Take the 2-minute tour ×
Database Administrators Stack Exchange is a question and answer site for database professionals who wish to improve their database skills and learn from others in the community. It's 100% free, no registration required.

I'm a PhD student in an academic research laboratory. One part of my project involves collecting samples and generating data. Another part involves establishing a research database to store the results. As the project has grown, a Masters student and two post-doctoral fellows are also working on the samples and generating data. All data relating to the samples I collected is to be added to the database, regardless of who generated the data. Each entry in the database has an "owner" field to keep track of this. Our supervisor would like the database to be used more extensively by people working on other projects as well.

I am meeting some resistance from members of our group (both those workings on "my" project and on their own/other's projects) regarding submitting data/results to "my" database. At a recent meeting, I explained once again the concepts of a (central) database, that queries can easily extract data, that the "ownership" field keeps track of whose data point it is, and reminded everyone of the benefits of a central database which stores all data generated. I highlighted the kinds of analyses which can be undertaken comparatively on different data sets, how meta-analyses can be done, and so on. I then offered to do some research around the issue of database management and ownership. Some people in the group do not want to submit data to the database until their papers are published, despite the fact that it is impossible for anyone else in the group to simply throw together a "competing" paper from the data alone and publish it "first". This is also impractical because it means that data which is never published is never added to the database. It also means that no-one else has access to it UNTIL the paper is published, thereby slowing down other research. Internal sharing of this kind of data should be routine.

I would appreciate any suggestions of resources or comments around these issues. Should everyone sign a "waiver" regarding their data going INTO the database and another one regarding the use of data FROM the database? We're not a particularly mistrustful group, but everyone else in the group are "wet lab" researchers, so the concepts of databases and sharing data is possibly a little foreign to them.

share|improve this question
    
Honestly this is something to be decided by the dean or whoever else is in charge of the department. This is internal politics and we can't do anything really to help you. –  jcolebrand Feb 11 '11 at 15:29
    
I was looking for more guidance regarding policies implemented by others in similar situations. With respect, I'm not sure my Dean would really understand fully what this is all about and what the implications are. We conduct medical research and electronic databases storing all the data of a research group is not something any of the researchers or academics encounter on a regular basis. –  SabreWolfy Feb 11 '11 at 15:44
    
~ So ze mods have discussed amongst themselves and it seems that your question is primarily a matter of what policy should be set, and how to go about achieving that goal. So if that's really the case, it should go on programmers.StackExchange.com, and if not, then it should be made more technically focused for it to stay here. I'm open to going either way, and am more than willing to help you sort things out if you need some help. Feel free to visit in the chat as well and we can discuss in realtime. –  jcolebrand Feb 17 '11 at 0:16
add comment

2 Answers

Welcome to office politics. Things like NDAs regarding the disclosure of data can help with this. Your adviser and the universities legal department should be able to help you get these papers drawn up.

share|improve this answer
add comment

I'm torn on this one ... yes, the issue is information policy, not necessarily database implementation, which is why I'd prefer this site not be called "Database Administrators", although even for "Database Professionals" it's iffy if it's on topic or not.

I can't answer your question directly, other than to say you're probably doing it wrong if you're starting from scratch.

Search for the terms 'data repository' + your discipline, and 'informatics' + your discipline. (although, if you search for 'medical informatics', you'd going to find a huge field).

You're in a pariticularly tricky case, as with medical information, you have to compy with HIPPA, so you need to make sure that you've removed any PII (Personally Identifiable Information) from the data.

Some of the 'small science' fields are using Fedora Repository or DSpace to manage their content, but I don't know if it has rules for your granular authorization needs. Most of the 'big science' fields tend to roll their own, but there are some more generic products out there like iRODS which can connect multiple backends and OODT (was JPL, now Apache) to stop you from having to design things from the ground up.

... but I'd look to the people doing informatics work in your field -- they could tell you if there are metadata standards for your discipline, and if there's any sort of standardization for storage and management, or even some sort of existing federation.

... and as for publishing -- it's a HUGE thing; many disciplines embargo the data until the research is published; some journals won't accept papers that are in publicly-accessible repositories as they consider it to be 'previously published', but most (not all) will make exemptions for 'institutional repositories' (those only accessible to people at the researcher's place of work), and after it's been accepted for publication, there are some loopholes that you can use if it's federally funded research.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.