunately the Wikipedia article "document management system" does not clearly tell the difference and relation to terms like institutional repository, publishing repository, and digital library. Don't trust Wikipedia definitions unless you have edited them! ;-)

share|improve this answer
What I was referring to re: semantic relations is an extension of metadata languages to show relations between documents and the metadata that describes the document. Common methods include using RDF as a foundation for creating taxonomy and ontologies which allow for simple inferencing. Lightweight means something that is not resource intensive and can happily live on VMs. I hope those terms aren't that buzzy. – AaronC Jul 25 '12 at 13:56
"Relations between documents" is one thing but "metadata that describes the document" is pleonasm. RDF is used in this context but the technical level is far too low, one can also have semantic relations with another technology (and map to RDF if needed). – Jakob Jul 25 '12 at 16:25
up vote 3 down vote accepted
+50

After much searching and deliberation I have settled down to testing four to five document management systems (DMS). I do agree with some of the replies that I have read about a DMS not being a monolithic solution but I do think that a DMS should, by default, imply some basic core functionality, such as faceted searching.

Now I know some people want would want me to explain this but I will summarize by just saying "library." In any case I am currently testing OpenKM, Knowledge Tree Community Edition, Epiware, Alfresco Community Edition (once I can get it to run) and I am thinking to add Fedora (formerly DuraSpace) to the mix of tests. Some of these use the Solr and Apache Lucene search technology, and Solr I know is an excellent search engine.

share|improve this answer
1  
Found Alfresco to be the best of anything I tried. – Dark Star1 Dec 27 '12 at 3:02
up vote 2 down vote

I'm going to give something of a contrarian answer - there isn't one and there shouldn't be! The problem is that the different parts of the stack change at different speeds. You generally want your backend store to change very slowly, so you know the data you've got is safe. Your ingest processes tend to be big phases of development followed by long phases of maintenance and use. Your access systems tend to need to change more rapidly and constantly, as you react to your users' needs. I suspect that this is why the open source solutions tend to focus on these three phases separately, so that each is modular and replaceable, rather than one monolithic system which can do everything and anything.

share|improve this answer
1  
I fully agree. A document management system (DMS) consist of several components, which can also be used independently. Good software should loosely couple independent component (service oriented architecture). Therefore there is not a single good DMS. There might be, however, a good software distribution that packages several DMS components! – Jakob Jul 25 '12 at 7:09
Great points. I think these are of course some of the foundational concepts which support more recent trends of curation microservices, and the build-your-own repository ideology. However there are certainly circumstances where the management and support of say 3 modular components would not be desirable for a particular project. I'll give two examples: the OP appears to be prototyping a web app, and may just want something quick n dirty; and my own case where I want something temporary for small internal projects. – AaronC Jul 25 '12 at 13:38
up vote 1 down vote

Check out the Fedora Repository Project. It's an open-source digital repository with significant support from a number of institutions. Their "About" page lists that they have users from all these industries:

  • broadcasting and media
  • consortia
  • corporations
  • government agencies
  • IT-related institutions
  • medical centers and libraries
  • museums and cultural organizations
  • national and public libraries and archives
  • professional societies
  • publishing
  • research groups and projects
  • semantic and virtual library projects
  • university libraries and archives
share|improve this answer
Thanks. I was just looking at Knowledge Tree which seems to fulfill most of my reqs and is php based, apparently it has an open source community edition which I can't get at so far. – Dark Star1 Jul 19 '12 at 15:44
@DarkStar1 to elaborate Fedora Commons is more of a middleware application. There is another project called Islandora which is a Drupal module that hooks up with Fedora Commons and can be customized to enable faceting (Solr), create simple and complex relationships (Fedora relationship ontology, RELS-INT/EXT... RDF essentially). – AaronC Jul 19 '12 at 20:58
up vote 1 down vote

Don't ask much, do you? ;)

It might make the most sense for you to look at curation microservices and incorporate the ones you think you need into your system. You'll still need to bolt on search/browse separately, but as suggested above, Blacklight should make that feasible.

share|improve this answer
up vote 1 down vote

Have you tried DocFinity from Optical Imaging Technology? I use it for my finance department. It includes work flow solutions to streamline AP and AR functions. It is easy to use.

share|improve this answer
Thanks for your answer. I already have committed to working with Alfresco. Which is quite a good system – Dark Star1 May 13 at 18:09
up vote 1 down vote

Initially we had tried OpenKM, but reached a number of documents (actually a few hundred thousand) we found ourselves in trouble. Fortunately we found LogicalDOC, a very similar software with regard to the graphical interface but more performant on large amounts of documents.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.