If it is possible, I consider myself to be both an idealist and a pragmatist with regard to scholarly publishing.
On the idealist side, I view scientific publishing as a natural extension of the necessity of sharing scientific results with others. Without communicating results to others, scientific knowledge remains stagnant as each researcher spends all their time re-discovering others work. From this perspective, open access publication is the best and really only way to go. Paywalls only retard the growth of scientific knowledge.
My pragmatic side knows that the money to fund this communication has to come from somewhere. The scientists need to be able to make a living, so institutions pay their salaries. The work of some editors and most reviewers is done for free as a service to the profession (that idealism starts sneaking in again here). But the final publications need some money to keep everything going. A bit of money for a couple of paid editors, typesetting, pretty web design, hosting, bandwidth, etc. Publishers add a bit of value to the publication (just not nearly as much as the for-profit companies would like you to believe). This money can come from author fees, from funding agencies (who pay a lot of author fees anyway) and from institutions as cash or in kind support. My least favorite method of supporting the work of scholarly publication comes from reader fees - from subscriptions, copyright clearance fees paid by interlibrary loan departments, or fees paid by individuals for access to individual articles.
So I must admit I'm not sure what to think about today's announcement from DeepDyve allowing users to "rent" complete journal articles for free for five minutes.
My pragmatic side has mixed feelings. I'm not fond of the rental concept for journal articles for several reasons. Importantly, I don't think that rentals work very well for the way that academics work. When reading an article, I want to underline things, jot down notes and circle important points. And chances are I will need to go back to that article in six months time when I'm actually putting together a paper or a presentation. So renting a 12 page article that I can't mark up, can't print, can't save to my computer and won't have access to in six months doesn't make much sense to me.
On the other hand, being able to view more than just the abstract to an article for free could be very helpful in deciding if you actually need to get your hands it. DeepDyve is hoping the free access (for registered users) will encourage you to rent or purchase a copy of the article through them. DeepDyve is also hoping that this fleeting access to articles will encourage sharing of information on social networks. For those who aren't affiliated with an academic institution, this may be the best of a series of bad options for accessing an article. But I worry that students or researchers will get suckered into paying for content their institution has already paid for (or can get for them via interlibrary loan).
My idealistic side is uncomfortable because I don't think this is helpful, in the long run, to the cause of openly sharing the results of scientific research. It's a money-grab as highly profitable publishing companies go after additional revenue streams. With always-greedy share-holders to accommodate, publishing companies are going after money from individuals (and grant funds) in order to increase profits. Traditional revenue streams (library subscriptions) feel less secure as library budgets are cut and funding agencies call for greater access to research results. Make no mistake - scientific publishing is big business and they want your money any way they can get it.
So have a look at the free offerings from DeepDyve. And then use just about any other legal way of getting your hands on the paper for free: go through your school or institution, use Google Scholar to locate a free pre-print or post-print in a repository, use interlibrary loan to request a copy or email the author to see if they can provide a copy.
As I tell my students, publishing companies are doing quite well for themselves. They don't need your money.