A great experience with Eprints software, and an open minded publisher
Eprints and open access
Eprints (www.eprints.org) is a mature and well supported software written by the Edimburgh University, to help self archiving and open access publishing.
There are many interesting resources about self-archiving and the importance of spreading scientific research results around the globe, a citation from an article published on eprint’s website:
£1.5 billion lost annually in potential return on British science UK Science Research in action
The UK is losing around £1.5 billion annually in the potential impact of its scientific research expenditure, according to one of the key figures in the global open access publishing movement. Professor Stevan Harnad, Moderator of the American Scientist Open Access Forum and Professor of Cognitive Science at the University of Southampton’s School of Electronics and Computer Science, has calculated the potential return on the investment in scientific research findings that are being lost to the UK each year through the limitations of the current academic publishing environment.
In other words, it is now very important that research results and, more widely speaking, culture, is available to everybody. This is now possible, thanks to the new media technologies.
Universities and other istitutions, have adoptet the self-archiving repository concept: researchers who wish to publish their results can upload, tag and classify their works and make them available through some specialized repository software.
How could we define Open Access?.
From the FAQ section of DOAJ (Directory of Open Access Journals):
How do we define Open Access, Research Journal, quality control?
Open Access Journal:
We define open access journals as journals that use a funding model that does not charge readers or their institutions for access. From the BOAI definition  of "open access" we take the right of users to "read, download, copy, distribute, print, search, or link to the full texts of these articles" as mandatory for a journal to be included in the directory.
Open Access series, books, journals or other media are a new way to deliver culture to the largest possible extent.
This new way of distributing media has a clear advantage over traditional hard-copy-only printed distribution:
- authors ave immediate access to the world’s largest public
- word’s progress is enhanced by access to a greater base of culture
- repetition and copying can be limited through a broader knowledge fo authors and their works
- printed copy sales increases (this is the most surpising fact, but can be easily explained)
I’ve recently got in touch with an open-minded publisher (Polimetrica Publisher) that launched the idea of combined traditional prints and open access series. This means that the books of this serie will be immediately available both in printed and digital copy, being the lattest free of charge.
After evaluating DSPACE and Eprints, we choosed Eprints as the code base for their repository.
A developer’s wiewpoint on Eprints
Eprints, is written in perl, and needs Apache’s mod_perl.
I must admit that even if Perl is not the programming language of my choice, I had’nt any difficult in understanding how Eprints is organized: it is well documented, well written and well supported.
Configuration has to be done through a bunch of XML files and through some generate perl files. This means you must know some perl if you want to configure the program.
Installation, graphics integration and configuration process can be quite long, it takes from 3 to 5 working days.
Keep in mind some limitations:
- after completing the definition of metadata (i.e. the set of fields that will be used to classify items in the repository), it will be very difficult to change this structure without altering directly both the code and the database structure
- every change on configuration files, language files, or other templates, will require re-running some scripts and restarting the webserver. This slow down the development process.
- there is no Delete function, items cannot be deleted from the repository without some handworks
- there is no parent-child hyerarchical relationship between items, it is not possibile (without hacking the code) to define collective books and insert its content as individual items
- Installation and configuration can be quite an hard job, additionally it is strongly suggested to have at least a virtual server with shell root access
Despite its limitations, the overall impression is that this is a great piece of software!