Authorship issues

I don't recall ever having a formal discussion on this topic in graduate school, but I thought I had a reasonable idea of who should be listed as authors on a manuscript:

• Individuals who contributed experimentally (i.e. conducted experiments or analyzed data)
• Individuals who made significant intellectual contributions to the project (i.e. conceived the project or conducted critical preliminary studies)
• And, of course, PIs of the labs involved

However, recent events in my current lab (not related to my work) caused me to wonder if I might be missing something... or rather someone, I suppose. Namely does the contribution of a reagent entitle an individual to authorship? If it's the first time a reagent (or a method, for that matter) has been reported in the literature, then you should definitely be a coauthor. But at what point do you move from author list to acknowledgements?

While I'm on authorship questions: How useful is co-first authorship? I've received "second billing" co-first authorship on a couple of papers, but I realize it's important to have sole first author papers as well. When people read that co-first author paper, they may or may not read that footnote "These authors contributed equally to this work", and co-first authorship doesn't show up on PubMed--you're just another name. One PI/journal editor from my graduate program feels that co-first authorship is total crap--either your did the majority of the work or you didn't. Is co-first authorship really valid? How many ways can you split authorship before it becomes ridiculous?