Some strategies (H/T Michelle). Unfortunately the efficacy of many of these tactics varies by discipline, and it is often really hard to get a hold of something without academic access when it comes to the biological sciences. There's arXiv and SSRN for physical and social scientists, but in the biosciences you have clusters of journals such as at PLoS and BioMed which are relatively exceptional archipelagos of easy access. If you follow one of the Eisenbrothers you'll also be made very conscious of the unfortunate fact that many a time papers which are supposed to be open access at mainstream publishing houses such as Nature are mistakenly gated because of the content management's default settings. This technical sloppiness is rather galling because these publishing houses are often very vigilant about scanning the web (I assume via crawlers) looking for people who have put up "their"* content for free (yes, I know this from personal experience). I have many smart and motivated readers who happen not to have academic access. It's definitely a consideration when I want to blog a paper whether it is open or gated. Obviously I blog gated papers, but it does change the nature of the task when many of your readers aren't going to be able to see the primary source themselves. * I hope I'm not getting anyone in trouble when I pass on the reality that many of the original content producers often are excited when they see unauthorized distribution,
because they're thrilled that people care enough about their research to do this.