If all you need to do is read an article (but not download), you can register for a free acount with jstor that allows you to read 100 articles a month: https:

If all you need to do is read an article (but not download), you can register for a free acount with jstor that allows you to read 100 articles a month: https://www.jstor.org/register?redirectUri=%2F https://www.jstor.org/register?redirectUri=%2F I have jstor access elsewhere (through a school) which allows me to actually download pdfs, but it doesn't carry all the journals that going straight through jstor does. And often I just need a single result from a paper or the article is short, and reading online without downloading is fine. arxiv.org is best if the article is recent and available there. Journals published by professional societies sometimes have free access - e.g. some, but not all, of the American Mathematical Society journals. Go to the organization's web site and search the journal there. If you know the author(s) of the paper and they're still active, a search on the author(s) will often turn up faculty web pages with download links. I've found that it's useful to search (ddg, google, etc.) on the author + title of the book/paper, as opposed to going to a site that aggregates works and then searching, because a given site might not have that book/paper. I've often found books, for instance, that were not on places like Project Gutenberg.