This list has been posted a number of times before [1], with this version having the loosest title. To be more precision, this is a list of best paper awards given at selected conferences based on reviewer scores at the time, and only since ~1996.
IIRC, expanding with programming languages (PLDI, CGO, ISMM, OOPSLA, etc.) was already discussed when this link was first mentioned on HN. Unfortunately, it seems to not have been done yet...
Anyways, PLDI has the "Best of PLDI" papers from 1979 to 1999 (http://www.informatik.uni-trier.de/~ley/db/conf/pldi/pldi200...) and they establish the most important paper 10 after its publication. AFAIK/IIRC OOPSLA/SPLASH does this now as well. I think it's good, but it would also be very interesting to know, whether there is an intersection between the set of "Best Paper" awards and the set of "Most important/impact" awards.
I guess "Lasting Impact" awards are a much better metric than "Best Paper" awards. There is evidence that best paper awards are not actually indicative of a paper's future impact (if defined as the number of citations):
Judging quality remains a difficult task for the initial reviewers, but also for the best paper award committee. Despite its honest efforts, the best paper award committee has not selected papers that are cited more often than other papers. In other words, the best paper award committee did not perform better than random chance. From this viewpoint, Desney Tan’s claim that “The Best of CHI awards represent the top one percent of research submissions to CHI” appears too optimistic.
True that vast information resources from research conducted with publicly(at times) funded money is hidden behind paywalls like the one from ACM. Here is an old discussion about that topic
Anyways, most of the researchers will have a pdf version of their paper on their site. Googling the first author's name or the paper title generally throws the desired result most of the times. Also, as mentioned in another comment Google scholar does show a PDF link on the right if it finds one.
I found this quite biased towards recent papers. Almost all papers was published after 2000, and not a single paper before 1990. Many of the best papers was published in early days of computer science, but not a single one of these was on this list.
Yes, like I don't know this little known paper?
Alan M. Turing - On Computable Numbers, with an Application to the Entscheidungsproblem
Proceedings of the London Mathematical Society. Second Series 42:230--265, 1936
[1]: http://news.ycombinator.com/item?id=2051437 http://news.ycombinator.com/item?id=1702977 http://news.ycombinator.com/item?id=1619156 http://news.ycombinator.com/item?id=2351416