I spent an hour on Friday pocking the brain of a senior colleague and among other things I realized how disorganized (comparing to his) is my ever growing collection of scientific papers, references, electronic textbooks, and monographs.
I started looking around for a a decent BibTex aware collection manager. By decent I mean a relatively simple, open source, (my OS of choice is OpenBSD), self hosted (no dark clouds please), collection manager which is capable of importing BibTeX references either from BibTeX files or online sources like MathSciNet or BibTeX search, with BibTeX syntax checker (an example would be biblean) which stores BibTeX records in searchable SQL database (again Nelson Beebe's bibsql and bibtosql come to mind). I also would like to attach these BibTeX records to full text PDFs as well as possibly TeX sources for the articles I wrote. In another words I would like to use BibTeX files as a metadata for PDFs and TeX sources.
Such things of course exist MathSciNet would be an example but with exception of Tellico bloatware I have not found anything that comes close to what I want. It seems that I should be able to hack something on my own using Pybtex as this interesting post reveals.
Any comments, references, or suggestions? I actually want to do mathematics not play with publishing tools but if I could get something workable in few evenings I would not mind getting my hands dirty. I would in particular appreciate the comments of Nelson Beebe and authors of Pybtex. I am also finding another interesting thoughts.
A possible approach to my problem is to tag files with BibTeX records and have tag aware file manager or even tag aware file system. I see interesting tools