I need some feedback about how useful would it be and how much support would I get if I invest some time and work effort in implementing a database solution for tagging, organizing and cross-referencing pieces of information and ideas from multiple (academic) sources. The tool/toolset shall be distributed to the open source community and (ideally) continuously be developed alongside willing collaborators.

Background & motivation

I will soon finish my first year as a PhD student in Electronic Systems, and I have trouble organizing important pieces of information I acquire from all the sources I encounter: articles, reports, manuals, web sites, etc. Apart from "having the memory span of a gnat", I constantly come with ideas and proposals which could easily be enforced if I could trace back everything that triggered a specific thought.

I already use an academic database like Zotero, and it works wonderfully, but unfortunately it does not have the idea granularity I need. And mind-map, billboard and other solutions of this type do not work at all for me. On the contrary, they just make me messier.

Ideally I would like a solution that connects my Zotero database or BibTeX references to organizable / traceable pieces of information.

[thread continued in the following post, due to limited amount of characters allowed]

What is already implemented

I already spent some free time implementing a software infrastructure in the form of a Python project. its current features are:

* Library-style Python classes and modules for separating database operations from interface, frontend and backend
* CLI environment
* Plain-text fontend parser. Currently the information is inputted via plain-text files. Each source is associated with a file which follows a convention. The parser uses keywords like TAG, AT, BIBREF, etc. At the moment, the only connection between the academic database and this one is the BibTeX references generated by Zotero.
* SQLite database, having a simple architecture that binds sources (e.g. articles) to references, tags, pieces of information, and offering multiple ways to query information in a useful manner.
* Plain-text formatted backend.

Future plans

Based on how much spare time I will have in the future or what impact it will have on the open-source community, these are some of the features I plan to introduce, in chronological order :

* a nicely-formatted LaTeX backend
* HTML simple backend
* HTML GUI and frontend, with dynamic queries and auto-complete tags
* HTML interactive backend (with hyperlinks)
* develop it as a Firefox plugin
* develop it as a Zotero extension

Feedback

As the title says, I would definitely appreciate any feedback you can give me. Is it worth my spare time to document and support his project and implement the proposed features? What other features would be useful? Am I reinventing the wheel? Does anyone else need such a tool/solution, or is it relevant only for me?

In any case, I will continue using it at least on my own since it works like a charm for me. I am really forced to synthesize information now, and can track it down to build stronger hypotheses. I will also publish it probably on GitHub, but not before documenting everything properly and writing a tutorial.

I thank you in advance for any feedback or opinion!