Looking for a Wiki or database intranet tool

[decafjava]

Ok, I got a request at work for a solution to a particular problem. I work at a research foundation that works in numerous countries around the globe (we are not part of the UN) mostly as web content manager. I also am charged with preparing a weekly report emailed to staff and our sponsors on our recent activities over the past week. They are in the form of a word document transformed to pdf which lists events like conferences, training sessions, new publications that are generally listed by region, country and city with dates. They are numbered by week in the current year and stored in a folder numbered by year.

The problem is if one wants to search all entries by a certain country the usual procedure is to combine the pdfs into one and search manually. You can see why this is less than ideal.

I need a system which is more efficient, it needs these:

1) Closed as this is only for internal consumption within the organization.

2) Easily searchable, perhaps with hyperlinks to each category.

3) Easy to enter data into.

Is there some sort of internal database/intranet solution? We have an internal wiki but that seems a bit clunky for what I am looking for.

Any suggestions appreciated.

 

[D.T.]

Seems like a couple of routes, from a high level architecture (i.e., not getting into platform specific details). I was involved in something along these lines for medical records for research, cool stuff.

Build an app (internal, web based intranet) that provides an interface for user based meta-data, uploading, and parsing/extracting PDF content. It sounds like there's a fairly standard structure to the documents, and it's about information availability (searching, keywords, simplified views).

Use a full text / semantic search engine, which would still require some of the above, less intricate coding since the engine would provide most of the heavy lifting - this is more geared towards large document volumes where the structures are more disparate, and the goal is to find matching documents.

Without knowing more details it _sounds_ like the first option might be your best. There are some outstanding opensource solutions for either, and some very specific libraries/packages/mods for just about every technology stack, so it would be easy to maintain consistency with your existing infrastructure.