What the tech geeks thought of my faculty database idea.

I didn't read your link. But, I am in the process and I only have access to PsycInfo when I'm working in the lab, not really time to cruise for mentors. I've been currently doing the research for mentors somewhat, it seems, inefficiently. I have the 2008/2009 Insider's so I look through programs that I think I'll meet their stats and that are funded. Then I go to their website and look at each faculty to see if there is someone I'm interested in.

I have to say some of the reseach at some of the schools is not at all what I expected it to be, based on some vague preconceived notion I had of the school or something.

Some off the top of my head ideas with your database idea, professors move. Plus, for it to be useful to me, it would have to include the other information about the program, like student stats, funding, avg. time to complete, etc.

The logistics make it tough.

1. Professors move.
2. Research shifts.
3. Accuracy of data and/or self-reports?\
4. Couldn't do on SDN if you hosted it 3rd party and/or had advertising (against TOS to essentially use SDN and/or membership for personal gain).

Did not read your link, but from your short description my questions would be as follows:

Who are you expecting to build this tool? It costs time and money to build something like this and I'm just wondering who is going to foot the bill? Not to mention who will maintain it, provide the infrastructure, provide user support, etc?

Secondly, who is going to enter in the data initially and who is going to keep it up to date as professors and projects change? If you could get professors excited about this tool and convince them to keep their own profile up to date then great! But I have a feeling that most profs don't have a problem getting enough students applying to their programs so this would be just extra "busy work" for them with no real benefit. Also many profs are not technologically proficient and couldn't be bothered to learn. This is a serious problem because without current information this tool is pretty much useless.

I'm not saying it's not a good idea, but there are some major hurdles that would need to be overcome in order to make it a worthwhile project.

I'm currently looking for professors to work with as well and I have wished for this database you speak of. I would be ecstatic if it was created. However, because of several of the points previously made, it seems that such a project is unlikely to become a reality.

I got bored, so here is my 2 minute brainstorm.....

Building a script to search existing databases would be easier. Database owners tend to frown on uncompensated data mining, particularly if it takes away "uniques" from them, so you'd have a better shot at targeting public databases.

The trickiest part is writing an effective script that mines applicable data. This would take some trial and error, and you'd better hope the source doesn't change their system. Then you take that raw data and dump it into your own custom database with pre-set sorts and filters to clean up the data. Once you get a decent amount of data, you can then write a script to compare your original scrubbed data against "new" scrubbed data. It isn't perfect, but it'd get you probably 70% of the way there if you do everything right. A few more iterations of your script and you'd eventually fill in the holes and have a workable setup.

The most obvious limitation is that you are assuming your source has access to enough publication sources....which tends to be hit and miss. You also lack a sufficient market to make this a viable option. The above isn't overly complex if you have a programming and database background, but it'd cost WAY too much if you had to hire someone.

A better approach would be to target specific research areas, identify an additional way to sell/market the data, and then use your initial funding to build out other subject databases.

Simple, no?

bsomerville said:

Agreed-- but before I get hung up on how to do it I'd like to know if it's wanted/needed. You're absolutely right that researchers have no motivation to update their profiles. But students could be bothered to enter updated info-- especially if that were a hoop to jump through before being granted access to the database. Or it could be a much more open, friendly, web 2.0 arrangement (Help us keep this thing up to date! Win kudos, badges, gold stars for doing so!) And so forth.

My point being, the feasibility of the project is separate from its viability-- right now I'm wondering how others are approaching the program research phase, and if having a searchable list of faculty research interests would be a significant timesaver or not.

Click to expand...

Absolutely a good idea. As an ug that would have been immensely helpful.
Good luck

bsomerville said:

Well, with that kind of attitude it will never become a reality!

Click to expand...

Haha sorry for the pessimism. Honestly, I think its a great idea. If I had that right now it would be awesome and I probably would pay money for it. But I wouldn't pay that much money for it lol. I think its possible that you could get some sort of database but it will have a lot of kinks. I just wonder if the "kinks" you have to face will outweigh the interest of the potential market (e.g. This is a good service but with its flaws I'm better off looking myself for free)

Psychology 76 said:

Haha sorry for the pessimism. Honestly, I think its a great idea. If I had that right now it would be awesome and I probably would pay money for it. But I wouldn't pay that much money for it lol. I think its possible that you could get some sort of database but it will have a lot of kinks. I just wonder if the "kinks" you have to face will outweigh the interest of the potential market (e.g. This is a good service but with its flaws I'm better off looking myself for free)

Click to expand...

This is why we haven't seen it yet. Good idea though.

You could just create a wiki where people can add info as they come across it, and edit as needed. This would obviously not generate revenue, and would be completely user driven. But it would be helpful. Plenty of disciplines do this for the academic job market.