Journal access is an anachronistic nightmare
July 29, 2011
There has been considerable discussion on the internet about academic journal publishing in the aftermath of the Aaron Swartz arrest. Aaron stands accused of downloading a vast array of articles from JSTOR, allegedly with the intention of uploading them to P2P filesharing sources (from Wired). A lot of the discussion surrounding the case has been about access to articles which may or may not be public domain, and the merits, or lack thereof of charging to access these. However, even as things stand, there are significant issues even for those who have paid subscription fees to major journals, which casts into sharp relief the need for modernisation of the system.
The most frustrating part of searching for scientific articles is the laborious process you have to perform to locate and access articles. This looks something like this (probably familiar to anyone using PubMed).
- Find the paper on Pubmed.
- Find Pubmed doesn't have a link to the paper with its abstract.
- Search on Google.
- Find reference to some distribution site, like Science Direct etc.
- Get to the distribution site, demands personal login. Athens login doesn't work, institution access doesn't work.
- Contact the library, find you have to be on site to get access.
- Spend several minutes calming down having wasted too much of your life trying to access said paper.
Yes, this is an extreme case, but even when you find the link first time on PubMed, you still have to go to a distributors site - Science Direct, Wiley, Springer Link, Wolters Kluwer etc. Every system handles access control in a different way - Athens, institutional login (shibboleth etc.), or IP based lookup to an institutional workstation. The last is clearly the most anachronistic - who on earth reads journal articles only from an institutional workstation? Please.
Why do publishers make accessing their works for paying customers so laborious? Ethically and arguably most in tune with the idea of scientific progress, all science should be published in an Open Access model, without restricting access to only the institutions who can afford the exorbitant subscription fees. Even if publishers remain inside their paywalled ivory towers, blocking out those in less fiscally able countries or institutions to partake in current scientific debate, there are ways to sort this out to make life easy for people from institutions who have paid substantial sums of money to get access.
- Remove IP based 'institution logins' - this relies on the failed assumption that IP address equates with location in a 1:1 relationship, and that people only read articles on institutional workstations. This isn't 1995.
- Make all Pubmed links go straight to where you can get the article, and make sure every Pubmed reference has a link. No link, no Pubmed index.
- Journals should make their own sites integrate with the distributer, rather than making you have to do the run around to find out where to get the article. A good example here is The Lancet - institutional access is not handled through their own site, but via sources such as Science Direct.
- Distribution should be simplified - each publisher having their own is immensely irritating, you have to know who publishes what, where to go, the idiosyncracies of their individual system etc.
In sum, the system is broken. Finding articles is obstructive, expensive, and needlessly complex. Paying customers (or their students) find accessing articles needlessly obtuse. Even if publishers want to charge exorbitant prices for access to their publications, they should make access for those who have paid simple. At present, the industry is immensely successful at barring access to those who have not paid, but at the cost of making it hard for those who have. This is a pattern seen the world over - the music industry, the film industry, the games industry. The fact that scientific publication signs up to the same methodology is quite disturbing.
Why does the industry remain so resistant to modernisation? We are stuck in some neolithic paper-focused publishing model appropriate for the early 20th century, despite living in an era where online, Open Access publication lowers the cost to the consumer and publisher. And that is without considering the concomitant broadening of access to modern scientific progress. This anachronistic system desperately needs a refresh. As novel journals such as Plos One gain traction, one can only hope that this will force the members of the old guard to update their business models and streamline this painful process.