Speaking from personal experience, I find tracking down a reference in an actual library to be much more satisfying. Plus you have the option of looking up additional papers that are referenced by the one you wanted (and finding an unexpected gem in the same journal issue that you didn't even know existed). It's only in the case of very rare or hard-to-get papers that I'd resort to begging for a PDF. Certainly for such common journals as Gaia or JVP I'd head to a library first...