If you’ve ever pushed past the daily grind of your undergraduate education into the beautiful world of the graduate student, this article is for you! Today, I expound to you, dearest reader, the virtues of open access research. “What?!”, you say. “That’s utter madness”, you say. “Academic suicide!” and “What about quality control?!”, I hear. Well … pish posh! And how will I convince you that open access research is the way to go? I shall employ that somewhat archaic notion of logical arguments my friends! That’s right, you heard it here first! Logic CAN be used for useful things! 🙂
Folks. The academic publishing system is broken. Its time to break the status quo – no pun intended! Yes, yes … gasp if you must – but I have legitimate complaints against the current system. And I strongly believe that in the modern context of the internet, open access and self-publishing is not only possible, but also easy! And yes it can meet peer-review quality standards too! But first a brief history lesson!
In the beginning there was man, he was curious and discovered many a great thing about the world and its physical laws. He used these laws to build great machines. But all was not well in the world of man. For he was alone. To discover even greater things he needed the help of other men. But alas, knowledge was power and man needed a way to pass knowledge on to other men. So it was that language was born. And from it stories came and then writing. And man was able to share knowledge across generations and civilization was born. With civilization greater and greater heights were reached … but alas there was a problem. For now there were many men telling many stories. And so man was overwhelmed. For he had only limited time on this world. He needed the stories to guide his life so that he may live long and prosper. He needed to trust the stories. And so it was that stories became linked. And “Citations” were born. Now man could trust, for he knew from whence his stories came. And if they came from men of repute and knowledge, must surely be true. And so stories became science, for man could now examine and compare stories. But for man progress existed naught without sharing. Man, alone, was poor he could not share easily. So discovery and science was slow and man’s life was short. Then came the “Publisher” and man was joyous. For now he could share easily with other men. And the “Publisher”, for fair recompense, took man’s work and with devotion passed it on to other men. And with that the modern world was born.
Alright, now that I’ve satisfied my desire for waxing poetic with elegant prose I shall get back to the point. If I’m going to be criticizing the publishing status quo, I should probably describe it first shouldn’t I? After that I’ll point out weaknesses and propose that we can do things much better in this day and age by utilizing open-access journals and self-publishing tools. I should add, that I’ve been a bit intellectually lazy in this article – you should examine everything I say with a careful critical eye. At some point I’ll enable comments so that you can openly (and politely) criticise my work. Oh and one additional disclaimer … all the images used in this article have been borrowed from google images, I hold no rights over them and claim no ownership. My apologies in advance to the people who worked hard to make these images – they aren’t currently properly referenced.
The Current Status Quo
There are four parties involved in the publishing process: the authors, the publishers, the reviewers, and the readers. Authors typically do research and come together to write either a conference paper or a journal paper. Conference papers present newer, in progress research. Journal papers tend to present more complete and in-depth research.
Subsequent to authorship, the written paper is submitted to a specific conference or journal. Independently organized or established journals exist. These are typically operated from a not-for-profit basis, though they often rely on external sponsorship from organizations to cover basic expenses. Expenses for conferences might include room bookings, copying and publishing fees, catering etc. Likewise for journals, these expenses fall mostly under the category of publishing. Most conferences do have an attendance fee for participants – and this can cover some organizational expenses. Journals however, typically don’t have a publishing fee. There are exceptions though … for instance journals might charge per page fees for extra pages in excess of the recommended page limit. Most reputable conferences or journals are sponsored and/or organized by for-profit publishing companies. The publishers take the submitted papers and nominate qualified (expert) peers to review them (peer review or refereeing).
Peer reviewers are volunteers and are not usually paid. Sometimes publishers will offer small token compensation. The reviewers provide comments about the particular paper and send them back, along with recommendations, to the specific editor associated with the journal or conference. Their reviews are considered and passed on to the author. The editor may reject the author’s paper outright or recommend revision and re-submission. Alternatively he/she may accept a paper without condition or conditional upon revision. Revised papers may or may not undergo another peer review process.
The entire peer review process for reputable journals can take a fair amount of time. Two to six months or even more can be quite typical depending on the journal (though this is not always the case). At the end of this time, the paper is published as part of either a conference proceeding or a journal. And this is where the reader/audience comes in. The reader then pays a subscription fee to access the particular article or journal. This is usually paid by the institution or organization to which the reader belongs. Some journals (usually those that are not-for-profit) provide open access to their publications. That is, the reader – any reader – can freely access articles at no cost.
Arguments for Open Access and Self-Publication
At this point, You’re probably asking yourself: “What could possibly be wrong with this system?” It seems emminently reasonable. Well, it does, or rather it was prior to the internet age. If you actually analyze the system critically, from the modern perspective it doesn’t look so good anymore! So let’s get on with and do just that by providing arguments for open access and self-publication.
Heather Morrison has done a detailed analysis in her 2012 Ph.D. thesis [1] of the potential for free scholarship in the internet age albiet from a social science perspective.
Who Makes Money?
Isn’t it funny how everything comes down to money? Often, authors of academic publications are researchers working in publicly funded institutions – these institutions are universities, research councils, and other not-for-profit organizations. There are of-course for-profit institutions who publish too. But, my suspicion is that most of these institutions prefer to maximize profits by commoditizing intellectual property (e.g. patents, trade secrets, etc.). Whatever the case, authors in these institutions submit their publicly (or privately) funded research to a for-profit publisher for free.
The publisher doesn’t pay royalties to the author or the institute. Nor does the publisher fund author research in anyway. Subsequently, all this public research that is given freely to the publisher is made available to the public for subscription fees. In recent years, with the advent of the internet, publishing costs have been rapidly dropping, yet subscriptions fees have been rapidly increasing. All this has meant that (mostly) publicly funded research which should be available to the public freely is becoming harder and harder to access.
Pay-per use online access to a single article ONCE from a single computer typically costs around 30 USD or more – my suspicion is that the cost of publishing this article, for the publisher, given the cost of web hosting, etc. is probably some tiny fraction of a dollar. If you’re in an institution, the subscription structure is hidden from you. But the situation is still bleak. Institutions pay a subscription fee, with fees for limited time online access to a single journal being in the 1000 USD range. Many journals cost much much more.
To make things worse, reviewers who go through and verify research quality of articles submitted to the publisher are mostly volunteers. So publishers essentially get almost all current research (after being reviewed by volunteers) for free. They then, subsequently, charge increasingly exhorbitant subscription fees to researchers who then produce more free research that publishers can profit from further. SOUNDS like a great racket to me! Sign me up – I want me in on the profiteering band wagon too!
WHY DO WE STAND FOR THIS? Web publishing is so cheap now that individual researchers can easily afford it. I’m paying around 70 CAD per year for this particular domain and host combined. Furthermore, your articles, published in this way, are completely free to access for anyone with a web connection. Not only that, but blogging platforms are now largely open-source, accessible and easy to use. “But Siraj, what about Peer Review??!” … Come now, you’re an intelligent audience, surely you don’t think publishing in its current form is the only solution to the peer review problem? … regardless … I’ll get to this question shortly!
Copyright & Control of Information
An interesting thing has happened recently. Over the decades, the plethora of for-profit publishers have slowly been bought out. Now most of journals/conference proceedings currently published are owned by just four or five publishing companies. Not only that, but these publishers gain copyright and hence distribution rights over the articles researchers publish.
Surely you see the problem. Publishing companies now control access to a significant part of the world’s scientific resources – most of which was paid for by taxpayer money. And they can continue to hold this control perpetually, holding exclusive distribution rights and charging subscription fees for articles whose copyrights have long since expired.
Most, if not all, open access journals place your articles under the creative commons license which allows researchers to share their research while still maintaining their rights. Combined with free open access, how can you go wrong?
Accessibility
Subscription fees charged by publishers limit accessibility. You might think, “Well, I’m an academic – my institution pays my subscription fees. That’s good enough.” But wouldn’t this mean that ordinary people are denied the opportunity to see how their taxes are used? More over, people even outside the academia may also have a desire to do research and study the sciences. And finally, what of countries where financial resources are limited? Institutions in these countries may have papers accepted by publishers but finances will deny them access to up-to-date science. Clearly, open-access addresses this issue.
But there are further advantages possible with self-publication and open access. For one, dissemination of research does not need to wait for a full peer review. Draft papers can be made available immediately. Of-course this begs the question of what happens to research quality. There may well still be a place for publishers in this regard. When talking about accessibility, I think we can also do better than the cryptic and incomplete papers that are the standard for academic publication – for instance, a paper in an electrical engineering journal should be understandable to someone of intelligence who has (at worst) masters level training in the same field – something that is NOT currently true for many papers.
Reproducibility
Currently, publishers still operate using the traditional paper media, albeit on a digital medium. A fundamental element of science is the ability to reproduce other people’s research. The current publishing status quo leaves us with a medium for information dissemination that is not sufficient for this purpose. Papers are often incomplete, difficult to understand even for someone with reasonable knowledge of the field.
Online approaches to publishing are not limited to only text and the occasional picture. Videos can be uploaded showing the exact steps required to reproduce and implement the given research. Interactive graphs, flash programs and most importantly algorithms, code, cad models, simulation results and schematics representing a complete implementation of what is presented in the paper can be made available. “But Siraj! THIS is too much information!”. I have only one response to this statement. Obviously the job of a paper is to summarize research results – having these additional pieces of information wouldn’t negate the ability to provide concise summaries of research.
The key is building an online portal designed to deliver information well. Moving into the realm of conjecture, the function of this portal may not necessarily be to host an author’s article, but simply to connect a series of blog articles of repute together through an easy to search interface. Currently most publishers require LaTex formatting as a means of separating style and content – and this can then be used to generate html for web use. But actually LaTex is still a paper format. We could do far better with existing and up coming web standards and protocols. For instance, CSS combined with HTML/XML, PHP and JAVASCRIPT provides an incredibly robust multimedia capable document platform that seperates style and structure and function very very well. Content management systems such as DRUPAL or JOOMLA likewise provide brilliant ways of integrating all of the various web protocols to produce seamless documents. Documents from individual blogs can be integrated into other portals by CSS styling. Furthermore, web publishing wouldn’t be limited to offline viewing – tools to convert CSS and HTML into more traditional PDF style documents have existed for years. Having a free repository of research publications in standardized open document standards would allow the use of statistical analysis tools. These statistical tools can be used to more effectively rank, sort and identify promising research areas.
Designing a portal of the nature I’ve just been describing wouldn’t be difficult – and given my own web design experience and hazarding a guess, I would say that this would be possible, within a year’s time frame. with a small group of dedicated and passionate designers working fulltime.
Peer Review
And so, with that, we come to the BIG question: “What about PEER REVIEW??” This is perhaps one of the most cited reasons for not self-publishing and sticking to the big publishing names. The argument goes that self-published and open access articles can’t be peer reviewed properly. How would we peer review without these big publishing names? Surely, people as highly educated as us can find solutions?
To begin with current peer review relies on a few experts. These experts unfortunately have limited time and can only go so far to review articles. Furthermore they can’t always know everything about the field. The mechanisms of modern web design for doing this same peer review already exist and can easily be implemented in open access systems. But, why stop at standard peer review?
We can go much further. Suppose I were to publish an online article as a draft. This draft could be added to a search portal as an article that needs peer review. People working on related research would be the most likely people to access this article anyway and can form competent reviewers – self-selection is likely to take place. They won’t necessarily have the same expertise as a vetted expert – but through crowd-sourcing could well prove to be more intelligent reviewers as a whole. A system of quality control may well still be needed – but flagging and up- or down-voting would quickly bring to light the most glaring inaccuracies in draft articles. This system of up-voting or down-voting isn’t without precedent, some coding forums use it quite successfully. These same forums also demonstrate that open peer review can work (though an appropriate interface would be required). A beautifully simple example of one such forum is stackexchange: https://stackexchange.com/
The added advantage of an open review process is that comments and concerns that people have with regard to your article are visible to readers. This added visibility, if implemented properly, can make it easier to critically analyze articles. Tools to add notes and pointers to specific points in an article are not unheard of either. The greatest problems occur when visitors are either unwilling to leave comments or the comments are not relevant, inaccurate or spam. With this respect the traditional publisher could still have a role in moderating any kind of review discussion (e.g. by flagging content). If no comments are left, the original peer review approach is still a valid approach.
With online systems, peer review approaches would require peers to register and login to a system. This can be quite tedious if the open access portal points to articles from different user blogs. However, recently efforts have been underway to uniformize registration and login using protocols such as OpenID (https://openid.net/).
There’s added benefit to making the review process more open. I have heard stories about professional jealousy, where reviewers deliberately delay reviewing publications relevant to their own research so that they can gain preferential treatment with their own articles. Sometimes, reviewers will preferentially select articles from people they know. By making draft publications openly available right away allows automatic logging and ensures that the original researcher retains credit for his/her original ideas. Furthermore it ensures that an article is accessible to a much larger pool of peers than traditional publishers can manage. This larger selection pool is significant, because under the current scheme, good peer reviewers tend to be nominated more often for additional work. This increases the volunteer work load of the few good reviewers out there. Traditional publishers, because of their centralized approach, simply cannot access enough qualified refereeing candidates.
You might go even further and ask: “But how would you ensure data longevity?” This is a valid question – individual blogs are subject to change or may be permanently shutdown. There are solutions where blog data can be preserved by third party portals. The way-back machine (https://archive.org/web/web.php) is one system that does this albeit for a different purpose than what we would want. Limitations can be put on the number of revisions an article can go through. In truth no article would necessarily be rejected – however, articles of higher quality could be re-positioned by some form of automated vetting process. Perhaps creating distinct “journals” based on the quality of the article. Here bias would exist, whereby perhaps some articles would gain more prominence due to the popularity of the field – but the current publishing system does quite badly in this department … probably worse than an online open system (that uses statistical algorithms for organizing data) might do.
What about Reputation?
Some of you might object at this point and complain: open-access journals have a terrible reputation. They are low quality and the same is true for random websites that individuals put up. No one reading a “real” paper would ever take it seriously if a website were listed as a reference – and especially if the website belonged to the very same author who wrote the paper. This is definitely true for the older generation. But to some extent true for us of the newer generation … I also often find myself looking at paper references that contain website links with contempt. Being able to trust an information source is critical – that’s why the reputable journals are reputable.
But let’s be honest, how many of you academics out there rely on random websites for information. Most Ph.D. students I’ve met use these websites to understand basic mathematics and even more complex physics and statistics every single day. Furthermore, why do you think open-access journals have a poor reputation? One reason is certainly because, academics perceive these journals as low quality – hence prefer to publish good quality research in high impact journals. And so things become a self-fulfilling prophesy. Open-access journals are low quality so reputable research isn’t published in them, but because reputable research isn’t published in them, open-access journals aren’t reputable. Can anyone say “catch-22”?
With personal websites, I agree, there can sometimes be doubt as to the quality of research – these sites aren’t vetted. But, enabling a proper commenting system alone, can greatly improve the ability to judge the quality of these sites. When it comes to technical research articles posted on blogs, I find that the quality can be high – the individuals who post tend to be more diligent than the average blogger and have a personal stake in producing good quality articles – comments from the readers also goes a long way to improving quality – of-course this only holds true for articles of a technical nature that don’t have political or social commentary.
I think the real barrier to open-access and self-publishing is a perception. There is no reason why articles written on blogs can’t be properly referenced. The standard chain of proof can be followed. The fact that people don’t take web references seriously in journal articles has nothing to do with the quality of these web references, but rather the intellectual laziness of researchers who aren’t willing to properly follow through and look at the actual reference. I use bibtex data for references on this blog – so there is no reason why references can’t easily be integrated into blogging platforms.
Perhaps part of the reason for the negative perception, associated with self-publishing, is because internet sites are in constant flux. A person cannot be certain whether they’ll still have access to the article in the state it was when they looked at it. This is a very real issue. But not without solutions. Many reputable ezines, mark out post-publishing changes to an article clearly for the reader to see. The same can be done on blogs. Likewise, facilities are available on web-based content management systems. These systems can display multiple revisions of a single document … so every change that is made to a particular document can actually be made visible and transparent to the reader or reviewer.
Conclusion
I’ve discussed some of the issues with the current publishing status quo. And I’ve offered a set of quick and dirty suggestions for alternative solutions involving open access and self-publication. I’m not making the claim that this process will be cost-less – indeed there are costs. What I am doing is suggesting that we can do much better. The idea that overthrowing the current regime of academic publishing is a revolution is false. It is a natural evolution! There are things each one of us can do right now – and these things are quite simple and easy to do. All you have to do is be willing to think ever so slightly outside the box – something that researchers should be good at, but generally aren’t.
I haven’t talked about everything but I hope I’ve at least inserted some small amount of doubt in your mind about the traditional publishing approach. At the time of writing, I haven’t gotten around to enabling comments. I promise I’ll enable them shortly … sorta … when I get time. Once I do, please leave comments and politely worded criticisms and I’ll try to update the article so that its more accurate and so that the tone is more appropriate. Thanks folks!
References
- Morrison H. Freedom for Scholarship in the Internet Age. Simon Fraser University, School of Communication, Doctoral Thesis [Internet]. 2012:177. Available from: https://pages.cmns.sfu.ca/heather-morrison/chapter-two-scholarly-communication-in-crisis/