Last week, OmniTI hosted a pretty awesome conference about scalability called Surge. I really enjoyed John Allspaw's keynote in which he spoke about the importance of looking to academia and different fields to help solve some of the problems we face with upcoming web technologies and in general. I completely agree and while I do a little of this already, I hope to delve into it more. I'd like to share some of the resources I am currently using and hope that others will find them useful and be able to point me in the direction of similar useful resources we can all use to find papers and articles pertaining to the problems we are trying to solve on the web.
ACM Digital Library
The ACM Digital Library has a vast amount of great articles, papers, courses, and more related to our field. You can gain limited trial access for free, ACM membership (cheaper if you're a student) grants you full access.
Another good tool, which helps to find specific articles. I have found it interesting to read older (1960s) white papers about implementations and algorithms in managing large datasets. You can find some really cool stuff with Google scholar.
MIT Press Journals : Computational Linguistics
Something that I find particularly interesting is the field of computational linguistics. The MIT Press Journals has free volumes online that have some cool articles, such as "Sorting Texts by Readability".
I also think it is important to engage the current ranks in academia. It isn't terribly difficult to find out which universities and professors have interest in the particular topics where you need help. Contact them, contact your alma mater, contact student chapters at universities like AITP and ACM. I'm sure there are plenty of professors out there with something helpful to contribute and an even larger amount of students out there wanting to help.
I do think that people have the impression that the academic help route is going to be a slow process, which is not ideal for those of us working on the web where things move fast. I'll admit that I felt that way too until I got involved with AITP during my time in college and realized that there are professors and students out there with the same interests who want to be involved. On the reverse side, there is an argument for a more lengthy process. We don't just want to solve these problems quickly, we want to take the care to solve them in the right way. This is something that I think way too many programmers believe is mutually exclusive, however I'll save that topic for a future post.